A systematic review of community-based interventions for emerging zoonotic infectious diseases in Southeast Asia

Executive Summary

Background

Southeast Asia has been at the epicentre of recent epidemics of emerging and re-emerging zoonotic diseases. Community-based surveillance and control interventions have been heavily promoted but the most effective interventions have not been identified.

Objectives

This review evaluated evidence for the effectiveness of community-based surveillance interventions at monitoring and identifying emerging infectious disease; the effectiveness of community-based control interventions at reducing rates of emerging infectious disease; and contextual factors that influence intervention effectiveness.

Inclusion criteria

Participants

Communities in Brunei, Cambodia, Indonesia, Laos, Malaysia, Myanmar, the Philippines, Singapore, Thailand and Viet Nam.

Types of intervention(s)

Non-pharmaceutical, non-vaccine, and community-based surveillance or prevention and control interventions targeting rabies, Nipah virus, dengue, SARS or avian influenza.

Types of outcomes

Primary outcomes: measures: of infection or disease; secondary outcomes: measures of intervention function.

Types of studies

Original quantitative studies published in English.

Search strategy

Databases searched (1980 to 2011): PubMed, CINAHL, ProQuest, EBSCOhost, Web of Science, Science Direct, Cochrane database of systematic reviews, WHOLIS, British Development Library, LILACS, World Bank (East Asia), Asian Development Bank.

Methodological quality

Two independent reviewers critically appraised studies using standard Joanna Briggs Institute instruments. Disagreements were resolved through discussion.

Data extraction

A customised tool was used to extract quantitative data on intervention(s), populations, study methods, and primary and secondary outcomes; and qualitative contextual information or narrative evidence about interventions.

Data synthesis

Data was synthesised in a narrative summary with the aid of tables. Meta-analysis was used to statistically pool quantitative results.

Results

Fifty-seven studies were included. Vector control interventions using copepods, environmental cleanup and education are effective and sustainable at reducing dengue in rural and urban communities, whilst insecticide spraying is effective in urban outbreak situations. Community-based surveillance interventions can effectively identify avian influenza in backyard flocks, but have not been broadly applied. Outbreak control interventions for Nipah virus and SARS are effective but may not be suitable for ongoing control. Canine vaccination and education is more acceptable than culling, but still fails to reach coverage levels required to effectively control rabies. Contextual factors were identified that influence community engagement with, and ultimately effectiveness of, interventions.

Conclusion

Despite investment in community-based disease control and surveillance in Southeast Asia, published evidence evaluating interventions is limited in quantity and quality. Nonetheless this review identified a number of effective interventions, and several contextual factors influencing effectiveness. Identification of the best programs will require comparative evidence of effectiveness acceptability, cost-effectiveness and sustainability.

Implications for practice

Interventions are more effective if there are high levels of community ownership and engagement. Linkages between veterinary and public health surveillance systems are essential. Interventions are not well accepted when they fail to acknowledge the importance of animals for economic activity in communities.

Implications for research

Evidence is needed on functioning and outcomes of current surveillance systems and novel low-cost methods of surveillance. Evaluations of control interventions should control for confounding and report measures of disease, cost and sustainability. Translational research is needed to assess generalisability and evaluate roll-out of effective interventions as regional or national programs.

Background

The 2004 WHO/FAO/OIE joint consultation on emerging zoonotic diseases defined such diseases as, “a zoonosis that is newly recognized or newly evolved, or that has occurred previously but shows an increase in incidence or expansion in geographical, host or vector range”.1 Avian influenza, severe acute respiratory syndrome (SARS), Nipah virus, monkeypox, Hendra virus, and the lentiviruses that cause Acquired Immunodeficiency Syndrome (AIDS) are a few examples of the growing number of diseases that humans can contract from animals.

The Asia Pacific Region has, unfortunately, been at the epicentre of such epidemics. Over 30 new infectious agents have been detected in the last three decades, 75% of which were zoonotic.2 A number of factors contribute to these circumstances. The absence of effective surveillance and control programs, prevailing socio-cultural practices and weak public health and veterinary services infrastructure exacerbates the vulnerability of these settings. Other factors including climate change, environmental degradation, encroachment of humans on areas where wildlife exists, cohabitation of humans and food animals within households, and the mixing of species in live animal markets play a role in increased disease transmission.

Influenza remains a global priority with the potential to cause large, global epidemics. Approximately 10% to 15% of people worldwide contract influenza annually, with attack rates as high as 50% during major epidemics.3 In 2003 the SARS epidemic affected around 8000 people and killed 780. In 2006 a new avian H5N1, and in 2009, a new H1N1 ‘swine’ influenza pandemic threat, caused widespread anxiety.4

In addition to mortality and morbidity, zoonotic diseases have and are predicted to cause huge economic losses. The economic cost of the major outbreaks of new epidemic zoonotic diseases over the past decade, including SARS and H5N1 influenza, has been estimated to be $200 billion.4

To prevent and control zoonotic infections in Southeast Asia (SE Asia), a multi-sectoral and multi-disciplinary approach, involving many levels of the health and non-health sector, is needed, which places a strong emphasis on both the early detection and early control of infectious disease outbreaks.

Surveillance activities

Early detection of disease outbreaks requires effective disease surveillance systems. Systems in developing countries face many operational challenges, including a lack of accurate and timely information exchange between local, provincial, national and regional levels, and inadequate human resource and laboratory capacity for speedy diagnosis. The WHO's Asia Pacific Strategy for Emerging Diseases 2010 highlights the need for community involvement in surveillance.2 Zoonotic disease detection and control also depends on effective veterinary surveillance and the ability to contain outbreaks amongst animal populations, systems that are often poorly developed or non-existent in developing countries.

Jones et al.5 suggest that local targeted surveillance of at-risk people may be the best way to prevent large-scale emergence. Brownstein et al.6 in their discussion of web surveillance suggest that the use of news media and other non-traditional sources of surveillance data such as web-accessible discussion sites and disease reporting networks could facilitate early outbreak detection and increase public awareness of disease outbreaks prior to their formal recognition. May et al.7 review the evidence for syndromic surveillance systems in developing countries (systems utilising existing clinical data prior to a diagnosis) and find that this may be a feasible and effective approach to infectious disease surveillance in developing countries.

Evaluating surveillance activities

The effectiveness of surveillance systems in responding generally to emerging infectious diseases has not been reviewed systematically. Reviews aimed at particular contexts (for example, prevention of bioterrorism8 and public health surveillance for trachoma2) have been undertaken, however, neither review was able to state whether surveillance systems are achieving the ultimate goal of detecting outbreaks early and providing an accurate picture of infection rates in the area covered by the surveillance program.2

Most evaluations of surveillance programs have been qualitative, and focused on evaluating the practical structure and operation of the system, rather than the impact on infectious disease transmission.9–11 Many researchers have used the Centers for Disease Control and Prevention (CDC) guideline which recommends how a surveillance system can be assessed to verify if it meets its objectives.12 This provides a framework for evaluating how well a system is functioning and determining reasons why it may or may not be functioning to detect and respond to infectious disease outbreaks and/or support ongoing control activities to tackle endemic diseases.

The CDC guideline recommends that reports of surveillance systems include the following:

• descriptions of the public health importance of the health event under surveillance; the system under evaluation; the direct costs needed to operate the system; the usefulness of the system;

• evaluations of the system's simplicity, stability (its ability to withstand external changes), flexibility (that is, “the system's ability to change as surveillance needs change”), acceptability (“as reflected by the willingness of participants and stakeholders to contribute to the data collection, analysis and use”), sensitivity to detect outbreaks, positive predictive value of system alarms for true outbreaks, representativeness of the population covered by the system, and timeliness of detection.

Prevention and control activities

Control of emerging infectious disease requires an effective response to surveillance data. Single measures such as the use of vaccines or antiviral drugs may be unavailable, unaffordable or not in sufficient quantity. The control of these infectious diseases in resource constrained settings is more likely to be influenced by community-based and behavioural change interventions as well as by strengthening of national and international commitment to their control.13 Over the last decade there have been increased efforts to promote community-based infectious disease control.2

For vector-borne infections, such as dengue, attention has been focused on interventions to reduce larval, and ultimately adult, vector populations. Programs have attempted to achieve this via a range of chemical, biological and physical interventions to reduce vector populations, as well as trying to initiate behavioural change at the community level to prevent contact with the mosquito vectors.14 Heintze et al. have previously reviewed the evidence for community-based dengue control programs.15 This systematic review completed in 2005 found at that time that the evidence for these activities was weak and inconclusive and suggested a number of priorities for future research in this area. However, the review has not since been updated.

Community-based interventions to control the spread of respiratory viruses, such as influenza, have focused on hygiene and respiratory etiquette to prevent human-to-human transmission. Many of these interventions have only been evaluated in a developed country context. Aledort et al.16 and Jefferson et al.13 undertook systematic reviews of physical interventions to interrupt or reduce the spread of respiratory viruses. Both reviews found handwashing was effective whilst there was no evidence to support school/workplace closure. However, these findings are from a predominantly North American context and may not be generalisable to countries with limited access to safe water and sanitation.

Evaluating control activities

To understand whether community-based control activities will be effective and why requires us to look at the behavioural mechanisms through which these interventions work and the context in which they are based. Behavioural mechanisms operate through the experiences, beliefs and values of groups and individuals. These mechanisms are therefore dependent in part on the context in which they are used. This framework was used in a recent synthetic review of water and sanitation projects.17 The framework is shown in Figure 1.

Figure 1: Framework for evaluating the impact of context and behavioural mechanisms on intervention outcomes.

Review objective/questions

The objective of this review is to identify the effectiveness of surveillance systems and community-based interventions in identifying and responding to emerging and re-emerging zoonotic infections in SE Asia.

It aims to provide a critical review of published evidence that evaluated the effectiveness of community-based surveillance and prevention and control interventions for emerging zoonotic infectious diseases. In addressing the three research questions outlined below we will summarise evidence for not only the effectiveness of community surveillance and prevention and control interventions in SE Asia in identifying and responding to these infectious diseases, but also explore the contextual factors that influenced their success.

More specifically the review questions were:

1. What is the effectiveness of community-based surveillance interventions designed to identify emerging zoonotic infectious diseases?

2. What is the effectiveness of non-pharmaceutical community-based interventions designed to prevent transmission of emerging zoonotic infectious diseases?

3. How do factors related to the emergence and management of emerging zoonotic infectious diseases impact the effectiveness of interventions designed to identify and respond to them?

Inclusion criteria

Types of participants

This review considered studies that evaluated interventions that are non-pharmaceutical, non-vaccine, and community-based. Community-based is defined as implemented outside a healthcare institution with at least one component of the intervention targeted directly at the community (e.g. educational meetings, involvement of local leaders). Interventions with no community participation (i.e. top-down vector control programs) were excluded as they were outside the scope of this project.

The review was limited to the ten member countries of the Association of Southeast Asian Nations (ASEAN)18: Brunei, Cambodia, Indonesia, Laos, Malaysia, Myanmar, the Philippines, Singapore, Thailand and Viet Nam, and the following diseases of interest developed from the list of emerging and re-emerging zoonotic infections published on the CDC website19 as commonly occurring in Southeast Asia:

• rabies

• Nipah virus

• dengue

• SARS, and

• avian influenza

Types of intervention(s)/phenomena of interest

Interventions of interest included, but were not limited to:

Surveillance Interventions: syndromic surveillance programs, communications programs, training/education of health workers and community workers to detect and/or prevent disease, local level surveillance & response teams, web surveillance. Following the One Health20 concept of a synergistic approach to health we will also include animal/livestock surveillance systems where they are specifically evaluated with respect to their impact on human health and disease outcomes.

Control Interventions (subcategorised into the following):

Health promotion interventions: self-reporting of suspected infections, promotion of voluntary self isolation, advocating use/provision of personal protective equipment, e.g. masks, public/community education on hygiene and respiratory etiquette, safe slaughter and preparation of animals and animal products (in particular poultry),

Physical interventions: contact tracing, isolation, quarantine, social distancing, barriers, school/workplace closure, movement restriction,

Environmental interventions: environmental cleaning, waste disposal, coverage or removal of water containers, vector control, larval control including larvivorous fish and copepods, destruction of potentially infected animals and animal products

Types of outcomes

Primary outcomes A range of different outcomes used in the studies were examined. For the purposes of this review, they can be broadly categorised into primary and secondary outcomes. Primary outcomes aim to measure the incidence of infection or disease in the community. We considered studies that reported any type of quantitative infection/disease/outbreak outcome data or morbidity and mortality rates attributable to the infectious disease. This includes the following types of primary outcome measures: rates of infection, numbers of cases of infection reported and confirmed mortality rates attributable to the infectious disease, rates of hospitalisation attributable to the infectious disease, number of outbreaks, time/size of epidemic peak, duration of outbreak/epidemic.

Secondary outcomes: To help contextualise our findings and address review question three, we also extracted any information on other indicators relating to the functioning of the surveillance and/or control program. These indicators can be used as intermediate outcomes to predict how the intervention might impact on infection or disease. For example, an intervention program may not show a reduction in disease but may result in an improved capacity for detection and containment of outbreaks or high levels of vector control. We categorised indicators based around the WHO framework for the monitoring and evaluation of surveillance and response systems for communicable disease21 and categorise these as secondary outcomes:

• Process indicators: Activities such as training sessions delivered, guidelines developed or number of sites monitored,

• Output indicators: The results of the activities conducted e.g. proportion of surveillance centres providing timely reporting, number of households with containers covered, proportion of the community attending education session,

• Outcome indicators: The extent to which the surveillance and response objectives are being achieved, including the quality of the surveillance systems and the appropriateness of any outbreak response e.g. proportion of outbreaks where appropriate control response initiated, incidence-reporting-response times, numbers of larvae/vectors, improvements in knowledge relating to hygiene education campaigns.

Types of studies

Following the recommendations of the Cochrane Effective Practice and Organisation of Care Review Group (EPOC),22 which is concerned with evaluating interventions in community healthcare settings, only studies that provide evidence that draws a comparison between an intervention setting and a non-intervention setting were included. A second inclusion criterion was that the study must report results as quantitative infection/disease/outbreak data (as described under types of outcomes). We aimed to include studies reporting original primary data or systematic reviews of this type of evidence (i.e. not theoretical model based studies).

Acceptable study designs included: systematic reviews/meta-analyses, randomised controlled trials, controlled clinical trials, controlled before and after trials, interrupted time series (we require only one time point before and after the intervention). We also accepted mixed-method studies that included one of the above, and systematic review and economic evaluations that were based on one of the above. Conference papers, clinical observations, program reports with only one time point and non-systematic overview articles were excluded.

The quantitative component of the review extracts data from included studies on all disease outcomes and process indicators measured. This information is used to address review questions 1 and 2.

The textual component of the review considers the textual information included in the introduction, methods and discussion of all papers included in this systematic review. This is used to supplement the quantitative information on process indicators and address review question 3.

Search strategy

Studies published in any language with an abstract available in the English language were considered for inclusion in this review. Studies were assessed for inclusion based on title and abstract only; with studies only translated if they met inclusion criteria. Studies published between 1980 and 2011 were considered for inclusion in this review, with a start date of 1980 chosen as surveillance programs in most SE Asian countries commenced in the early 1990s. By including data from 1980, we hoped to capture any information on community-based surveillance and intervention programs that may have contributed to the development of formal surveillance programs.

The databases searched included: PubMed, CINAHL, ProQuest, EBSCOhost, Web of Science, Science Direct, the Cochrane Library of systematic reviews, the WHO library database (WHOLIS), British Development Library, LILACS, World Bank (East Asia) and the Asian Development Bank. Further details on the search strategy are given in Appendix I.

Methods of the review

Assessment of methodological quality

Quantitative papers selected for retrieval were critically appraised by two independent reviewers prior to inclusion in the review using standardised instruments from the Joanna Briggs Institute Meta Analysis of Statistics Assessment and Review Instrument (JBI-MAStARI) (Appendix II). Any disagreements arising between the reviewers were resolved through discussion, or in consultation with a third reviewer.

Data collection

Quantitative data was extracted from papers included in the review using a data extraction tool specifically developed for this review that is shown in Appendix III. The tool was drafted during protocol development and piloted on a subset of studies across the five diseases. Based on this, a number of modifications were made to the tool to facilitate comparison across the diverse study types included in the review. The final tool still captured key details about the interventions evaluated and the methods and outcomes used in the evaluations, but to make extraction of contextual information easier for reviewers, the prescriptive categories used in the tool presented in the protocol were removed and replaced with three broad categories: contextual factors, behavioural mechanisms and program structure and delivery. These modifications allowed reviewers to capture the diverse range of factors reported in the studies and aided with categorisation of studies for the narrative analysis of findings.

Data was extracted on details about the interventions, populations, and study methods, program context and other outcomes of significance to the review question and specific review objectives. This included both disease outcomes and process indicators as described above to enable us to look at both the effectiveness and function of the programs.

To enable us to comment better on why programs have been (un)successful, we collected both quantitative data (i.e. process indicators) and qualitative data constituting narrative evidence or speculation by the authors on why interventions have been effective or not and any comment on sustainability. Textual data was extracted from the papers included in the quantitative review to capture the following specific details about the context and mechanisms of the program relevant to the review question and specific objectives:

• Features of the study setting, i.e. the geographical setting, the social, cultural and political context, the season,

• Features of the interventions i.e. what was done, how it was delivered, who was targeted, where it was delivered and by whom, funding organisation, technical and financial program details and any behavioural mechanisms targeted by the intervention,

• Level of participants i.e. communities, households, individuals, details on age and gender.

Data synthesis

Data extracted on the effectiveness of interventions and regarding the factors that aided or impeded effectiveness was synthesised in a narrative summary with the aid of tables and figures. We used the frameworks for evaluating infectious disease surveillance systems and behavioural interventions outlined in the background section to guide categorisation in our synthesis of this evidence where the evidence allowed us to, for surveillance activities we grouped abstracted information according to the CDC criterion for evaluating surveillance activities and for control programs we used a behavioural change framework to look at mechanisms and context for change.

Meta-analysis

Comparable study findings from individual studies were combined statistically in a meta-analysis. This approach allowed us to increase the power of the analysis, improve the precision of our estimates of an intervention and assess whether an intervention was similar in similar situations. The relative homogeneity in results across the different types of intervention supported this decision. Upon review of the data from the included studies, we elected not to use the Frequentist meta-analysis built into the JBI-MAStARI statistical software (as originally outlined in the review protocol). We decided against this standard approach, which calculates odds ratios and 95% confidence intervals using the Mantel-Haenszel test as the default meta-analytical method for dichotomous data, in favour of a Bayesian approach that calculated odds or rate ratios with 95% credible intervals, for several reasons.

The Bayesian meta-analysis adjusts for multiple individual or repeated results from the same study (for example, results from two or more villages or from the same village at two or more time points) by using a random study effect, rather than combining the results across villages (for example, summing the total incidences and samples across all control villages and across all intervention villages). This accounts for instances where combining villages might be problematic (e.g., villages in high and low risk areas) and also adjusts for the fact that multiple within-study results are likely to be correlated and should not be entered as independent studies in a meta-analysis, nor should they be combined to give a study average, as this ignores the potentially valuable between-result heterogeneity (eg. villages in high and low and high risk areas). Standard meta-analysis software cannot model repeated results from the same study and is therefore likely to give a less accurate estimate. The Bayesian meta-analysis also easily copes with zero cells, for example, no positive results from a control village, which was not uncommon in studies of small sample sizes. Information about the underlying statistical assumptions and full set of equations and priors for the meta-analysis of binomial data has been included in the full meta-analysis report in Appendix VII, as well as the raw numbers used to generate the odds and rate ratios presented in the results section.

There were two types of dependent data in our analysis:

1. Counts of the number of successes and failures, for example, the number of containers that tested positive for mosquitoes and the total number of containers tested. These were modelled using a binomial distribution. The meta-analysis of count data weights all results by study size. Results were expressed as odds ratios.

2. Failure rates, for example the number of containers that tested positive for mosquitoes per 100 sampled. These data were modelled using a Poisson distribution. These data often did not provide information on the denominator used to calculate the rate so studies were unable to be weighted by study size in the meta-analysis. Results were expressed as rate ratios.

The meta-analysis was easily fitted in R using the R2WinBUGS software version 1.4.323 and a Bayesian model with a random intercept for each study. We plotted the means and 95% credible intervals for the odds or rate ratios using the ‘forestplot’ function in the ‘rmeta’ library of the R software.24 We generated plots at both the study and result level to visually show both the between-study and between-result variability. Odds ratios or rate ratios under one meant the intervention was effective; odds or rate ratios over one meant the intervention was not effective.

Uncertainty in estimates is expressed as a 95% credible interval, a standard approach in Bayesian statistical analyses, which has an interpretation similar to a 95% confidence interval (as would be calculated in the standard Frequentist meta-analysis orginally outlined in the protocol). A credible interval contains a 95% probability of containing the true estimate, in comparison to 95% confidence intervals, whose correct interpretation relies on imagining repeating the study multiple times, calculating multiple confidence intervals, and then counting the number of times the true estimate is contained in the intervals. Credible intervals therefore have a far simpler interpretation.

Due to the small numbers of studies reporting common outcomes, we were unable to carry out the planned sub-group analyses (which were by intervention type, urban/rural context and country). Instead, we pooled results across intervention types for meta-analysis, and a “leave one study out” sensitivity analysis was used to show the influence of each study on the summary odds or rate ratio when there were more than two studies. The relative homogeneity in results supported this decision.

Definitions

The following definitions were used to classify outcome measures in the meta-analysis:

• Household index (HI) is the proportion of households positive for Aedes aegypti larvae.

• Container index (CI) is the proportion of containers positive for Aedes aegypti larvae.

• Breteau index (BI) is the number of containers positive for Aedes aegypti larvae per 100 households.

• Larval population number (LPN) is the number of Aedes aegypti larvae counted in the survey.

• Larval density index (LDI) is the average number of larvae per house.

• Mosquito bite rate (MBR) is the average number of mosquito bites per person per hour.

• Rate of dengue haemorrhagic fever (R_DHF) is the number of cases of dengue haemorrhagic fever per 100,000 population

Review results

Description of included studies

A total of 5131 potentially relevant titles were identified by the search. Of these, 538 abstracts of potentially relevant papers identified by the literature search were examined, and 417 papers were excluded after evaluation of the abstract (Figure 2). A more detailed examination was conducted of 121 short-listed papers and 57 papers were excluded after review of the full paper. Sixty-four papers were then assessed for methodological quality, after which, seven studies were excluded leaving a final list of 57 papers to be included in the systematic review.

Figure 2: Flowchart of number of citations identified, retrieved, included and excluded.

Of the 57 papers included in the systematic review, 19 studies looked at surveillance interventions25–43 and 44 studies presented data on prevention and control interventions.29, 30, 36–38, 40, 44–81 Data was available evaluating prevention and control interventions for all 5 emerging infectious diseases included in the review, with the most evidence available for Dengue interventions and the least for Nipah virus interventions. Evaluations of surveillance activities were available for all diseases except Rabies, where only descriptive studies were retrieved. Details of all the included studies, and the information extracted from these papers, can be found as Appendix IV in Tables 9–20 and 28–35, and a list of excluded studies can be found as Appendix V.

The studies were conducted in a range of Southeast Asian countries, with the exception of studies on interventions for SARS, where five of the six included studies were from Singapore, and studies for interventions on the Nipah virus outbreak, which were all conducted in Malaysia. With the exception of one study that was a cluster randomised trial,76 none of the other included studies were randomised trials. Study designs used to evaluate surveillance systems were predominantly retrospective and based on analysis of case series or surveillance data,28, 31, 32, 41, the exception used a prospective evaluation.41The study designs used to evaluate prevention and control activities included:experimental before and after studies,49, 50, 60, 78, 80, 81 observational prospective comparable cohorts,38, 39, 44, 45, 47, 48, 52, 53, 55–62, 64–66, 71, 74, 77 prospective cohort studies,67, 70, 72, 79 retrospective cohorts,34, 51, 63, 69, 73, 75 and retrospective analysis of interventions using outbreak or surveillance data.25–27, 29, 30, 33, 35–37, 40, 42, 43, 54

Results for interventions targeted at rabies

114 papers were short-listed for comprehensive examination from the original list of 1694 (Figure 2). The majority of the original articles (103) and six systematic reviews were excluded after reading the abstract and an overview of the contents of the paper. The full text articles of 16 studies were retrieved. Reasons for exclusion are outlined in Appendix V and included narrative reviews or descriptive analyses that did not present any data, model-based studies or cross-sectional knowledge, attitudes and practices (KAP) or seroprevalence surveys with no intervention evaluated. Of the 16 full text articles, six studies were assessed for methodological quality and one was excluded at this stage82 as it only measured one time point. Five studies have been included in the review (Table 9), two studies from the Philippines on oral canine vaccination in owned dog populations50, 67 and studies from Thailand54, Malaysia68 and Indonesia81 all describing retrospective data from rabies control programs in response to an outbreak of rabies or an increase in the number of human rabies deaths.

Methodological quality of the studies

Overall the quality of the studies was low. None of the studies presented randomised groups, and criteria for inclusion in the study were sometimes not defined, as were confounding factors. The study by Estrada et al.50 only included a small proportion (10.5%) of the vaccinated group that were tested pre- and post-intervention for seroconversion to assess vaccination coverage. Otherwise, vaccination coverage was estimated by the number of animals vaccinated directly and the number of dogs that accepted bait and subsequently punctured the container. Measurement of the success of the vaccination campaign in the study by Robinson was assessed using dog collars, paint marks or both.67 Vaccination coverage is summarised as an odds ratio with 95% confidence intervals. It is worth noting that 18% of dogs included in this estimate as vaccinated did not have a vaccination marker but were self-reported by their owners as vaccinated during the campaign. Robinson et al. also provide data on the likelihood of vaccination following receipt of campaign information as an odds ratio but fail to provide confidence intervals for their estimate.

The study by Kamoltham et al.54 reports on a five-year rabies control program in the Phetchabun province of Thailand between 1997 and 2001. The authors use the number of human deaths during the program to measure success of the vaccination program, but this is confounded by the increased uptake of post-exposure prophylaxis (PEP) as a result of expansion of the existing treatment regimen. They also mention increasing awareness of rabies through advocacy in provincial schools, television programs, and newspapers, but do not assess these educational initiatives with knowledge surveys before and after the educational campaign.

The data published in Soon et al.68 are from a retrospective case series and presents the number of confirmed cases of rabies in animals in Malaysia from 1946 to 1987, and information about a rabies control program initiated in 1952. Data on evaluation of the vaccination campaign is not presented, other than to report on the decline in the number of cases of rabies in animals, although the denominator of this main outcome data and how it was sampled is not mentioned.

Finally, the study by Windiyaningsih et al.81 also describes rabies control measures in response to an outbreak on Flores Island in Indonesia. Control measures implemented included mass culling and canine vaccination, and post-exposure prophylaxis for exposed cases who had suffered an animal bite. It was difficult to calculate the vaccination coverage as the number of dogs vaccinated for each region was not always provided. It was also difficult to assess the success of the campaign as it was confounded by post-exposure prophylaxis administered to exposed cases.

Rabies – Review findings

Surveillance interventions

From the included papers, data were available for surveillance activities in only two studies54, 68 Details of the interventions evaluated and the main findings from each included study are presented in Table 10 and Table 11. The study by Kamoltham et al.54 presents the number of potentially exposed cases who received treatment from rabies treatment centres, hospitals and clinics in Phetchabun province, and report that rabies is a notifiable disease in Thailand. A census of the dog population and canine vaccination coverage was also carried out by the Livestock Department of Phetchabun during the program, although this appears to have been on an ad hoc basis collected specifically for the elimination program. The study by Soon et al.68 presents veterinary surveillance data carried out by the Ministry of Agriculture and also mentions surveillance of human cases of rabies infection by the Ministry of Health, as part of the Malaysian National Rabies Control Program. They do not mention how the outbreak was detected or whether any form of surveillance was in place prior to the outbreak. Furthermore, neither of these studies present an analysis or evaluation of their surveillance programs, other than to show a decrease in the number of deaths from rabies infection54 or to say there was a “decrease in the number of rabies deaths” after the interventions.68

The studies in the Philippines50, 67 and Indonesia81 do not mention ongoing rabies surveillance. The outbreak on Flores Island was reported by word of mouth by local fishermen when three dogs died.

Prevention and control interventions

Control interventions discussed in the five included studies included canine vaccination,50, 54, 67, 68, 81 sterilisation,54 culling of the dog population,81 public health education,54, 67, 68 movement restriction of infected dogs68 and quarantine of newly introduced dogs.68 Details of the interventions evaluated and the main findings from each included study are presented in Table 10 and

Table 11. All evidence on control interventions looked at effectiveness of canine vaccination, administered either through oral baits50 or by direct injection.50, 54, 67, 81 The study by Kamoltham et al.54 mentions canine sterilisation, but no data is presented on the latter. Although some interventions included a health education/awareness component,54, 67, 68 this was only evaluated by Robinson et al. The study by Windiyaningsih et al.81 does present data on culling of the dog population in an outbreak setting, although data for all provinces where the intervention was carried out was not recorded. Overall, the available evidence is low quality, and outbreak data is based predominantly on the analysis of a case series using historical controls. The outcome measure used in three studies is the proportion of dogs vaccinated.50, 67, 81 Kamoltham et al.54 and Windiyaningsih et al.81 also report on human disease indicators such as the number of cases of human rabies. In both studies, attempts to evaluate the impact of the intervention on the numbers of rabies exposures in humans are confounded by concurrent expansion of a cheaper and safer rabies treatment regimen. In the Thai study54, despite the aggressive vaccination campaign from 1996 – 2001, the number of exposures to suspected and proven rabid animals continued to increase from 1992 – 2000. Inability to enforce movement restrictions of animals, culling of diseased animals and incomplete vaccination coverage contributed to failure to control the outbreak on Flores Island.81

Contextual factors

A number of studies reported how behavioural mechanisms by community members and the public impacted on the success of the study. Details of contextual factors extracted from each included study are presented in Table 12. Robinson et al.67 observed that pre-campaign education and advertisements contributed to the success of the program and that good information dissemination impacted on the likelihood to vaccinate. Dogs were more likely to be vaccinated if the household had received campaign information from more than one source (OR=4.45, CI not provided, p=0.04, statistical test not stated), and less likely to be vaccinated if the household had learned of the campaign primarily through posters (OR=0.30, CI not provided, p=0.015, statistical test not stated). Poor understanding of vaccination also contributed to refusal to participate (the perception that the vaccine altered the meat if a dog was kept for consumption).

Good engagement of the community was also vital for the uptake of vaccination, particularly for the owned dog population, as owners had right to refuse. Other reasons for refusing to participate included the owner not wanting to cause injury to the dog from vaccination. Estrada et al.50 also reported a reluctance of owners to have dogs repeatedly bled. The study may also have been compromised as a result of dog owners demanding financial compensation for dogs handed over for rabies diagnosis. Reluctance by members of the public to kill dogs in the Flores Island outbreak81 perpetuated the outbreak, as some owners moved their dogs to rabies-free districts or sold them at markets to avoid killing them. The practice of fishermen travelling with their dogs and subsequently visiting other islands also aided the spread of outbreak.

Both Kamoltham et al.54 and Windiyaningsih et al.81 mention decentralisation of services in Thailand and Indonesia impacting on the ability to obtain complete data in the former and to control the spread of the outbreak in the latter. Lack of coordination between local authorities made it difficult to contain the infected dog population and prolonged the outbreak. Higher level support and the involvement of the authorities was essential in the success of outbreak control measures because some form of law enforcement was required,68 particularly where no one claimed ownership such as the stray dog and common dog population.

Summary

In summary, no evidence was available for routine human or veterinary surveillance activities, nor an analysis or evaluation of an existing surveillance program for Rabies. All evidence on control interventions looked at the effectiveness of canine vaccination. Although some interventions included a health education/awareness component, this was only evaluated in one study. Overall, the available evidence is of low methodological quality, and outbreak data is based predominantly on the analysis of a case series using historical controls. The outcome measure used was the proportion of dogs vaccinated in three studies. In one study, attempts to evaluate the impact of the intervention on the number of rabies exposures in humans are confounded by concurrent expansion of the rabies treatment regimen. A number of studies highlighted the importance of pre-campaign education and advertisements to impact on the success of the program, as well as coordination of local services and higher-level support to conduct a successful campaign.

Results for interventions targeted at Nipah virus

89 papers were short-listed for more comprehensive examination from the original list of 534 (Figure 2) of which seven studies were critically appraised for methodological quality and subsequently included in the review (Table 13). Reasons for exclusion are outlined in Appendix V and include papers which were: review articles which did not present any data on interventions, prevalence surveys, risk factor studies, and clinical and outbreak reports.

Two studies were outbreak reports of the Malaysian outbreak epidemic29, 30 and present an epidemic curve of the number of human cases by the date of onset of their illness. Four studies also discussed the National Swine Surveillance Program and subsequent control measures initiated by the Malaysian Government in response to the outbreak.26, 36, 37, 40 In the final study,25 authors discuss an active surveillance initiative for the detection of Nipah virus infected swine in Indonesia.

Methodological quality of the studies

The quality of the data in the outbreak reports29, 30 was poor and based primarily on a case series with historical controls. Neither study outlined when the control measures were initiated in relation to the progression of the outbreak. Bunning29 presented the number of human and swine cases of Nipah virus infection (an epidemic curve) over the period that the interventions were initiated but do not provide a denominator for this data.

The studies by Ozawa et al., Muniandy et al., Mohd Nor et al. and Arshad et al.26, 36, 37, 40 describe the National Sero-surveillance program initiated post-outbreak to detect any remaining infected pig farms and abattoirs not already depopulated. The number of abattoirs and farms tested and proportion positive for Nipah virus is presented. All studies discuss the sampling strategy of the program, the results of the laboratory testing and subsequent control measures taken.

Nipah virus – Review findings

All published evidence included in this systematic review on surveillance and control interventions is based on the outbreak response in Malaysia and Singapore in 1999. Subsequent to 1999, Nipah virus was identified as causing clusters of disease in humans in India and Bangladesh,83 countries which are outside the scope of this review.

Surveillance interventions

Details of the surveillance interventions evaluated and the main findings from each included study are presented in Table 14 and Table 15. From the included studies, no ongoing routine veterinary or human surveillance appears to have been in place in this region prior to the commencement of the outbreak. Human and swine surveillance was instigated in Malaysia as a measure of active case finding to guide outbreak control measures (the National Swine Surveillance Program).26, 36, 37, 40 Swine surveillance was also carried out in Indonesia25 in response to restrictions on the export of Indonesian pork by other Asian countries.

The National Swine Surveillance Program in Malaysia on farms was carried out till the end of December 2000 to detect and cull additional infected herds, and abattoir surveillance was continued in 2001 and 2002 of all pigs entering abattoirs.37 While there is evidence that the swine surveillance and subsequent control measures were effective, as Malaysia achieved a Nipah virus –free status by the end of December 2001, no evaluation was carried out and no comment has been made on the feasibility and on–going sustainability of this program.

Despite descriptive reports of on-going surveillance activities such as animal tracking systems (coding of farms, ear tagging and tattooing) to aid trace back,36, 40 educational programs for farmers and health promotion campaigns,36 there are no studies that report quantitative data on the functioning of these systems and there has been no evaluation of these systems. Strategies of herd health monitoring and improved farm management practices were also briefly discussed by Mohd Nor.36

Prevention and control interventions

A number of control measures were discussed in the seven included studies, which included mass culling, quarantining, movement restrictions, education about contact with pigs and the use of personal protective equipment (PPE). Details of the interventions evaluated and the main findings from each included study are presented in Table 14 and Table 15. All interventions were government driven, with involvement from volunteers from non-governmental organisations, farmers and members of the public.37

Evidence is indicative that infected pigs were required to sustain transmission, based on the decline in the outbreak following movement restrictions on the farmed pig population, culling of infected herds and in the case of the Singaporean outbreak, bans on the importation of pigs from Malaysia by the Singaporean government. However, the evidence included in this review is low quality and based predominantly on the analysis of a case series using historical controls.

While there is mention of the total estimated loss to the swine industry in terms of cost,37 there is no information on the cost of the interventions. The sustainability and feasibility of using these interventions outside of an outbreak situation has not been discussed. The study by Muniandy outlines some future challenges to the swine industry in Malaysia, and makes recommendations for long term reform.37

Contextual factors

Details of contextual factors extracted from each included study are presented in Table 16. Several of the studies discussed the enormous impact this outbreak had on the pig industry in Malaysia.26, 29, 36, 37, 40 The eradication of 1.1 million swine represented about 40% of the swine population in Malaysia in 199929 and the number of farms were reduced from 1885 to 829 farms.36 Many pig farmers lost their livelihoods with the culling of their entire pig farm.

The outbreak caused dramatic changes in the pig industry, with pig farming only allowed in pig farming areas designated by the government. Restrictions on pork products and live pig exportation of Indonesian pigs by governments of Thailand, Singapore and the Philippines prompted the Indonesian government to initiate swine surveillance in Indonesia to restore faith in the Indonesian swine industry.

The transmission of Nipah virus was thought to be related to the movement of fruit bat populations in farming areas with the risk of greater exposure of pig farms to foraging fruit bats, although this has not been confirmed.37 The authors have suggested that intensification of traditional farming systems, particularly of pigs and poultry, has contributed to environments that enhance transmission of diseases from wildlife reservoirs. Furthermore, while intensification and expansion have been on-going, biosecurity measures, the lack of environmental impact assessments, inadequate pollution and waste management practices have left much to be desired.

Muniandy et al. also recommends changes locally.37 They refer to a traditional practice amongst pig farmers in Malaysia of sharing boars and moving sows from farm to farm and recommend that this practice be discontinued. In fact, it was the fire sale of sick pigs from one farm in Perak that was thought to be responsible for the initial spread of the outbreak.30

Ozawa et al. also discussed difficulties encountered by the trace back system in abattoir surveillance, pointing to irregularities with the tattooing system.40 Farm codes were tattooed on the back of the animals stamped by the butchers themselves. Ear notching was later introduced to circumvent fraud.

Summary

In summary, there was no evidence of a surveillance system in place to provide early warning of the outbreak. The only surveillance activities described were initiated in response to the outbreak to guide control measures in Malaysia, and as an active case finding exercise in Indonesia. All evidence on control interventions is based on the outbreak response in Malaysia, which included mass culling, quarantining, and movement restrictions. There is no information on the cost of the interventions. The sustainability and feasibility of using these interventions outside of an outbreak situation is likely to be low, and there is no evidence for more sustainable ongoing activities such as animal tracking systems or health promotion campaigns. Despite many reports of ongoing surveillance activities there has been no evaluation of these systems.

Results for interventions targeted at dengue

The original searches retrieved 1302 potentially relevant articles. This included 15 systematic reviews which were identified during this stage and the reference lists for these were examined for additional references. From this list, 118 papers were short-listed for more comprehensive examination. 87 articles were rejected after perusal of the abstract and full text. Reasons for exclusion are outlined in Appendix V and include papers based on: cross-sectional surveys or reports of surveillance activity with no intervention, narrative reviews with no original data and model-based studies. Thirty-five papers were critically appraised for methodological quality and based on these, 31 papers were included in the review (Figure 2). The studies were categorised by country, urban/rural setting and type of intervention to facilitate analysis. Characteristics of the included studies are detailed in Appendix IV (Table 17 to Table 20).

Six studies evaluated the effectiveness of community-based surveillance programs for dengue. These studies came from different countries (Thailand, Indonesia, Singapore, Viet Nam, Cambodia and Malaysia). 28, 31, 32, 38, 39, 41 Two studies evaluated established national level systems,28, 32 whilst the remaining four were evaluations of novel or improved systems undertaken at a more local/regional level.31, 38, 39, 41

Twenty-six studies report on 28 evaluations of dengue prevention and control activities. Much of the evidence for community based dengue interventions comes from studies undertaken in Thailand47, 49, 58, 64–66, 71, 72, 74, 76, 79 (n=11) and Viet Nam38, 52, 53, 55–57, 61, 62 (n=8). Within these countries, studies have been undertaken in a wide range of regions so evidence is available from northern, central and southern provinces. The remaining studies were undertaken in Singapore44 (n=1), Malaysia48 (n=1), Myanmar76 (n=1), Cambodia69 (n=1), the Philippines59, 76 (n=2) and Indonesia45, 70, 77 (n=3).

Three studies evaluated community dengue health education and disease awareness campaigns.45, 64, 74 A further thirteen studies44, 47, 49, 55–57, 61, 62, 66, 69–72 used educational components in conjunction with a combination of environmental, biological and occasionally chemical vector control. Five studies48, 52, 70, 71, 77 evaluated environmental control strategies (including use of screens, covering of water containers, and community clean up to reduce larval breeding sites), the majority included an educational component. A further eleven studies44, 49, 55–59, 61, 62, 66, 72 included environmental control activities alongside chemical or biological control interventions. Seven studies55–58, 61, 62, 65 looked at biological vector control strategies (including introduction of either copepods or other biological control agents to water containers); in all studies this was in combination with dengue education and environmental control activities. Eleven studies38, 44, 47, 49, 53, 59, 66, 69, 72, 76, 79 looked at chemical vector control strategies (including larvicide/insecticide distribution and fogging), either in isolation or occasionally combined with either dengue education or environmental control activities.

Methodological quality of the studies

To summarise the general level of evidence for each type of intervention:

• Surveillance – medium quality (generally appropriate study design (cross sectional studies) and analysis but do not control for confounding factors such as epidemic pattern of disease, changes in population structure, changes in patterns of urbanisation and concurrent disease control initiatives)

• Environmental control – medium quality (small sample size, insufficient follow up periods, inappropriate or no control groups, do not control for confounding factors such as seasonality and epidemic pattern of disease, focus on vector rather than disease outcomes)

• Biological control – high quality (adequate sample size, control for seasonality, appropriate control groups, full description of intervention, full description of baseline characteristics of intervention and control groups, report vector and disease outcomes)

• Chemical control- medium quality for national studies (sample size and follow up periods adequate, do not control for confounding factors such as seasonality), low quality for local level studies (small sample size, no or inappropriate control groups, report vector and disease outcomes)

• Educational interventions – low quality (small sample size, insufficient follow up periods, inappropriate or no control groups, do not control for confounding factors including seasonality and concurrent disease surveillance and control activities, focus on process outcomes)

A major limitation of the body of evidence evaluating prevention and control activities is the reliance on entomological indices to evaluate program effectiveness, as the correlation of these indicators with clinical indicators is relatively weak.84 Of these studies, eighteen47, 49, 52, 56–59, 61, 62, 64–66, 70–72, 74, 76, 77 use larval indices as a main outcome measure, and ten47, 48, 53, 55, 57, 58, 61, 62, 65, 77 report adult mosquito indices as a primary outcome. Fourteen studies38, 44, 49, 55–57, 61, 64, 66, 69, 71, 72 used numbers of dengue cases or dengue incidence as an outcome. Two studies53, 58 reported that clinical indicators of dengue could not be used as no cases of dengue were reported from either the intervention or control site but were able to report outcomes in terms of number of positive dengue serology results. Five studies45, 56, 57, 61, 74 reported data on knowledge, attitude and practice indicators, five53, 56, 71, 76, 79 presented data on the uptake or acceptability of the intervention and four57, 69, 76, 79 presented measures of cost or sustainability of the program.

Seven studies had a follow up of 6 months or less,44, 49, 59, 64, 70, 76, 77 four of less than one year,38, 48, 53, 65 and eight had a follow up of less than two years45, 47, 52, 58, 71, 72, 74, 79. These studies are unable to evaluate the impact of the intervention as fluctuations may reflect the seasonal and epidemic trends in vector and disease indices seen with dengue rather than any effect of the intervention. Short duration of follow-up also limits interpretation of the sustainability of results. Six studies did not have a control group45, 49, 69, 70, 72, 79 and one study picked inappropriate control groups48 in that the intervention was evaluated based on entomological indices, but the vector was not present in control sites at baseline.

Where activities have been evaluated in outbreak situations it is not easy to demonstrate effectiveness as reduction in the incidence of infection may simply reflect the natural pattern of peak and decline seen in epidemics. This problem is demonstrated in the study by Ang et al.44 which evaluates the use of “carpet combing” insecticide spraying exercises during a dengue epidemic in Singapore. Although results suggest that this activity was able to reduce the duration and peak of the epidemic, they do not control for other confounders such as changes in personal protective behaviour during the epidemic.

Dengue – Review findings

Surveillance interventions

Six studies looked at different aspects of the functioning of dengue surveillance activities. Details of the interventions evaluated and the main findings from each included study are presented in Table 18 and Table 19. One study evaluated a passive surveillance system in terms of its ongoing functioning for monitoring endemic dengue,31 three studies looked at active surveillance of suspected dengue cases,38, 39, 41 whilst the other two evaluated the ability of surveillance data to predict or provide early warning of outbreaks or epidemics of dengue which occur periodically in endemic areas.28, 32 Outcomes used in these studies include the number of dengue cases and/or incidence rates, predictive ability of the system (generally in terms of outbreak or epidemic warning), and sensitivity and specificity of the system. One study41 reported data on the cost of the surveillance system.

The evaluation by Chairulfatah et al. was of the local surveillance system in Bandung, Indonesia.31 The authors reported significant underreporting of hospital cases to the local Municipal Health office (only 31% reported). Poor record keeping impacted on assessment of the system's timeliness. No other qualities of the system were evaluated (representativeness, positive predictive value). The studies on active surveillance systems evaluated systems based on community reporting38, 39 and a sentinel GP surveillance pilot to detect suspected dengue cases.41 The study by Osaka et al.38 was inconclusive, as it seemed to be set up to look at the impact of concurrent interventions (done in conjunction with active surveillance) rather than improved surveillance, as both the intervention and control group received the active surveillance component. No information was provided on the increased cost of active surveillance. Oum et al.39 used syndromic surveillance definitions to conduct community-based surveillance on a number of diseases, including ‘haemorrhagic fever’ (HF). Their evaluation showed value in their approach, as only 33% of cases of HF had contacted a health facility, with 67% of them being treated at home, although they do not estimate a predictive value positive (PPV) for their definition of HF. The majority of deaths (80%) also occurred at home. The surveillance system also detected two clusters of HF reported in one commune. Other system attributes were not evaluated. The sentinel GP pilot41 compared a sensitive versus more specific case definition of suspected dengue cases presenting to two clinics. The more specific case definition uses diagnostic criteria for DHF outlined by the WHO, so it is not surprising that a higher proportion of patients were positive by serology (33% vs 7%) and virus isolation (50% vs 15%) using the latter case definition.

Barbazan et al. used retrospective surveillance data to show that the spatial analysis would allow focusing control activities on 5% of the months to control 37% of cases, and early warning of epidemics could have been done in advance.28 Chan et al. used web search query data to build a model that estimated ‘true dengue activity’.32 They showed good correlation of their predictive model with retrospective data using datasets from Indonesia and Singapore.

Prevention and control interventions

26 studies report on 28 evaluations of dengue control activities. Interventions evaluated were based on a variety of methods including environmental, biological and chemical vector control, as well as dengue disease awareness campaigns and health education activities. Details of the interventions evaluated and the main findings from each included study are presented in Table 18 and Table 19.

Education interventions

Three studies evaluated purely dengue education programs which did not include any other control activities.45, 64, 74 Two of these studies, undertaken with textile factory workers in Indonesia45 and a rural community in Thailand,64 focused on evaluating measures of community engagement with the program as opposed to changes in vector indices or disease outcomes. Both showed an increase in knowledge amongst participants about dengue symptoms and transmission, and awareness about how to reduce vector breeding habitats. Therawiwat et al.74 (also undertaken in a rural Thai community) measured both knowledge and larval indicators. They showed significant increases in knowledge and self efficacy in control of dengue, along with a 90% reduction in larval indices by the end of the study. No data are presented by any of the studies on whether effects translated into any impact on the number of cases of dengue.

Environmental control

Four studies evaluated environmental control strategies. One study looked at the provision of new water tanks with solid covers;52 this study showed a reduction in larval indices (average number of larvae per container) in the new tanks, but no impact on larval indices in old existing containers at the study site. Overall there was not a significant reduction in larval indices. The new tanks showed high levels of acceptability amongst the community. The other three studies looked at the effectiveness of environmental cleanup campaigns in combination with dengue education and awareness activities.48, 70, 71 Crabtree et al. used a strategy based in schools and the general community with use of mass media and targeted activities to promote community awareness.48 They found a 60% reduction in the number of households in the intervention area positive for Aedes aegypti, however, the vector was not present in the control area at baseline, which effectively meant that there was no control group. The number of households positive for Aedes aegypti increased toward the end of the study period. Suroso et al. used a predominantly school based strategy to promote clean up amongst the wider community.70 They found a 50% reduction in larval indices in households, but only a 35% reduction in households with school children and school buildings. They concluded the program had been less successful amongst school children. Suwanbamrung et al. conducted their study in three semi-urban communities and used targeted community education activities to promote clean up campaigns.71 They showed a 50% reduction in household index and an 80% reduction in container index in the village with high levels of community engagement and dengue control capacity. In the two villages where capacity and engagement were lower they demonstrated only a 15% reduction in these larval indices. None of these studies reported data on the cost of the intervention or provided follow up beyond one year to look at sustainability.

Biological control

Seven studies evaluated a dengue control program that included use of biological control agents. One study65 is a pilot study of the use of Larvitab^© (larvicidal bacteria) in a rural location in Thailand and includes no educational or environmental activities. The study reported a 70-85% reduction in larval indices and a 75% reduction in adult mosquito indices (as compared to only a 10% and 35% reduction respectively in the control group). The other six studies evaluated the use the copepod Mesocyclops as a biological control agent in water containers and also included health education and disease awareness, and environmental cleanup activities as part of the control program. One of these was a study undertaken in urban Thailand58 that found that the percent of containers positive for larvae went from around 38% to close to 0%, mosquito landing numbers went from around 1 to close to 0, and the percentage of children screened who were dengue sero positive went from 13.5% to 0%. In contrast the percentage of positive containers and children increased in the control area.

The other five studies evaluating copepods came from Viet Nam and were conducted by the same research group over a period of 15 years in a variety of rural and urban settings across North and Central Viet Nam. This group includes four original trials conducted in different communes55, 56, 61, 62 and a follow up study looking at the cost and sustainability of the interventions up to nine years post-intervention.57 The intervention was very comprehensive including use of copepods, environmental cleanup campaigns, the use of microcredit schemes to encourage development of recycling business, and broad community education activities and awareness campaigns. In all four original studies the intervention achieved a reduction in vector indices, reducing larval populations by over 97% 12 months post introduction of copepods and achieving 99% reduction or elimination with the addition of community education and environmental cleanup activities. The studies also reported a reduction in dengue incidence and this was also maintained with no project communes reporting local cases of disease (only a handful of imported cases) since 2003. In contrast larval population and dengue numbers remained present in control areas with figures fluctuating with the seasonal and epidemic nature of the disease. Participants reported a 99.5% rate of willingness to participate and a 97.8% acceptance of copepods.56 The average cost per person per year of the original program was estimated at $US2, with a marginal cost of expansion of 20c.55 In the follow up study, the average cost of the program was calculated to be 61c per person per year (equivalent to a total cost of $6,134 annually) in International Dollars.57 Using a self-developed tool to measure sustainability they found that the project rated 4.42 out of 5 indicating it was highly sustainable.57 Rates of both vector indices and cases of dengue remained at zero in the original project communes over five years after the end of the original research study.57

Chemical control

Eleven studies evaluated some form of chemical control (including insecticides, larvicides, fogging or spraying programs, or use of impregnated nets or curtains). Five of the studies which evaluated chemical methods of control also had an environmental cleanup component to the intervention to reduce vector breeding habitats.47, 49, 59, 66, 76 There was no systematic difference in effectiveness between these studies and those not including this component. Three studies looked at the use of impregnated nets or curtains.53, 59, 79 The two studies evaluating nets showed a significant reduction in adult vector indices53 and larval vector indices59 but were not able to show a reduction in dengue infections as measured by dengue serology.53 Madarieta et al. reported a significant increase in the number of people using non-intact nets or not using nets consistently over the six month study period.59 Vanlerberghe et al. showed that curtains were well accepted by the community79 but correct use was not sustainable – follow up observations noted nets with tears/holes, nets not hung and nets being used for other purposes (e.g. storage). No information was presented on dengue infection numbers or costs.

Four studies evaluated the use of a chemical larvicide (Temephos©) in water containers,47, 66, 69, 76 one study evaluated the effectiveness of fogging or spraying of insecticide targeted at the adult vector44 and two studies evaluated a combination of larviciding and spraying.49, 72 Of these seven studies, five measured outcomes in terms of larval indices, of which four studies showed that use of larvicides or insecticides reduced larval indices by between 50-80%.47, 66, 72, 76 The remaining study, which evaluated use of this method of control during an outbreak as opposed to within an ongoing control program, was not able to show any impact on larval indices.49 Only one study47 measured the impact of larviciding on mosquito landing rates and was not able to demonstrate any impact for the intervention. Three studies reported outcomes as numbers of dengue infections.44, 66, 69 All showed a reduction in dengue cases, of around 50% in urban areas and 80% reduction in the one rural area studied.66 Ang et al. reported that the impact of dengue spraying on dengue notifications was greatest during an outbreak than under endemic conditions.44 Two studies provided data on cost and sustainability. Suaya et al. estimated the cost of an annual larviciding program in Cambodia at 11c per person covered.69 Phantumacinda et al. reported that they only achieved 70-86% coverage in their larviciding program.66

Comparisons of control methods

Three studies compared different types of dengue control. Umniyati et al. compared a program based on environmental cleanup with repeat insecticidal fogging in an urban setting in Indonesia.77 They found that the environmental cleanup intervention was more effective than the chemical control program in reducing larval and mosquito indices. Osaka et al. compared use of insecticidal aerosol cans with ultra low volume (ULV) fogging in an urban region of Viet Nam.38 They found that use of aerosol cans for household spraying was more effective (a 71% v. a 52% reduction in dengue cases) and less costly (US$393 v. $US553) than the fogging program.

Tun-Lin et al.76 looked at the use of vector control programs targeted at the most productive container types versus untargeted control programs in urban settings in Thailand, Myanmar, and the Philippines (note that data from Viet Nam was excluded from the review as there was no follow up at this site). In the Philippines, two forms of environmental control were compared. Tyre splitting, water drum cleaning and waste management was compared to a general community clean up and awareness campaign. In Myanmar, introduction of biological agents (dragon-fly nymphs and fish) to the most productive water containers was compared to a blanket approach where all containers were targeted and chemical control (Temephos©) was used intermittently. In Thailand introduction of a biological larvicide (Bti) to the most productive containers was compared to use of chemical control (Temephos©) in productive containers, plus regular emptying of all other containers and occasional insecticide spraying. In all three countries both the targeted approach and the non-targeted approach were equally effective at reducing entomological indices (Breteau index) by 80% in Myanmar and the Philippines and 50% in Thailand. Implementation costs were reported for Myanmar and the Philippines. In Myanmar, the targeted vector control program had lower implementation costs ($4.47 per year per household covered) than the non-targeted campaign to which it was compared ($6.45 per year per household). In the Philippines, the targeted intervention had higher implementation costs compared to the non-targeted campaign ($9.32 v. $2.19 per year per household). In the Philippines high levels of acceptance of the interventions were reported. Coverage of 70% and 73.5% was achieved in the Philippines and Myanmar respectively.

Contextual information

Given the larger number of studies identified that evaluated dengue surveillance or control activities, contextual information extracted from each study were grouped under the following headings for discussion: contextual factors, behavioural mechanisms, and program design. Within each of these broad headings, subcategories have been used to draw conclusions across studies. Details of contextual factors extracted from each included study are presented in Table 20.

Contextual factors

Study context

There were geographical differences in the types of prevention and control interventions evaluated and the location of study sites. Sixteen of the 26 studies included study sites in rural locations. These studies came from Viet Nam52, 53, 55–57, 61, 62, Thailand47, 49, 58, 64–66, 74, Indonesia70 and Malaysia48, and looked at interventions that used a combination of health education, environmental vector control strategies (mainly focusing on reducing vector breeding sites) and biological vector control strategies (predominantly the introduction of copepods to both public and private water containers). Fourteen studies looked at the effectiveness of control programs in urban settings. This evidence came from Singapore44, Myanmar76, Cambodia69, Indonesia45, 77, Vietnam38, 55–57, 85, Thailand66, 71, 72, 79 and the Philippines59, 76 and was more likely to be evaluations of chemical forms of vector control (including use of insecticides in public water sources, fogging of dwellings and public buildings and promotion of use of insecticide treated nets).

Four interventions were evaluated in both urban and rural settings; these were impregnated bed nets (a chemical intervention), larviciding (a chemical intervention) introduction of Mesocyclops to water containers (a biological intervention) and community cleanup campaigns (an environmental intervention). Similar levels of effectiveness for all of these interventions were achieved in both settings except larviciding which appeared to be more effective in rural area at reducing both vector indices and dengue rates. Phantumacinda undertook their study in both urban and rural areas of Thailand and had higher levels of volunteer participation in the urban areas.66

Seasonality

Umniyati et al. compared source reduction of larval habitats with insecticide fogging in both the wet and dry season.77 Source reduction out-performed fogging at reducing larval numbers in both seasons. In relation to reducing mosquito numbers, source reduction was better than fogging in the dry season but in the wet season the two methods were equivalent. Swaddiwuhipong et al. found that a health education and temephos larviciding program was more effective in epidemic than inter-epidemic years.72 Ang et al. reported that the impact of dengue spraying on dengue notifications was greatest during an outbreak as opposed to under endemic conditions.44

Water supply infrastructure and environmental management

Crabtree et al. reported a lack of piped water supply and significant problems managing refuse due to lack of infrastructure in the rural coastal Malaysian villages as barriers to their environmental and educational program.48 Butraporn et al. reported poor wastewater management and a lack of affordability of piped water in their study as hindering the effectiveness of their chemical and environmental control program.47

The six studies 55–58, 61, 62 which included the use of copepods in their intervention commented that use of this method of control is applicable where the major breeding habitats for the vector are large water storage containers (which cannot be easily emptied and cleaned), that are used as stores from which smaller containers are refilled (thereby transferring copepods).

Models of behaviour change

None of the included studies refer to specific models of behavioural change being used to design the intervention programs, however all of the studies make reference to the fact that the programs were designed to result in changes in practices to prevent dengue infection and transmission.

Dengue knowledge

Beckett et al. evaluated an education program and showed that improvement in knowledge scores was strongly correlated with educational level.45 Therawiwat et al. found that education level and being male were predictive of high knowledge scores.74 Kay et al.57 showed a direct link between the frequency of household visits by dengue program volunteers, household knowledge of dengue prevention and the practice of dengue control activities. They also found that use of copepods as a biological method of dengue control was less successful when not combined with health education and awareness building activities. Kittayapong found that dengue education was needed to ensure that the water container covers distributed in their intervention were used properly.58 Vanlerberghe et al., who evaluated use of insecticide treated curtains, found that disease knowledge was not correlated with uptake or correct use of the curtains79 and Butraporn et al. found that increased knowledge and awareness did not translate to increased use of Temephos© in household water containers or improved waste management.47

Perceived importance of dengue

The three studies authored by Kay et al.55–57 report that health volunteers in their interventions were paid a stipend of $US 2-4 per month (approximately 4 days of work) plus given a uniform. In the follow up study57 they report that these stipends were not motivation for the volunteers. This in fact stemmed from the prestige of the position which derived from the value assigned to these roles by the community based on the severity of dengue as a public health problem. In contrast, the village health volunteers in the surveillance system evaluated by Oum et al. were reported to be motivated because they were financially rewarded.39

Chairulfatah et al. states that the doctors in their surveillance system often wished to postpone reporting until a diagnosis of dengue was confirmed and health municipality officials were often asked to report only patients with obvious dengue haemorrhagic fever or dengue septic shock.31

Perceived effectiveness of the intervention

Both Butraporn et al.47 and Vanlerberghe et al.79 reported a link between perceived effectiveness of the intervention amongst community members and continued engagement in program activities, with a decline in the use of Temephos© plus waste management and the use of impregnated curtains over the study period as participants failed to see reductions in mosquito and dengue rates.

Community input, ownership and involvement

In many studies, key community members were identified as leaders or champions for the program and these individuals were involved in developing the intervention and mobilising activities in the wider community. Kay et al.57 compared vector indices and dengue rates in project communes to those in communes which received a rollout of the intervention but which didn't offer communities the opportunity for local input and modification to the program prior to implementation. They report continued absence of the dengue vector Aedes aegypti and reduced rates of dengue in the original project communes whilst the non-project communes had higher rates of both outcomes.

In the chemical control programs, community involvement in the intervention was generally passive, i.e. they received the program. The interventions were designed, coordinated and run by centralised agencies or teams. Community members' involvement was usually restricted to uptake of household strategies such as use of impregnated nets or larvicides, or allowing access to the household for spraying activities. In contrast, community involvement was active in the education, environmental and biological control programs. Community members were involved in the design and planning of environmental cleanup strategies, involved in development and delivery of the health education and disease awareness components of the program and trained and used for the distribution of biological agents and ongoing monitoring and evaluation of the programs.

High levels of community ownership and involvement are consistently reported as an important factor in the success of these control programs; Crabtree et al. state that the grass roots community action promoted success of their environmental cleanup program48 and Suwanbamrung et al.71 reported that their community education and environmental cleanup campaign resulted in a significant reduction in vector indices in all three villages. However, the village with the highest community capacity for dengue control amongst leaders and the general community recorded the lowest entomological and epidemiological indicators. Nam et al.62, Nam et al.61, and Kay et al.55, all report that community leaders mobilised the whole community to take high levels of ownership of the program, which enabled a multi-level community approach to control. Nam61 report that continuous community input is required into their intervention based around use of copepods and environmental clean up to prevent reinfestation with the dengue vector. Vanlerberghe found that active engagement of the community in promoting continued use of impregnated curtains was more important in increasing uptake than continued educational messages about dengue.79

Crabtree et al. report spin-off benefits to the community from participating in their intervention.48 These were increased civic pride, well-being, and more effective networking and self-advocacy with government agencies as a result of their environmental and education based program.

Use of schools to deliver education activities

Suroso et al. delivered their educational and environmental intervention primarily through schools.70 They reported lower reduction in vector indices in school premises and households with schoolchildren, relative to households with no school children, and attribute this to a lack of motivation amongst school children. In contrast, Phantumacinda et al. reported that students were better volunteers in their larviciding intervention than village participants,66 and Swaddiwudhipong et al. found that their education program was more effective in schools compared to private households and other public buildings.72 Kay et al. used schools as a key platform for delivering education and awareness activities to both school children and the wider community,55 highlighted the importance of school children in providing an important service to the community in the inoculation of copepods to water sources as necessary,57 and reported that teachers and school children were particularly important in the success of clean up campaigns.56

Community attitudes toward government responsibilities

Crabtree et al. reports that sustainability of their environmental control program in Malaysia is threatened by local attitudes that place responsibility with government agencies to address and enact positive changes in local environment and infrastructure.48 These attitudes undermine sustainability of changes to attitudes and behaviour regarding environmental clean up to reduce vector breeding habitats. This is in contrast to Nam et al.'s evaluation of a biological and environmental control program in Viet Nam which cites the importance of recycling as an economic activity as one of the reasons why clean up campaigns were so successful in reducing entomological indices.62

Program structure and delivery

Cost of the intervention

Financial support for all the surveillance systems came, at least in part, from international donors or commercial organisations. The study which evaluated the use of internet sources for rumour surveillance was sponsored by Google who provided access to the data and technical support.32Two studies commented on financial constraints to their surveillance system; Chairulfatah et al. reported that they could only perform serology tests as opposed to the more definitive recovery of virus tests in their system,31 whilst Osaka et al. reports that an extension of their system beyond a pilot would likely require use of cheaper serology testing.38

Adequate investment and resourcing for both start up and maintenance was also frequently identified as an important factor for prevention and control interventions. Beckett et al. states that their budget was too low to enable them to reach all community members with their educational intervention.45 Eamchan et al. reports the high price of larvicidal agents (specifically Temephos©) as a potential barrier for ongoing use.49 Swaddiwudhipong et al. which had included this as part of their intervention was forced to drop it during the final (and epidemic) year of their intervention due to a lack of funds.72 Phantumacinda states that ongoing supply of Temephos© is necessary as periodic mass campaigns are less effective and therefore not economical or practical.66 Phan-Urai et al. notes that Larvitab© (like Temephos©) requires repeat dosing at regular intervals.65 In contrast, the six studies55–58, 61, 62 which included the use of copepods in their intervention stated that all control tools were locally produced including the copepods which can be farmed locally for minimal cost.

The cost to participants of the intervention may facilitate or inhibit success of the program. Hien et al. studied the use of new containers with solid lids that were provided free to the community.52 Whilst larval indices in these containers were low, the community continued to use many old containers and there was no overall reduction in larval indices. Kay et al.55–57 report their interventions included microcredit schemes for small businesses that were based around recycling and waste removal. These acted as catalysts for sustained environmental cleanup and some of the profits from these activities are reinvested into other dengue control activities.

Acceptability of the intervention

None of the studies evaluating educational, environmental or biological control programs reported issues with acceptability of the program in community participants. Several studies specifically reported high levels of acceptability by the community.48, 55, 71 The use of copepods in water containers was also well accepted and reported as requiring minimal time and effort to sustain.57 Phan-Urai et al. reports how participants had no complaints about the use of Bti (a biological larvicide) in water supplies.65 The product was perceived as safe and preferable to Temephos (a chemical larvicide) which was thought to be oily and raised concerns about the use of chemicals in drinking water. Eamchan et al. and Phantumacinda et al. also reported issues with acceptability of Temephos, including smell, taste and not wanting to place the agent in drinking water.49, 66 Madarieta et al. and Vanlerberghe et al. both reported a decline in use or correct use of impregnated curtains over their study to less than 50%.59, 79 Igarashi et al. found that 100% of households found impregnated bed nets simple, convenient and comfortable to use.53

Technical support

Pengvanich et al. states that a long-term version of their program would need support from authorities not just volunteers. Kay et al.57 state that communes in Viet Nam which received a rollout of their biological, environmental and educational intervention without support from a technical program team achieved lower reductions in entomological and epidemiological indices than the original project sites. In their evaluation of an educational intervention, Therawiwat et al. report that interaction between key stakeholders and researchers enhanced reflection and dialogue amongst stakeholders. However, Oum et al. reports that there was tension between the village health volunteers (VHV) used in their surveillance system and official health staff who were often not receptive to the VHVs efforts.39

Use of targeted v. blanket strategies

Tun-Lin et al. found that the targeted intervention used in Myanmar was less costly but equally effective as a non-targeted strategy, whilst in the Philippines, where a strong social intervention component was included in the program, the targeted intervention cost almost five times more than the non-targeted intervention for comparable levels of effectiveness in regards to reduction of vector indices.76 Kittyayapong et al. report that whilst targeted vector control could feasibly be rolled out beyond a research program in their Thailand setting it was likely to be too costly to implement.58

Meta-analysis

We included a number of outcome measures including household index, container index, Breteau index, the larval population number, larval density index, the mosquito biting rate and the number of cases of dengue infection. Definitions for each of these outcome measures are given in the methods section and further description of the meta analysis can be found in Appendix VII. The first five outcome measures measure Aedes aegypti larval populations in a number of settings (for example, in houses, in containers, as a total population number), the sixth measures the presence of adult mosquitoes, and the final is a measure of clinical infection with dengue virus. The studies varied in size from 61 to 6341 households and 1163 to 2.9 million people (represented by the size of the square in the forest plot), and covered a range of interventions, including environmental, educational, biological and chemical interventions, as well as a combination of more than one intervention.

Household index

Ten studies measured household index as an outcome measure. The meta-analysis showed that the dengue control interventions resulted in a statistically significant reduction in the household index giving a summary mean odds ratio of 0.21 (95% credible interval 0.05, 0.68) (

Table 21). Despite the forest plot showing heterogeneity between studies, with Crabtree et al. and Madarieta et al. reporting that the intervention increased household index (Figure 3), the sensitivity analysis showed that the summary mean odds ratio was stable to the influence of each individual study.

Table 21: Mean odds ratio and 95% credible intervals from studies reporting household index.

Figure 3: Forest plot of odds ratios from ten studies reporting household index; dengue control interventions led to a significant reduction in household index.

Container index

Six studies measured container index as an outcome measure. The meta-analysis showed that the dengue control interventions resulted in a statistically significant reduction in the container index, giving a summary mean odds ratio of 0.38 (95% credible interval 0.15, 0.94) (Table 22). The forest plot shows heterogeneity between the individual study findings, with Madarieta et al. reporting the intervention had no impact on container index (odds ratio=1) (Figure 3). However, the sensitivity analysis showed that the summary mean odds ratio was stable to the influence of each individual study.

Table 22: Mean odds ratio and 95% credible intervals from studies reporting container index.

Figure 4: Forest plot of odds ratios from six studies reporting container index dengue control interventions led to a significant reduction in container index.

Breteau index

Twelve studies measured Breteau index as an outcome measure. The meta-analysis showed that the dengue control interventions resulted in a statistically significant reduction in the Breteau index giving a summary mean rate ratio of 0.40 (95% credible interval 0.26, 0.61) (Table 23). The forest plot shows homogeneity between studies (Figure 3), and the sensitivity analysis showed that the summary mean odds ratio was stable to the influence of each individual study.

Table 23: Mean rate ratio and 95% credible intervals from studies reporting Breteau index.

Figure 5: Forest plot of rate ratios from twelve studies reporting Breteau index; dengue control interventions led to a significant reduction in Breteau index.

Larval population number

Only two studies measured the larval population number. The meta-analysis showed that the dengue control interventions resulted in a non-significant reduction in the larval population numbers giving a summary mean rate ratio of 0.21 (95% credible interval 0, 156.9) (Table 24). The small number of studies is the reason for the wide credible interval. The forest plot shows homogeneity between the individual study findings (Figure 6). As there were only two studies, a leave one out sensitivity analysis was not performed.

Table 24: Mean rate ratio and 95% credible intervals from studies reporting larval population number.

Figure 6: Forest plot of rate ratios from two studies reporting larval population number; dengue control interventions led to a non-significant reduction in larval population number.

Larval density index

Three studies measured the larval density index. The meta-analysis showed that the dengue control interventions resulted in a non-significant reduction in the larval density index giving a summary mean rate ratio of 0.09 (95% credible interval 0, 11.51) (Table 25). The small number of studies is the reason for the wide credible interval. The forest plot shows homogeneity between studies (Figure 7), and the sensitivity analysis showed that the summary mean odds ratio was stable to the influence of each individual study.

Table 25: Mean rate ratio and 95% credible intervals from studies reporting larval density index.

Figure 7: Forest plot of rate ratios from three studies reporting larval density index; dengue control interventions led to a non-significant reduction in larval desntiy index.

Mosquito bite rate

Two studies measured the presence of adult mosquitoes by recording the mosquito bite rate. The meta-analysis showed that the dengue control interventions resulted in a non-significant reduction in the mosquito bite rate giving a summary mean rate ratio of 0.68 (95% credible interval 0, 634.83) (Table 26). The small number of studies may be the reason for the wide credible interval, as the forest plot shows homogeneity between studies (Figure 8). As there were only two studies a leave one out sensitivity analysis was not performed.

Table 26: Mean rate ratio and 95% credible intervals from studies reporting mosquito bite rate.

Figure 8: Forest plot of rate ratios from two studies reporting mosquito bite rate; dengue control interventions led to a non-significant reduction in mosquito bite rate.

Rate of dengue haemorrhagic fever

Seven studies measured the rate of dengue haemorrhagic fever. The meta-analysis showed that the dengue control interventions resulted in a non-significant reduction in the infection rate giving a summary mean rate ratio of 0.22 (95% credible interval 0.02, 1.32) (Table 27). The forest plot shows homogeneity between studies (Figure 9), and the sensitivity analysis showed that the summary mean odds ratio was stable to the influence of each study.

Table 27: Mean rate ratio and 95% credible intervals from studies reporting the rate of dengue haemorrhagic fever.

Figure 9: Forest plot of rate ratios from seven studies reporting the rate of dengue haemorrhagic fever; dengue control interventions led to a non-significant reduction in Dengue haemorrhagic fever.

Summary of results for meta-analyses for dengue control interventions

The results of the meta-analysis showed that overall, the interventions included in this review were able to show a statistically significant impact on larval indices; including approximately an 80% reduction in the proportion of positive households, approximately a 60% reduction in the proportion of containers positive for Aedes aegypti larvae and approximately a 60% reduction in the Breteau index, when results are pooled across type of intervention. Although we anticipated being able to draw indirect comparisons of effectiveness between intervention types, the small number of studies for any one intervention type precluded formal sub-analyses to look at relative effectiveness. However, a narrative interpretation of the forest plots shows no trend in levels of effectiveness by type of intervention, country, urban versus rural context or study size.

Two studies48, 59 showed inconsistent results (an increase in larval indices as opposed to a decrease), but this difference does not appear related to the type of intervention. The study by Crabtree et al.48 trialled an environmental cleanup intervention; it is a low quality study that is weakened by its inappropriate choice of control area (the mosquito vector was not present in the control area at baseline). The study by Madarieta et al.59 was a small, short, low quality trial of insecticide impregnated bednets. A feature common to both studies that may partially explain their contradictory findings is that each reported problems with the sustainability of the intervention. After 6 months, 52% of nets were no longer in use and 60% of nets in use had been washed (reducing their insecticidal properties). At the end of the environmental cleanup study, the authors report ongoing waste management issues, and a failure to alter ingrained attitudes that the government should address these issues, rather than seeing them as a community responsibility.

Pooling of results across intervention types estimates that vector control results in approximately an 80% reduction in the rate of dengue haemorrhagic fever, however this result does not achieve statistical significance. Effectiveness does not appear to vary by country or urban/rural context. There is the suggestion of a slight trend to greater levels of effectiveness for interventions incorporating biological vector control methods56, 58, 61 versus chemical vector control methods38, 66, 69, 72, but this is not statistically significant. Only three studies56, 66, 72 included in the meta-analysis reported data on both larval and dengue outcomes. All showed consistency in the direction of the effect across outcomes, however all showed a bigger reduction in larval indices than number of dengue infections. This provides support for use of larval indicators as an intermediate outcome in evaluations of dengue control interventions, but suggests that they cannot be used to directly estimate the impact of the intervention on disease outcomes.

Summary

A substantial body of evidence is available evaluating the effectiveness of dengue control interventions and surveillance activities. The available evidence comes from countries across SE Asia providing confirmation these programs work in a diverse range of geographical and social contexts. A wide range of vector control interventions have been evaluated, including chemical, biological and environmental methods of control. These have been evaluated both in isolation and in conjunction with health education and disease awareness campaigns. The majority of this evidence relies on entomological indices to evaluate programs (as opposed to disease outcomes). Duration of follow-up varied from one month49 to five years.55, 61

A review of included studies evaluating surveillance interventions show that well-functioning surveillance systems can be successfully used to spatially and temporally predict dengue epidemics in Thailand. This result has not been replicated in other countries. Community based surveillance methods appear to offer improvements over hospital/clinical surveillance in terms of sensitivity of the system, particularly in settings where there are significant financial barriers to accessing healthcare that results in under-reporting of dengue case numbers.

The dengue vector is amenable to many forms of chemical, biological and environmental control. The meta-analysis showed that overall, the interventions included in this review were able to show a statistically significant 80% reduction in the proportion of positive households and a 60% reduction in the proportion of containers positive for Aedes aegypti larvae, regardless of the type of intervention. There was also a non-significant 80% reduction in the rate of dengue haemorrhagic fever.

Interventions based on health education, environmental and biological vector control appear to be effective, low cost, well accepted, and sustainable in both urban and rural settings. Interventions based on chemical control in urban settings appear to be well accepted and there is evidence for effectiveness but they are expensive and there is limited evidence on their sustainability. A single study comparing environmental clean-up with repeat fogging found that environmental cleanup was more effective at reducing mosquito numbers than the chemical control program. A study reporting on evaluations in a range of countries showed that targeted environmental and larviciding interventions are as effective at reducing vector indices as blanket interventions and have lower implementation costs. Sub-group meta-analyses by intervention type or rural and urban settings were not possible because of the small number of eligible studies.

Key factors for success in interventions that have shown sustained reductions in entomological indices and disease incidence are the use of behavioural change strategies within their education and awareness programs, combined with support and investment in ongoing environmental management, high levels of community ownership of the program, and sufficient investment and resourcing for both start up and maintenance. There have been few evaluations comparing types of control and the cost-effectiveness of these programs has not been evaluated.

Results for interventions targeted at SARS

The search strategy identified 854 potentially relevant titles, of which 110 were shortlisted (Figure 2), based on details in the title and abstract. 96 original references were excluded (reasons are given) in Appendix V and papers which were outbreak reports with no data included on interventions, narrative reviews with no original data, cross-sectional KAP surveys and model-based studies. Nine systematic reviews that were identified were also excluded, as these focused on either a geographical region outside of the scope of this review(either China, Hong Kong and/or North America), the use of pharmaceutical or clinical interventions (vaccines and lab assays for clinical diagnosis), or the prevention of nosocomial (as opposed to community) transmission. Full text of twenty studies was retrieved, of which five studies were critically appraised and subsequently included in the review (Table). Four of the five were from Singapore34, 51, 63, 73 with the remaining study originating from Viet Nam75.

Methodological quality of the studies

All five studies were of low quality and did not control for confounding in their assessment of the effectiveness of the interventions studied. The study evaluating workplace-based surveillance for febrile disease was conducted over too short a timeframe to capture seasonal fluctuation in the incidence of this illness and no data is given on the sensitivity and time-sensitivity of the system.34 The results from the study are also unlikely to be generalisable to a wide range of workplaces as the study was conducted in a tertiary hospital setting where there were well-established reporting hierarchies and electronic documentation of staff sick leave.

The evidence for prevention and control interventions is derived from descriptive studies based on outbreak data from the 2003 global outbreak.51, 63, 73, 75 Given the high profile of this outbreak and the laboratory resources available in Singapore and Viet Nam, the datasets used are likely to be comprehensive and capture all symptomatic infections, giving an accurate picture of the epidemic and any impact of prevention and control interventions. However, this type of data also presents major limitations. Firstly, it reduces the ability to determine the impact of these individual interventions from amongst the range of community-based and government strategies that were implemented at that time. Secondly, it is unclear whether any impact is generalisable to future outbreaks, as the studies are unable to control for features unique to the 2003 outbreak. These include epidemiological features of SARS (such as the fact individuals were symptomatic whilst infectious and the relatively low risk of transmission compared to an infectious organism such as measles). Thirdly, they are based on retrospective data and are unable to obtain data on confounders or contextual factors if these were not collected at the time.

SARS – Review findings

Surveillance interventions

Of the papers included in this review, only a single study was identified that reported on ongoing surveillance systems for SARS.34 Details of the intervention and the main findings from the study are presented in Table & Table. The study evaluates the practicality of post-SARS surveillance recommendations in Singapore. The study focuses on the use of staff electronic medical records for early detection of outbreaks of febrile illness. Although the study is conducted in medical staff at a large general hospital, it is included here as it is being used as an early detection system for outbreaks rather than solely to prevent nosocomial transmission. The study finds that as documented fever is rare in sick leave amongst staff, passive surveillance in healthcare workers would be efficient in identifying outbreaks of febrile illness. Effective markers were found to be clustering of illness, prolonged or repeated absence from work, or the incidence of abnormally high fevers. The authors conclude that such a system is practical and likely to be sensitive in this setting should appropriate indicators be chosen, but that currently the system is not specific. This may lead to many false alarms and ultimately to outbreak “fatigue” whereby people fail to respond to early warning signals. The authors also note that surveillance is time-consuming and resource intensive.

Tan et al.73 provides descriptive information only about the use of temperature screening amongst school children during the epidemic. The study reports that none of the children diagnosed with SARS were detected through this system despite the extensive effort and resources this system required. They describe the benefit of this type of surveillance as psychological, with the purpose being to reassure parents and the public that schools were safe during the outbreak.

Prevention and control interventions

Four studies reported on community-based interventions to prevent and control SARS. Details of the interventions evaluated and the main findings from each included study are presented in Table and Table. Three of the studies are from Singapore51, 63, 73 and evaluated the effectiveness of home isolation and quarantine protocols and its contact tracing policy. All of the studies were based on retrospective analyses of the same outbreak dataset. The three studies reach the conclusion that the system was effective as it was able to reduce the time from onset of SARS symptoms to isolation from nearly one week to just over a day. They showed that a wide-net approach (i.e. pre-emptive isolation of exposed individuals using a broad definition of exposure) to surveillance and isolation of suspected cases was effective in ensuring progressively earlier isolation of probable SARS cases as the outbreak progressed. They also saw a reduction in the number of secondary infections per case. Only 0.3% of those quarantined broke quarantine. One quarter of all SARS cases had been on quarantine orders prior to diagnosis. Only 0.5% of those isolated went on to develop SARS. It is noteworthy that these interventions were evaluated within an outbreak setting and that a range of other community-focused strategies were also put into place.

A single study from Viet Nam75 is a risk factor analysis for SARS transmission in contacts of SARS cases in Viet Nam. The study looked at the effectiveness of masks in preventing transmission of SARS from index cases to known contacts (in particular household members and carers). Unfortunately this was a small observational study based on retrospective data and 95% of SARS contacts reported never wearing a mask, so no conclusions could be drawn about their effectiveness.

Contextual factors

Details of contextual factors extracted from each included study are presented in Table31. Escudero et al.'s study of a work-based surveillance system acknowledges that electronic documentation of staff medical certificates were important in enabling the system to function in a timely manner and at low cost.34 The authors also highlight that the study coincided with admission of an isolated case of SARS contracted due to a laboratory accident, which may have both increased awareness of febrile illness amongst staff and improved participation and acceptance rates amongst staff members.

All four studies of prevention and control activities highlighted that there were particular epidemiological features of SARS that made it more amenable to control – namely that patients were only infectious whilst symptomatic.51, 63, 73, 75 This makes it easier to identify when and where interventions need to be put in place for infected individuals to prevent transmission. It may also increase uptake of interventions amongst non–infected individuals as people are better able to judge their risk of infection.

The three studies evaluating isolation and quarantine policies in Singapore all stated that strong government/political leadership and high levels of community support were important factors in successful implementation of quarantine measures.51, 63, 73 Other factors identified by these studies as contributing to success in both implementing isolation and quarantine measures, and in halting the epidemic, were good and timely communication both between agencies and outward to the general public and substantial investment to develop information technology systems and laboratory systems capable of providing accurate and timely information over the course of the outbreak. Ooi et al. also highlights that Singapore has particular features (small population, high GDP, urban setting) that facilitate the ability to implement large-scale quarantine and states that “imposition of large-scale quarantine should be implemented only under specific situations in which it is legally and logistically feasible”.63

Ooi et al. gives useful information about the public and individual response to the isolation and quarantine policy, stating that stigmatisation of quarantined individuals was reported.63 Those quarantined were generally agreeable to being confined at home whilst the response was less positive to potential confinement in an institution such as a health centre. Finally, the study indicates that substantial resources were directed toward quarantined individuals, including repeat visits by nurses to deliver health education, the installation of electronic surveillance systems in each household to monitor compliance to quarantine orders, and the use of financial incentives to compensate individuals for lost income. All of these factors are likely to have contributed to the low rate of non-compliance reported in the study.

The study by Tuan et al.75 did not provide any contextual information about why people did not wear masks.

Summary

Five studies were included in the review, one looking at a work-based surveillance system in a Singapore hospital, three evaluating the effectiveness of isolation and quarantine in Singapore as a response to the 2003 SARS outbreak, and one study from Viet Nam reporting on the effectiveness of masks in reducing risk of SARS in people exposed to SARS patients. Overall the evidence is low quality and based predominantly on analysis of case series data from the 2003 outbreak. All three studies that evaluated the impact of isolation and quarantine found this intervention to be effective, however, the major limitation in these studies is that they are all based on analysis of the same routine dataset, and none can control for the impact of the multiple other interventions that were put in place in Singapore at the time of the outbreak. The study reporting on use of masks was an observational study and was unable to comment on whether the intervention was effective as 95% of participants reported never wearing a mask. Important factors contributing to the success of isolation and quarantine policies were good organisation, adequate resources, good communication and public support.

Results for interventions targeted at avian Influenza

Unlike Nipah virus infection, where, to date, the outbreak in 1999 has been an isolated event in SE Asian countries, several outbreaks of avian influenza have been recorded in a number of SE Asian countries from 2004 to 2008. In line with our inclusion criteria, outbreak control measures for each country-specific outbreak were included as long as they had a significant component of community involvement and engagement, even if in concert with “top-down” government-driven initiatives.

The search identified 737 potentially relevant titles, of which 107 papers were short-listed for comprehensive examination (Figure 2). Of these, eleven studies were critically appraised and nine studies were subsequently included in the review (Table 32). Reasons for exclusion are outlined in Appendix V, and include papers based on: outbreak reports with no data on interventions, narrative reviews with no original data, risk factor analyses, model-based studies and cross-sectional KAP and prevalence surveys.

Methodological quality of the studies

Of the nine included studies, five evaluated an existing or newly established surveillance system,27, 33, 35, 42, 43 four studies evaluated prevention and control interventions in the form of education,46, 60, 78, 80 which was combined, in one study, with behaviour modelling.80 The surveillance program conducted in Indonesia also incorporated a prevention and control component.

With the exception of the evaluation carried out by Perry et al.42 all studies were of poor quality, with the most common limitation in most studies being inadequate evaluation and assessment of the effectiveness of the intervention, or if the assessment was carried out, the authors failed to present the results of the evaluation. In the study by Bhandari et al., 100 farmers participated in an educational intervention about proper biosecurity measures for the prevention and control of highly pathogenic avian influenza (HPAI).46 However, no information is provided about the knowledge of the participants on this subject matter prior to the intervention, and it is difficult to attribute the results to the intervention.

The study by Desvaux et al. conducted poultry market monitoring, surveillance of broilers and hens and surveillance of sentinel villages for the presence of HPAI.33 The authors admit several constraints identified during the implementation of the program impacted on the quality of the study: insufficient training of field staff (collecting the wrong swabs), biased selection of market places and small sample sizes resulting in the study not being representative. Evaluation of the performance of the system was also needed.

The study by Manabe et al. looked at how educational interventions in an intervention and control village influenced awareness relating to H5N1 and the accessibility of healthcare.60 There were some differences in the intervention and control groups (the control group reported a higher proportion of farmers) and also differences in participants pre- and post-intervention in the intervention commune (greater proportion of participants reported a higher economic level post-intervention). The educational intervention was evaluated by a qualitative survey using face-to-face interviews with a relatively small sample of only 16 participants from the intervention commune.

In an educational intervention in Cambodia, the study by Van Kerkhove et al. looked at training programs for village animal health workers following domestic poultry outbreaks in the area.78 The study evaluated changes in poultry handling behaviours before and after educational campaigns. The study had some limitations. There were differences in sampling methods in the 2006 survey (pre-intervention) compared to the 2007 survey (post-intervention). There were also some demographic differences between the two study populations, and poultry handling behaviours were self-reported, not observed independently.

Educational initiatives were also run in three countries in the Mekong region (Viet Nam, Cambodia and Lao PDR).80 KAP surveys were conducted pre- and post-intervention. Unfortunately, other than one pre- and post-intervention score on the effectiveness of the intervention in Viet Nam, no other assessment was carried out (or presented) for the Cambodian and Laotian studies, so there is no data presented on disease outcomes.

The study by Samaan et al. evaluated a rumour surveillance system based on information from internet news and public health mailing lists and chat rooms.43 The study covered all countries covered by the WHO Western Pacific Regional Office (WPRO), which includes several countries outside the scope of this review; however, the findings were heterogeneous across country settings.

The final three studies conducted in Indonesia all relate to the same program of participatory epidemiology for the surveillance, prevention and control of HPAI in backyard poultry. The studies by Jost et al. and Azhar et al. are interim reports of the program.27, 35 A comprehensive evaluation of the program was conducted by Perry et al. for the FAO Evaluation Service.42 The study is limited by changes in the form of data collection used during the period of evaluation, but aside from this the evaluation is wide-ranging and includes an assessment of the role of all stakeholders involved in the surveillance and prevention components of the program, the quality, clarity and adequacy of the program design, the quality of the data, program outputs and sustainability of the program.

Avian Influenza – Review findings

Surveillance interventions

Five papers detailed surveillance initiatives, three of them27, 35, 42 presenting data for the same active surveillance program established in Indonesia in 2006. Details of the interventions evaluated and the main findings from each included study are presented in Table33 and Table 34. The studies by Jost and Azhar are early reports of the establishment of the system and a three year report of results, respectively.27, 35 A comprehensive evaluation was conducted by Perry et al. in 2009 as part of the FAO Evaluation Service.42 The evaluation detailed the establishment of “participatory disease surveillance”(PDS), based on principles of participatory epidemiology, i.e. the application of participatory methods to disease surveillance. Participatory epidemiology recognizes that local people have very rich and detailed knowledge about the animals they keep and the infectious and zoonotic diseases that affect their livelihoods and endanger human health. The system focused on the detection of HPAI in the backyard poultry sector (defined by the FAO as sector 4) on a village-wide basis by veterinary surveillance officers, where it was commonly believed the majority of HPAI virus was harboured. It was later expanded to have a prevention and control component. The program was successful in training up a number of Master Trainers, who subsequently delivered training to more than 2,000 surveillance officers. In its three years of operation, it was operational in 76% (341) districts in Indonesia, 27/33 provinces, covering 25,525 villages where surveillance activities had been completed,27 1455 of these resulting in diagnosis of HPAI. As of March 2009, infection status of villages was determined as ‘infected with HPAI’ (2.5%, 490/19,673), 8.1% (1598) suspected infected, 3.1% (612) controlled and 86.3% (16,973) apparently free of infection. The authors comment that the surveillance system enhanced existing passive surveillance systems and addressed a gap in surveillance.

The study by Desvaux et al. focused on targeted surveillance of markets, semi-commercial poultry farms located in former outbreak areas, sentinel village monitoring to strengthen surveillance at village level, and serological surveillance of domestic duck farms.33 The study did not detect HPAI in the market (0/712) or farm (0/51) samples. Market monitoring: samples were collected in seven provinces. Interviews were conducted in 52 villages and on 23 farms, which were subsequently classified according to their risk of having faced an HPAI outbreak. 14/70 (20%) premises were not suspected, 3/70 (4%) were classified as low probability, 18/70 (26%) were classified moderate probability, 35/70 (50%) high probability. The authors identified several constraints during the implementation of program that impacted on the success of the study – lack of motivation of provincial staff, limited capacity of the central team to compile and analyse the data generated, weak diagnostic capabilities and the reluctance of farmers to have animals sampled. They also state that selection of animals in market places was biased and that sample sizes were below defined levels and hence not representative, which may explain the zero detection rate of HPAI in markets and on commercial farms.

The study by Samaan et al. used rumour surveillance to analyse rumours generated primarily by the media and email and evaluate if the rumours could offer timely assistance to potentially affected nations, prompt countries to undertake preparedness measures, and inform public and international community about relevant events.43 Rumours were followed up by email or telephone request to the relevant WHO country office to investigate their veracity. A total of 40 rumours were identified from 20 countries and one Special Administrative Region. 23% of the rumours were confirmed to be true. The authors conclude that this type of surveillance was successful in informing public health action, and was relatively inexpensive to conduct.

Prevention and control interventions

Five studies evaluated education based avian influenza prevention and control interventions. Details of the interventions evaluated and the main findings from each included study are presented in Table 33 and Table 34.

Four studies were identified where education and training were the main component of the prevention and control interventions, all of them were conducted in the Mekong region, which comprises Viet Nam, Cambodia and Lao PDR. The programs were aimed at communities in rural settings, focusing on increasing awareness of HPAI,46, 60 motivating people to access healthcare earlier60 and encouraging a change in hygiene and poultry handling behavior.78, 80

The study by Bhandari et al. conducted training to 100 farmers who then served as demonstrators for a model of proper biosecurity measures for the prevention and control of HPAI.46 The authors report that no outbreaks have been reported in the communities in the project areas since the intervention. The program was evaluated in more depth using the funder's own model, the Participatory Self-Review and Planning Toolkit, but the authors do not give details of the tool kit or the evaluation process.

The studies by Manabe et al.60 and Van Kerkhove et al.78 are both before and after intervention studies in comparable cohorts, the former in two agricultural communities in Viet Nam, the latter in two southern provinces in Cambodia. Baseline and post-intervention KAP scores were calculated by Manabe; they also reported differences in the frequency of health and hygiene behaviours between the intervention and control groups.60 The authors state a greater proportion of participants reported receiving information from a health care worker or a friend after the intervention, and more people were likely to seek early access to healthcare after the intervention. Habits such as touching and eating dead or sick poultry were reported both pre- and post-intervention. Van Kerkhove et al. also describes increased reporting to village chiefs, but not to the animal health officer.78 Awareness of HPAI was high, but understanding of transmission was still low. While there were some improvements to basic hygiene practices and reduction in risky poultry handling behaviours, some risky behaviours still persisted (allowing children to play with poultry, proper treatment of poultry in the household environment).

The study by Waisbord et al. was a large undertaking, with training delivered to 3840 district and commune women's union officers in Viet Nam, 810 village promoters in Cambodia and 93 reporters and editors in Lao PDR.80 The authors provide the number of people, districts, farmers, trained, as process measures. Only in the Viet Namese study do they report pre- and post-intervention KAP scores, reporting an increase from 54% to 92%. Nine percent of farmer households in Cambodia set up model farms after participating in the study, and in Lao PDR, AI coverage on TV and radio improved in both quality and quantity.

The HPAI program in Indonesia began as separate PDS and PDR (participatory disease response) teams, but was later rolled into combined surveillance and response officers. The surveillance component of the program was successful in detecting HPAI, and the PDSR education component also achieved good coverage (29,476 education meetings held with community leaders, 10,093, 6,804, 103,832 and 9,971 meetings held with groups of community members, other organizations, individual households and persons from commercial enterprises, respectively).27, 35, 42

Contextual factors

Details of contextual factors extracted from each included study are presented in Table 35. Some recurring themes emerged in several of the studies analysed. Several studies reported changing behaviours and customs was difficult,33, 42, 60, 78 particularly for residents of a rural area with a one-time educational intervention.60 Van Kerkhove et al. report that educational efforts that succeeded in raising awareness and knowledge about the disease did not always succeed in increasing the likelihood of reporting of suspected disease to the authorities (only to community leaders).78 Perry et al. also reported the difficulty in implementing poultry movement control in Indonesia in general, but particularly in the backyard poultry sector.42 While database recorded movement control was implemented for all HPAI confirmed cases, discussions held with farmers in field visits showed clearly that selling of surviving chickens was widely practiced. Lack of cooperation was also reported in the study by Desvaux et al. from farmers who were reluctant to have animals bled for studies.33

Nevertheless, they also reported other benefits and strengths of the programs, such as better collaborative networks both at a local level as well as between agencies, sometimes enhancing existing national systems already in place.42 The PDSR program in Indonesia had very positive impacts on revitalising veterinary services in Indonesia, and in particular in strengthening the local animal health services (Dinas), as well as empowering communities' access to public services. Manabe et al. acknowledged the importance of the involvement of local healthcare workers and administrators in H5N1 education and outreach, and that the main impact of the educational intervention was to increase people's trust in local health care providers.60 Waisbord et al. commented that training brought commune council people together and provided the opportunity to network and cooperate more closely in the future.80 However, it became apparent that the PDSR response alone was insufficient and unlikely to contain and eliminate the disease for a number of reasons: inability to offer compensation to encourage culling, inability by the officers to enforce movement control, inability of the farmers to buy cages and feed to restrain poultry. The program evaluators advocated the need for transition into more sustainable and responsive animal health services.42

Summary

Evidence for surveillance interventions of HPAI was identified in programs in Indonesia and Cambodia. The PDSR program in Indonesia has been very successful in training surveillance officers and detecting HPAI in backyard poultry. It has also added value to existing veterinary health services in Indonesia. Conversely, results from the surveillance interventions in Cambodia were equivocal because of several constraints that impacted on the success of the study. Prevention and control initiatives were identified in Cambodia, Viet Nam, Lao PDR and Indonesia. Several programs were not evaluated in terms of final outcomes (only process outcomes were used), or if evaluated, the results have not been published. Several studies identified risky poultry behaviour despite the educational intervention and efforts by disease control staff to contain and eliminate disease for a number of reasons. The need to transition to more sustainable, long –term animal health services was also discussed. Despite this, a benefit of these programs has been to strengthen local collaborative networks and bring people together.

Discussion

Rabies

Surveillance interventions

None of the five studies included in this section of the review present any ongoing community-based human or animal surveillance interventions for rabies, and we did not find other evidence for community-based surveillance interventions. Both of the studies with successful rabies control programs required coordination and cooperation among government and provincial services. Kamoltham comments that rabies is reportable in Thailand, and this is also true of other countries in SE Asia.86 It is possible that surveillance of rabies is not suited at a community level. With the advent of cheaper and safer human vaccines and the development of more economical regimes for human post-exposure treatment (PET), most Southeast Asian countries are able to administer PET through rabies treatment centres, hospitals and clinics similar to those discussed in the paper by Kamoltham,54 and surveillance of human cases of rabies through these health provision settings would be a reasonable and feasible approach.

In recent years, the WHO has taken the initiative to develop a regional strategy for the elimination of human rabies transmitted by dogs and advocate for rabies control programs in SEA. Rabies control activities in a number of SE Asian countries are now government-driven with the involvement of government officials, health workers and community members.86

Prevention and control interventions

The majority of human rabies is transmitted by dogs through human-animal bite injuries. Models for rabies control programs summarised in this review were based on use of a number of control interventions, including vaccination of animals, restriction of movement of animals, removal of unrestricted animals (culling) and health education. However, the results of the studies included in this review would suggest that mass canine vaccination is the mainstay of successful canine rabies control programs. This has been shown to be the case in a number of other countries throughout the world 87. Estrada showed oral baits to be an acceptable82 and successful50 method of vaccine delivery to vaccinate dogs that was easier to administer than injection. Studies conducted in other countries support this evidence, particularly in the stray and ownerless (common) dog population.88

These interventions require high level support and coordination for their implementation.54, 68, 81 Inability to implement these strategies properly contributed to failure to control the outbreak on Flores Island.81 Legislation to enforce these interventions is also an essential component of rabies control strategies but in recent years, the WHO has also developed and standardised innovative control tools and techniques that may help support future control programs.89

The reduction in the number of deaths from rabies in the study by Kamoltham is noteworthy, and is likely a result of a combined effect of expansion of the PET regimen in humans as well as the dog vaccination campaign.54 This reduction in the number of human deaths due to the increased uptake of the PET for rabies has also been documented in other Asian countries.90 However, the number of rabies exposures are increasing in many countries, which may be explained by the finding that the use of effective dog control programs for dog rabies elimination has become rarer in developing countries.91

None of the studies evaluated the cost benefits and cost-effectiveness of rabies control interventions, particularly in comparison to the cost of patient expanded treatment (PET) regimen used in these countries. Canine vaccination has been shown to be a comparatively inexpensive and ethical way to control the disease in animals and prevent human exposure and illness in model-based studies, especially in resource-limited countries,92 more so than the use of tissue-culture vaccines used in post-exposure prophylaxis.

Contextual factors

A number of studies showed the importance of education and good information dissemination, as well as the form of campaign information, on the likelihood of owners to vaccinate their pets. This has been backed by other studies,92, 93 who have shown that 70-75% of dogs are accessible to control measures, particularly vaccination, if the approach is adapted to the dog-man relationship and the community is fully involved in the rabies elimination program.

Higher level support and the involvement of the authorities was also essential in the success (or failure) of both outbreak control measures and routine canine vaccination, because some form of law enforcement was required, particularly where no one claimed ownership such as the stray dog and common dog population. Lack of coordination between local authorities made it difficult to contain the infected dog population and prolonged the outbreak.

Nipah virus

Surveillance interventions

The Nipah virus outbreak in Malaysia was initially thought to have been illness due to Japanese encephalitis (JE), a mosquito-borne illness, and early control efforts focussed on mosquito source reduction and administration of a JE vaccine.29 An epidemiological trace-back study conducted by scientists from the CDC and the AAHL with the collaboration of local veterinarians later identified Nipah virus as the causative agent. The lack of an established early warning system that incorporated some form of ongoing monitoring of herd health hampered the prompt identification and control of the outbreak, and would certainly have impacted on the magnitude of the outbreak.

Swine surveillance implemented during the outbreak and after the outbreak ceased was shown to be effective in detecting infected herds. None of the studies discussed the cost of the surveillance system, or the feasibility of an ongoing system. A sustainable, ongoing and structured monitoring system for Nipah virus as well as other animal diseases would reduce the impact of any further outbreaks of zoonotic disease. We found no evaluations of surveillance initiatives post the 1999 outbreak; nor any assessment of costs or other attributes such as the functionality of these systems.

The study by Ozawa et al. presented trace back systems in several Asian countries.40 The study comments that trace back systems are not well developed and marking of animals for trace back is practised only in a limited number of countries in specific areas or zones and for specific purposes only. A comprehensive herd monitoring system would need to incorporate some form of identification system to be able to trace back and isolate an infected animal from a particular farm.

Prevention and control interventions

All control measures discussed in the studies were emergency measures used in response to the Malaysian Nipah virus outbreak. They included culling, movement restrictions, quarantine, PPE for farmers and all persons coming into contact with infected pigs (the military, healthcare staff), health education and practices of farm-gate biosecurity (disinfection, isolation). The sustainability and feasibility of using these interventions outside of an outbreak situation has not been discussed and it is unlikely that some of the more extreme interventions are appropriate for routine use.

Our review found some evidence of proposed long-term sustainable prevention and control measures. New guidelines proposed by the Department of Veterinary Services, Malaysian Ministry of Agriculture, to restructure the industry in line with designated pig farming areas and “good animal husbandry concepts” were to be implemented in each State.37 Muniandy et al. and Aziz et al. also outline recommendations for future reform in their paper, which were subsequently discussed at a regional seminar on Nipah virus infection held in Kuala Lumpur in 2001 and jointly organised by the OIE and the Department of Veterinary Services in Malaysia.26 These include policies and protocols for sound farm management practices, which would incorporate farm-gate biosecurity (i.e. quarantine of new animals brought onto the farm, exclusion testing to establish disease status) and would require the engagement of the pig farming industry. Other preventative measures include outbreak preparedness plans for the management of future disease outbreaks and laboratory diagnostic capability. It is unclear how much progress there has been in this area as our review did not find any evidence to show the implementation of any of these measures.

Contextual factors

The outbreak came at enormous political and social cost to Malaysia. The importance of Nipah virus as a newly emerging viral disease in the SE Asian region cannot be understated. While the disease was eradicated from pigs in Malaysia, its natural history suggests there is an on-going need for preparedness for the potential of further outbreaks of Nipah virus in the region. The challenge for Malaysia and other countries in the Asian and Oceanic regions will be to implement a herd monitoring system and control strategies that are acceptable and sustainable, and the need to develop their own preparedness plans.

Dengue

Surveillance interventions

Some evidence was available evaluating surveillance activities for dengue, although we would agree with the recommendations from an earlier systematic review94 that more prospective studies are required to determine the most appropriate dengue surveillance system capable of providing early warning of epidemics. Six studies looked at dengue surveillance at a number of levels, ranging from community level to reporting to the provincial health services. From the results presented in Oum et al., it appears that a considerable proportion of people with symptoms consistent with dengue haemorrhagic fever do not access healthcare and are treated at home.39 Furthermore Chairulfatah et al. found significant underreporting to the local public health office of cases that do seek healthcare.31 These results have implications for the estimation of the burden of disease of dengue fever as well as actioning of control activities in response to hyper endemic activity.

The study by Pang et al. shows the usefulness in incorporating GP sentinel surveillance utilising ‘point of care’ testing to assess suspected cases of dengue haemorrhagic fever in a timely fashion for those cases that do access healthcare.41 However, this study was trialled with two GP clinics in an urban setting within the Kuala Lumpur area of Malaysia. It is unlikely that this approach will be applicable in a rural setting where people are less likely to go to a doctor, and confirmatory testing is costly and logistically challenging.

The two studies utilising surveillance data were useful in predicting outbreak activity and spatial clustering of outbreaks.28, 32 However, the spatial analysis has not taken into account other factors accounting for spatial clustering of outbreaks. Such approaches are also only as good as the underlying data they use. In countries such as Singapore, with well-established surveillance in place and good government support for health services, surveillance information is likely to be robust. Other developing countries in need of surveillance improvements could not use this approach.

Prevention and control interventions

The majority of studies reported outcomes in terms of larval or mosquito indices rather than disease outcomes. There has not been much attempt to look at the correlation between dengue vector and disease indicators. Long term absence (or low rates) of the vector does appear to translate into reduced disease incidence, however, in the short-term vector and disease outcomes do not appear to be well correlated.84 As such reliance on larval or vector indices as the primary outcome measures poses a limitation in terms of evaluating the impact of these strategies on dengue control and the burden of dengue illness in this region given that the majority of studies had follow up periods of less than two years. Short duration of follow-up also means results can be confounded by seasonal and epidemic trends in vector populations and dengue incidence.

Two studies reported the use of serology to measure rates of dengue of infection and showed that rates of dengue seropositivity were higher than rates of clinical dengue infection. This provides interesting evidence for a high incidence of subclinical infection and supports the idea that there is a silent reservoir of disease. This has large implications for the evaluation of both dengue surveillance and control activities, as evaluations based on clinical reporting will underestimate rates of dengue.

The dengue vector, Aedes aegypti responds to control via a variety of methods, and successful programs are fairly homogenous in the extent to which they are able to reduce larval indices over short time periods (less than two years). This finding concurs with the results of three earlier systematic reviews that have looked at the effectiveness of dengue vector control interventions on reducing entomological indicators.15, 95, 96 Choice between methods of vector control for a given setting may rest on factors such as feasibility, cost and sustainability, as well as contextual factors such as cultural and community acceptance (see next section), factors which have been poorly explored in the included studies.

Chemical options for vector control appear to be better suited to epidemic or outbreak situations. The higher cost of chemical control relative to environmental and biological vector control, plus the need for repeat dosing makes them a less sustainable option for ongoing vector control, particularly in rural areas. However evidence from Cambodia showed that a program based on twice yearly larviciding and dengue awareness activities prevented an increase in dengue incidence over five years, indicating that their use may be more relevant in urban areas where water and waste removal infrastructure is better developed. Larviciding (targeting breeding sites) has been studied more than insecticiding (where adult mosquitoes are targeted), the latter has mainly been used in outbreak scenarios indicating it is unlikely to be an option for long-term control. Trials of insecticide treated curtains indicate that this intervention is unlikely to be sustainable for a number of reasons, including poor use and maintenance of the curtains.

High quality studies conducted in Viet Nam showed that control interventions based around biological, environmental and education components can maintain their effectiveness in reducing entomological indices to the point of local elimination of the vector and in reducing cases of dengue infection over sustained periods of time (10 years).57 Copepods (natural predators of mosquito larvae) were introduced into water containers to reduce larvae numbers. They can be locally produced, are low cost and have a higher level of acceptability as compared to Temephos©. They have been shown to be effective in both rural and urban areas. However, their use seems best suited to contexts where water is sourced predominantly from large communal water containers, and these containers represent the major breeding habitat for the mosquito. Use of copepods has always been evaluated in conjunction with environmental and waste management activities. These activities have been a core component of most dengue control interventions that have been evaluated and are highly successful if high levels of community involvement can be achieved. They are appropriate for use in both urban and rural settings, and clean-up targeted at the most productive vector breeding sites is as effective as a blanket approach at lower cost.

Unfortunately there are few direct comparisons of dengue control programs. A narrative interpretation of the forest plots generated in the meta-analysis suggests that there is relative homogeneity of effectiveness across types of vector control intervention, country, and urban v. rural context. Based on a single study77 that compared environmental cleanup to a fogging intervention, the environmental cleanup intervention was more effective during the dry season at reducing larval and vector numbers, however the interventions were comparable during the wet season. This supports the suggestion that chemical interventions may be most suited to outbreak/epidemic situations44, 49. For ongoing control, targeted interventions (where specific containers, buildings or areas are identified to receive the intervention, rather than trying to achieve blanket coverage) appear to offer comparable levels of effectiveness but at a lower cost.38, 76 Targeted strategies obviously rely on having good epidemiological data to ensure that the right sites are identified.

There are few studies that have tried to replicate findings from successful programs in other contexts, and few evaluations of interventions that have been rolled out as regional or national programs. This inhibits the ability to comment on the feasibility and sustainability or even likely effectiveness of interventions outside a pilot study or research context. An earlier systematic review of the functioning of vector control operations found a number of limitations to current programs including a lack of personnel, expertise and budgets, difficulties engaging communities and almost no monitoring and evaluation.96 There is an urgent need for evidence on how findings for successful interventions can be better translated into effective practice.

Contextual factors

The studies included in the review reported a range of contextual factors, behavioural mechanisms or intervention features that either improved or inhibited the effectiveness of the program. High levels of community engagement are necessary for dengue control interventions to be effective. Barriers to community engagement that are reported in the studies include the perception that dengue is not an important public health issue or that the proposed intervention is not effective. This highlights the importance of education and communication about both dengue and the intervention prior to roll-out in the community. It was also shown that knowledge alone did not automatically translate into improved dengue control behaviours and centrally coordinated environmental cleanup or temephos distribution activities were required to reinforce use of control methods. A barrier specific to community involvement in environmental cleanup activities is a belief that these activities are a government responsibility. Conversely, establishment of these tasks as economic activities, through use of microcredit schemes for small recycling businesses, promoted engagement with this activity. Recycling is not a new concept in Viet Nam; it is not clear whether this strategy would be successful settings where rates of recycling are currently low.

Factors that promote high levels of community engagement include the use of multiple methods of communication and education, repetition of education and awareness activities (rather than one-off sessions), use of existing community groups to promote and deliver intervention activities (in particular schools), and engagement of community members at all stages of the interventions; planning, delivery and evaluation. High levels of community ownership and responsibility for ongoing control activities (in particular environmental management) also have spin off benefits for the community not related to dengue control, including greater advocacy skills and an increase in civic pride. Unfortunately none of the studies provided clear descriptions or rationale for how they selected key community groups or leaders and none provided information on the content of their education and awareness activities or any models of behaviour change on which these had been based. This limits the ability to generalise findings to other social and cultural contexts or adapt successful programs for trial in other locations. It should also be noted that the highest quality evidence was undertaken in Viet Nam which has a fairly hierarchical culture that may have facilitated dissemination of information and increased social compulsion to engage with project activities.

There was only limited evidence for the sustainability of interventions. However, in studies with more than two year follow up periods, factors that promoted sustainability included broad community involvement across different levels (rather than isolated groups), and a sense of community ownership of and pride in the control program. Where activities were embedded into the economic activity of the community (e.g. support for recycling businesses) this also improved sustainability. However, it was also noted that community based programs still need support from authorities and cannot be solely based on the efforts of volunteer individuals and community groups. We would also add that it is unclear the extent to which being part of a research project with access to a highly skilled, motivated and engaged research team contributes to levels of effectiveness.

Kay et al.57 reported a lower level of effectiveness for their intervention when it was rolled out as part of a regional program. This has implications for the use of any of these methods of dengue control as part of a larger program. It is well recognised that adequate resourcing for both start up and maintenance of interventions is important to ensure that the program functions well and communities maintain engagement as without this effectiveness is compromised.94 Although some interventions are low cost per person covered, the total cost may still be large. It would be worth exploring mechanisms to offset these costs as was done in the Viet Namese studies where some of the profits from recycling businesses were put back into community dengue awareness activities.

SARS

Surveillance interventions

We found little evidence of evaluations of ongoing laboratory, animal or human surveillance systems set up in many countries in response to the SARS outbreak in 2003, nor any assessments of regional surveillance networks or linkages between countries in the Southeast Asian region, given the geographical restrictions placed on the scope of this review.

The single study by Escudero et al. of work-place surveillance within a hospital in Singapore was tested within a very structured work context, within which there was access to electronic staff leave records.34 It is unclear whether this system would work in a more loosely structured work environment. The absence of electronic systems would also increase the labour resources required for such a system. The lack of specificity and the possibility of “false outbreaks” which could eventually lead to fatigue amongst the staff and agencies involved in responding to potential outbreak situations limit its applicability as an ongoing surveillance system in its current form.

The study reporting on use of temperature screening amongst school children during the epidemic describe the benefits of this type of surveillance as psychological.73 A similar argument has been made for the use of temperature screening at airports97 which was also costly and had a very low yield in terms of detecting SARS cases. Whilst it is important to avoid negative reactions and panic amongst the community during outbreaks, these screening systems are an expensive (and ongoing) investment and it is unclear if they could be implemented in more resource constrained settings.

Prevention and control interventions

Three of the four studies51, 63, 73 included in this section of the review focused on contact tracing and quarantine protocols. While the fourth study75 looked at the effectiveness of personal protective equipment to prevent infection, the quality of the study was poor and the results of the study were of limited value. We did not find other studies reviewing other prevention and control measures (for example, handwashing, temperature screening, closure of workplaces and schools, education campaigns).

The three studies from Singapore all used the same outbreak dataset, and reach similar conclusions that the system was effective in ensuring progressively earlier isolation of probable SARS cases as the outbreak progressed. They also saw a reduction in the number of secondary infections over time. As SARS cases are only infectious whilst symptomatic, and they become more infectious over time, it is logical that this strategy would have been successful in helping to contain the outbreak in Singapore. Indeed the outbreak was brought under control. However, a range of other community-focused strategies were also put in place, including entry and exit screening at airports, market closures, temperature screening in school children and a variety of media health education campaigns, alongside the host of strategies put in place within healthcare facilities. Because of this, the studies are unable to estimate the independent effect of this particular intervention in stopping the outbreak, and no attempt has been made to analyse the size of the effect of confounding and interaction on the authors' results. Furthermore, although the system was sensitive (a quarter of all SARS cases had been on quarantine orders prior to diagnosis) it was not specific (only 0.5% of those isolated went on to develop SARS) making it highly resource intensive per SARS case detected.

Results from the study by Tuan are of limited value. As 95% of the contacts reported never using a mask, the study is underpowered to detect any beneficial impact from using this intervention. As such no conclusions about the effectiveness of this intervention can be drawn. Earlier systematic reviews of the use of masks and other personal protective equipment to prevent transmission of infectious agents have also been unable to draw firm conclusions about the effectiveness of this approach to control.98

Contextual factors

We are limited in the ability to generalise the findings about these interventions to detect and/or prevent spread of SARS because four of the five studies that were included were from Singapore, considered an “economically advanced country” with high GDP and level of education, and other particular features such as an urban setting and a small population. It has strong government and political leadership and good levels of community support. A significant proportion of its public are proficient with information technology. The challenges with implementing surveillance of SARS (and other respiratory diseases) in other Southeast Asian countries with less capable national agencies and healthcare institutions will be to engage with community-level healthcare workers and clinics to implement some form of symptomatic or sentinel non-confirmatory surveillance system.

The success of quarantine and home isolation measures in Singapore was in part due to the capability of the Singaporean government to commit significant financial resources to enforce this policy with random phone checking, electronic camera surveillance, nurse visits and financial incentives. It is unlikely these strategies would work in a resource-challenged country, either due to a lack of financial commitment from the government, the lack of technology (telephones, cameras), or a less well developed infrastructure. This is likely to be particularly true for rural and remote areas. There are also particular social and cultural features of Singaporean society that may have contributed to the high levels of acceptability and compliance with quarantine and isolation measures. Finally, tolerance for this approach outside a high profile outbreak scenario is likely to be low.

Avian Influenza

Surveillance Interventions

The PDSR program in Indonesia has been very successful in detecting HPAI in backyard poultry and allowed a clear and accurate picture of the disease status of HPAI in this sector.42 It has also added value to existing veterinary health services in Indonesia and proved to be a good investment, not just for AI, but also for other animal diseases. Perry et al.'s evaluation found a disproportionate focus on the backyard poultry sector 4, as farmers have considerable interaction with small-scale commercial farms (sector 3). For this reason, surveillance efforts need to cast a broader net and greater engagement of the commercial poultry sector is required. It is unfortunate that the study in Cambodia by Desvaux et al. had several methodological limitations, as it has been the only one to present information on surveillance of commercial poultry and duck farms.33 Both studies were able to classify villages according to their risk of either having faced an HPAI outbreak33 or the probability of HPAI infection,42 allowing prioritisation of control activities. Perry's report also showed that the majority of visits were scheduled or ‘active’ surveillance (87%) as opposed to passive surveillance (13%), but were more effective in detecting disease.42 Surveillance also identified sources of infection (traders, unsafe disposal of carcasses and contaminated vehicles).

The intensive surveillance program in Indonesia required considerable financial investment from external donors and it is unlikely that resource-challenged countries such as Cambodia and Lao PDR would be able to roll out a similarly extensive program. Surveillance programs would have to be setting-specific and tailored to the needs and funds available of the host country. A further challenge for surveillance of HPAI is to transition the achievements gained in the program into a sustainable national system that continues to be accepted. Other than the studies presented in Cambodia and Indonesia, no evidence was found for animal or human surveillance at a community level in the other countries included in this review.

The study by Samaan et al. indicated that rumour surveillance based on internet sources is timely and low cost but although it's sensitivity has been demonstrated it is not clear whether the system is specific enough to be of use.43 Detection of too many “false” outbreaks will limit the credibility of the system. Nevertheless, pilot studies of low cost systems such as this are an important avenue of research to try and extend surveillance coverage to areas with lower levels of information technology, laboratory and healthcare infrastructure.

Prevention and control interventions

Evidence for prevention and control interventions were reported in programs in Cambodia, Viet Nam, Lao PDR and Indonesia. Awareness of HPAI and education regarding risky poultry handling behaviours were common themes in the educational interventions. While the training and education seems to have been well-received, it did not always translate into behaviour modification or change. As some programs were not evaluated, it is difficult to say which components of the intervention have failed and why. An internet rumour based surveillance system represents a potential low cost and timely form of surveillance to inform immediate public health action, but may be limited in its applicability, across the region as it depends highly on the level of engagement of local public health professionals with the chat forums and mailing lists searched, and journalists awareness of AI and quality of reporting. It needs to be demonstrated whether it is capable of detecting outbreaks in resource limited areas where citizens may not have access to the technology on which the system relies.

The prevention and control interventions in the PDSR program in Indonesia had limited success in controlling and eliminating HPAI, for a variety of reasons. Veterinary officers have no legal mandate to enforce culling or movement restrictions of dead or infected poultry. Furthermore, in the absence of financial compensation for loss of livelihood, farmers are under no obligation to report mortality or sickness in animals and will not comply with the requests of disease control officers, and therefore interventions need to be setting-specific. This comes back to the theme of “participatory epidemiology” – engagement of the farmer in decision–making through education and training, but the corollary, that is, recognition of the needs of the community are also essential.

Contextual factors

The studies stressed the importance of engaging local people and civil societies, who can provide rich institutional resources to support difficult changes in health and animal husbandry practices. It was also recognised that local people have rich and detailed knowledge about the animals they keep and the diseases that affect them (termed “existing veterinary knowledge”), whilst researchers often do not know or understand the local context. In SE Asian countries, the poultry industry involves an enormous and diverse set of small entrepreneurs, linked in a number of business relationships and with a wide range of players. Effective HPAI control will require engagement at all levels of the industry. Issues such as financial compensation (or the lack of) for control activities will have to be addressed before governments can enforce policies around culling, movement restrictions and quarantining.

Conclusion

Several common themes emerged when reviewing the literature for the five diseases examined in this systematic review. On the whole, the quality of the studies was low to medium, with evidence on evaluations of surveillance and prevention and control programs not always identified. Evidence on the costs, cost benefits, feasibility and sustainability of these programs was also scarce. Interventions tended to have been evaluated as research or pilot projects rather than as ongoing activities. Most interventions had only been trialled in a single context and durations of follow up were short, limiting the evidence for generalisability and sustainability of findings. Given the limited quantity and quality of information on surveillance and control programs for emerging infectious disease in this region, the findings and conclusions drawn from this review should be interpreted with caution. Absence of evidence for an intervention should not be interpreted as it being ineffective or less effective in specific contexts, rather there is no available published evidence. Similarly, absence of evidence for contextual factors should not be taken to reflect their influence, or otherwise, over the functioning of programs, but rather a lack of reporting.

Appraisal of effective programs showed that sensitivity to local context, attitudes and mores is essential. Many studies were identified where the intervention was not successful or partly successful because of local cultural or social factors. The need for adequate resourcing was also a common theme. Finally, investment in national veterinary and local animal health services appears to have been either absent, insufficient or not given enough priority. Linkages between this sector and human health need to be strengthened. The framework of ‘One Health’, proposed by the FAO, WHO and the OIE, to expand interdisciplinary collaborations to address the animal-human-ecosystem interface, needs further investment for these diseases in this region.

Rabies

Evidence evaluating both veterinary and public health surveillance systems for rabies was not identified; this lack of evidence may in fact reflect the fact that there are no or poorly functioning systems in place. Canine vaccination appears to be the most promising strategy for control, but investment in education is essential for a successful vaccination campaign. Rabies control was more likely to be successful when canine vaccination was used in conjunction with other control strategies. However, canine control activities (including vaccination, sterilisation and culling) are not always popular with the public, and country-specific cultural attitudes can be important. Treatment programs for exposed cases continue to be expanded which has helped to reduce mortality rates, but not rates of exposure to rabies.

Nipah virus

While the concepts of farm-gate biosecurity and herd health monitoring were discussed at the OIE/DVS meeting, there has not been any further progress on recommendations set out at the meeting or any recent publications discussing progress in this area. All evidence on control activities has been in response to the outbreak. Data from targeted and ongoing surveillance as well as the cost and feasibility of the interventions will be essential to guide future prevention and control efforts outside of an outbreak setting, both of which have been absent from the literature. Local traditional farming practices will have to be considered when drafting policies and protocols for sound farm management practices.

Dengue

The dengue vector is amenable to short term control via a variety of vector control methods. Currently there are not enough studies of the same intervention type, nor many direct comparisons of interventions to be able to assess whether one form of intervention is more successful than another at reducing larval indices.

Where vector control is sustained, this appears to result in a reduction in dengue cases. However, many of the existing studies evaluating dengue control interventions do not have a long enough follow-up period to enable an assessment of the sustainability. Environmental management to reduce larval habitats is an effective way at reducing vector numbers and can be used in both urban and rural areas. It is often supplemented with the use of either biological (such as copepods) or chemical (such as Temephos©) larvicidal agents. The latter is reported as being less acceptable due to problems with smell and taste. The former has good evidence for sustainability and is low cost, but the suitability of this control method in settings where water is not obtained from large centralised tanks has not been evaluated. Disease education is important but in the absence of other coordinated activities does not result in improved control practices.

There is limited evidence for the use of interventions outside research projects. Studies that have evaluated roll out of interventions to regional programs indicate that effectiveness may be reduced, possibly due a lower level of access to technical expertise and lack of involvement of communities in the program planning stages. Sustainability requires communities to take ownership of ongoing control activities. High levels of community engagement require multiple methods of communication and activities.

SARS

Little evidence was available on evaluations of ongoing laboratory, animal or human surveillance systems implemented after the outbreak in 2003, nor any assessments of regional surveillance networks or linkages between countries in the SE Asian region. The single study of hospital records-based surveillance was conducted within a very specific setting and not generalisable. No community-based surveillance interventions were identified.

The majority of studies examining control interventions were based on the analysis of outbreak data to review contact tracing and quarantine protocols. While control measures were shown to be effective, they did not control for confounding from other community-focused control strategies. No other studies reviewing other prevention and control measures were identified. Most included studies were from Singapore, an advanced country that is very urbanized, and with a small population. Control measures were costly and cannot be applied in resource-challenged settings. The geographical scope of the review posed a limitation on the evaluation of control interventions for SARS as it excluded studies from China, Taiwan and Canada, countries that were most impacted by the outbreak of SARS in 2003.

Avian influenza

There appear to be large investments in several countries in SE Asia on training, educational and surveillance initiatives, but evidence on the evaluation of these programs was not always identified. Surveillance in the backyard poultry sector has been successful in identifying HPAI in backyard poultry flocks, but needs to be broadened to include other sectors of the commercial poultry industry. Surveillance interventions have had added spin-off benefits of strengthening the local animal health services. Prevention and control efforts have proved more challenging for a number of reasons. Successful educational campaigns have not always translated into behavior modification and change. Involvement of and recognition of the needs of the community are essential in addressing these barriers to change. Community programs have been largely reliant on external funding, and the challenge will be to incorporate them into the national process where the programs can become institutionalised in a sustainable way.

Limitations of the review

A major limitation of this review was that our literature search was limited to studies published in the English language, which will have excluded studies conducted in local languages and published in local non-English journals. The geographical scope of this review (the ten member countries of the ASEAN) also poses a limitation that had a great impact on the analysis of interventions for SARS as it excluded studies from Hong Kong, Taiwan, China and Canada; countries that were most impacted by the outbreak of SARS in 2003. Transmission of SARS in countries within the geographical scope of this review was mainly within the hospital setting,99 thus much of the available evidence concentrates on the prevention of nosocomial transmission and protection of healthcare workers, rather than evaluation of community-based strategies, and was excluded from the review.

A further limitation is that the review only included studies with empirical data and therefore, at least for some of the diseases, this resulted in a small number of included studies. Practical implications may also need to consider data from mathematical modelling studies, which were excluded from this review. Although these studies are essentially hypothetical, inferences from these studies may provide useful insights into the epidemiology and transmission of disease and can be used to predict the likely coverage, effectiveness and cost-effectiveness of different possible interventions under a range of scenarios.

Another limitation is that cross-sectional surveys of knowledge, attitude and practice regarding disease and prevention and control activities were excluded as outside the scope of this review and are listed in Appendix V. Although they do not test the effectiveness of an intervention they may provide information useful to plan the successful implementation of interventions. Similarly, our exclusion of purely qualitative studies limits our analysis of behavioural contextual factors that may affect the effectiveness of interventions. Our review has shown that these factors strongly influence a person's decision to act contrary to the clear health messages being delivered, or to engage with a program's messages and activities. The reasons that govern such behaviour and the decision-making process may be better elicited through an appraisal of qualitative research.

Some of the studies excluded from the review were done so on the basis that they were only available in abstract form and a full copy of the study could not be obtained for review. As such we are aware that there is potentially more evaluations that have been conducted than have been fully reported. If there is a systematic bias in which studies are published, this will bias review results.

Finally, countries such as Thailand (a developing country) and Singapore (considered an ‘advanced country’), have well established national health agencies and healthcare institutions. We excluded from the review studies that evaluated purely government-driven national health institutions or systems, however, some “top-down” government-driven initiatives may include local level community involvement and engagement in concert with the centrally coordinated response, especially in outbreak situations. As many studies do not or cannot provide a detailed description of every element of the intervention it is possible that we did not identify some activities with community-based elements. Exclusion of these studies also meant we were unable to undertake a broader analysis of a country's health systems which is relevant in the context of community-based health interventions transitioning into a national approach.

Implications for practice

Several implications for practice can be derived from the findings of this review. There are a number of general recommendations that relate to all five emerging or re-emerging infectious diseases included in the review, as well as recommendations specific to each disease. Each recommendation is assigned a level of evidence according to JBI criteria for evaluation of effectiveness studies (see Appendix VI). Where information on interventions was not available we are unable to comment on whether they are likely to be effective or what contextual factors may influence this. This should not be taken as evidence that these strategies are ineffective but rather represents a gap in our current knowledge.

The studies included in the review provide no detailed evidence for risk assessment in development of the interventions trialled (or indeed any program planning tools/frameworks utilised). This limits the ability to draw conclusions about which interventional approach may be most appropriate for a given setting where no situational analysis has or can be conducted.

Linkage of animal and human health systems for detection and control of disease is essential for zoonotic infectious diseases as animals represent the main reservoir of infection. We found only limited evidence for programs based on the framework of ‘One Health’. Some contextual information is available showing linkages need to be multi-level and be compatible with economic activity to be successful. The evidence for this is discussed under each disease. However, no information was provided in the studies on how linkage of these systems is best achieved.

General

• Community based prevention and control strategies are more effective if they have access to central coordination and support. (Level 3).

• Surveillance data can be used to build predictive models of outbreaks and transmission that can be helpful for planning control activities. (Level 3)

• Linkage between veterinary and public health surveillance systems improves timely detection of outbreaks. (Level 3).

• Higher levels of effectiveness are achieved where the community is involved in all stages of the program (planning, delivery and evaluation). (Level 2). Program activities can be delivered and coordinated through existing community groups. The use of schools is a good channel as long as school children are fully engaged in the program. (Level 3).

• Community participation in programs is higher where people perceive the disease as an important public health problem and are well informed about the control program and perceive it as likely to be effective (Level 3).

• Interventions that conflict with the economic activity of communities are poorly tolerated, whilst those that are compatible, or offer new economic opportunities, are well received (Level 2).

Rabies

• Rabies vaccine to stray and common dog populations is effective at reducing rates of rabies in canine or human populations if high levels of coverage are reached. (Level 3). Vaccination coverage is increased if comprehensive and effective education campaigns are used that are delivered via multiple channels. (Level 3). Owned dog populations tend to be easier to vaccinate as their owners have a vested interest in their well-being. (Level 3). Canine vaccination without incentives/enforcement is not as effective in achieving good coverage of vaccination. (Level 3). Vaccination coverage may increase if financial compensation for destroyed animals was offered (Level 3). Improvements in communication between public health and veterinary systems would make it easier to monitor canine vaccination. (Level 3).

• Culling of dogs is not socially acceptable in countries in this region and there is active resistance to this strategy. (Level 3).

• Improved treatment protocols have been effective in reducing the number of rabies deaths in humans potentially exposed to rabies through animal bites. (Level 3)

• Evaluations benefit from human rabies being notifiable, even where monitored via a passive surveillance system based in PET distribution clinics and hospital cases. (Level 3).

Nipah virus

• Advance planning for disease outbreaks (outbreak management plans) improves timeliness of response. (Level 3)

• Swine surveillance and culling of infected animals and herds is effective at detecting and halting outbreaks of Nipah virus (Level 3). Financial compensation for destroyed animals makes culling very expensive but increases tolerance and support of the strategy amongst farmers (Level 3).

• Effective monitoring of herd health requires adequate number of laboratory submissions as an early warning system, laboratory capacity, and a high level of farmer and veterinarian awareness of disease. (Level 4).

• Animal tracking systems used in swine surveillance required permanent forms of animal marking (such as ear notching) to reduce attempts to defraud the system. (Level 3).

• A cross-regional plan of trace back systems between Asian countries makes sense as much as local and national systems to track the movement of pigs. (Level 4).

• Pig farming industry should be managed differently including:

  • Policies and protocols for sound farm management practices, which incorporate farm-gate biosecurity (i.e. quarantine of new animals brought onto the farm, exclusion testing to establish disease status) and would require the engagement of the pig farming industry. (Level 4).

  • Traditional practice of sharing boars or moving sows from farm to farm should not be practiced. (Level 4).

  • Separation of animal farms from orchards and fruit and vegetable growing areas where fresh food is grown for human consumption. (Level 4).

Dengue

• The dengue vector is amenable to short term control via a variety of vector control methods. Where control is sustained this results in a reduction in dengue cases. (Level 2).

• Interventions using copepods, environmental clean-up and education activities are effective at reducing larval indices, mosquito indices and incidence of dengue to the point of local elimination. The strategy is low cost and sustainable. It is effective in urban and rural settings but is most suitable in settings where water is obtained from large central tanks. (Level 2).

• A combination of environmental vector control and education without biological agents is also effective at reducing larval and mosquito indices. (Level 3).

• Environmental vector control and larviciding of breeding habitats is effective at reducing larval indices, incidence of dengue and dengue morbidity in both urban and rural areas. (Level 3). Targeted control where the most productive habitats are targeted for environmental clean-up or introduction of copepods is equally effective but less costly than a blanket program which includes all containers. (Level 2). The use of chemical larvicides in water supplies is less well tolerated by communities than biological control agents. (Level 3).

• Chemical vector control based on fogging is equally effective but more costly than environmental vector control and there is no evidence for its sustainability. (Level 3). The use of chemical fogging is well suited to the control of outbreaks. (Level 3).

• The use of impregnated curtains is not sustainable. (Level 3).

• Educational interventions offered without coordinated environmental clean-up activities or distribution of chemical or biological larvicidal agents do not reduce vector indices or dengue incidence. (Level 3). Education needs targeting and repeat sessions to improve knowledge and awareness, but this alone does not translate to improvements in control practices. (Level 3).

• Environmental and waste management are important to the success of interventions. To ensure the intervention is sustainable communities need to take ownership and responsibility for these activities. Providing opportunities for economic activity in this area can support this process. (Level 2).

SARS

• Passive surveillance in healthcare workers can be used to detect outbreaks of febrile illness. The cost and timeliness of the system will depend on the extent to which staff medical records are electronic. (Level 3)

• Contact tracing and large scale isolation and quarantine is effective in reducing the time to isolation of suspected cases and reducing the number of potential contacts in urban areas, however its use is best suited to outbreak situations and should be restricted to situations where it is economically, logistically and legally feasible. (Level 3). Successful implementation of contact tracing and isolation requires good organization, good communication and high levels of public support. (Level 3).

Avian Influenza

• Communities have veterinary knowledge that can be successfully tapped to identify high risk areas or potential outbreaks amongst poultry. (Level 3). Village (backyard or farm) based surveillance is successful at identifying high risk areas and potential outbreaks. (Level 3).

• A multi-country rumour surveillance system based on web sources was successful at identifying outbreaks in a low cost and timely manner. (Level 3).

• Educational programs are successful at increasing awareness and knowledge about HPAI but have a lower impact at improving basic hygiene and risky poultry practices. (Level 3). More successful programs offer multiple opportunities for people to engage and access the program; one-off educational interventions do not work as well. (Level 3). Increased knowledge and awareness of HPAI increases rates of identification of sick poultry amongst community members, but not reporting to local authorities. (Level 3).

• Culling of sick poultry is effective at preventing spread of HPAI but is not well received by community members in the absence of financial compensation. (Level 3).

Implications for research

The review identified many important gaps in the available evidence. Suggestions for future areas of research which would benefit from methodically sound quantitative studies are discussed below. General comments applicable to all five emerging or re-emerging infectious diseases have been listed as well as suggestions specific to each disease.

General

• Better evidence is needed on the structure, functioning and outcomes of current local and national surveillance systems for emerging infectious diseases. Ideally these should be long-term studies capable of assessing sustainability rather than short pilots. Further evaluations of promising novel methods of surveillance utilising technology such as mobile phones or the internet should also be undertaken.

• Minimum datasets should be designed for use in outbreaks, to ensure more comprehensive data collection that will allow for more rigorous evaluation of the impact of outbreak control measures.

• Evaluations of prevention and control programs need to be longitudinal rather than cross-sectional and report on the impact on disease outcomes, health knowledge and practices, as well as information on the acceptability, cost and sustainability of programs.

• Education and awareness programs should be designed and evaluated against models of behaviour change to facilitate extrapolation of findings to other contexts.

• Evaluations of successful prevention and control interventions require replicating in other countries to test the generalisability of findings across different social, cultural and geographic contexts.

• More evidence is needed from a wider range of countries, in particular resource-constrained settings with less well developed infrastructure.

• To improve generalisability, more comprehensive descriptions of the community engagement strategies and activities used, and information on the acceptability and uptake of the program by different sectors of the community would be useful.

• There is a need for translational research to look at how findings from successful interventions can be translated into effective practice.

Rabies

• Long term evaluations of established rabies surveillance systems are needed that report both process and outcome measures. Researchers should investigate novel methods for passive surveillance for this disease and investigate linking data from veterinary and public health sources.

• There is a need for more evaluation of alternative prevention and control activities, such as canine sterilisation. Outcomes should include measures of human and canine disease, cost and sustainability as well as process indicators such as uptake.

• The use of education and awareness programs should be evaluated to investigate whether they are able to improve uptake of canine intervention strategies.

Nipah virus

• Research should try to identify a suitable animal sentinel for Nipah virus that could be placed under surveillance, as well as identifying animal reservoirs and factors that increase the likelihood of host-animal transmission that could be targeted by control programs.

• There is a need for evaluations of educational programs for farming communities. Studies should report data on process outcomes such as farm management practices and human disease outcomes, as well as rates of disease in pigs.

Dengue

• Evaluations that compare different methods of dengue prevention and control are required. This will allow identification of the independent contribution of specific components of the program to overall effectiveness and identify the most effective strategies. Evaluations should also include measures of cost and sustainability to allow identification of the most efficient long term interventions to reduce the incidence of dengue.

• Evaluations need longer follow-up periods to control for seasonality and the epidemic pattern of disease. They should also provide data on the cost and sustainability of programs.

• Evaluations should report data on dengue incidence and dengue mortality rather than relying on vector indices or process measures for the intervention such as KAP scores or percentage uptake. Further investigation of the correlation (if any) between vector indices (particularly larval indices) and dengue incidence would be useful.

• Researchers should provide more comprehensive descriptions of the community engagement strategies and activities used, and information on the acceptability and uptake of the program by different sectors of the community.

• A description of the role of the research team in projects under evaluation would also be useful to understand how this might impact the success of interventions offered via routine dengue programs rather than in a research environment.

SARS

• A wider range of interventions aiming to prevent or control spread of viral respiratory illness needs to be studied, including the effectiveness of masks and other personal protective equipment, hygiene promotion and disease awareness campaigns, in both close patient contacts and the wider community.

Avian Influenza

• Evaluations of veterinary surveillance systems for avian influenza should look for any evidence of correlation with incidence of influenza in human populations.

Citation details of included studies by disease

Rabies

50Estrada R, Vos A, De Leon R, Mueller T. Field trial with oral vaccination of dogs against rabies in the Philippines. BMC Infectious Diseases. 2001; 1:23

54Kamoltham T, Singhsa J, Promsaranee U, Sonthon P, Mathean P, Thinyounyong W. Elimination of human rabies in a canine endemic province in Thailand: five-year program. Bulletin of the World Health Organization. 2003;81:375-81.

67Robinson LE, Miranda ME, Miranda NL, Childs JE. Evaluation of a canine rabies vaccination campaign and characterization of owned-dog populations in the Philippines. Southeast Asian J. Of Tropical Medicine and Public Health. 1996;27:250-6.

68Soon TY. Rabies in Malaysia. South East Asian Journal of Tropical Medicine and Public Health. 1988;19:535-6.

81Windiyaningsih C, Wilde H, Meslin FX, Suroso T, Widarso HS. The rabies epidemic on Flores Island, Indonesia (1998-2003). Journal Of The Medical Association Of Thailand = Chotmaihet Thangphaet. 2004;87(11):1389-93.

Nipah virus

25Arjoso S, Wuryadi S, Windyaningsih C, Winoto IL, Heriyanto A, Ksiazek TG, et al. The economic imperative of Nipah virus surveillance in Indonesia. Transactions Of The Royal Society Of Tropical Medicine And Hygiene. 2001;95(4):368-9.

26Arshad MM. Phase three surveillance program for Nipah virus infection in pigs in Malaysia, in: Report of the Regional Seminar on Nipah virus infection, OIE and Department of Veterinary Services, Malaysia, Kuala Lumpur, Malaysia, 9-12 April, 2001.

29Bunning M. Nipah virus outbreak in Malaysia, 1998

30CDC, Update: outbreak of Nipah virus — Malaysia and Singapore, 1999. MMWR: Morbidity & Mortality Weekly Report. 1999;48(16):335-7.

36Mohd Nor MN, Gan CH, Ong BL. Nipah virus infection of pigs in peninsular Malaysia. Revue Scientifique et Technique (International Office Of Epizootics). 2000; 19(1):160-5.

37Muniandy N, Aziz JA. Effects of intensification of the traditional farming system on the environment and bio-safety of the human population: Nipah virus outbreak in Malaysia. In: Furukawa H, Mitsuaki N, Yasuyuki K, Yoshihiro K, editors. Ecological Destruction, Health, and Development: Advancing Asian Paradigms. 2004. p. 303-17.

40Ozawa Y, Ong BL, An SH. Traceback systems used during recent epizootics in Asia. Revue Scientifique Et Technique (International Office Of Epizootics). 2001;20(2):605

Dengue

44Ang LW, Foong BH, Ye T, Chow A, Chew SK. Impact of “carpet-combing” vector control operations in terminating the 2005 dengue outbreak in Singapore. Epidemiological News Bulletin 2007;33: 31-36.

28Barbazan P, Yoksan S, Gonzalez J-P. Dengue haemorrhagic fever epidemiology in Thailand: description and forecasting of epidemics. Microbes and Infection/Institut Pasteur. 2002;4:699-705.

45Beckett CG, Kosasih H, Tan R, Widjaja S, Listianingsih E, Ma'roef C, et al. Enhancing knowledge and awareness of dengue during a prospective study of dengue fever. The Southeast Asian Journal of Tropical Medicine And Public Health. 2004;35(3):614-7.

47Butraporn P, Saelim W, Sitapura P, Tantawiwat S. Establishment of an environmental master team to control dengue haemorrhagic fever by local wisdom in Thailand. Dengue Bulletin. 1999;23(99-104).

31Chairulfatah A, Setiabudi D, Agoes R, van Sprundel M, Colebunders R. Hospital based clinical surveillance for dengue haemorrhagic fever in Bandung, Indonesia 1994-1995. Acta Tropica. 2001;80(2):111-5.

32Chan EH, Sahai V, Conrad C, Brownstein JS. Using web search query data to monitor dengue epidemics: a new model for neglected tropical disease surveillance. Plos Neglected Tropical Diseases. 2011;5(5):e1206-e.

48Crabtree SA, Wong CM, Mas'ud F. Community participatory approaches to Dengue prevention in Sarawak, Malaysia. Human Organization. 2001;60(3):281-7.

49Eamchan P, Nisalak A, Foy HM, Chareonsook OA. Epidemiology and control of dengue virus infections in Thai villages in 1987. The American Journal of Tropical Medicine and Hygiene. 1989;41(1):95-101.

52Hien Tran V. Application of mosquito—proof water containers in the reduction of dengue mosquito population in a dengue endemic province of Viet Nam. Asian Pacific Journal of Tropical Disease. 2011;1(4):270-4.

53Igarashi A. Impact of dengue virus infection and its control. FEMS Immunology and Medical Microbiology. 1997;18(4):291-300.

55Kay B and Nam VS. New strategy against Aedes aegypti in Viet Nam. Lancet. 2005;365(9459):613-7.

56Kay BH, Nam VS, Tien TV, Yen NT, Phong TV, Diep VTB, et al. Control of aedes vectors of dengue in three provinces of Viet Nam by use of Mesocyclops (Copepoda) and community-based methods validated by entomologic, clinical, and serological surveillance. The American Journal of Tropical Medicine And Hygiene. 2002;66(1):40-8.

57Kay BH, Tuyet Hanh TT, Le NH, Quy TM, Nam VS, Hang PVD, et al. Sustainability and cost of a community-based strategy against Aedes aegypti in northern and central Viet Nam. The American Journal of Tropical Medicine And Hygiene. 2010;82(5):822-30.

58Kittayapong P, Yoksan S, Chansang U, Chansang C, Bhumiratana A. Suppression of dengue transmission by application of integrated vector control strategies at sero-positive GIS-based foci. The American Journal of Tropical Medicine And Hygiene. 2008;78(1):70-6.

59Madarieta SK, Salarda A, Benabaya MRS, Bacus MB, Tagle JR. Use of permethrin-treated curtains for control of Aedes aegypti in the Philippines. Dengue Bulletin 1999; 23

62Nam VS, Yen NT, Kay B, Marten GG, Reid JW. Eradication of Aedes aegypti from a village in Viet Nam using copepods and community participation. The American Journal of Tropical Medicine and Hygiene. 1998;59:657-60.

61Nam VS, Nguyen TY, Tran VP, Truong UN, Le QM, Le VL, et al. Elimination of dengue by community programs using Mesocyclops (Copepoda) against Aedes aegypti in central Viet Nam. The American Journal of Tropical Medicine and Hygiene. 2005;72(1):67-73.

38Osaka K, Ha DQ, Sakakihara Y, Khiem HB, Umenai T. Control of dengue fever with active surveillance and the use of insecticidal aerosol cans. The Southeast Asian Journal Of Tropical Medicine And Public Health. 1999;30(3):484-8.

39Oum S, Chandramohan D, Cairncross S. Community-based surveillance: a pilot study from rural Cambodia. Tropical Medicine & International Health. 2005 Jul;10(7):689-97.

41Pang T, Lam SK, Kok ML, Kok KY, Tho YC. A practical community-based approach to the diagnosis of dengue virus infections. Singapore Medical Journal. 1989;30(6):525-7.

64Pengvanich V. Family leader empowerment program using participatory learning process for dengue vector control. Journal Of The Medical Association Of Thailand = Chotmaihet Thangphaet. 2011;94(2):235-41.

65Phan-Urai P, Kong-ngamsuk W, Malainual N. Field trial of Bacillus thuringiensis H-14 (Larvitab) against Aedes aegypti larvae in Amphoe Khlung, Chanthaburi Province, Thailand. Journal of Tropical Medicine and Parasitology. 1995;16:35-41.

66Phatumachinda B, Phanurai P, Samutrapongse W, Chareonsook OA. Studies on community participation in Aedes aegypti control at Phanus Nikhom district, Chonburi Province, Thailand. Mosquito-Borne Diseases Bulletin. 1985;2:1-8.

69Suaya JA, Shepard DS, Caram M, Hoyer S, Nathan MB. Cost-effectiveness of annual targeted larviciding campaigns in Cambodia against the dengue vector Aedes aegypti. Tropical Medicine and International Health. 2007;12(9):1026-36.

70Suroso H, Suroso T. Aedes aegypti, control through source reduction by community efforts in Pekalongan, Indonesia. Mosquito-Borne Diseases Bulletin. 1990;7:59-62.

71Suwanbamrung C, Dumpan A, Thammapalo S, Sumrongtong R, Phedkeang P. A model of community capacity building for sustainable dengue problem solution in Southern Thailand. Health. 2011;3(9):584-601.

72Swaddiwudhipong W, Chaovakiratipong C, Nguntra P, Koonchote S, Khumklam P, Lerdlukanavonge P. Effect of health education on community participation in control of dengue hemorrhagic fever in an urban area of Thailand. The Southeast Asian Journal of Tropical Medicine And Public Health. 1992;23(2):200-6.

74Therawiwat M, Fungladda W, Kaewkungwal J, Imamee N, Steckler A. Community-based approach for prevention and control of dengue hemorrhagic fever in Kanchanaburi Province, Thailand. The Southeast Asian Journal of Tropical Medicine and Public Health. 2005;36(6):1439-49.

76Tun-Lin W, Lenhart A, Nam VS, Rebollar-Téllez E, Morrison AC, Barbazan P, et al. Reducing costs and operational constraints of dengue vector control by targeting productive breeding places: a multicountry non-inferiority cluster randomized trial. Tropical Medicine & International Health: TM & IH. 2009;14(9):1143-53. (but exclude data on Thailand as intervention is purely biological)

77Umniyati SR, Umayah SS. Evaluation of community-based Aedes control program by source reduction in Perumnas Condong Catur, Yogyakarta, Indonesia. Dengue Bulletin. 2000;24:1-3.

79Vanlerberghe V, Villegas E, Jirarojwatana S, Santana N, Trongtorkit Y, Jirarojwatana R, et al. Determinants of uptake, short-term and continued use of insecticide-treated curtains and jar covers for dengue control. Tropical Medicine & International Health: TM & IH. 2011;16(2):162-73.

SARS

34Escudero IHG, Chen MI, Leo YS. Surveillance of severe acute respiratory syndrome (SARS) in the post-outbreak period. Singapore Medical Journal. 2005;46(4):165-71.

51Goh K-T, Cutter J, Heng B-H, Ma S, Koh BKW, Kwok C, et al. Epidemiology and control of SARS in Singapore. Annals of the Academy Of Medicine, Singapore. 2006;35(5):301-16.

63Ooi PL, Lim S, Chew SK. Use of quarantine in the control of SARS in Singapore. American Journal of Infection Control. 2005;33(5):252-7.

73Tan C-C. SARS in Singapore-key lessons from an epidemic. Annals of The Academy Of Medicine, Singapore. 2006;35(5):345-9.

75Tuan PA, Horby P, Dinh PN, Mai LTQ, Zambon M, Shah J, et al. SARS transmission in Viet Nam outside of the health-care setting. Epidemiology and Infection. 2007;135(3):392-401.

Avian Influenza

27Azhar, M., A. S. Lubis, et al. (2010). “Participatory disease surveillance and response in Indonesia: strengthening veterinary services and empowering communities to prevent and control highly pathogenic avian influenza.” Avian Diseases 54(1 Suppl): 749-753.

46Bhandari, D. P., T. S. Wollen, et al. (2011). “Preventing highly pathogenic avian influenza (HPAI) at the rural community level: a case study from Cambodia.” Tropical Animal Health and Production 43(6): 1071-1073.

33Desvaux, S., S. Sorn, et al. (2006). HPAI surveillance program in Cambodia: results and perspectives. Developments In Biologicals 124: 211-224.

35Jost CC, Mariner JC, Roeder PL, Sawitri E, Macgregor-Skinner GJ. Participatory epidemiology in disease surveillance and research. Revue Scientifique Et Technique (International Office of Epizootics). 2007;26(3):537-49.

60Manabe, T., T. P. T. Pham, et al. (2011). “Impact of educational intervention concerning awareness and behaviors relating to avian influenza (H5N1) in a high-risk population in Viet Nam.” Plos One 6(8): e23711-e23711.

42Perry B, Isa KM, Tarazona C. Independent evaluationof FAO's participatory disease surveillance and response program in Indonesia. FAO Evaluation Service, July 2009.

43Samaan G, Patel M, Olowokure B, Roces MC, Oshitani H. Rumor surveillance and avian influenza H5N1. Emerging Infectious Diseases. 2005;11(3):463-6.

78Van Kerkhove MD, Ly S, Guitian J, Holl D, San S, Mangtani P, et al. Changes in poultry handling behaviour and poultry mortality reporting among rural Cambodians in areas affected by HPAI/H5N1. Plos One. 2009;4(7):e6466-e.

80Waisbord SR, Michaelides T, Rasmuson M. Communication and social capital in the control of avian influenza: lessons from behaviour change experiences in the Mekong Region. Global Public Health. 2008;3(2):197-213.

Appendix I – Search strategy

Only journal articles and reviews dating from January 1980 to December 2011 and published in the English language were considered for inclusion in the review. The following databases were searched: PubMed and CINAHL (via EBSCoHost), ProQuest, Web of Science, ScienceDirect and the Cochrane database of systematic reviews. Only studies that met the inclusion criteria of randomised controlled trials, controlled before-after trials or interrupted time series were evaluated22.

A two step search strategy was utilised in these databases, as specified below.

• Primary search strategy: Country of interest + Disease of interest + the terms: “surveillance”, “prevention and control”, and “outbreaks”.

• Secondary search strategy: Disease of interest + Search terms: “surveillance” OR “prevention and control” OR “outbreaks” + Search terms: “community” OR “intervention” or “effectiveness” OR “education”. In addition, two more search terms were included that were disease-specific (Table 8).

Table 8: Disease-specific search terms included in the secondary search strategy.

These search terms were selected on the basis of preliminary searches to determine the most commonly used intervention keywords or subject headings.

Searches based only on the disease of interest were undertaken of the following databases: the WHO library database (WHOLIS), British Development Library, LILACS, World Bank (East Asia) and the Asian Development Bank.

Finally, we examined the reference list of all shortlisted reports, existing systematic reviews and included articles for additional relevant studies.

Appendix II – Critical appraisal instruments

Appendix III – Data extraction instrument

Appendix IV – Characteristics of included studies

Table 9: Rabies – Study characteristics of included studies.

Table 10: Rabies – Details of interventions and outcomes evaluated in included studies.

Table 11: Rabies – Main findings and limitations of included studies.

Table 11: Rabies – Main findings and limitations of included studies (Continued).

Table 12: Rabies – Contextual information extracted from included studies.

Table 13: Nipah virus – Study characteristics of included studies.

Table 14: Nipah virus – Details of interventions and outcomes evaluated in included studies.

Table 14: Nipah virus – Details of interventions and outcomes evaluated in included studies (Continued).

Table 14: Nipah virus – Details of interventions and outcomes evaluated in included studies (Continued).

Table 15: Nipah virus – Main findings and limitations of included studies.

Table 15: Nipah virus – Main findings and limitations of included studies (Continued).

Table 16: Nipah virus – Contextual information extracted from included studies.

Table 16: Nipah virus – Contextual information extracted from included studies (Continued).

Table 17: Dengue – Study characteristics of included studies.

Table 17: Dengue – Study characteristics of included studies (Continued).

Table 17: Dengue – Study characteristics of included studies (Continued).

Table 17: Dengue – Study characteristics of included studies (Continued).

Table 17: Dengue – Study characteristics of included studies (Continued).

Table 17: Dengue – Study characteristics of included studies (Continued).

Table 18: Dengue – Details of interventions and outcomes evaluated in included studies.

Table 18: Dengue – Details of interventions and outcomes evaluated in included studies (Continued).

Table 18: Dengue – Details of interventions and outcomes evaluated in included studies (Continued).

Table 18: Dengue – Details of interventions and outcomes evaluated in included studies (Continued).

Table 18: Dengue – Details of interventions and outcomes evaluated in included studies (Continued).

Table 18: Dengue – Details of interventions and outcomes evaluated in included studies (Continued).

Table 18: Dengue – Details of interventions and outcomes evaluated in included studies (Continued).

Table 18: Dengue – Details of interventions and outcomes evaluated in included studies (Continued).

Table 19: Dengue – Main findings and limitations of included studies.

Table 19: Dengue – Main findings and limitations of included studies (Continued).

Table 19: Dengue – Main findings and limitations of included studies (Continued).

Table 19: Dengue – Main findings and limitations of included studies (Continued).

Table 19: Dengue – Main findings and limitations of included studies (Continued).

Table 19: Dengue – Main findings and limitations of included studies (Continued).

Table 19: Dengue – Main findings and limitations of included studies (Continued).

Table 19: Dengue – Main findings and limitations of included studies (Continued).

Table 19: Dengue – Main findings and limitations of included studies (Continued).

Table 19: Dengue – Main findings and limitations of included studies (Continued).

Table 19: Dengue – Main findings and limitations of included studies (Continued).

Table 19: Dengue – Main findings and limitations of included studies (Continued).

Table 19: Dengue – Main findings and limitations of included studies (Continued).

Table 20: Dengue – Contextual information extracted from included studies.

Table 20: Dengue – Contextual information extracted from included studies (Continued).

Table 20: Dengue – Contextual information extracted from included studies (Continued).

Table 20: Dengue – Contextual information extracted from included studies (Continued).

Table 20: Dengue – Contextual information extracted from included studies (Continued).

Table 20: Dengue – Contextual information extracted from included studies (Continued).

Table 21: SARS – Study characteristics of included studies.

Table 21: SARS – Study characteristics of included studies (Continued).

Table 22: SARS – Details of interventions and outcomes evaluated in included studies.

Table 22: SARS – Details of interventions and outcomes evaluated in included studies (Continued).

Table 30: SARS – Main findings and limitations of included studies.

Table 30: SARS – Main findings and limitations of included studies (Continued).

Table 31: SARS – Contextual information extracted from included studies.

Table 32: Avian influenza – Study characteristics of included studies.

Table 32: Avian influenza – Study characteristics of included studies (Continued).

Table 32: Avian influenza – Study characteristics of included studies (Continued).

Table 33: Avian influenza – Details of interventions and outcomes evaluated in included studies.

Table 33: Avian influenza – Details of interventions and outcomes evaluated in included studies (Continued).

Table 33: Avian influenza – Details of interventions and outcomes evaluated in included studies (Continued).

Table 34: Avian influenza – Main findings and limitations of included studies.

Table 34: Avian influenza – Main findings and limitations of included studies (Continued).

Table 34: Avian influenza – Main findings and limitations of included studies (Continued).

Table 34: Avian influenza – Main findings and limitations of included studies (Continued).

Table 34: Avian influenza – Main findings and limitations of included studies (Continued).

Table 34: Avian influenza – Main findings and limitations of included studies (Continued).

Table 35: Avian influenza – Contextual information extracted from included studies.

Table 35: Avian influenza – Contextual information extracted from included studies (Continued).

Appendix V – List of excluded studies by disease and reasons for exclusion

Rabies – articles

Akoso BT. Rabies in animals in Indonesia. In: Rabies control in Asia, Dodet B, Meslin FX, editors, 2001. p 219.

Reason for exclusion: Descriptive analysis, no intervention

Ali M, Canh DG, Clemens JD, Park JK, von Seidlein L, Thiem VD, et al. The vaccine data link in Nha Trang, Viet Nam: a progress report on the implementation of a database to detect adverse events related to vaccinations. Vaccine. 2003 Apr; 21(15):1681-6.

Reason for exclusion: Vaccine safety surveillance

Anonymous. WHO strategies for the control and elimination of rabies in Asia. Report of a WHO interregional consultation, Geneva, Switzerland, 17-21 July 2001. World Health Organization Technical Report Series 2002;2001 (WHO/CDS CSR/EPH/2002.8)

Reason for exclusion: No data on interventions

Arámbulo PV, 3rd. Veterinary public health: perspectives at the threshold of the 21st century. Revue Scientifique Et Technique (International Office Of Epizootics). 1992; 11(1):255-62.

Reason for exclusion: Narrative review, no data

Atienza VC. Epidemiology of rabies in animals in the Philippines. In: Rabies control in Asia, Dodet B, Meslin FX, editors, 2001.

Reason for exclusion: Results of microbiological survey of canine samples

Aye Y. Myanmar – Human aspects of rabies prevention and control. In: Rabies control in Asia, Dodet B, Meslin FX, editors 1997.

Reason for exclusion: Countiry report, descriptive epidemiology of rabies

Barboza P, Tarantola A, Lassel L, Mollet T, Quatresous I, Paquet C. Viroses émergentes en Asie du Sud-Est et dans le Pacifique. Médecine Et Maladies Infectieuses. 2008;38(10):513-23.

Reason for exclusion: Narrative review, no data

Beran. Ecology of dogs in the Central Philippines in relation to rabies control efforts. Comparative Immunology, Microbiology and Infectious Diseases. 1982.5: 265-270.

Reason for exclusion: Survey of dog population prior to rabies control program, no outcome data

Beran GW, Frith M. Domestic animal rabies control: an overview. Reviews Of Infectious Diseases. 1988;10 Suppl 4:S672-S7.

Reason for exclusion: Development of a model using a city in Ecuador

Bingham J. Rabies on Flores Island, Indonesia: is eradication possible in the near future? Dodet B, Meslin FX, editors2001.

Reason for exclusion: Narrative review, no data

Bögel K and Meslin FX. Economics of human and canine rabies elimination: guidelines for program orientation. Bulletin Of The World Health Organization. 1990;68(3):281-91.

Reason for exclusion: Model-based cost-effectiveness study Burki T. The global fight against rabies. Lancet. 2008; 372(9644):1135-6.

Reason for exclusion: News article

Cabello C C, Cabello C F. [Zoonoses with wildlife reservoirs: a threat to public health and the economy]. Revista Médica De Chile. 2008; 136(3):385-93.

Reason for exclusion: Narrative review, no data

Camba RA. Philippines – Update of rabies control program. In: Rabies control in Asia, Dodet B, Meslin FX, editors, 1997.

Reason for exclusion: Narrative review of rabies contol program

Childs JE, Robinson LE, Sadek R, Madden A, Miranda ME, Miranda NL. Density estimates of rural dog populations and an assessment of marking methods during a rabies vaccination campaign in the Philippines. Preventive Veterinary Medicine. 1998;33(1-4):207-18.

Reason for exclusion: Tracking methodology paper

Cleaveland S, Kaare M, Knobel D, Laurenson MK. Canine vaccination—providing broader benefits for disease control. Veterinary Microbiology. 2006;117(1):43-50.

Reason for exclusion: Narrative review, no data

Cleaveland S, Meslin FX, Breiman R. Dogs can play useful role as sentinel hosts for disease. Nature. 2006;440(7084):605-.

Reason for exclusion: Letter, no data

Clements ACA, Pfeiffer DU. Emerging viral zoonoses: Frameworks for spatial and spatiotemporal risk assessment and resource planning. The Veterinary Journal. 2009;182(1):21-30.

Reason for exclusion: Narrative review, no data

Coker RJ, Hunter BM, Rudge JW, Liverani M, Hanvoravongchai P. Health in Southeast Asia 3: Emerging infectious diseases in Southeast Asia: regional challenges to control. The Lancet. 2011;377(9765):599-609.

Reason for exclusion: Narrative review, no data

Coleman PG, Fevre, EM, Cleaveland, S. Estimating the public health impact of rabies. Emerging Infectious Diseases. 2004;10:140-2.

Reason for exclusion: Burden of disease using DALY

Dalla Villa P, Kahn S, Stuardo L, Iannetti L, Di Nardo A, Serpell JA. Free-roaming dog control among OIE-member countries. Preventive Veterinary Medicine. 2010;97(1):58-63.

Reason for exclusion: Description of activities but no data on outcomes

Dang Vung N. Animal-Human Health Interface and community based surveillance in Viet Nam-a strategy under Mekong Basin Disease Surveillance Cooperation (MBDS). BMC Proceedings. 2011;5 (Suppl 1):P113.

Reason for exclusion: Poster presentation at conference, only abstract available

DaSilva E and Iaccarino M. Emerging diseases: a global threat. Biotechnology Advances. 1999;17(4-5):363-84.

Reason for exclusion: Narrative review, no data

Denduangboripant J, Wacharapluesadee S, Lumlertdacha B, Ruankaew N, Hoonsuwan W, Puanghat A, et al. Transmission dynamics of rabies virus in Thailand: implications for disease control. BMC Infectious Diseases. 2005;5:52-.

Reason for exclusion: Planning using genetic epidemiology, no intervention

Dodet B, Goswami A, Gunasekera A, de Guzman F, Jamali S, Montalban C, et al. Rabies awareness in eight Asian countries. Vaccine. 2008;26(50):6344-8.

Reason for exclusion: Survey of awareness, no intervention

Douangmala S, Inthavong P. Laos – Report on medical and veterinary aspects of prevention and control of rabies, In: Rabies control in Asia, Dodet B, Meslin FX, editors 1997, pages 165–166.

Reason for exclusion: Narrative of rabies control program

Estrada R, Vos, A and De Leon, RC. Acceptability of local made baits for oral vaccination of dogs against rabies in the Philippines. BMC Infectious Diseases. 2001b; 1: article no. 19.

Reason for exclusion: Trial of acceptability of different baits; only one time point measured.

Fishbein DB, Miranda NJ, Merrill P, Camba RA, Meltzer M, Carlos ET, et al. Rabies control in the Republic of the Philippines: benefits and costs of elimination. Vaccine. 1991;9(8):581-7.

Reason for exclusion: Model-based cost-effectiveness study

Fu ZF. The rabies situation in Far East Asia. Developments In Biologicals. 2008;131:55-61.

Reason for exclusion: Cross-sectional survey of rabies epidemiology in Far East Asian countries

Gongal G, Wright AE. Human Rabies in the WHO Southeast Asia Region: Forward Steps for Elimination. Advances In Preventive Medicine. 2011;2011:383870-.

Reason for exclusion: Narrative review, no data

Grace D, Gilbert J, Lapar ML, Unger F, Fèvre S, Nguyen-viet H, et al. Zoonotic Emerging Infectious Disease in Selected Countries in Southeast Asia: Insights from Ecohealth. Ecohealth. 2011;8(1):55-62.

Reason for exclusion: No control/intervention data

Gummow B. Challenges posed by new and re-emerging infectious diseases in livestock production, wildlife and humans. Livestock Science. 2010; 130(1-3):41-6.

Reason for exclusion: Narrative review, no data

Hensel A, Neubauer H. Human pathogens associated with on-farm practices – Implications for control and surveillance strategies. Smulders FJM, Collins JD, editors2002.

Reason for exclusion: Narrative review, no data

Hernandez JA, Krueger TM, Robertson SA, Isaza N, Greiner EC, Heard DJ, et al. Education of global veterinarians. Preventive Veterinary Medicine. 2009;92(4):275-83.

Reason for exclusion: Description of veterinary degree

Hirayama N, Jusa ER, Noor MAR, Sakaki K, Ogata M. Immune state of dogs injected with rabies vaccines in the West-Java, Indonesia. Japanese Journal of Veterinary Science. 1990 Oct;52(5):1099-101.

Reason for exclusion: Experimental immune response study

Hoonsuwan W, Puanghat A. [Rabies control in Thailand]. Journal Of The Medical Association Of Thailand = Chotmaihet Thangphaet. 2005; 88(10):1471-5.

Reason for exclusion: In Thai

Hussin AA. Malaysia – Veterinary aspects of rabies control and prevention. In: Rabies control in Asia, Dodet B, Meslin FX, editors, 1997, pages 167–170.

Reason for exclusion: Country report, descriptive epidemiology of rabies

Huy BQ. Viet Nam – Rabies control in the dog population. In: Rabies control in Asia, Dodet B, Meslin FX, editors, 1997, pages 202–203.

Reason for exclusion: Country report, descriptive epidemiology of rabies

Jackson AC. Rabies. Neurologic Clinics. 2008;26(3):717-26.

Reason for exclusion: Clinical progression of disease

John TJ, Samuel R, Balraj V, John R. Disease surveillance at district level: A model for developing countries. The Lancet. 1998;352(9121):58-61.

Reason for exclusion: Setting India

Joshi DD. Organisation of veterinary public health in the south Asia region. Revue Scientifique Et Technique (International Office Of Epizootics). 1991;10(4):1101-2.

Reason for exclusion: Narrative review, no data. Rationale for veterinary public health office in WHO regional offices

Joshi DD, Bogel K. Role of lesser developed nations in rabies research. Reviews of infectious diseases. 1998;10,S4:S600-2.

Reason for exclusion: Nepal study, voluntary participation survey Kamoltham T, Tepsumethanon V, Wilde H. Rat rabies in Phetchabun Province, Thailand. Journal Of Travel Medicine. 2002 Mar-Apr;9(2):106-7.

Reason for exclusion: Case report

Kasempimolporn S, Jitapunkul S, Sitprija V. Moving towards the elimination of rabies in Thailand. Journal Of The Medical Association Of Thailand = Chotmaihet Thangphaet. 2008;91(3):433-7.

Reason for exclusion: Narrative review, no data

Kasempimolporn S, Sichanasai B, Saengseesom W, Puempumpanich S, Chatraporn S, Sitprija V. Prevalence of rabies virus infection and rabies antibody in stray dogs: A survey in Bangkok, Thailand. Preventive Veterinary Medicine. 2007;78(3-4):325-32.

Reason for exclusion: Dog seroprevalence study, no intervention

Kasempimolporn S, Sichanasai B, Saengseesom W, Puempumpanich S, Sitprija V. Stray dogs in Bangkok, Thailand: Rabies virus infection and rabies antibody prevalence. In: Dodet B, Fooks AR, Miller T, Tordo N, editors. Towards the Elimination of Rabies in Eurasia2008. p. 137-43.

Reason for exclusion: Dog seroprevalence study, no intervention

Kauffman FH, Goldmann BJ. Rabies. The American Journal of Emergency Medicine. 1986;4(6):525-31.

Reason for exclusion: Review article of treatment of rabies, no data

King AA, Turner GS. Rabies: A Review. Journal of Comparative Pathology. 1993;108(1):1-39.

Reason for exclusion: Review article of clinical management, no data

Kingnate D, Sagarasaeranee P, Choomkasien P. Thailand – Rabies control (human side). In: Rabies control in Asia, Dodet B, Meslin FX, editors, 1997, pages 194–6.

Reason for exclusion: Country report, mostly statistics on PET

Knobel D. Cleaveland S et al. Re-evaluating the burden of rabies in Asia and Africa. 2005. Bull World Health Organisation 83, 360-368.

Reason for exclusion: No intervention, model-based burden of disease estimation

Kongkaew W, Coleman P, Pfeiffer DU, Antarasena C, Thiptara A. Vaccination coverage and epidemiological parameters of the owned-dog population in Thungsong District, Thailand. Preventive Veterinary Medicine. 2004; 65(1-2):105-15.

Reason for exclusion: Knowledge and vaccine coverage survey, no intervention

Ksiazek TG, Rota PA, Rollin PE. A review of Nipah virus and Hendra viruses with an historical aside. Virus Research. 2011;162(1-2):173-83.

Reason for exclusion: Narrative review, no data

Li VC, Goethals PR, Dorfman S. A Global Review of Training of Community Health Workers. International Quarterly Of Community Health Education. 2006;27(3):181-218.

Reason for exclusion: Review article of community health worker training, no data

Loke YK, Murugesan E, Suryati A, Tan MH. An outbreak of rabies in dogs in the state of Terengganu 1995-1996. The Medical Journal Of Malaysia. 1998;53(1):97-100.

Reason for exclusion: Case study, no data

Ly S, Buchy P, Heng NY, Ong S, Chhor N, Bourhy H, et al. Rabies situation in Cambodia. Plos Neglected Tropical Diseases. 2009;3(9):e511-e.

Reason for exclusion: Descriptive epidemiology, no intervention

Mackenzie JS. Emerging zoonotic encephalitis viruses: Lessons from Southeast Asia and Oceania. Journal Of Neurovirology. 2005 Oct;11(5):434-40.

Reason for exclusion: Review of emergence and clinical features

Mackenzie JS, Chua KB, Daniels PW, Eaton BT, Field HE, Hall RA, et al. Emerging viral diseases of Southeast Asia and the Western Pacific. Emerging Infectious Diseases. 2001;7(3 Suppl):497-504.

Reason for exclusion: Review of epidemiology and emergence, no data

Mai LTP, Dung LP, Tho NTT, Quyet NT, Than PD, Mai NDC, et al. Community knowledge, attitudes, and practices toward rabies prevention in North Viet Nam. International Quarterly Of Community Health Education. 2010 2010-2011;31(1):21-31.

Reason for exclusion: Knowledge survey, no intervention

Meltzer MI and Rupprecht CE. A review of the economics of the prevention and control of rabies. Part 2: Rabies in dogs, livestock and wildlife. Pharmacoeconomics. 1998;14(5):481-98.

Reason for exclusion: Model-based cost-effectiveness and cost-benefit analysis

Meslin FX, Fishbein DB, Matter HC. Rationale and prospects for rabies elimination in developing countries. Current Topics in Microbiology and Immunology. 1994;187:1-26.

Reason for exclusion: Narrative review, no data

Miranda MEG. Rabies in humans in the Philippines. In: Rabies control in Asia, Dodet B, Meslin FX, editors, 2001, page 244.

Reason for exclusion: Short country report, no intervention

Mitmoonpitak C, Tepsumethanon V, Wilde H. Rabies in Thailand. Epidemiology And Infection. 1998;120(2):165-9.

Reason for exclusion: No useful data

Mitmoonpitak C, Wilde H, Tepsumetanon W. Current status of animal rabies in Thailand. The Journal Of Veterinary Medical Science / The Japanese Society Of Veterinary Science. 1997;59(6):457-60.

Reason for exclusion: No useful data on change or intervention, observational

Nara PL, Nara D, Chaudhuri R, Lin G, Tobin G. Perspectives on advancing preventative medicine through vaccinology at the comparative veterinary, human and conservation medicine interface: Not missing the opportunities. Vaccine. 2008;26(49):6200-11.

Reason for exclusion: Narrative review, no data

Nicholson, K.G. Rabies. The Lancet. 1990;335(8699):1201-2.

Reason for exclusion: Letter, no data

Oum S. Cambodia – Rabies control. In: Rabies control in Asia, Dodet B, Meslin FX, editors,1997, pages 128–9.

Reason for exclusion: Country report, mostly PET

Pearson JE. Worldwide risks of animal diseases: introduction. Veterinaria Italiana. 2006;42(4):293-.

Reason for exclusion: Narrative review, no data

Perry B, McDermott J, Randolph T. Can epidemiology and economics make a meaningful contribution to national animal-disease control? Preventive Veterinary Medicine. 2001;48(4):231-60.

Reason for exclusion: Narrative review, no data

Puanghat A. Human rabies in Thailand. In: Rabies control in Asia, Dodet B, Meslin FX, editors, 2001, pages 252-3.

Reason for exclusion: Descriptive anlaysis of cases, deaths, PET and laboratory capability

Reynes JM, Soares JL, Keo C, Ong S, Heng NY, Vanhoye B. Characterization and observation of animals responsible for rabies post-exposure treatment in Phnom Penh, Cambodia. Onderstepoort Journal of Veterinary Research. 1999 Jun;66(2):129-33.

Reason for exclusion: Prevalence survey, no intervention

Robertson K, Lumlertdacha B, Franka R, Petersen B, Bhengsri S, Henchaichon S, et al. Rabies-related knowledge and practices among persons at risk of bat exposures in Thailand. Plos Neglected Tropical Diseases. 2011;5(6):e1054-e.

Reason for exclusion: Knowledge survey, no intervention

Rupprecht CE, Barrett J, Briggs D, Cliquet F, Fooks AR, Lumlertdacha B, et al. Can rabies be eradicated? Developments In Biologicals. 2008;131:95-121.

Reason for exclusion: Narrative review, no data

Rupprecht CE, Hanlon CA, Slate D. Control and prevention of rabies in animals: paradigm shifts. Developments In Biologicals. 2006;125:103-11.

Reason for exclusion: Narrative review, no data

Sagarasaeranee P, Puanghat A, Kasempimolparn S, Khawplod P. Efficacy of oral rabies vaccine in dogs in Thailand. Dodet B, Meslin FX, editors2001.

Reason for exclusion: Microbiological vaccine study

Salman, MD. The role of veterinary epidemiology in combating infectious animal diseases on a global scale: The impact of training and outreach programs. Preventive Veterinary Medicine. 2009;92(4):284-7.

Reason for exclusion: Descriptive overview only, no data

Salva EP. Philippines – Human rabies: prevention and control. In: Rabies control in Asia, Dodet B, Meslin FX, editors,1997.

Reason for exclusion: Statistics on PET doses

Sen S. Rabies in animals in Cambodia. In: Rabies control in Asia, Dodet B, Meslin FX, editors, 2001.

Reason for exclusion: Country report, descriptive statistics

Shahirudin S. Rabies in animals in Malaysia. In: Rabies control in Asia, Dodet B, Meslin FX, editors, 2001.

Reason for exclusion: Country report, descriptive statistics

Shaughnessy A. Rabies. Evidence-Based Practice. 1999;2(8):11, insert 2p.

Reason for exclusion: Treatment guidelines

Simanjuntak GM, Suroso T. Indonesia – Rabies elimination: the national program and its impact. In: Rabies control in Asia, Dodet B, Meslin FX, editors1997.

Reason for exclusion: Descriptive statistics on cases, number of bites, PET

Singhchai C. Dog rabies control in Bangkok metropolitan area: why has rabies not been eliminated from Bangkok? In: Rabies control in Asia, Dodet B, Meslin FX, editors2001.

Reason for exclusion: Narrative of control interventions

Slater, MR. The role of veterinary epidemiology in the study of free-roaming dogs and cats. Preventive Veterinary Medicine. 2001;48(4):273-86.

Reason for exclusion: Narrative review, no data

Soeung SC, Grundy J, Morn C, Samnang C. Evaluation of Immunization Knowledge, Practices, and Service-delivery in the Private Sector in Cambodia. Journal of Health, Population and Nutrition. 2008;26(1):95-104.

Reason for exclusion: Vaccine evaluation

Sok T. Rabies in humans in Cambodia. In: Rabies control in Asia, Dodet B, Meslin FX, editors, 2001.

Reason for exclusion: Country report, descriptive statistics

Sornnuwat J. Animal rabies and animal rabies control in Thailand. In: Rabies control in Asia, Dodet B, Meslin FX, editors, 2001.

Reason for exclusion: Narrative of control interventions

Sriaroon C, Sriaroon P, Daviratanasilpa S, Khawplod P, Wilde H. Retrospective: animal attacks and rabies exposures in Thai children. Travel Medicine And Infectious Disease. 2006;4(5):270-4.

Reason for exclusion: Epidemiology of bites, no intervention

Srisongmuang W. Thailand – Structure of veterinary services of rabies disease control. In: Rabies control in Asia, Dodet B, Meslin FX, editors, 1997.

Reason for exclusion: Country report, descriptive statistics

Stahl JP, Mailles A, Dacheux L, Morand P. Epidemiology of viral encephalitis in 2011. Médecine Et Maladies Infectieuses. 2011;41(9):453-64.

Reason for exclusion: Clinical review

Suroso T, Ganefa W, Wilfried C, Tato T, Endang. Rabies in humans in Indonesia. In: Rabies control in Asia, Dodet B, Meslin FX, editors2001.

Reason for exclusion: Descriptive statistics of cases and PET

Suroso T, Simanjuntak GM. Indonesia – Rabies elimination: policy implementation. In: Rabies control in Asia, Dodet B, Meslin FX, editors,1997.

Reason for exclusion: Narrative of rabies control activities

Swe TB, Hla T. Rabies control in Myanmar. Dodet B, Meslin FX, editors,2001.

Reason for exclusion: Descriptive statistics of cases, deaths, outbreaks and PET

Tepsumethanon W, Polsuwan C, Lumlertdaecha B, Khawplod P, Hemachudha T, Chutivongse S, et al. Immune response to rabies vaccine in Thai dogs: A preliminary report. Vaccine. 1991;9(9):627-30.

Reason for exclusion: No community data

Vaillancourt JP. A regional approach to biosecurity: the poultry example. Bulletin De L Academie Veterinaire De France. 2009 Jul-Sep;162(3):257-64.

Reason for exclusion: Narrative review, no data

Voelckel J. [Diseases arising from contact with animals in an urban tropical milieu]. Bulletin De La Société De Pathologie Exotique Et De Ses Filiales. 1983;76(3):293-9.

Reason for exclusion: Narrative review, no data

Wallerstein C. Rabies cases increase in the Philippines. BMJ (Clinical Research Ed). 1999;318(7194):1306-.

Reason for exclusion: News article

Warrell MJ, Warrell DA. Rabies and other lyssavirus diseases. The Lancet. 2004;363(9413):959-69.

Reason for exclusion: Clinical review

Wasi C, Chaiprasithikul P, Thongcharoen P, Choomkasien P, Sirikawin S. Progress and achievement of rabies control in Thailand. Vaccine. 1997;15, Supplement(0):S7-S11.

Reason for exclusion: No community or outbreak data

Wilde H, Chutivongse S, Hemachudha T. Rabies and its prevention. The Medical Journal Of Australia. 1994;160(2):83-7.

Reason for exclusion: Narrative review, no data

Wilde H, Hemachudha T, Khawplod P, Tepsumethanon V, Wacharapluesadee S, Lumlertdacha B. Rabies 2007: perspective from Asia. Asian Biomedicine. 2007 Dec;1(4):345-57.

Reason for exclusion: Narrative review, no data

Wilde H, Khawplod P, Khamoltham T, Hemachudha T, Tepsumethanon V, Lumlerdacha B, et al. Rabies control in South and Southeast Asia. Vaccine. 2005;23(17-18):2284-9.

Reason for exclusion: Narrative review, no data

Xuyen DK. Viet Nam – Rabies control. In: Rabies control in Asia, Dodet B, Meslin FX, editors, 1997.

Reason for exclusion: Country report, descriptive statistics on cases and PET

Zinsstag J, Schelling E, Roth F, Bonfoh B, de Savigny D, Tanner M. Human benefits of animal interventions for zoonosis control. Emerging Infectious Diseases. 2007;13(4):527-31.

Reason for exclusion: Narrative review, no data

Rabies – systematic reviews

Iara Marques Medeiros, Humberto Saconato. Antibiotic use of mammalian bites. Cochrane Library of Systematic Reviews. July 2008.

Reason for exclusion: Pharmaceutical intervention

Shim E, Hampson K, Cleaveland S, Galvani AP. Evaluating the cost-effectiveness of rabies post-exposure prophylaxis: a case study in Tanzania. Centre for Reviews and Dissemination. 2009.

Reason for exclusion: Setting is Africa, pharmaceutical intervention

Zinsstag J, Durr S, Penny MA, Mindekem R, Roth F, Menendez Gonzalez S, Naissengar S, Hattendorf J. Transmission dynamics and economics of rabies control in dogs and humans in an African city. Centre for Reviews and Dissemination. 2009

Reason for exclusion: Setting is Africa

Kaare M, Lembo T, Hampson K, Ernest E, Estes A, Mentzel C, Cleaveland S. Rabies control in rural Africa: evaluating strategies for effective domestic dog vaccination (Provisional abstract) Centre for Reviews and Dissemination. 2009

Reason for exclusion: Setting is Africa

Knobel DL, Cleaveland S, Coleman PG, Fevre EM, Meltzer MI, Miranda ME, Shaw A, Zinsstag J, Meslin FX. Re-evaluating the burden of rabies in Africa and Asia (Structured Abstract) Centre for Reviews and Dissemination. 2005.

Reason for exclusion: Setting is Africa

Dhankhar P, Vaidya SA, Fishbien DB, Meltzer MI. Cost-effectiveness of rabies post-exposure prophylaxis in the United States (Provisional abstract) Centre for Reviews and Dissemination. 2008.

Reason for exclusion: Pharmaceutical intervention

Nipah virus – articles

Ahmad K. Malaysia culls pigs as Nipah virus strikes again. Lancet. 2000;356(9225):230-.

Reason for exclusion: News article

Ali R, Mounts AW, Parashar UD, Sahani M, Lye MS, Isa MM, et al. Nipah virus among military personnel involved in pig culling during an outbreak of encephalitis in Malaysia, 1998-1999. Emerging Infectious Diseases. 2001;7(4):759-61.

Reason for exclusion: Letter, no data

Barboza P, Tarantola A, Lassel L, Mollet T, Quatresous I, Paquet C. Viroses émergentes en Asie du Sud-Est et dans le Pacifique. Médecine Et Maladies Infectieuses. 2008;38(10):513-23.

Reason for exclusion: Review article, no information on control and no data

Bellini WJ. Commentary: Paramyxoviruses, pigs and abattoirs. International Journal Of Epidemiology. 2001;30(5):1020-.

Reason for exclusion: Comment only, no data

Bellini WJ, Harcourt BH, Bowden N, Rota PA. Nipah virus: an emergent paramyxovirus causing severe encephalitis in humans. Journal Of Neurovirology. 2005;11(5):481-7.

Reason for exclusion: Review article, focussing on genetics of henipaviruses, no data

Breed AC, Field HE, Epstein JH, Daszak P. Emerging henipaviruses and flying foxes – Conservation and management perspectives. Biological Conservation. 2006 Aug;131(2):211–20.

Reason for exclusion: Review article, no data

Butler D. Fatal fruit bat virus sparks epidemics in southern Asia. Nature. 2004;429(6987):7-.

Reason for exclusion: News article only

Caplan CE. Update on the new virus in Malaysia. Canadian Medical Association Journal. 1999; 160(12):1697.

Reason for exclusion: Short outbreak report, same data presented elsewhere

Chan KP, Rollin PE, Ksiazek TG, Leo YS, Goh KT, Paton NI, et al. A survey of Nipah virus infection among various risk groups in Singapore. Epidemiology And Infection. 2002;128(1):93-8.

Reason for exclusion: Prevalence survey, no intervention

Chastel C. Emergence of new viruses in Asia: is climate change involved? Médecine Et Maladies Infectieuses. 2004 Nov;34(11):499-505.

Reason for exclusion: Review article, no data

Chew MH, Arguin PM, Shay DK, Goh KT, Rollin PE, Shieh WJ, et al. Risk factors for Nipah virus infection among abattoir workers in Singapore. The Journal Of Infectious Diseases. 2000;181(5):1760-3.

Reason for exclusion: Risk factor study, no intervention

Choi C. Nipah virus's return. The lethal “flying fox” virus may spread between people. Scientific American. 2004;291(3):21A.

Reason for exclusion: Comment only, no data

Choi CQ. Going to bat. Scientific American. 2006;294(3):24.

Reason for exclusion: Comment only, no data

Chong HT, Abdullah S, Tan CT. Nipah virus and bats. Neurology Asia. 2009 Jun;14(1):73-6.

Reason for exclusion: Review article, no data

Chong HT, Kunjapan R, Thayaparan T, Tong JMG, Petharunam V, Jusoh MR, et al. Nipah virus encephalitis outbreak in Malaysia, clinical features in patients from Seremban. Canadian Journal of Neurological Sciences. 2002 Feb;29(1):83-7.

Reason for exclusion: Clinical reports

Chua KB. Nipah virus outbreak in Malaysia. Journal of Clinical Virology: The Official Publication Of The Pan American Society For Clinical Virology. 2003;26(3):265-75.

Reason for exclusion: Epidemiological report on outbreak but no data on intervention

Chua KB. Epidemiology, surveillance and control of Nipah virus infections in Malaysia. The Malaysian Journal Of Pathology. 2010a;32(2):69-73.

Reason for exclusion: Epidemiological report on outbreak, no data before and after interventions

Chua KB. Risk factors, prevention and communication strategy during Nipah virus outbreak in Malaysia. The Malaysian Journal Of Pathology. 2010b;32(2):75-80.

Reason for exclusion: Description of control measures but no data on interventions

Chua KB, Bellini WJ, Rota PA, Harcourt BH, Tamin A, Lam SK, et al. Nipah virus: a recently emergent deadly paramyxovirus. Science (New York, NY). 2000;288(5470):1432-5.

Reason for exclusion: Review article, no data

Chua KB, Goh KJ, Wong KT, Kamarulzaman A, Tan PSK, Ksiazek TG, et al. Fatal encephalitis due to Nipah virus among pig-farmers in Malaysia. The Lancet. 1999;354(9186):1257-9.

Reason for exclusion: Clinical outcome, no intervention

Chua KB, Lam SK, Goh KJ, Hooi PS, Ksiazek TG, Kamarulzaman A, et al. The presence of Nipah virus in respiratory secretions and urine of patients during an outbreak of Nipah virus encephalitis in Malaysia. The Journal Of Infection. 2001;42(1):40-3.

Reason for exclusion: No intervention trialled

Easton A. New virus is identified in Malaysia epidemic. British Medical Journal. 1999;318(7193):1232-.

Reason for exclusion: News article

Eaton BT, Broder CC, Wang L-F. Hendra and Nipah virus es: pathogenesis and therapeutics. Current Molecular Medicine. 2005;5(8):805-16.

Reason for exclusion: Microbiology and clinical management, no intervention

Enserink M. New virus fingered in Malaysian epidemic. Science. 1999;284(5413):407-10.

Reason for exclusion: News article

Enserink M. Emerging diseases. Malaysian researchers trace Nipah virus outbreak to bats. Science (New York, NY). 2000;289(5479):518-9.

Reason for exclusion: News article

Enserink M. Malaysian researchers trace Nipah virus outbreak to bats. Science. 2000;289(5479):518-9.

Reason for exclusion: News article

Enserink M. Emerging infectious diseases. Nipah virus (or a cousin) strikes again. Science (New York, NY). 2004;303(5661):1121-.

Reason for exclusion: News article

Epstein JH, Field HE, Luby S, Pulliam JRC, Daszak P. Nipah virus: impact, origins, and causes of emergence. Current Infectious Disease Reports. 2006;8(1):59-65.

Reason for exclusion: Review article, no data

Epstein JH, Rahman SA, Pulliam JRC, Hassan SS, Halpin K, Smith CS, et al. The Emergence of Nipah virus in Malaysia: The Role of Pteropus Bats as Hosts and Agricultural Expansion as a Key Factor for Zoonotic Spillover. International Journal of Infectious Diseases. 2008 Dec;12:E46-E.

Reason for exclusion: Discusses transmission, factors in emergence, no intervention

Epstein JH, Rahman SA, Smith CS, Halpin K, Sharifah SH, Jamaluddin AA, et al. The emergence of Nipah virus in Malaysia: Epidemiology and host ecology of Pteropus bats. American Journal of Tropical Medicine and Hygiene. 2007 Nov;77(5):272-.

Reason for exclusion: Ecology of bats, no intervention

Farrar JJ. Nipah virus-virus encephalitis-investigation of a new infection. Lancet. 1999;354(9186):1222-3.

Reason for exclusion: Reference to Pro-MED, no data

Field H, Kung N. Henipaviruses – unanswered questions of lethal zoonoses. Current Opinion in Virology. 2011;1(6):658–61.

Reason for exclusion: Review of epidemiology, no information on control

Field H, Mackenzie J, Daszak P. Novel viral encephalitides associated with bats (Chiroptera) – host management strategies. Archives of Virology. 2004:113–21.

Reason for exclusion: Review article on Henipaviruses and host management strategies, no data

Field H, Young P, Yob JM, Mills J, Hall L, Mackenzie J. The natural history of Hendra and Nipah virus es. Microbes and Infection. 2001;3(4):307-14.

Reason for exclusion: Review article, no data

Field HE, Mackenzie JS, Daszak P. Henipaviruses: emerging paramyxoviruses associated with fruit bats. Current Topics In Microbiology And Immunology. 2007;315:133-59.

Reason for exclusion: Review article, no data

Gurley ES, Luby SP. Nipah virus transmission in south Asia: exploring the mysteries and addressing the problems. Future Virology. 2011 Aug;6(8):897-900.

Reason for exclusion: Editorial review

Halpin K, Mungall BA. Recent progress in henipavirus research. Comparative Immunology, Microbiology and Infectious Diseases. 2007;30(5-6):287-307.

Reason for exclusion: Review of antibody research, no information on control activities

Henrich TJ, Hutchaleelaha S, Jiwariyavej V, Barbazan P, Nitatpattana N, Yoksan S, et al. Geographic dynamics of viral encephalitis in Thailand. Microbes and Infection. 2003 Jun;5(7):603-11.

Reason for exclusion: Spatial model of vaccine effectiveness, no intervention

Heymann DL. Social, behavioural and environmental factors and their impact on infectious disease outbreaks. Journal of Public Health Policy. 2005;26(1):133-9.

Reason for exclusion: Commentary only, no data

Heymann DL, Rodier GR. Hot spots in a wired world: WHO surveillance of emerging and re-emerging infectious diseases. The Lancet Infectious Diseases. 2001;1(5):345-53.

Reason for exclusion: Discusses WHO's Global Outbreak Alert and Response Network (GOARN)

Kai C. [Nipah virus infections]. Nihon Naika Gakkai Zasshi The Journal Of The Japanese Society Of Internal Medicine. 2004;93(11):2341-6.

Reason for exclusion: Review article, no data

Kolomytsev AA, Kurinnov VV, Mikolaľchuk SV, Zakutskiľ NI. [Nipah virus encephalitis]. Voprosy Virusologii. 2008;53(2):10-3.

Reason for exclusion: Review article, no data

Ksiazek TG, Rota PA, Rollin PE. A review of Nipah virus and Hendra viruses with an historical aside. Virus Research. 2011;162(1-2):173-83.

Reason for exclusion: Review article, no data

Lam, SK. Nipah virus —a potential agent of bioterrorism? Antiviral Research. 2003;57(1-2):113-9.

Reason for exclusion: Description of control measures but no data on interventions

Lam, SK and Chua, KB. The Nipah virus outbreak and control response in Malaysia. Emergence and control of zoonotic ortho and paramyxovirus diseases. In: Dodet B, Vicari M, editors, John Libbey Eurotext, Paris, pp. 199-203. 2001.

Reason for exclusion: Review article, no data

Lam SK and Chua KB. Nipah virus encephalitis outbreak in Malaysia. Clinical Infectious Diseases: An Official Publication of the Infectious Diseases Society Of America. 2002;34 Suppl 2:S48-S51.

Reason for exclusion: Review article, no data

Ling AE. Lessons to be learnt from the Nipah virus outbreak in Singapore. Singapore Medical Journal. 1999;40(5):331-2.

Reason for exclusion: Description of outbreak only

Looi L-M, Chua K-B. Lessons from the Nipah virus outbreak in Malaysia. The Malaysian Journal Of Pathology. 2007;29(2):63-7.

Reason for exclusion: Narrative of outbreak progression, clinical features of infection. No data on interventions

Luby SP, Gurley ES, Hossain MJ. Transmission of human infection with Nipah virus. Clinical Infectious Diseases: An Official Publication Of The Infectious Diseases Society Of America. 2009;49(11):1743-8.

Reason for exclusion: Human to human transmission, Bangladesh outbreak

Madić J. [Zoonoses caused by new viruses in the Paramyxoviridae family]. Liječnički Vjesnik. 2001;123(5-6):141-5.

Reason for exclusion: Review article, no data

McCormack JG. Hendra and Nipah virus es: new zoonotically-acquired human pathogens. Respiratory Care Clinics Of North America. 2005;11(1):59-66.

Reason for exclusion: Clinical review, no data

Mills JN, Alim ANM, Bunning ML, Lee OB, Wagoner KD, Amman BR, et al. Nipah virus infection in dogs, Malaysia, 1999. Emerging Infectious Diseases. 2009;15(6):950-2.

Reason for exclusion: Prevalence survey in dogs

Nahar N, Sultana R, Gurley ES, Hossain MJ, Luby SP. Date Palm Sap Collection: Exploring Opportunities to Prevent Nipah virus Transmission. Ecohealth. 2010;7(2):196-203.

Reason for exclusion: Setting is Bangladesh

Ng CW, Choo WY, Chong HT, Dahlui M, Goh KJ, Tan CT. Long-term socioeconomic impact of the Nipah virus encephalitis outbreak in Bukit Pelanduk, Negeri Sembilan, Malaysia: A mixed methods approach. Neurology Asia. 2009 Dec;14(2):101-7.

Reason for exclusion: Socio-economic outcomes for Nipah virus patients

Okabe N, Morita K. [Nipah virus outbreak in Malaysia, 1999]. Uirusu. 2000;50(1):27-33.

Reason for exclusion: Review article, no data

Olival KJ, Daszak P. The ecology of emerging neurotropic viruses. Journal Of Neurovirology. 2005 Oct;11(5):441-6.

Reason for exclusion: Risk factor study, no intervention

Parashar UD, Sunn LM, Ong F, Mounts AW, Arif MT, Ksiazek TG, et al. Case-control study of risk factors for human infection with a new zoonotic paramyxovirus, Nipah virus, during a 1998-1999 outbreak of severe encephalitis in Malaysia. Journal of Infectious Diseases. 2000 May;181(5):1755-9.

Reason for exclusion: Case control study of risk factors. Study did not guide an intervention, was confirmatory of outbreak control measures

Paton NI, Leo YS, Zaki SR, Auchus AP, Lee KE, Ling AE, et al. Outbreak of Nipah virus-virus infection among abattoir workers in Singapore. Lancet. 1999;354(9186):1253-6.

Reason for exclusion: Clinical report, no outbreak data

Premalatha GD, Lye MS, Ariokasamy J, Parashar UD, Rahmat R, Lee BY, et al. Assessment of Nipah virus transmission among pork sellers in Seremban, Malaysia. The Southeast Asian Journal Of Tropical Medicine And Public Health. 2000;31(2):307-9.

Reason for exclusion: Transmission estimates, no intervention

Pulliam JR, Dushoff J, Field HE, Epstein JH, Dobson AP, Daszak P, et al. Understanding Nipah virus emergence in peninsular Malaysia: The role of epidemic enhancement in domestic pig populations. American Journal of Tropical Medicine and Hygiene. 2007 Nov;77(5):273-.

Reason for exclusion: Abstract only (American Society for Tropical Medicine and Hygiene 56th Annual Meeting), no data

Redington JJ, Tyler KL. Viral infections of the nervous system, 2002: Update on diagnosis and treatment. Archives of Neurology. 2002;59(5):712-8.

Reason for exclusion: Clinical review

Rollin PE, Rota P, Zaki S, Ksiazek TG. Hendra and Nipah virus es. Baron EJ, Jorgensen JH, Landry ML, Pfaller MA, editors2007.

Reason for exclusion: Review article, no data

Roth C. Crises, Challenges and Response Epidemic and Pandemic Alert and Response. Refugee Survey Quarterly. 2006;25(4):100-3.

Reason for exclusion: Review article, no data

Sahani M, Parashar UD, Ali R, Das P, Lye MS, Isa MM, et al. Nipah virus infection among abattoir workers in Malaysia, 1998-1999. International Journal Of Epidemiology. 2001;30(5):1017-20.

Reason for exclusion: Prevalence estimates (sero-survey), cross-sectional study, no intervention

Sendow I, Field HE, Adjid A, Ratnawati A, Breed AC, Darminto, et al. Screening for Nipah virus infection in West Kalimantan province, Indonesia. Zoonoses And Public Health. 2010;57(7-8):499-503.

Reason for exclusion: Prevalence survey in bats

Sohayati AR, Hassan SS, Hassan L, Epstein JH, Arshad SS, Mohamed R, et al. Endemicity of Nipah virus in Pteropus Bats Over Wide Geographical Areas in Peninsular Malaysia. International Journal of Infectious Diseases. 2008 Dec;12:E138-E.

Reason for exclusion: Prevalence survey in bats

Solomon T. Exotic and emerging viral encephalitides. Current Opinion In Neurology. 2003;16(3):411-8.

Reason for exclusion: Review article, no data

Taha M 1999. An outbreak of Nipah virus in Malaysia. A working paper for WHO meeting on zoonotic paramyxoviruses, Kuala Lumpur, Malaysia, 19-21 July, 1999.

Reason for exclusion: Outbreak report to MOH, Malaysia. Case series of no. of cases, no. of deaths

Tambyah PA. The Nipah virus outbreak-a reminder. Singapore Medical Journal. 1999;40(5):329-30.

Reason for exclusion: Comment, no data

Tan CT, Wong KT. Nipah virus encephalitis outbreak in Malaysia. Annals Of The Academy Of Medicine, Singapore. 2003;32(1):112-7.

Reason for exclusion: Clinical summary, no control data

Tee KK, Takebe Y, Kamarulzaman A. Emerging and re-emerging viruses in Malaysia, 1997-2007. International Journal of Infectious Diseases. 2009;13(3):307-18.

Reason for exclusion: Review article, no data

Ternhag A, Penttinen P. [Nipah virus –another product from the Asian “virus factory”]. Läkartidningen. 2005;102(14):1046–7.

Reason for exclusion: News article

Uppal PK. Emergence of Nipah virus in Malaysia. Annals Of The New York Academy Of Sciences. 2000;916:354-7.

Reason for exclusion: Narrative of outbreak, but no data on interventions

Wacharapluesadee S, Boongird K, Wanghongsa S, Ratanasetyuth N, Supavonwong P, Saengsen D, et al. A Longitudinal Study of the Prevalence of Nipah virus in Pteropus lylei Bats in Thailand: Evidence for Seasonal Preference in Disease Transmission. Vector-Borne and Zoonotic Diseases. 2010 Mar;10(2):183-90.

Reason for exclusion: Prevalence in bats, seasonal surveillance, no intervention

Wacharapluesadee S, Lumlertdacha B, Boongird K, Wanghongsa S, Chanhome L, Rollin P, et al. Bat Nipah virus, Thailand. Emerging Infectious Diseases. 2005;11(12):1949-51.

Reason for exclusion: Prevalence in bats

Watts J. Hendra-like virus responsible for epidemic in Malaysia. The Lancet. 1999;353(9161):1335-.

Reason for exclusion: News article

Wild TF. Henipaviruses: a new family of emerging Paramyxoviruses. Pathologie-Biologie. 2009;57(2):188-96.

Reason for exclusion: Review of virus, no information on surveillance or control

WHO. Nipah virus —information from the World Health Organization. Journal of Environmental Health. 2002;64(6):54-.

Reason for exclusion: Review article, no data

WHO. Nipah virus. Relevé Épidémiologique Hebdomadaire / Section D'hygiène Du Secrétariat De La Société Des Nations = Weekly Epidemiological Record / Health Section Of The Secretariat Of The League Of Nations. 2011;86(41):451-5.

Reason for exclusion: Review only, no evidence or data

Wong KT. Emerging and re-emerging epidemic encephalitis: a tale of two viruses. Neuropathology and Applied Neurobiology. 2000 Aug;26(4):313-8.

Reason for exclusion: Clinical outcomes

Wong KT. Nipah virus and Hendra viruses: recent advances in pathogenesis. Future Virology. 2010;5(2):129-31.

Reason for exclusion: Report of vaccine research, no intervention

Yob JM, Field H, Rashdi AM, Morrissy C, van der Heide B, Rota P, et al. Nipah virus infection in bats (order Chiroptera) in peninsular Malaysia. Emerging Infectious Diseases. 2001;7(3):439-41.

Reason for exclusion: Prevalence in animals, no intervention

Dengue – articles

Dengue control program in Malaysia. Gaoxiong Yi Xue Ke Xue Za Zhi = The Kaohsiung Journal Of Medical Sciences 1994;10 Suppl:S113-S5.

Reason for exclusion: Abstract only

Anonymous. Dengue fever/dengue haemorrhagic fever surveillance. 1991. Relevé Épidémiologique Hebdomadaire / Section D'hygiène Du Secrétariat De La Société Des Nations = Weekly Epidemiological Record / Health Section Of The Secretariat Of The League Of Nations 1992;67:296-7.

Reason for exclusion: report of surveillance activity – no evaluation

Anonymous. World Health Organization: Strengthening implementation of the global strategy for dengue fever/dengue haemorrhagic fever prevention and control. Report of the Informal Consultation, 18-20 October 1999, WHO, Geneva. 2000.

Reason for exclusion: Report of consultation, no data or intervention

Arunachalam N, Tana S, Espino F, Kittayapong P, Abeyewickreme W, Wai KT, et al. Eco-bio-social determinants of dengue vector breeding: a multicountry study in urban and periurban Asia. Bulletin Of The World Health Organization 2010;88:173-84.

Reason for exclusion: cross-sectional survey of risk factor, no intervention

Barbazan P, Tuntaprasart W, Souris M, Demoraes F, Nitatpattana N, Boonyuan W, et al. Assessment of a new strategy, based on Aedes aegypti (L.) pupal productivity, for the surveillance and control of dengue transmission in Thailand. Annals Of Tropical Medicine And Parasitology 2008;102:161-71.

Reason for exclusion: cross-sectional survey of feasibility, no intervention

Beauté J, Vong S. Cost and disease burden of Dengue in Cambodia. BMC Public Health 2010;10:521-.

Reason for exclusion: burden of disease study, no estimates of cost or effectiveness of control

Burattini MN, Chen M, Chow A, Coutinho FAB, Goh KT, Lopez LF, et al. Modelling the control strategies against dengue in Singapore. Epidemiology and Infection 2008; 136:309-319.

Reason for exclusion: Model based

Chaikoolvatana A, Chanruang S, Pothaled P. A comparison of dengue hemorrhagic fever control interventions in North-eastern Thailand. The Southeast Asian Journal of Tropical Medicine And Public Health. 2008;39(4):617-24.

Reason for exclusion: Only cross-sectional data, not community-based intervention

Chansang C, Kittayapong P. Application of mosquito sampling count and geospatial methods to improve dengue vector surveillance. American Journal of Tropical Medicine and Hygiene. 2007;77(5):897-902.

Reason for exclusion: Research study comparing methods, not a community-based intervention

Charuai S. Community capacity for sustainable community-based dengue prevention and control: domain, assessment tool and capacity building model. Asian Pacific Journal of Tropical Medicine 2010;3:499-504.

Reason for exclusion: theoretical model for developing interventions, no evaluation data

Chunsuttiwat S, Wasakarawa S. Dengue vector control in Thailand: development towards environmental protection. Gaoxiong Yi Xue Ke Xue Za Zhi = The Kaohsiung Journal Of Medical Sciences 1994;10 Suppl:S122-S3.

Reason for exclusion: abstract only

Coutard B, Canard B. The VIZIER project: Overview; expectations; and achievements. Antiviral Research 2010;87:85-94.

Reason for exclusion: vaccination research

Dominguez NN. National dengue prevention and control program in the Philippines. Gaoxiong Yi Xue Ke Xue Za Zhi = The Kaohsiung Journal Of Medical Sciences 1994;10 Suppl:S118-S21.

Reason for exclusion: Only abstract available

Egger JR, Ooi EE, Kelly DW, Woolhouse ME, Davies CR, Coleman PG. Reconstructing historical changes in the force of infection of dengue fever in Singapore: implications for surveillance and control. Bulletin Of The World Health Organization 2008;86:187-96.

Reason for exclusion: Model-based study predicting epidemiological changes in disease, no intervention

Eisen L, Beaty BJ, Morrison AC, Scott TW. ProactiveVector control strategies and improved monitoring and evaluation practices for dengue prevention. Journal Of Medical Entomology 2009;46:1245-55.

Reason for exclusion: no evaluation data

Elder JP, Ballenger-Browning K. Community involvement in dengue vector control. BMJ (Clinical Research Ed) 2009;338:b1023-b.

Reason for exclusion: editorial, no data

Goh KT, Ng SK, Chan YC, Lim SJ, Chua EC. Epidemiological aspects of an outbreak of dengue fever/dengue haemorrhagic fever in Singapore. The Southeast Asian Journal Of Tropical Medicine And Public Health 1987;18:295-302.

Reason for exclusion: outbreak report

Gratz NG. Lessons of Aedes aegypti control in Thailand. Medical And Veterinary Entomology 1993;7:1-10.

Reason for exclusion: narrative review article, no original data

Gubler DJ. Aedes aegypti and Aedes aegypti-borne disease control in the 1990s: top down or bottom up. Am J Trop Med Hyg 1989;40:571-8.

Reason for exclusion: narrative review article, no original data

Gubler DJ, Clark GG. Community-based integrated control of Aedes aegypti: a brief overview of current programs. The American Journal of Tropical Medicine And Hygiene 1994;50:50-60.

Reason for exclusion: narrative review article, no original data

Gubler DJ, Clark GG. Community involvement in the control of Aedes aegypti. Acta Tropica 1996;61:169-79.

Reason for exclusion: narrative review article, no original data

Guzman MG, Halstead SB, Artsob H, Buchy P, Farrar J, Gubler DJ, et al. Dengue: a continuing global threat. Nature Reviews Microbiology 2010;8:S7-S16.

Reason for exclusion: narrative review article, no original data

Hairi F, Ong C-HS, Suhaimi A, Tsung T-W, bin Anis Ahmad MA, Sundaraj C, et al. A knowledge, attitude and practices (KAP) study on dengue among selected rural communities in the Kuala Kangsar district. Asia-Pacific Journal Of Public Health / Asia-Pacific Academic Consortium For Public Health 2003;15:37-43.

Reason for exclusion: cross-sectional KAP survey, no intervention

Halstead SB. Dengue in the health transition. Gaoxiong Yi Xue Ke Xue Za Zhi = The Kaohsiung Journal Of Medical Sciences 1994;10 Suppl:S2-S14.

Reason for exclusion: abstract only

Hanh TTT, Hill PS, Kay BH, Quy TM. Development of a framework for evaluating the sustainability of community-based dengue control projects. The American Journal of Tropical Medicine and Hygiene. 2009;80(2):312-8.

Reason for exclusion: Development of an assessment framework, results are duplicates of those in Kay 2010

Hien, Takano T, Seino K, Ohnishi M, Nakamura K. Effectiveness of a capacity-building program for community leaders in a healthy living environment: a randomized community-based intervention in rural Viet Nam. Health Promotion International 2008;23:354-64.

Reason for exclusion: evaluation of community healthworker program but not focused on dengue

Hotez PJ, Remme JHF, Buss P, Alleyne G, Morel C, Breman JG. Combating tropical infectious diseases: report of the Disease Control Priorities in Developing Countries Project. Clinical Infectious Diseases: An Official Publication Of The Infectious Diseases Society Of America 2004;38:871-8.

Reason for exclusion: narrative review article, no original data

Hsieh Y-H, Ma S. Intervention measures, turning point, and reproduction number for dengue, Singapore, 2005. The American Journal of Tropical Medicine And Hygiene. 2009;80(1):66-71.

Reason for exclusion: Model fitting exercise

Huy R, Buchy P, Conan A, Ngan C, Ong S, Ali R, et al. National dengue surveillance in Cambodia 1980-2008: epidemiological and virological trends and the impact of vector control. Bulletin of the World Health Organisation. 2010;88(9):650-7.

Reason for exclusion: No actual before and after data given

Jennings CD, Phommasack B, Sourignadeth B, Kay BH. Aedes aegypti control in the Lao People's Democratic Republic, with reference to copepods. The American Journal Of Tropical Medicine And Hygiene. 1995;53(4):324-30.

Laboratory study

Kantachuvessiri A. Dengue hemorrhagic fever in Thai society. The Southeast Asian Journal Of Tropical Medicine And Public Health 2002;33:56-62.

Reason for exclusion: descriptive study, no intervention

Kauffman KS, Myers DH. The changing role of village health volunteers in Northeast Thailand: an ethnographic field study. International Journal Of Nursing Studies 1997;34:249-55.

Reason for exclusion: evaluation of community healthworker program but not focused on dengue

Kay BH. Intersectoral approaches to dengue vector control. Gaoxiong Yi Xue Ke Xue Za Zhi = The Kaohsiung Journal Of Medical Sciences 1994;10 Suppl:S56-S61.

Reason for exclusion: Only abstract available

Kay B, Nam VS, Yen NT, Tien TV, Holynska M. Successful dengue vector control in Viet Nam: A model for regional consideration. Arbovirus Research Australia. 2001;8:187-93.

Reason for exclusion: Duplicate of results in Kay 2002 Am J Trop Med Hyg

Kenyon G. Scientists try new strategy to eradicate dengue fever. BMJ (Clinical Research Ed) 1999;318:555-.

Reason for exclusion: news article

Khun S, Manderson L. Community and school-based health education for dengue control in rural Cambodia: a process evaluation. Plos Neglected Tropical Diseases 2007;1:e143-e.

Reason for exclusion: no quantitative data (qualitative only)

Khun S, Manderson LH. Abate distribution and dengue control in rural Cambodia. Acta Tropica 2007;101:139-46.

Reason for exclusion: cross-sectional KAP survey, no intervention

Kittayapong P, Chansang U, Chansang C, Bhumiratana A. Community participation and appropriate technologies for dengue vector control at transmission foci in Thailand. Journal Of The American Mosquito Control Association. 2006;22(3):538-46.

Reason for exclusion: Same data/results as Kittayapong, 2008

Kittigul L, Suankeow K, Sujirarat D, Yoksan S. Dengue hemorrhagic fever: knowledge, attitude and practice in Ang Thong Province, Thailand. The Southeast Asian Journal Of Tropical Medicine And Public Health 2003;34:385-92.

Reason for exclusion: cross-sectional KAP survey, no intervention

Koenraadt CJM, Tuiten W, Sithiprasasna R, Kijchalao U, Jones JW, Scott TW. Dengue knowledge and practices and their impact on Aedes aegypti populations in Kamphaeng Phet, Thailand. The American Journal Of Tropical Medicine And Hygiene 2006;74:692-700.

Reason for exclusion: cross-sectional KAP survey, no intervention

Koh BKW, Ng LC, Kita Y, Tang CS, Ang LW, Wong KY, et al. The 2005 dengue epidemic in Singapore: epidemiology, prevention and control. Annals Of The Academy Of Medicine, Singapore 2008;37:538-45.

Reason for exclusion: outbreak report

Kumarasamy V. Dengue fever in Malaysia: time for review? The Medical Journal Of Malaysia 2006;61:1-3.

Reason for exclusion: editorial, no data

Lavery JV, Tinadana PO, Scott TW, Harrington LC, Ramsey JM, Ytuarte-Nuñez C, et al. Towards a framework for community engagement in global health research. Trends In Parasitology 2010;26:279-83.

Reason for exclusion: theoretical model for developing interventions, no evaluation data

Lines J, Harpham T, Leake C, Schofield C. Trends, priorities and policy directions in the control of vector-borne diseases in urban environments. Health Policy And Planning 1994;9:113-29.

Reason for exclusion: narrative review article, no original data

Luz PM, Vanni T, Medlock J, Paltiel AD, Galvani AP. Dengue vector control strategies in an urban setting: an economic modelling assessment. Lancet 2011;377:1673-80.

Reason for exclusion: Brazil, not SE Asia

Mahilum MM, Ludwig M, Madon MB, Becker N. Evaluation of the present dengue situation and control strategies against Aedes aegypti in Cebu City, Philippines. Journal Of Vector Ecology: Journal Of The Society For Vector Ecology. 2005;30(2):277-83.

Reason for exclusion: Only before/after data is for field trial, community study is cross-sectional data

Moodie R, Borthwick C, Galbally R. Health promotion in South-East Asia: Indonesia, DPR Korea, Thailand, the Maldives and Myanmar.* (*This paper was commissioned as part of a WHO-sponsored initiative utilizing a common presentation framework.). Health Promotion International 2000;15:249-.

Reason for exclusion: description of infrastructure for health promotion, no evaluation data

Nagao Y, Thavara U, Chitnumsup P, Tawatsin A, Chansang C, Campbell-Lendrum D. Climatic and social risk factors for Aedes infestation in rural Thailand. Tropical Medicine & International Health: TM & IH 2003;8:650-9.

Reason for exclusion: cross-sectional survey of risk factor, no intervention

Naing C, Ren W, Man C, Fern K, Qiqi C, Ning C, et al. Awareness of Dengue and Practice of Dengue Control Among the Semi-Urban Community: A Cross Sectional Survey. Journal of Community Health 2011;36:1044-9.

Reason for exclusion: cross-sectional KAP survey, no intervention

Nalongsack S, Yoshida Y, Morita S, Sosouphanh K, Sakamoto J. Knowledge, attitude and practice regarding dengue among people in Pakse, Laos. Nagoya Journal Of Medical Science 2009;71:29-37.

Reason for exclusion: cross-sectional KAP survey, no intervention

Nam VS, Yen NT, Holynska M, Reid JW, Kay BH. National progress in dengue vector control in Viet Nam: survey for Mesocyclops (Copepoda), Micronecta (Corixidae), and fish as biological control agents. The American Journal Of Tropical Medicine And Hygiene 2000;62:5-10.

Reason for exclusion: cross-sectional survey of vector control prevalence, no intervention. Not a before/after study, more a feasibility study

Nathan MB, Focks DA, Kroeger A. Pupal/demographic surveys to inform dengue-vector control. Annals Of Tropical Medicine And Parasitology 2006;100 Suppl 1:S1-S3.

Reason for exclusion: narrative review article, no original data

Nguyen LAP, Clements ACA, Jeffery JAL, Yen NT, Nam VS, Vaughan G, et al. Abundance and prevalence of Aedes aegypti immatures and relationships with household water storage in rural areas in southern Viet Nam. International Health 2011;3:115-25.

Reason for exclusion: cross-sectional survey of risk factor, no intervention

Nitatpattana N, Chaimarin A, Barbazan P. SILENT TRANSMISSION OF VIRUS DURING A DENGUE EPIDEMIC, NAKHON PATHOM PROVINCE, THAILAND 2001. Southeast Asian Journal of Tropical Medicine and Public Health 2006;37:899-903.

Reason for exclusion: clinical outbreak report

Ole S. Microbial control in Southeast Asia. Journal of Invertebrate Pathology 2007;95:168-74.

Reason for exclusion: narrative review article, no original data

Ong DQR, Sitaram N, Rajakulendran M, Koh GCH, Seow ALH, Ong ESL, et al. Knowledge and practice of household mosquito breeding control measures between a dengue hotspot and non-hotspot in Singapore. Annals Of The Academy Of Medicine, Singapore 2010;39:146-9.

Reason for exclusion: cross-sectional KAP survey, no intervention

Ooi E-E, Goh K-T, Gubler DJ. Dengue prevention and 35 years of vector control in Singapore. Emerging Infectious Diseases 2006;12:887-93.

Reason for exclusion: description of control activities but no evaluation data

Ortiz-Quesada F, Méndez-Galván JF, Ritchie-Dunham J, Rosado-Muñoz FJ. [Organizational decision making in health: the case of dengue]. Salud Pública De México 1995;37 Suppl:S77-S87.

Reason for exclusion: Mexico, not SE Asia

Paupy C, Delatte H, Bagny L, Corbel V, Fontenille D. Aedes albopictus, an arbovirus vector: from the darkness to the light. Microbes And Infection / Institut Pasteur 2009;11:1177-85.

narrative review article, no original data

Pérez D, Lefèvre P, Sánchez L, Van der Stuyft P. Comment on: What do community-based dengue control programs achieve? A systematic review of published evaluations. Transactions Of The Royal Society Of Tropical Medicine And Hygiene 2007;101:630-1.

Reason for exclusion: letter, no data

Phuanukoonnon S, Mueller I and Bryan JH. Effectiveness of dengue control practices in household water containers in northeast Thailand. Tropical Medicine and International Health. 2005;10:755-63.

Cross-sectional survey

Phuanukoonnon S, Brough M, Bryan JH. Folk knowledge about dengue mosquitoes and contributions of health belief model in dengue control promotion in Northeast Thailand. Acta Tropica 2006;99:6-14.

Reason for exclusion: cross-sectional KAP survey, no intervention

Phuong HL, De Vries PJ, Boonshuyar C, Binh TQ, Nam NV, Kager PA. Dengue risk factors and community participation in Binh Thuan Province, Viet Nam, a household survey. The Southeast Asian Journal Of Tropical Medicine And Public Health 2008;39:79-89.

Reason for exclusion: cross-sectional KAP survey, no intervention

Poovaneswari S, Lam SK. Problems in dengue control: a case study. The Southeast Asian Journal of Tropical Medicine and Public Health. 1992;23(4):723-5.

Reason for exclusion: No before/after data, surveillance not carried out prior to interventions

Pylypa J. Local perceptions of dengue fever in northeast Thailand and their implications for adherence to prevention campaigns. Anthropology & Medicine 2009;16:73-83.

Reason for exclusion: cross-sectional KAP survey, no intervention

Rodriguez-Tan RS, Weir MR. Dengue: a review. Texas Medicine 1998;94:53-9.

Reason for exclusion: narrative review article, no original data

Seleena P, Lee HL, Chiang YF. Thermal application of Bacillus thuringiensis serovar israelensis for dengue vector control. Journal Of Vector Ecology: Journal Of The Society For Vector Ecology 2001;26:110-3.

Reason for exclusion: field trial of intervention, not community based

Seng CM, Setha T, Nealon J, Socheat D, Chantha N, Nathan MB. Community-based use of the larvivorous fish Poecilia reticulate to control the dengue vector Aedes aegypti in domestic water storage containers in rural Cambodia. Journal of Vector Ecology: Journal of the Society for Vector Ecology. 2008; 33(1):139-44.

Reason for exclusion: No baseline data

Soedarmo SP. The epidemiology, prevention and control of dengue hemorrhagic fever in Indonesia. Gaoxiong Yi Xue Ke Xue Za Zhi = The Kaohsiung Journal Of Medical Sciences 1994;10 Suppl:S109-S12.

Reason for exclusion: abstract only

Sørensen E. [Dengue hemorrhagic fever. Experiences from Thailand]. Tidsskrift For Den Norske Lægeforening: Tidsskrift For Praktisk Medicin, Ny Række 1992;112:2194-5.

Reason for exclusion: narrative review article, no original data

Sorge F. [Prevention with repellent in children]. Archives De Pédiatrie: Organe Officiel De La Sociéte Française De Pédiatrie 2009;16 Suppl 2:S115-S22.

Reason for exclusion: clinical review, no intervention

Spiegel J, Bennett S, Hattersley L, Hayden MH, Kittayapong P, Nalim S, et al. Barriers and Bridges to Prevention and Control of Dengue: The Need for a Social-Ecological Approach. Ecohealth 2005;2:273-90.

Reason for exclusion: narrative review article, no original data

Sulaiman S, Pawanchee ZA, Othman HF. Field evaluation of cypermethrin and cyfluthrin against dengue vectors in a housing estate in Malaysia. Journal Of Vector Ecology: Journal Of The Society For Vector Ecology. 2002;27:230-4.

Reason for exclusion: Field trial – no community component

Susan B R. Paradigms lost: Toward a new understanding of community participation in health programs. Acta Tropica 1996;61:79-92.

Reason for exclusion: review of community participation, no dengue intervention

Sutton RN. Why bother with arboviruses? The Journal Of Infection 1985;11:99-102.

Reason for exclusion: narrative review article, no original data

Suwanbamrung C, Nukan N, Sripon S, Somrongthong R, Singchagchai P. Community capacity for sustainable community-based dengue prevention and control: study of a sub-district in Southern Thailand. Asian Pacific Journal of Tropical Medicine 2010;3:215-9.

Reason for exclusion: cross-sectional KAP survey, no intervention

Suwannapong N, Howteerakul N, Pacheun O. Strengthening the capability of family health leaders for sustainable community-based health promotion. Southeast Asian Journal of Tropical Medicine and Public Health 2005;36:1039-47.

Reason for exclusion: review of community participation, no dengue intervention

Swaddiwudhipong W, Lerdlukanavonge P, Khumklam P, Koonchote S, Nguntra P, Chaovakiratipong C. A survey of knowledge, attitude and practice of the prevention of dengue hemorrhagic fever in an urban community of Thailand. The Southeast Asian Journal Of Tropical Medicine And Public Health 1992;23:207-11.

Reason for exclusion: cross-sectional KAP survey, no intervention

Sweeney AW. Prospects for control of mosquito-borne diseases. Journal Of Medical Microbiology 1999;48:879-81.

Reason for exclusion: narrative review article, no original data

Thomas SJ, Strickman D, Vaughn DW. Dengue epidemiology: virus epidemiology, ecology, and emergence. Advances In Virus Research 2003;61:235-89.

Reason for exclusion: narrative review article, no original data

Thompson F, Caltabiano ML. The Health Belief Model and dengue fever preventative behaviours: a pilot program. International Journal of Health Promotion & Education 2010;48:9-19.

Reason for exclusion: Australia, not SE Asia

Tien NTK, Ha DQ, Tien TK, Quang LC. Predictive indicators for forecasting epidemic of dengue/dengue haemorrhagic fever through epidemiological, virological and entomological surveillance. Dengue bulletin. 1999;23.

Reason for exclusion: No evaluation or intervention, more a descriptive analysis of surveillance data

Tourdjman M, Rekol H, Sirenda V. Evaluation of the dengue surveillance system in Cambodia. National Dengue Control Program. Phnom Penh: Ministry of Health Cambodia and Institute Pasteur du Cambodge 2005.

Reason for exclusion: Library unable to locate report

Tran HP, Adams J, Jeffery JAL, Nguyen YT, Vu NS, Kutcher SC, et al. Householder perspectives and preferences on water storage and use, with reference to dengue, in the Mekong Delta, southern Viet Nam. International Health 2010;2:136-42.

Reason for exclusion: cross-sectional KAP survey, no intervention

Tsuzuki A, Huynh T, Tsunoda T, Luu L, Kawada H, Takagi M. Effect of existing practices on reducing Aedes aegypti pre-adults in key breeding containers in Ho Chi Minh City, Viet Nam. The American Journal Of Tropical Medicine And Hygiene 2009;80:752-7.

Reason for exclusion: cross-sectional KAP survey, no intervention

Tsuzuki A, Thiem VD, Suzuki M, Yanai H, Matsubayashi T, Yoshida L-M, et al. Can daytime use of bed nets not treated with insecticide reduce the risk of dengue hemorrhagic fever among children in Viet Nam? The American Journal Of Tropical Medicine And Hygiene 2010;82:1157-9.

Reason for exclusion: cross-sectional survey of risk factor, no intervention

Uma Deavi A, Gan Chong Y, Ooi Guat S. A knowledge attitude and practice (KAP) study on dengue/dengue haemorrhagic fever and the Aedes mosquitoes. The Medical Journal Of Malaysia 1986;41:108-15.

Reason for exclusion: cross-sectional KAP survey, no intervention

Van Benthem BHB, Khantikul N, Panart K, Kessels PJ, Somboon P, Oskam L. Knowledge and use of prevention measures related to dengue in northern Thailand. Tropical Medicine & International Health: TM & IH 2002;7:993-1000.

Reason for exclusion: cross-sectional KAP survey, no intervention

Vong S, Khieu V, Glass O, Ly S, Duong V, Huy R, et al. Dengue incidence in urban and rural Cambodia: results from population-based active fever surveillance, 2006-2008. Plos Neglected Tropical Diseases 2010;4:e903-e.

Reason for exclusion: report of surveillance activity – no evaluation

Wallace HG, Lim TW, Rudnick A, Knudsen AB, Cheong WH, Chew V. Dengue hemorrhagic fever in Malaysia: the 1973 epidemic. The Southeast Asian Journal Of Tropical Medicine And Public Health 1980;11:1-13.

Reason for exclusion: outbreak report

Wang NC. Control of dengue vectors in Singapore. Gaoxiong Yi Xue Ke Xue Za Zhi = The Kaohsiung Journal Of Medical Sciences 1994;10 Suppl:S33-S8.

Reason for exclusion: abstract only

Wilder-Smith A, Lover A, Kittayapong P, Burnham G. Hypothesis: Impregnated school uniforms reduce the incidence of dengue infections in school children. Medical Hypotheses 2011;76:861-2.

Reason for exclusion: hypothesis only, no data

Winch PJ, Lloyd LS, Hoemeke L, Leontsini E. Vector control at the household level: an analysis of its impact on women. Acta Tropica 1994;56:327-39.

Reason for exclusion: discussion piece, no data

Wong HB. Dengue haemorrhagic fever in Singapore—the future. The Journal Of The Singapore Paediatric Society 1986;28:210-5.

Reason for exclusion: narrative review article, no original data

Wongbutdee J, Chaikoolvatana A, Saengnill W, Krasuaythong N, Phuphak S. Geo-database use to promote dengue infection prevention and control. The Southeast Asian Journal Of Tropical Medicine And Public Health 2010;41:841-57.

Reason for exclusion: cross-sectional KAP survey, no intervention

Yap HH, Chong NL, Foo AE, Lee CY. Dengue vector control: present status and future prospects. Gaoxiong Yi Xue Ke Xue Za Zhi = The Kaohsiung Journal Of Medical Sciences 1994;10 Suppl:S102-S8.

Reason for exclusion: abstract only

Yoshimura N, Jimba M, Poudel KC, Chanthavisouk, Iwamoto A, Phommasack, et al. Health promoting schools in urban, semi-urban and rural Lao PDR. Health Promotion International 2009;24:166-76.

Reason for exclusion: review of community participation, no dengue intervention

Dengue – systematic reviews

Al-Muhandis N, Hunter PR. The value of educational messages embedded in a community-based approach to combat dengue Fever: a systematic review and meta regression analysis. Plos Neglected Tropical Diseases 2011;5:e1278-e.

Reason for exclusion: Not specific to SE Asia

Anderson KB, Chunsuttiwat S, Nisalak A, Mammen MP, Libraty DH, Rothman AL, Green S, Vaughn DW, Ennis FA, Endy TP. Burden of symptomatic dengue infection in children at primary school in Thailand: a prospective study (Provisional abstract) Centre for Reviews and Dissemination. 2007

Reason for exclusion: Cross-sectional study, no intervention

Ballenger-Browning KK, Elder JP. Multi-modal Aedes aegypti mosquito reduction interventions and dengue fever prevention. Tropical Medicine & International Health: TM & IH 2009;14:1542-51.

Reason for exclusion: Not specific to SE Asia

Blacksell SD, Doust JA, Newton PN, Peacock SJ, Day NP, Dondorp AM. A systematic review and meta-analysis of the diagnostic accuracy of rapid immunochromatographic assays for the detection of dengue virus IgM antibodies during acute infection (Structured abstract) Centre for Reviews and Dissemination. 2006

Reason for exclusion: Review of diagnostic method, not an intervention

Erlanger TE, Keiser J, Utzinger J. Effect of dengue vector control interventions on entomological parameters in developing countries: a systematic review and meta-analysis. Medical And Veterinary Entomology 2008;22:203-21.

Reason for exclusion: Not specific to SE Asia

Esu E, Lenhart A, Smith L, Horstick O. Effectiveness of peridomestic space spraying with insecticide on dengue transmission; systematic review. Tropical Medicine & International Health: TM & IH 2010;15:619-31.

Reason for exclusion: Not community based, evaluation of national dengue control programs

Heintze C, Garrido MV, Kroeger A. What do community-based dengue control programs achieve? A systematic review of published evaluations. Transactions Of The Royal Society Of Tropical Medicine And Hygiene 2007;101:317-25.

Reason for exclusion: Not specific to SE Asia

Horstick O, Runge-Ranzinger S, Nathan MB, Kroeger A. Dengue vector-control services: how do they work? A systematic literature review and country case studies. Transactions Of The Royal Society Of Tropical Medicine And Hygiene 2010;104:379-86.

Reason for exclusion: Not specific to SE Asia

Lee, DL Connor, SB Kitchen, KM Bacon, M Shah, ST Brown, RR Bailey, Y Laosiritaworn, DS Burke, DA Cummings. Economic value of dengue vaccine in Thailand (Provisional abstract). Centre for Reviews and Dissemination. 2011

Reason for exclusion: Pharmaceutical intervention

Luz PM, Vanni T, Medlock J, Paltiel AM, Galvani AP. Dengue vector control strategies in an urban setting: an economic modelling assessment (Provisional abstract) Centre for Reviews and Dissemination. 2011

Reason for exclusion: Model-based

Runge-Ranzinger S, Horstick O, Marx M, Kroeger A. What does dengue disease surveillance contribute to predicting and detecting outbreaks and describing trends? Tropical Medicine and International Health 2008;13:1022-41.

Reason for exclusion: Not specific to SE Asia

Schiøler KL, McCarty CW. Vaccines for preventing dengue infection. Cochrane Library of Systematic Reviews. January 2009

Pharmaceutical intervention

Shepard DS, Suaya JA, Halstead SB, Nathan MB, Gubler DJ, RTMahoney, Wang DN, Meltzer MI. Cost-effectiveness of a pediatric dengue vaccine (Structured abstract) Centre for Reviews and Dissemination. 2004

Reason for exclusion: Pharmaceutical intervention

Squizzato A, Ageno W. Recombinant activated factor VII as a general haemostatic agent: evidence-based efficacy and safety (Provisional abstract) Centre for Reviews and Dissemination. 2007

Reason for exclusion: Pharmaceutical intervention

Panpanich R, Sornchai P, Kanjanaratanakorn K. Corticosteroids for the treatment of dengue shock syndrome. Cochrane Library of Systematic Reviews. February 2010

Reason for exclusion: Pharmaceutical intervention

SARS – articles

Anonymous. SARS timeline of an epidemic: special report. Canadian Journal of Infection Control. 2003; 18(2):51-5.

Reason for exclusion: Review article

Anonymous. Singapore distributes wrist tags to track SARS patients. Medical Letter on the CDC. 2003:37-8.

Reason for exclusion: News article

Singapore eases its anti-SARS measures. Medical Letter on the CDC. 2003:30-

Reason for exclusion: News article

Singapore removed from World Health Organization list of SARS-affected areas. Medical Letter on the CDC. 2003:37-.

Reason for exclusion: News article

Mask or tissues? SARS ushers in an age of ‘respiratory etiquette’: some see universal masking as unworkable. Hospital Infection Control. 2003;30(11):137-41.

Reason for exclusion: Editorial on workability of masks

Anderson RM, Fraser C, Ghani AC, Donnelly CA, Riley S, Ferguson NM, et al. Epidemiology, transmission dynamics and control of SARS: the 2002-2003 epidemic. Philosophical Transactions Of The Royal Society Of London Series B, Biological Sciences. 2004;359(1447):1091-105.

Reason for exclusion: Model-based study

Arguin PM, Navin AW, Steele SF, Weld LH, Kozarsky PE. Health communication during SARS. Emerging Infectious Diseases. 2004;10(2):377-80.

Reason for exclusion: Description of CDC web visits during outbreak

Arita I, Kojima K, Nakane M. Transmission of severe acute respiratory syndrome. Emerging Infectious Diseases. 2003;9(9):1183-4.

Reason for exclusion: Letter only, no data

Bauch CT, Lloyd-Smith JO, Coffee MP, Galvani AP. Dynamically modeling SARS and other newly emerging respiratory illnesses: past, present, and future. Epidemiology (Cambridge, Mass). 2005;16(6):791-801.

Reason for exclusion: Model based

Becker NG, Glass K, Li Z, Aldis GK. Controlling emerging infectious diseases like SARS. Mathematical Biosciences. 2005;193(2):205-21.

Reason for exclusion: Model-based study

Bell DM. Public health interventions and SARS spread, 2003. Emerging Infectious Diseases. 2004;10(11):1900-6.

Reason for exclusion: Review article, no data

Bellika JG, Hasvold T, Hartvigsen G. Propagation of program control: A tool for distributed disease surveillance. International Journal Of Medical Informatics. 2007;76(4):313-29.

Reason for exclusion: Development of surveillance system in Norway

Bener A, Al-Khal A. Knowledge, attitude and practice towards SARS. Journal of the Royal Society for the Promotion of Health. 2004;124(4):167-70.

Reason for exclusion: Qatar KAP survey

Bensimon CM, Upshur REG. Evidence and effectiveness in decision making for quarantine. American Journal Of Public Health. 2007;97:S44-8.

Reason for exclusion: Ethics discussion piece

Cai Q-c, Jiang Q-w, Xu Q-f, Cheng X, Guo Q, Sun Q-w, et al. [To develop a model on severe acute respiratory syndrome epidemics to quantitatively evaluate the effectiveness of intervention measures]. Zhonghua Liu Xing Bing Xue Za Zhi = Zhonghua Liuxingbingxue Zazhi. 2005;26(3):153-8.

Reason for exclusion: Model-based

Castillo-Salgado C. Trends and directions of global public health surveillance. Epidemiologic Reviews. 2010;32(1):93-109.

Narrative review

Cava MA, Fay KE, Beanlands HJ, McCay EA, Wignall R. Risk perception and compliance with quarantine during the SARS outbreak. Journal Of Nursing Scholarship: An Official Publication Of Sigma Theta Tau International Honor Society Of Nursing / Sigma Theta Tau. 2005;37(4):343-7.

Reason for exclusion: Setting is Toronto

CDC, Outbreak of severe acute respiratory syndrome—worldwide, 2003. MMWR Morbidity and Mortality Weekly Report. 2003;52:226-8.

Reason for exclusion: Outbreak update

CDC, Preliminary clinical description of severe acute respiratory syndrome. MMWR Morbidity and Mortality Weekly Report. 2003;52(12):255-6.

Reason for exclusion: Clinical description

CDC. Severe acute respiratory syndrome—Singapore, 2003. MMWR Morbidity and Mortality Weekly Report. 2003;52(18):405-11.

Reason for exclusion: Case study report on superspreaders

Chan GCT, Koh D. Reviewing lessons learnt of SARS in Singapore during planning for influenza pandemic. International Maritime Health. 2006;57(1-4):163-76.

Reason for exclusion: Lists interventions, but no estimate of effectiveness

Chan EA, Chung JWY, Wong TKS. Learning from the severe acute respiratory syndrome (SARS) epidemic. Journal Of Clinical Nursing. 2008;17(8):1023-34.

Reason for exclusion: Handwashing study in Hong Kong nurses

Chan-Yeung M, Ooi GC, Hui DS, Ho PL, Tsang KW. Severe acute respiratory syndrome. The International Journal Of Tuberculosis And Lung Disease: The Official Journal Of The International Union Against Tuberculosis And Lung Disease. 2003;7(12):1117-30.

Reason for exclusion: Review article, no data

Chan-Yeung M, Xu R-H. SARS: epidemiology. Respirology (Carlton, Vic). 2003;8 Suppl:S9-S14.

Reason for exclusion: Review article, no data

Chen Y-C, Chang S-C, Tsai K-S, Lin F-Y. Certainties and Uncertainties Facing Emerging Respiratory Infectious Diseases: Lessons from SARS. Journal of the Formosan Medical Association. 2008;107(6):432-42.

Reason for exclusion: Review article, clinician focussed

Cherry JD. The chronology of the 2002-2003 SARS mini pandemic. Paediatric Respiratory Reviews. 2004;5(4):262-9.

Review article, no data

Chng SY, Chia F, Leong KK, Kwang YP, Ma S, Lee BW, et al. Mandatory temperature monitoring in schools during SARS. Archives Of Disease In Childhood. 2004;89(8):738-9.

Reason for exclusion: No intervention

Chowell G, Castillo-Chavez C, Fenimore PW, Kribs-Zaleta CM, Arriola L, Hyman JM. Model parameters and outbreak control for SARS. Emerging Infectious Diseases. 2004;10(7):1258-63.

Reason for exclusion: Model-based

Chu C-M, Cheng VCC, Hung IFN, Chan K-S, Tang BSF, Tsang THF, et al. Viral load distribution in SARS outbreak. Emerging Infectious Diseases. 2005;11(12):1882-6.

Reason for exclusion: Epidemiological study, no intervention

Cleri DJ, Ricketti AJ, Vernaleo JR. Severe Acute Respiratory Syndrome (SARS). Infectious Disease Clinics Of North America. 2010;24(1):175-202.

Reason for exclusion: Clinical review, no data

Cooke FJ, Shapiro DS. Global outbreak of severe acute respiratory syndrome (SARS). International Journal Of Infectious Diseases: IJID: Official Publication Of The International Society For Infectious Diseases. 2003;7(2):80-5.

Reason for exclusion: Review only, no data

Deng JF, Olowokure B, Kaydos-Daniels SC, Chang HJ, Barwick RS, Lee ML, et al. Severe acute respiratory syndrome (SARS): knowledge, attitudes, practices and sources of information among physicians answering a SARS fever hotline service. Public Health. 2006;120(1):15-9.

Reason for exclusion: KAP survey, Taiwanese doctors

Deurenberg-Yap M, Foo LL, Low YY, Chan SP, Vijaya K, Lee M. The Singaporean response to the SARS outbreak: knowledge sufficiency versus public trust. Health Promotion International. 2005;20(4):320-6.

Reason for exclusion: KAP survey

Diamond B. SARS spreads new outlook on quarantine models. Nature Medicine. 2003;9(12):1441-.

Reason for exclusion: News article

Donnelly CA, Fisher MC, Fraser C, Ghani AC, Riley S, Ferguson NM, et al. Epidemiological and genetic analysis of severe acute respiratory syndrome. The Lancet Infectious Diseases. 2004;4(11):672-83.

Reason for exclusion: Review of model-based studies

Drake JM, Chew SK, Ma S. Societal learning in epidemics: intervention effectiveness during the 2003 SARS outbreak in Singapore. Plos One. 2006;1:e20-e.

Reason for exclusion: Model-based

Dye C, Gay N. Epidemiology. Modeling the SARS epidemic. Science (New York, NY). 2003;300(5627):1884-5.

Reason for exclusion: Overview of modelling studies

Franas A, Plusa T. [Severe acute respiratory syndrome]. Polski Merkuriusz Lekarski: Organ Polskiego Towarzystwa Lekarskiego. 2006;21(123):205-10.

Reason for exclusion: Clinical review

Fu Y-C, Chen M-Y, Feng H-C. [The community health team: roles and responsibilities in infection control]. Hu Li Za Zhi The Journal Of Nursing. 2011;58(4):21-7.

Reason for exclusion: Setting is Taiwan

Gensini GF, Yacoub MH, Conti AA. The concept of quarantine in history: from plague to SARS. The Journal Of Infection. 2004;49(4):257-61.

Reason for exclusion: Review of quarantine practice and ethics

Greaves F. What are the most appropriate methods of surveillance for monitoring an emerging respiratory infection such as SARS? Journal of Public Health. 2004;26(3):288-92.

Reason for exclusion: Focussed on US, Australia, Europe

Gumel AB, Ruan S, Day T, Watmough J, Brauer F, van den Driessche P, et al. Modelling strategies for controlling SARS outbreaks. Proceedings Biological Sciences / The Royal Society. 2004;271(1554):2223-32.

Reason for exclusion: Model-based

Han B, Leong T-Y. We did the right thing: an intervention analysis approach to modeling intervened SARS propagation in Singapore. Studies in Health Technology And Informatics. 2004;107(Pt 2):1246-50.

Reason for exclusion: Model-based interventional analysis

Handel A, Longini IM, Jr., Antia R. What is the best control strategy for multiple infectious disease outbreaks? Proceedings Biological Sciences / The Royal Society. 2007;274(1611):833-7.

Reason for exclusion: Model-based

Hazreen AM, Myint Myint S, Farizah H, Abd Rashid M, Chai CC, Dymna VK, et al. An evaluation of information dissemination during the severe acute respiratory syndrome (SARS) outbreak among selected rural communities in Kuala Kangsar. The Medical Journal Of Malaysia. 2005;60(2):180-7.

Reason for exclusion: KAP survey

Healy M. Are quarantines back?… SARS. Los Angeles Times – Southern California Edition (Front Page). 2003:F1.

Reason for exclusion: News article

Henley E. SARS: lessons learned thus far. The Journal Of Family Practice. 2003;52(7):528-30.

Reason for exclusion: Review article, US focussed

Hui DSC, Chan PKS. Severe Acute Respiratory Syndrome and Coronavirus. Infectious Disease Clinics Of North America. 2010;24(3):619-38.

Reason for exclusion: Clinical presentation

James L, Shindo N, Cutter J, Ma S, Chew SK. Public health measures implemented during the SARS outbreak in Singapore, 2003. Public Health. 2006;120(1):20-6.

Reason for exclusion: Narrative of control measures. No data

Klinkenberg D, Fraser C, Heesterbeek H. The effectiveness of contact tracing in emerging epidemics. Plos One. 2006;1:e12-e.

Reason for exclusion: Model-based

Knight-Jones TJD, Hauser R, Matthes D, Stärk KDC. Evaluation of effectiveness and efficiency of wild bird surveillance for avian influenza. Veterinary Research. 2010;41(4):50-.

Reason for exclusion: Setting is Europe, not Southeast Asia

Krumkamp R, Duerr H-P, Reintjes R, Ahmad A, Kassen A, Eichner M. Impact of public health interventions in controlling the spread of SARS: Modelling of intervention scenarios. International Journal Of Hygiene And Environmental Health. 2009;212(1):67-75.

Reason for exclusion: Model-based

Kullberg BJ, Voss A. Severe acute respiratory syndrome: lessons and uncertainties. The Netherlands Journal Of Medicine. 2003;61(7):235-7.

Reason for exclusion: Review article

Lam WK, Zhong NS, Tan WC. Overview on SARS in Asia and the world. Respirology (Carlton, Vic). 2003;8 Suppl:S2-S5.

Reason for exclusion: Review article

Lateef F, Lim SH, Tan EH. New paradigm for protection: the emergency ambulance services in the time of severe acute respiratory syndrome. Prehospital Emergency Care: Official Journal Of The National Association Of EMS Physicians And The National Association Of State EMS Directors. 2004;8(3):304-7.

Reason for exclusion: Descriptive, no data

Lee N, Rainer TH, Ip M, Zee B, Ng MH, Antonio GE, et al. Role of laboratory variables in differentiating SARS-coronavirus from other causes of community-acquired pneumonia within the first 72 h of hospitalization. European Journal Of Clinical Microbiology & Infectious Diseases: Official Publication Of The European Society Of Clinical Microbiology. 2006;25(12):765-72.

Reason for exclusion: Clinical diagnosis

Lee PJ, Krilov LR. When animal viruses attack: SARS and avian influenza. Pediatric Annals. 2005;34(1):42-52.

Reason for exclusion: Review, clinical

Leidner DE, Pan G, Pan SL. The role of IT in crisis response: Lessons from the SARS and Asian Tsunami disasters. The Journal of Strategic Information Systems. 2009;18(2):80-99.

Reason for exclusion: No quantitative data

Leo YS, Chen M, Heng BH, Lee CC. Severe Acute Respiratory Syndrome—Singapore, 2003. JAMA. 2003;289(24):3231-4.

Reason for exclusion: No data

Leung GM, Quah S, Ho L, Ho S, Hedley AJ, Lee H, et al. A tale of two cities: community psychobehavioral surveillance and related impact on outbreak control in Hong Kong and Singapore during the severe acute respiratory syndrome epidemic. Infection Control & Hospital Epidemiology. 2004;25(12):1033-41.

Reason for exclusion: KAP survey

Liang W-N, Liu M, Chen Q, Liu Z-J, He X, Xie X-Q. Specific features of the contact history of probable cases of severe acute respiratory syndrome. Biomedical And Environmental Sciences: BES. 2005;18(2):71-6.

Reason for exclusion: Epidemiology, not control

Liao CM, Chen SC, Chang CF. Modelling respiratory infection control measure effects. Epidemiology And Infection. 2008;136(3):299-308.

Reason for exclusion: Hypothetical model

Lipsitch M, Cohen T, Cooper B, Robins JM, Ma S, James L, et al. Transmission dynamics and control of severe acute respiratory syndrome. Science (New York, NY). 2003;300(5627):1966-70.

Reason for exclusion: Model-based

Lloyd-Smith JO, Galvani AP, Getz WM. Curtailing transmission of severe acute respiratory syndrome within a community and its hospital. Proceedings Biological Sciences / The Royal Society. 2003;270(1528):1979-89.

Reason for exclusion: Model-based

Losavio K. SARS alert. JEMS: Journal of Emergency Medical Services. 2003;28(5):26-.

Reason for exclusion: Setting is US/Canada

Mackay B. SARS: “a domino effect through entire system”. CMAJ: Canadian Medical Association Journal = Journal De L'association Medicale Canadienne. 2003;168(10):1308-.

Reason for exclusion: News article

Menon KU. SARS revisited: managing “outbreaks” with “communications”. Annals of the Academy of Medicine, Singapore. 2006;35(5):361-7.

Reason for exclusion: No data provided, descriptive only

Mubayi A, Zaleta CK, Martcheva M, Castillo-Chávez C. A cost-based comparison of quarantine strategies for new emerging diseases. Mathematical Biosciences and Engineering: MBE. 2010;7(3):687-717.

Reason for exclusion: Model-based

Murphy C. The 2003 SARS outbreak: global challenges and innovative infection control measures. Online Journal Of Issues In Nursing. 2006;11(1):6-.

Reason for exclusion: Review article, no data

Ng EYK. Is thermal scanner losing its bite in mass screening of fever due to SARS? Medical Physics. 2005;32(1):93-7.

Sensitivity/specificity of thermal scanners

Nishiura H, Kuratsuji T, Quy T. Rapid awareness and transmission of severe acute respiratory syndrome in Hanoi French Hospital, Viet Nam. American Journal of Tropical Medicine and Hygiene. 2005;73:17-25.

Reason for exclusion: Focus on nosocomial transmission, not community

Ong EHM. War on SARS: a Singapore experience. Canadian Journal of Emergency Medicine. 2004;6(1):31-7.

Reason for exclusion: Review article, no data

Ooi GL, Phua KH. SARS in Singapore-challenges of a global health threat to local institutions. Natural Hazards. 2009;48(3):317-27.

Reason for exclusion: Review article, no data

Parashar UD, Anderson LJ. Severe acute respiratory syndrome: review and lessons of the 2003 outbreak. International Journal Of Epidemiology. 2004;33(4):628-34.

Reason for exclusion: Review article, no data

Peiris JSM, Yuen KY, Osterhaus ADME, Stöhr K. The severe acute respiratory syndrome. The New England Journal Of Medicine. 2003;349(25):2431-41.

Reason for exclusion: Review article, no data

Quah SR, Hin-Peng L. Crisis prevention and management during SARS outbreak, Singapore. Emerging Infectious Diseases. 2004;10(2):364-8.

Reason for exclusion: KAP survey

Scott RD, 2nd, Gregg E, Meltzer MI. Collecting data to assess SARS interventions. Emerging Infectious Diseases. 2004;10(7):1290-2.

Reason for exclusion: Description of minimum datasets for evaluation of SARS outcomes

Seng SL, Lim PS, Ng MY, Wong HB, Emmanuel SC. A study on SARS awareness and health-seeking behaviour – findings from a sampled population attending National Healthcare Group Polyclinics. Annals Of The Academy Of Medicine, Singapore. 2004;33(5):623–9.

Reason for exclusion: KAP survey

Shamian J, Petch T, Lilius F, Shainblum E, Talosi R. What's the plan? The unique challenges facing the home and community care sector in preparing for a pandemic. Healthcare Papers. 2007;8(1):38-42.

Reason for exclusion: Discussion of home care in Canada

Slaughter L, Keselman A, Kushniruk A, Patel VL. A framework for capturing the interactions between laypersons' understanding of disease, information gathering behaviors, and actions taken during an epidemic. Journal of Biomedical Informatics. 2005;38(4):298-313.

Reason for exclusion: No quantitative data, setting Hong Kong, Taiwan and Canada

Speakman J, González-Martín F, Perez T. Quarantine in severe acute respiratory syndrome (SARS) and other emerging infectious diseases. The Journal Of Law, Medicine & Ethics: A Journal Of The American Society Of Law, Medicine & Ethics. 2003 2003 Winter;31(4 Suppl):63-4.

Reason for exclusion: Debate on ethics of quarantine

Tambyah PA. SARS: responding to an unknown virus. European Journal Of Clinical Microbiology & Infectious Diseases: Official Publication Of The European Society Of Clinical Microbiology. 2004;23(8):589-95.

Review, mainly focussed on hospital health care workers

Tambyah PA. SARS: two years on. Singapore Medical Journal. 2005;46(4):150-2.

Reason for exclusion: Editorial

Tan B-H, Leo Y-S, Chew S-K. Lessons from the SARS crisis—more relevant than ever. Annals Of The Academy Of Medicine, Singapore. 2006;35(5):299-300.

Reason for exclusion: Editorial

Tan C-C. SARS in Singapore—key lessons from an epidemic. Annals of The Academy Of Medicine, Singapore. 2006;35(5):345-9.

Reason for exclusion: Descriptive only

Tan CC. SARS in Singapore: looking back, looking forward. Annals of The Academy Of Medicine, Singapore. 2003;32(5 Suppl):S4-S5.

Reason for exclusion: Review article, no data

Tan NC, Goh LG, Lee SS. Family physicians' experiences, behaviour, and use of personal protection equipment during the SARS outbreak in Singapore: do they fit the Becker Health Belief Model? Asia-Pacific Journal Of Public Health / Asia-Pacific Academic Consortium For Public Health. 2006;18(3):49-56.

Reason for exclusion: Health care worker use of masks

Tan Y-M, Chow PKH, Soo K-C. Severe acute respiratory syndrome: clinical outcome after inpatient outbreak of SARS in Singapore. BMJ (Clinical Research Ed). 2003;326(7403):1394-.

Reason for exclusion: Clinical case report

Teleman M, Boudville IC, Heng BH, Zhu D, Leo YS. Factors associated with transmission of severe acute respiratory syndrome among health-care workers in Singapore. Epidemiology and Infection. 2004;132:797-803.

Reason for exclusion: Focus on nosocomial transmission, not community

Teo P, Yeoh BSA, Ong SN. SARS in Singapore: surveillance strategies in a globalising city. Health Policy (Amsterdam, Netherlands). 2005;72(3):279-91.

Reason for exclusion: Attitudes to surveillance and KAP survey

Tsang T, T Lai-Yin LP-Y, Lee M. Update: Outbreak of severe acute respiratory syndrome-worldwide, 2003. JAMA. 2003;289(15):1918-20.

Reason for exclusion: Outbreak update report, no data

Van Bever P, Hia CPP, Quek SC. Childhood SARS in Singapore. Archives Of Disease In Childhood. 2003;88(8):742-.

Reason for exclusion: Clinical comment

Vu TH, Cabau J-F, Nguyen NT, Lenoir M. SARS in Northern Viet Nam. The New England Journal Of Medicine. 2003;348(20):2035-.

Reason for exclusion: Clinical report

Wallinga J, Teunis P. Different epidemic curves for severe acute respiratory syndrome reveal similar impacts of control measures. American Journal Of Epidemiology. 2004;160(6):509-16.

Reason for exclusion: Model-based

Wang M, Jolly AM. Changing virulence of the SARS virus: the epidemiological evidence. Bulletin Of The World Health Organization. 2004;82(7):547-8.

Reason for exclusion: Review of epidemiology

WHO, Global surveillance for severe acute respiratory syndrome (SARS). Relevé Épidémiologique Hebdomadaire / Section D'hygiène Du Secrétariat De La Société Des Nations = Weekly Epidemiological Record / Health Section Of The Secretariat Of The League Of Nations. 2003;78(14):100-19.

Reason for exclusion: Surveillance definitions and reporting standards

WHO, Severe acute respiratory syndrome (SARS): over 100 days into the outbreak Relevé Épidémiologique Hebdomadaire / Section D'hygiène Du Secrétariat De La Société Des Nations = Weekly Epidemiological Record / Health Section Of The Secretariat Of The League Of Nations. 2003;78:217-20.

Reason for exclusion: Review article, no data

WHO, SARS outbreak in the Philippines. Relevé Épidémiologique Hebdomadaire / Section D'hygiène Du Secrétariat De La Société Des Nations = Weekly Epidemiological Record / Health Section Of The Secretariat Of The League Of Nations. 2003;78(14):189-92.

Reason for exclusion: Outbreak report

Wilder-Smith A, Goh KT, Paton NI. Experience of severe acute respiratory syndrome in Singapore: importation of cases, and defense strategies at the airport. Journal Of Travel Medicine. 2003;10(5):259-62.

Reason for exclusion: No intervention

Wong ML, Koh D, Iyer P, Seow A, Goh LG, Chia SE, et al. Online health education on SARS to university students during the SARS outbreak. International Electronic Journal of Health Education. 2005;8:1-13.

Reason for exclusion: KAP survey

Yu ITS, Li Y, Wong TW, Tam W, Chan AT, Lee JHW, et al. Evidence of airborne transmission of the severe acute respiratory syndrome virus. The New England Journal Of Medicine. 2004;350(17):1731-9.

Reason for exclusion: Model of transmission during 2003 outbreak

Zapp R, Krajden M, Lynch T. SARS: a quality management test of our public health safety net. Quality Management in Health Care. 2004;13(2):120-9.

Reason for exclusion: Setting is Canada

Zhang S-x, Jiang L-j, Zhang Q-w, Pan J-j, Wang W-y. [Role of mass media during the severe acute respiratory syndrome epidemic]. Zhonghua Liu Xing Bing Xue Za Zhi = Zhonghua Liuxingbingxue Zazhi. 2004;25(5):403-6.

Reason for exclusion: Impact on knowledge, China

Zhong N-S, Wong GWK. Epidemiology of severe acute respiratory syndrome (SARS): adults and children. Paediatric Respiratory Reviews. 2004;5(4):270-4.

Reason for exclusion: Review of 2003 global events

Zhou Z-X, Jiang C-Q. [Effect of environment and occupational hygiene factors of hospital infection on SARS outbreak]. Zhonghua Lao Dong Wei Sheng Zhi Ye Bing Za Zhi = Zhonghua Laodong Weisheng Zhiyebing Zazhi = Chinese Journal Of Industrial Hygiene And Occupational Diseases. 2004;22(4):261-3.

Reason for exclusion: Hospital risk factors, China

SARS – systematic reviews

Gupta AG, Moyer CA, Stern DT. The economic impact of quarantine: SARS in Toronto as a case study (Structured abstract). Centre for Reviews and Dissemination. 2005

Reason for exclusion: Setting is Toronto

Fung IC, Cairncross S. How often do you wash your hands? a review of studies of hand-washing practices in the community during and after the SARS outbreak in 2003. International Journal of Environmental Health Research. 2007;17(3):161-83.

Reason for exclusion: Review of studies of hand-washing practice.

Fung IC-H, Cairncross S. Effectiveness of handwashing in preventing SARS: a review. Tropical Medicine & International Health: TM & IH. 2006;11(11):1749-58.

Reason for exclusion: Settings mainly China, Hong Kong and North America. Single study from Singapore looked at preventing nosocomial infection

Jefferson T, Foxlee R, Del Mar C, Dooley L, Ferroni E, Hewak B, et al. Interventions for the interruption or reduction of the spread of respiratory viruses. Cochrane Database Of Systematic Reviews (Online). 2007(4):CD006207.

Reason for exclusion: Settings for included studies are China, Hong Kong, Taiwan and North America. Focus on nosocomial transmission

Jefferson T, Foxlee R, Del Mar C, Dooley L, Ferroni E, Hewak B, et al. Interventions for the interruption or reduction of the spread of respiratory viruses: systematic review BMJ (Clinical Research Ed). 2008;336(7635):77-80.

Reason for exclusion: Settings for included studies are China, Hong Kong, Taiwan and North America. Focus on nosocomial transmission

Liu X, Zhang M, He L, Li Y. Chinese herbs combined with Western medicine for severe acute respiratory syndrome (SARS). Cochrane Library of Systematic Reviews. October 2010.

Reason for exclusion: Pharmaceutical interventions

Liu J P, Manheimer E, Shi Y. Systematic review and meta-analysis on the integrative traditional Chinese and Western medicine in treating SARS (Provisional abstract). Centre for Reviews and Dissemination. 2005

Reason for exclusion: Pharmaceutical interventions

Mahony JB, Petrich A, Louie L, Song XY, Chong S, Smieja M, Chernesky M, Loeb M, Richardson S. Performance and cost evaluation of one commercial and six in-house conventional and real-time reverse transcription-PCR assays for detection of severe acute respiratory syndrome coronavirus (Structured abstract). Centre for Reviews and Dissemination. 2004

Reason for exclusion: No intervention, evaluation of laboratory assay

Stockman LJ, Bellamy L, Garner P. SARS: systematic review of treatment effects (Structured abstract). Centre for Reviews and Dissemination. 2006

Reason for exclusion: Pharmaceutical interventions

Avian Influenza – articles

Avian influenza should be ruffling our feathers. The Lancet Infectious Diseases. 2004;4(10):595-.

Reason for exclusion: Editorial

Avian influenza, Thailand. Weekly Epidemiological Record. 2004;79(42):377-8.

Reason for exclusion: Update of Thai 2004 outbreak. No data

Assessment of risk to human health associated with outbreaks of highly pathogenic H5N1 avian influenza in poultry-situation as at 14 May 2004. Relevé Épidémiologique Hebdomadaire / Section D'hygiène Du Secrétariat De La Société Des Nations = Weekly Epidemiological Record / Health Section of The Secretariat Of The League Of Nations. 2004;79(21):203-4.

Reason for exclusion: Outbreak update and comment on transmission

Avian influenza – situation in Viet Nam at of 18 August 2004. Weekly Epidemiological Record. 2004;79(34):309–.

Reason for exclusion: Outbreak update

Avian influenza A(H5N1) in humans and poultry, Viet Nam. Weekly Epidemiological Record. 2004;79(3):13-4.

Reason for exclusion: Outbreak update

Current concepts: avian influenza A (H5N1) infection in humans. New England Journal of Medicine. 2005;353(13):1374.

Review. No data

Best defence against avian flu is to fight the virus in Asia. Bulletin Of The World Health Organization. 2005;83(12):887-9.

Reason for exclusion: Discusses issues around control, but no data

Bird Flu; Cost-effective disease prevention includes closing or regulating wildlife markets. Virus Weekly. 2007(15316424):14-.

Reason for exclusion: News article

Alders RG, Bagnol B, Brum E, Lubis AS, Young MP. Continuing education in the prevention and control of HPAI: a case study on Indonesia. 2009; 65:529-31.

Reason for exclusion: Discussion paper, no intervention

Amonsin, A., C. Choatrakol, et al. (2008). Influenza virus (H5N1) in live bird markets and food markets, Thailand. Emerging Infectious Diseases 14(11): 1739-1742.

Reason for exclusion: Sero-surveillance program 2006-7; no intervention

Areechokchai, D., C. Jiraphongsa, et al. (2006). Investigation of avian influenza (H5N1) outbreak in humans-Thailand, 2004. MMWR. Morbidity And Mortality Weekly Report 55 Suppl 1: 3-6.

Reason for exclusion: Matched case-control study to study risk factors of infection

Auewarakul, P., W. Hanchaoworakul, et al. (2008). Institutional responses to avian influenza in Thailand: control of outbreaks in poultry and preparedness in the case of human-to-human transmission. Anthropology & Medicine 15(1): 61-67.

Reason for exclusion: All Thai government-driven initiatives

Beltran-Alcrudo D, Bunn DA, Sandrock CE, Cardona CJ. Avian flu school: a training approach to prepare for H5N1 highly pathogenic avian influenza. Public Health Reports (Washington, DC: 1974). 2008; 123(3):323-32.

Reason for exclusion: Not Southeast Asia focused

Bhatia R, Narain JP. Preventing avian influenza in humans: the role of simple public health interventions. The Southeast Asian Journal of Tropical Medicine and Public Health. 2006; 37 (6): 1229-36.

Reason for exclusion: Narrative review, no data.

Buranathai, C, Amonsin A, et al. (2007). “Surveillance activities and molecular analysis of H5N1 highly pathogenic avian influenza viruses from Thailand, 2004-2005.” Avian Diseases 51(1 Suppl): 194-200.

Reason for exclusion: Clinical and laboratory surveillance following outbreak. Not community-based

Capua I and Alexander DJ. The challenge of avian influenza to the veterinary community. Avian Pathology. 2006; 35(3):189-205.

Reason for exclusion: Narrative review, no data

Cardona CJ, Byarugaba D, Mbuthia P, Aning G, Sourou S, Bunn DA et al. Detection and prevention of highly pathogenic avian influenza in communities with high poultry disease burdens. Avian Diseases. 2010;54(1 Suppl):754-6.

Reason for exclusion: Setting Africa, focuses on Newcastle disease

Chantong, W. and J. B. Kaneene (2011). “Poultry raising systems and highly pathogenic avian influenza outbreaks in Thailand: the situation, associations, and impacts.” The Southeast Asian Journal Of Tropical Medicine And Public Health 42(3): 596-608.

Reason for exclusion: Review of structure of poultry raising systems

Chuengsatiansup, K. (2008). “Ethnography of epidemiologic transition: avian flu, global health politics and agro-industrial capitalism in Thailand.” Anthropology & Medicine 15(1): 53-59.

Reason for exclusion: Not at community level

Chiu, D. An informatics and epidemiological evaluation of infectious disease surveillance ad reporting practices in Thailand: a case study of Suphanburi province and Avian Influenza. AMIA 2008 Symposium Proceedings page 904.

Reason for exclusion: Surveillance evaluation of national government-established systems

Chunsuttiwat, S. (2008). “Response to avian influenza and preparedness for pandemic influenza: Thailand's experience.” Respirology (Carlton, Vic.) 13 Suppl 1: S36-S40.

Reason for exclusion: Prevention programs, pandemic preparedness. Not at community level

Chutinimitkul S, Payungporn S, Chieochansin T, Suwannakarn K, Theamboonlers A, Poovorawan Y. The spread of avian influenza H5N1 virus; a pandemic threat to mankind. Journal of Medical Association of Thailand= Chotmaihet Thangphaet. 2006;89 Suppl 3:S218-S33.

Reason for exclusion: Narrative review, interventions discussed are pharmaceutical only

Clague B, Chamany S. A household survey to assess the burden of influenza in rural Thailand. Southeast Asian Journal of Tropical Medicine and Public Health. 2006;37(3):488-93.

Reason for exclusion: No intervention

Cristalli, A. and I. Capua (2007). “Practical problems in controlling H5N1 high pathogenicity avian influenza at village level in Viet Nam and introduction of biosecurity measures.” Avian Diseases 51(1 Suppl): 461-462.

Reason for exclusion: Comment on inclusion of backyard flocks in preventative measures

Dauphin, G., K. Hamilton, et al. (2010). Main achievements of the World Organisation for Animal Health/United Nations Food and Agriculture Organization network on animal influenza. Avian Diseases 54(1 Suppl): 380-383.

Reason for exclusion: Overview of OFFLU

de Sa, J., S. Mounier-Jack, et al. (2010). Responding to pandemic influenza in Cambodia and Lao PDR: challenges in moving from strategy to operation. The Southeast Asian Journal Of Tropical Medicine And Public Health 41(5): 1104-1115.

Reason for exclusion: Review of government and international health organisation-driven initiatives

Eagles, D., E. S. Siregar, et al. (2009). H5N1 highly pathogenic avian influenza in Southeast Asia. Revue Scientifique Et Technique (International Office Of Epizootics) 28(1): 341-348.

Reason for exclusion: Narrative review.

Ellis TM, Leung CYHC, Chow MKW, Bissett LA, Wong W, Guan Y et al. Vaccination of chickens against H5N1 avian influenza in the face of an outbreak interrupts virus transmission. Avian Pathology: Journal of the WVPA/ 2004;33(4):405-12.

Reason for exclusion: Vaccine trial in field setting. Not community based, Hong Kong study

Farnsworth, M. L., C. Hamilton-West, et al. (2010). Comparing national and global data collection systems for reporting, outbreaks of H5N1 HPAI. Preventive Veterinary Medicine 95(3-4): 175-185.

Reason for exclusion: Evaluation of EMPRES (FAO) surveillance data. Not community focused

Farnsworth M, Fitchett S, Hidayat MM, Lockhart C, Hamilton-West C et al. Metapopulation dynamics and determinants of H5N1 highly pathogenic avian influenza outbreaks in Indonesian poultry. Preventative Veterinary Medicine, 2011;102:206-217.

Reason for exclusion: Model-based analysis to assess probability of HPAI occurrence

Figuie M, Fournier T. Global Health risks and National Policies. Avian Influenza Risk Management in Viet Nam. Review of Agricultural and Environmental Studies/Revue d'Etudes en Agriculture et Environnement. 2010;91(3):327-43.

Reason for exclusion: Narrative review, no intervention

Fleming D. Influenza pandemics and avian flu. BMJ: British Medical Journal (International Edition). 2005; 331(7524):1066-9.

Reason for exclusion: Narrative, very general

Hampson AW, Mackenzie JS. The influenza viruses. The Medical Journal of Australia. 2006; 185(10 Suppl):S39-S43.

Reason for exclusion: General, no intervention

Hanvoravongchai, P., W. Adisasmito, et al. (2010). “Pandemic influenza preparedness and health systems challenges in Asia: results from rapid analyses in 6 Asian countries.” BMC Public Health 10: 322-322.

Reason for exclusion: Analysis of pandemic preparedness programs

Häsler B, Howe KS, Stärk KDC. Conceptualising the technical relationship of animal disease surveillance to intervention and mitigation as a basis for economic analysis. BMC Health Services Research. 2011; 11:225-

Reason for exclusion: AI used as example to illustrate framework

Henning, J., D. U. Pfeiffer, et al. (2009). “Risk factors and characteristics of H5N1 Highly Pathogenic Avian Influenza (HPAI) post-vaccination outbreaks.” Veterinary Research 40(3): 15-15.

Reason for exclusion: Risk factor analysis, but no intervention

Henning, K. A., J. Henning, et al. (2009). “Farm- and flock-level risk factors associated with Highly Pathogenic Avian Influenza outbreaks on small holder duck and chicken farms in the Mekong Delta of Viet Nam.” Preventive Veterinary Medicine 91(2-4): 179-188.

Reason for exclusion: Risk factor analysis

Ifft J, Roland-Holst D, Zilberman D. Production and Risk Prevention Response of Free Range Chicken Producers in Viet Nam to Highly Pathogenic Avian Influenza Outbreaks. American Journal of Agricultural Economics. 2011; 93(2):490-7.

Reason for exclusion: Behavioural analysis, model-based

Imperato PJ. The Growing Challenge of Avian Influenza Journal of Community Health. 2005;30(5):327-30.

Reason for exclusion: Surveillance in Asia. Very general

Indriani, R., G. Samaan, et al. (2010). “Environmental sampling for avian influenza virus A (H5N1) in live-bird markets, Indonesia.” Emerging Infectious Diseases 16(12): 1889-1895.

Reason for exclusion: Survey of markets, risk and protective factors

Jost CC. Immediate assistance for strengthening community-based early warning and early reaction to avian influenza in Indonesia. In 5^th Quarter Report (October – December 2006) and Chief Technical Advisor End of Contract Report (1 February 2007). Tufts University School of Veterinary Medicine International Program for the Food and Agriculture Organisation of the United Nations, Rome.

Reason for exclusion: Project proposal, no data

Juckett G. Avian influenza: preparing for a pandemic. American Family Physician. 2006;74(5):783-90.

Reason for exclusion: Narrative review

Kelly TR, Hawkins MG, Sandrock CE, Boyce WM. A review of highly pathogenic avian influenza in birds, with an emphasis on Asian H5N1 and recommendations for prevention and control. Journal of Avian Medicine and Surgery. 2008;22(1):1-16.

Reason for exclusion: Narrative review, US focused

Khanna, M., P. Kumar, et al. (2008). “Emerging influenza virus: A global threat.” Journal of Biosciences 33(4): 475-482.

Reason for exclusion: Review of epidemiology

Kimball AM, Moore M, French HM, Arima Y, Ungchusak K, Wilbulpolprasert S et al. Regional infectious disease surveillance networks and their potential to facilitate the implementation of the international health regulations. Medical Clinic of North America. 2008;92(6):1459-71.

Reason for exclusion: Discusses Mekong Basin Disease Surveillance Network, but no data

Kitphati R, Apisarnthanarak A, Chittaganpitch,M, Tawatsupha P, Auwanit W et al. A nationally coordinated laboratory system for human avian influenza A (H5N1) in Thailand: program design, analysis and evaluation. Clinical Infectious Diseases, 2008;46:1394-1400.

Reason for exclusion: Review of government and international health organisation-driven initiatives

Kleinman, A. M., B. R. Bloom, et al. (2008). “Asian flus in ethnographic and political context: a biosocial approach.” Anthropology & Medicine 15(1): 1-5.

Reason for exclusion: Analysis of local context. No intervention

Koh, G., T. Wong, et al. (2008). “Avian Influenza: a global threat needing a global solution.” Asia Pacific Family Medicine 7(1): 5-5.

Reason for exclusion: Very general review

Kruy, S. L., Y. Buisson, et al. (2008). “[Asia: avian influenza H5N1].” Bulletin De La Société De Pathologie Exotique (1990) 101(3): 238-242.

Reason for exclusion: Narrative review

Kuiken T, Leighton FA, Fouchier RAM, LeDuc JW, Peiris JSM et al. Pathogen surveillance in animals. Science, 2005; 309:1680-81.

Reason for exclusion: Public health policy forum.

Lazzari S and Stohr K. Avian influenza and influenza pandemics. Bulletin of the World Health Organization. 2004;82(4)242-242A.

Reason for exclusion: Editorial

Leibler JH, Otte J, Roland-Holst D, Pfeiffer DU, Soares-Magalhaes R, Rushton J et al. Industrial Food Animal Production and Global Health Risks: Exploring the Ecosystems and Economics of Avian Influenza. Ecohealth. 2009;6(1):58-70.

Reason for exclusion: Narrative review

Leong, H. K., C. S. Goh, et al. Prevention and control of avian influenza in Singapore. Annals Of The Academy Of Medicine, Singapore. 2008; 37(6): 504-509.

Reason for exclusion: Review of government driven interventions

Ligon BL. Avian influenza virus H5N1: a review of its history and information regarding its potential to cause the next pandemic. Seminars in Pediatric Infectious Diseases. 2005;16(4):326-35.

Reason for exclusion: Narrative review, no intervention

Longini, I. M., Jr., A. Nizam, et al. (2005). “Containing pandemic influenza at the source.” Science (New York, N.Y.) 309(5737): 1083-1087.

Reason for exclusion: Model to estimate effectiveness of quarantine, anti-virals

Loth, L., M. Gilbert, et al. (2011). Identifying risk factors of highly pathogenic avian influenza (H5N1 subtype) in Indonesia. Preventive Veterinary Medicine 102(1): 50-58.

Reason for exclusion: Identification of risk factors for presence of HPAI; no intervention

Lubroth J. Control strategies for highly pathogenic avian influenza: a global perspective. Developments In Biologicals. 2007;130:13-21.

Reason for exclusion: Prevention programs for detection and control of HPAI. First world focused

Lugnér AK, Postma MJ. Mitigation of pandemic influenza: a review of cost-effectiveness studies. Expert Review of Pharmacoeconomics & Outcomes Research. 2009;9(6):547-58.

Reason for exclusion: Model-based cost-effectiveness studies

Marris E. Despite doubts, containment plans for pandemic take shape. Nature Medicine. 2006;12(5):485-

Reason for exclusion: Describes WHO draft containment plan March 2004

McLeod A. Economics of avian influenza management and control in a world with competing agendas. Avian Diseases. 2010;54(1 Suppl):374-9.

Reason for exclusion: Review of economic issues, no data

Moore M, Dausey DJ. Response to the 2009-H1N1 influenza pandemic in the Mekong Basin: surveys of country health leaders. BMC Research Notes. 2011;4(1):361.

Reason for exclusion: Swine influenza epidemic

Morris SK. H5N1 avian influenza, Kampot Province, Cambodia. Emerging Infectious Diseases. 2006;12(1):170-1.

Letter

Morse, SS. (2007). Global Infectious Disease Surveillance And Health Intelligence. Health Affairs 26(4): 1069-1077.

Reason for exclusion: Narrative review

Nicholson KG, Wood JM, Zambon M. Influenza. The Lancet. 2003;362(9397):1733-45.

Reason for exclusion: Narrative review

Normile D. Avian influenza. WHO proposes plan to stop pandemic in its tracks. Science (New York, NY). 2006;311(5759):315-6.

Reason for exclusion: Magazine article, no data

Normile D. Indonesia Taps Village Wisdom to Fight Bird Flu. Science. 2007;315(5808):30-3.

Reason for exclusion: Magazine article, no data

Normile, D. and M. Enserink (2005). “Lapses Worry Bird Flu Experts.” Science 308(5730): 1849-1849,1851.

Reason for exclusion: Magazine article, no data

Normile D. Avian influenza. Warning of H5N1 resurgence surprises community. Science (New York, NY). 2011;333(6048):1369-.

Reason for exclusion: Magazine article, no data

Olsen S, Ungchusak K, Birmingham M, Bresee, J, Dowell SF, Chunsuttiwat S. Surveillance for avian influenza in human beings in Thailand. The Lancet Infectious Diseases, 2006; 6:757-8.

Reason for exclusion: Short commentary

Oshitani H. Potential benefits and limitations of various strategies to mitigate the impact of an influenza pandemic. Journal of Infection And Chemotherapy: Official Journal Of The Japan Society Of Chemotherapy. 2006;12(4):167-71.

Reason for exclusion: Narrative review

Otto, J. L., P. Baliga, et al. Training initiatives within the AFHSC-Global Emerging Infections Surveillance and Response System: support for IHR (2005). BMC Public Health 2011;11(Suppl 2): S5.

Reason for exclusion: Not community-based

Padmawati S, Nichter M. Community response to avian flu in Central Java, Indonesia. Anthropology & Medicine. 2008;15(1):31-51.

Reason for exclusion: Overview of poultry system, survey of perceptions of AI, no intervention

Parry, J. (2004). “Mortality from avian flu is higher than in previous outbreak.” British Medical Journal 328(7436): 368-368.

Reason for exclusion: News article, no data

Parry, J. (2004). “WHO investigates possible human to human transmission of avian flu.” British Medical Journal 328(7435): 308-308.

Reason for exclusion: Article, no data

Parry, J. (2004). “WHO confirms avian flu outbreak in Hanoi.” British Medical Journal 328(7432): 123-123.

Reason for exclusion: News article, no data

Parry, J. Ten years of fighting bird flu. Bulletin of the World Health Organization. 2007;85:3-4.

Reason for exclusion: News article

Paul M, Wongnarkpet S, Gasqui P, Poolkhet C, Thongratsakul S et al. Risk factors for highly pathogenic avian influenza (HPAI) H5N1 infection in backyard chicken farms, Thailand. Acta Tropica. 2011;118:209-16.

Reason for exclusion: Risk factor identification through case control study. No intervention

Pfeiffer DU, Minh PQ, Martin V, Epprecht M, Otte MJ. An analysis of the spatial and temporal patterns of highly pathogenic avian influenza occurrence in Viet Nam using national surveillance data. Veterinary Journal (London, England: 1997). 2007;174(2):302-9.

Reason for exclusion: Surveillance data to inform outbreak risk

Pitsuwan S. Challenges in infection in ASEAN. The Lancet. 2011;377(9766):619-21.

Reason for exclusion: Government driven initiatives

Rushton J, Viscarra R, Guerne Bleich E, McLeod A. Impact of avian influenza outbreaks in the poultry sectors of five South East Asian countries (Cambodia, Indonesia, Lao PDR, Thailand, Viet Nam): outbreak costs, responses and potential long term control. World's Poultry Sci J. 2005; 61:491-514.

Reason for exclusion: Narrative review, no data on community interventions

Samaan G, Gultom A, Indriani R, Lokuge K, Kelly PM. Critical control points for avian influenza A H5N1 in live bird markets in low resource settings. Preventive Veterinary Medicine. 2011;100(1):71-8.

Reason for exclusion: Cross-sectional and KAP surveys, microbiological survey

Sambhara S, Poland GA. H5N1 Avian influenza: preventive and therapeutic strategies against a pandemic. Annual Review of Medicine. 2010;61:187-98.

Reason for exclusion: Narrative review, focusing on anti-virals and vaccines

Sims LD. Lessons learned from Asian H5N1 outbreak control. Avian Diseases. 2007;51(1 Suppl):174-81.

Reason for exclusion: Narrative review, no data

Soares Magalhães RJ, Ortiz-Pelaez A, Thi KLL, Dinh QH, Otte J, Pfeiffer DU. Associations between attributes of live poultry trade and HPAI H5N1 outbreaks: a descriptive and network analysis study in northern Viet Nam. BMC Veterinary Research. 2010;6:10-

Reason for exclusion: Cross-sectional survey, social network analysis of poultry traders. No intervention

Soares Magalhães RJ, Pfeiffer DU, Otte J. Evaluating the control of HPAIV H5N1 in Viet Nam: virus transmission within infected flocks reported before and after vaccination. BMC Veterinary Research. 2010;6:31-.

Reason for exclusion: Model-based study to assess vaccination and depopulation measures

Stone R. Avian influenza. Combating the bird flu menace, down on the farm. Science (New York, NY). 2006;311(5763):944-6.

Reason for exclusion: News article

Subbarao K. Evaluation of novel influenza A viruses and their pandemic potential. Pediatric Annals. 2000;29(11):712-8.

Reason for exclusion: Narrative review

Swayne DE, Suarez DL. Highly pathogenic avian influenza. Revue Scientifique Et Technique (International Office Of Epizootics). 2000;19(2):463-82.

Reason for exclusion: Narrative review

Thomas N. The Regionalization of Avian Influenza in East Asia: Responding to the Next Pandemic(?). Asian Survey. 2006;46(6):917-36.

Reason for exclusion: Overview of H5N1 in Asia, but no data

Thorson A, Petzold M, Nguyen TKC, Ekdahl K. Is exposure to sick or dead poultry associated with flulike illness?: a population-based study from a rural area in Viet Nam with outbreaks of highly pathogenic avian influenza. Archives Of Internal Medicine. 2006;166(1):119-23.

Reason for exclusion: Risk factor analysis, no intervention

Tiensin T, Chaitaweesub P, Songserm T, Chaisingh A, Hoonsuwan W, Buranathai C, et al. Highly pathogenic avian influenza H5N1, Thailand, 2004. Emerging Infectious Diseases. 2005;11(11):1664-72.

Reason for exclusion: Control measures all government-driven

Tiensin T, Nielen M, Songserm T, Kalpravidh W, Chaitaweesub P, Amonsin A, et al. Geographic and temporal distribution of highly pathogenic avian influenza A virus (H5N1) in Thailand, 2004-2005: an overview. Avian Diseases. 2007;51(1 Suppl):182-8.

Reason for exclusion: Thai outbreak, 2004. All government driven control measures

Trevennec K, Chevalier V, Grosbois V, Garcia JM, Thu HH, Berthouly-Salazar C, et al. Looking for avian influenza in remote areas. A case study in Northern Viet Nam. Acta Tropica. 2011;120(3):160-6.

Reason for exclusion: Serosurvey, interviews, but no intervention; risk factors for infection

Van Kerkhove MD, Vong S, Guitian J, Holl D, Mangtani P, San S, et al. Poultry movement networks in Cambodia: implications for surveillance and control of highly pathogenic avian influenza (HPAI/H5N1). Vaccine. 2009;27(45):6345-52.

Reason for exclusion: Cross-sectional survey of trading practices, no intervention

Watts J. Viet Nam needs cash to stave off future outbreaks of bird flu. The Lancet. 2005;365(9473):1759-60.

Reason for exclusion: Article, no data

Williams JR, Chen P-Y, Cho CT, Chin TDY. Influenza: prospect for prevention and control. The Kaohsiung Journal of Medical Sciences. 2002;18(9):421-34.

Reason for exclusion: Narrative review, no data

Yee KS, Carpenter TE, Cardona CJ. Epidemiology of H5N1 avian influenza. Comparative Immunology, Microbiology and Infectious Diseases. 2009;32(4):325-40.

Reason for exclusion: Narrative review, no data

Yupiana Y, de Vlas SJ, Adnan NM, Richardus JH. Risk factors of poultry outbreaks and human cases of H5N1 avian influenza virus infection in West Java Province, Indonesia. International Journal Of Infectious Diseases: IJID: Official Publication Of The International Society For Infectious Diseases. 2010;14(9):e800-e5.

Reason for exclusion: Risk factor analysis, no intervention

Avian Influenza – systematic reviews

Manzoli L, Salanti G, De Vito G, Boccia A, Ioannidis JP, Villari P. Immunogenicity and adverse events of avian influenza A H5N1 vaccine in healthy adults: multiple-treatments meta-analysis (Structured abstract). Centre for Reviews and Dissemination. 2009

Reason for exclusion: Pharmaceutical intervention

Appendix VI –Joanna Briggs Institute levels of evidence

Taken from the Joanna Briggs Institute – Reviewers' Manual 2011 Edition: Appendix VI – JBI Levels of evidence p. 150

Appendix VII – Extended meta–analysis report

Appendix VIII – Abbreviations used in this review

AAHL Australian Animal Health Laboratory

AI Avian Influenza

AIDS Acquired Immunodeficiency Syndrome

ASEAN Association of Southeast Asian Nations

AusAID Australian Agency for International Development

BI Breteau index

Bti Bacillus thurigensis

CDC Centers for Disease Control and Prevention

CEDAC Centre d'Etude et de Developement Agricole Cambodgien

CI Container index

DF Dengue fever

DfID Department for International Development

DHF Dengue haemorrhagic fever

DI Density Index

DVS Department of Veterinary Services, Malaysia

ELISA Enzyme-linked immunosorbent assay

EMPRESFAO Emergency Prevention System

EPOC Cochrane Effective Practice and Organisation of Care Review Group

FAO Food and Agriculture Organisation of the United Nations

GDP Gross Domestic Product

GPS Global positioning satellite

HI House Index

HPAI Highly pathogenic avian influenza

JBI Joanna Briggs Institute

JBI-MAStARI JBI-Meta-Analysis of Statistics Assessment and Review Instrument

KAP Knowledge, attitudes and practices

Lao PDR Lao People's Democratic Republic

LDCC Local Disease Control Centres

LJA Lao Journalists' Association

MOH Ministry of Health

NaVRI National Veterinary Research Institute, Cambodia

NGO Non-government organisation

OFFLU OIE/FAO network of expertise on avian influenza

OI Ovitrap Index

OIE International Office of Epizootics

OR Odds ratio

PDSR Participatory Disease Surveillance and Response

PEP Post-exposure prophylaxis

PET Post-exposure treatment

PHU Public health unit

PPE Personal protective equipment

PPI Pupae per person index

PPV Positive predictive value

RNAS (+)Regional Network on Asian Schistosomiasis and Other Helminth Zoonoses

RR Rate ratio

SARS Severe acute respiratory syndrome

SE Asia Southeast Asia

SEARO South-east Asia Regional Office (of WHO)

SES Socio-economic status

SNT Serum neutralisation test

IT Insecticide-treated

ULV Ultra-low volume

USAID United States Agency for International Development

VHV Village health volunteer

VWU Viet Nam Women's Union

WHO World Health Organization

WHOLISWHO library database

WRPO Western Regional Pacific Office (of WHO)