Building operational research capacity in the Pacific

Abstract

Operational research (OR) in public health aims to investigate strategies, interventions, tools or knowledge that can enhance the quality, coverage, effectiveness or performance of health systems. Attention has recently been drawn to the lack of OR capacity in public health programmes throughout the Pacific Islands, despite considerable investment in implementation. This lack of ongoing and critical reflection may prevent health programme staff from understanding why programme objectives are not being fully achieved, and hinder long-term gains in public health. The International Union Against Tuberculosis and Lung Disease (The Union) has been collaborating with Pacific agencies to conduct OR courses based on the training model developed by The Union and Médecins Sans Frontières Brussels-Luxembourg in 2009. The first of these commenced in 2011 in collaboration with the Fiji National University, the Fiji Ministry of Health, the World Health Organization and other partners. The Union and the Secretariat of the Pacific Community organised a second course for participants from other Pacific Island countries and territories in 2012, and an additional course for Fijian participants commenced in 2013. Twelve participants enrolled in each of the three courses. Of the two courses completed by end 2013, 18 of 24 participants completed their OR and submitted papers by the course deadline, and 17 papers have been published to date. This article describes the context, process and outputs of the Pacific courses, as well as innovations, adaptations and challenges.

Operational research (OR) in public health has been defined as research into strategies, interventions, tools and knowledge that can enhance the quality, coverage, effectiveness or performance of the health system or disease programme in which research is being conducted.¹ We view OR as a spectrum of activities that encompasses reviews of data already collected in patient registers, treatment cards or patient files, and evaluations of operational practices and the implementation of new strategies, interventions and technologies.²

Since 2009, the International Union Against Tuberculosis and Lung Disease (The Union), in collaboration with Médecins Sans Frontières (MSF), has been running courses designed to build the capacity of national health staff in performing OR. These courses have been convened by leveraging funding that is predominantly for National Tuberculosis Programmes (NTPs). However, as the OR programme has evolved, we have been able to attract additional funds to expand the focus of OR projects to a range of health issues. The purpose of the course is to teach practical skills for conducting and publishing OR that will contribute to changes in policy and practice that are relevant to local needs.³ The modular-based course emphasises 1) careful selection of participants according to pre-defined criteria; 2) logical progression through the steps of the whole research process, i.e., from creating a research question through to publishing a paper and formulating recommendations for changes in policy or practice; 3) the achievement of milestones to progress from one module to the next; 4) support from employers of participants to allow time and resources for the research; 5) support from experienced public health practitioners who act as mentors; 6) the development of a final product, i.e., a research paper submitted to a peer-reviewed scientific journal (which constitutes the final milestone); and 7) potential for trained participants to become mentors on future courses.⁴ Course graduates are followed up by The Union and MSF after completing the course, and ongoing OR activity is monitored.⁵

There are many reasons why publication is a key focus of the course.³ First, participants learn the process and rigour needed to produce well-argued, concise work for peer review, their work benefits from the quality control and intellectual input provided by peer review, and they gain credibility as first author of a published paper. Second, research results need to be published for greater credibility and national and international dissemination. Third, national and international policy making increasingly uses and cites published scientific papers to underpin statements in policies and guidelines.

The term ‘operational research’ is often used interchangeably with implementation research. Depending on the definitions used, there may be little difference between the two. The World Health Organization (WHO) defines implementation research, as ‘the scientific inquiry into questions concerning implementation’ taking into account contextual factors.⁶ OR may differ slightly from implementation research, as OR explicitly seeks to build capacity in local health programme staff to conduct the research. In addition, implementation research is often broader in scope than OR and may use a broader range of research methodologies.⁷ The OR courses have placed an emphasis on the publication of study results so that policy and practice can be influenced, whereas implementation research may tend to focus more on report writing for policy makers. However, both seek to conduct research that is embedded in a ‘real world’ setting, with due regard for contextual factors, including the structure of the health system.^1,6–8

THE PACIFIC CONTEXT

The Pacific Islands region is composed of 22 Pacific Island countries and territories (PICTs) with diverse populations, cultures, economies and politics. The region is divided into three sub-regions—Micronesia, Melanesia and Polynesia—based on linguistic, cultural and ethnic characteristics (Figure).⁹ The total population is estimated at 10 million. Approximately 50% live in rural areas and outer islands, although there is an ongoing process of urbanisation towards the Pacific Islands' capitals.¹⁰

FIGURE.

Map of the Pacific Island countries and territories (Courtesy Secretariat of the Pacific Community, Nouméa, New Caledonia).

Challenges in delivering health care

Health care in PICTs is primarily provided by national governments, often supplemented by considerable amounts of international support and with small contributions from the private sector.¹¹ Health care systems are largely based on a primary health care model that is similar in most low- and middle-income countries (LMICs). However, they are often fragile and poorly funded, and at times struggle to meet the health care needs of their populations.¹¹ Curative and tertiary services continue to receive most of the allocated health resources.¹¹ Provision of health care to remote and rural areas is often expensive and logistically challenging. Most PICTs suffer from a shortage of health care workers and other health care commodities.^11,12

Disease burden

Rapid changes in the social, economic and environmental situation have significantly impacted the health and well-being of Pacific Island populations. As health determinants and risk factors have changed over time, there has been an increase in non-communicable diseases (NCDs) across the region.¹³ Diabetes mellitus is of particular concern, due to its relatively high prevalence, debilitating complications and economic costs to governments and society. The prevalence of diabetes among adults (aged 20–79 years) was estimated at 7.6% in 2012, and is expected to increase to 8.6% by 2030.¹⁴ There are large regional differences in diabetes prevalence,^15–19 with estimated prevalence figures from as low as 6.1% in the most populous country in the Pacific, Papua New Guinea (PNG), to as high as 47.3% in American Samoa.¹⁴

In addition, communicable diseases such as lower respiratory infections and diarrhoea, and neglected tropical diseases such as tuberculosis (TB), leprosy, leptospirosis and lymphatic filariasis, persist in the Pacific, and in some cases are re-emerging. TB case notification rates in some PICTs are often among the highest in the World Health Organization (WHO) Western Pacific Region. In 2011, a total of 16 534 TB cases were notified in the Pacific, with the majority of cases notified in PNG (n = 14 893, 90%). In 2011, Kiribati reported the region's highest TB case notification rate, at 334 cases per 100 000 population.²⁰ The burden of TB and other communicable diseases varies greatly by PICT, with some countries moving towards the elimination of TB and other communicable diseases, and others still recording high rates of TB and other communicable diseases.¹²

The evidence for interactions and synergies between communicable diseases and NCDs is growing in the Pacific, with countries at various stages of the epidemiological transition.^21,22 The association between TB and diabetes has long been a concern for NTPs in the Pacific. In 2010, it was estimated that approximately 42% of all TB cases in the northern Pacific were attributable to diabetes.²³ Although some NTPs are now implementing TB-diabetes collaborative activities, evidence about the effectiveness, cost and yield of these activities is lacking, and OR may assist PICTs in evaluating and improving TB-diabetes interventions before scale-up.

Rationale for operational research in the region

International public health strategies and plans recommend that OR should become a key component of health programmes.^24–28 This is reflected in regional disease-specific strategies such as the Regional Strategy to Stop TB in the Western Pacific: 2011–2015, which states that ‘operational research is a crucial tool to evaluate and study the application of various mechanisms, interventions and tools.’²⁹ Our collective experience in international public health has also convinced us that OR is essential. Well-designed, programme-based OR can identify constraints in meeting health programme objectives, investigate them and recommend improvements to enable better functioning of the programme. However, OR capacity in the Pacific has not been well developed, and additional focus is needed to build and sustain this capacity. We believe that OR capacity building should begin with national health programme staff,³⁰ while collaborations are forged between national governments, technical agencies, local and regional academic institutions and others, to support and promote OR in the region.

Operational research: needs and challenges

The adage ‘data rich, information poor’ seems to fit many of the PICTs. Health programmes are collecting a wealth of data on various health indicators.^31–33 Although these data are used for annual reports by Ministries of Health and regional agencies, there is often little evidence of any further analysis by the local health services themselves, or of use of the data to improve policy and practice. These data are seldom used for carrying out OR, and there appears to be a lack of a research culture within and between countries in the Pacific in general. There may be many reasons for this, including the fact that research is not regarded as high priority; programme staff having limited time, research skills and lacking motivation to do research; difficulty obtaining the necessary resources; and variable quality in the design of studies and in the recording, analysis and reporting of data.³⁴ Additional factors believed to contribute to the failure among programme staff to publish and/or to disseminate the results of any research conducted to a wider audience include limited ethics review procedures, meaning that researchers have little guidance about how to conduct and write up ethical research or transparency about how decisions about ethics approval and accountability are made, and limited scientific paper writing skills and peer review opportunities.³⁴ Cuboni et al. noted that research and the sharing of information between countries is further complicated, among other things, by the Pacific region's cultural and linguistic diversity, geographical distances and the range of skills among health workers, while highlighting that one fundamental element of getting research done is still ‘the determination to see it done.’³⁵

The lack of a research culture³⁶ may also have been influenced by what has been described in a key paper from the year 2000 as ‘research imperialism,’ in which health research is driven by external agendas, leaving local personnel and communities feeling that research is owned and conducted by ‘outsiders’ who are interested in specific issues, regardless of their relevance to the region's needs.³⁴

Related to this is the concern that research efforts by international agencies and organisations in the Pacific have not concentrated enough on training local researchers.³⁷ A study examining the number of papers authored by Pacific researchers on reproductive health in 14 PICTs from 2000 to 2011 found that only 19 (16%) of 122 multi-authored papers had first authors of Pacific ethnicity.³⁷ A similar review of health research publications in Fiji found that, of 298 papers published between 1965 and 2002, only 95 (31.8%) had one or more Fijians involved in the authorship. Of the total of 815 instances of authorship (including sole, first and co-authorship), Fijian authorship represented only 17.7% (n = 144), and of these 144 Fijian authorships, only 38 (26.4%) were first or sole authors.³⁵

Also of concern is the limited availability of personnel with appropriate research skills or qualifications in specific disciplines to conduct the type of research required for evidence-informed policy making. One project that sought to facilitate the achievement of higher academic degrees for Pacific Islanders suggested that being employed in a key role within a Ministry was a major barrier to undertaking further study and research.³⁸ Other barriers were potential financial losses, job insecurity and the need to support family on a doctoral stipend. A project involving collaboration between Samoa and New Zealand included a collaborative health research methods course to strengthen health research capacity in Samoa.³⁹ The authors reported that although Samoan citizens who are sent to be trained in overseas institutions do return, many return with minimal skills in health research, and all return to a work environment that is not yet suited to undertaking or supporting locally led health research.³⁹

In fact, as most Pacific Island nations are not able to provide postgraduate medical training or health research training, many of their health care staff have to go overseas to train. Limited academic preparation is recognised as a barrier to pursuing such training. Geographical isolation continues to pose a challenge for their ongoing training and skill development.⁴⁰ Open and distance learning approaches are often portrayed as low-cost ways to overcome issues of distance. However, these approaches do not necessarily meet local cultural expectations or preferred learning styles.⁴¹

Various factors have been identified to support learning among Pacific people during in-country or overseas training, including having a social structure during learning that allows for group activities and positive peer pressure,^40,42 and incorporating a hands-on and oral learning style as compared with the more individualised ‘Western’ written educational and testing style.⁴¹

Translating research evidence into policy has been recognised as another challenge facing the region.³⁴ Seeing that few evidence-informed policy initiatives were being undertaken to address the problem of overweight and obesity in the Pacific, one recent research project implemented and evaluated an innovative knowledge exchange intervention between researchers and policy makers in Fiji.⁴³ This resulted in the presentation of 20 policy briefs to high-level decision makers, and the improvement in skills of researchers and policy makers in the research-to-policy processes. Such initiatives, however, appear to be rare, and the research-to-policy-to-practice paradigm does not yet appear to be integrated throughout the general health services, universities and non-government organisations (NGOs) in the Pacific.

OPERATIONAL RESEARCH COURSES IN THE PACIFIC

Since 2011, three Union/MSF OR courses have been organised in the Pacific: Fiji September 2011–August 2012, Fiji May 2013–February 2014, and the Pacific course September 2012–August 2013. The two courses in Fiji included participants from Fiji only, while the Pacific course included participants from seven PICTs. The Fiji courses were collaborative efforts between the Fiji National University, the Fiji Ministry of Health, the Secretariat of the Pacific Community (SPC), the WHO, The Union and MSF, while the Pacific course represented a collaboration between the SPC, The Union and the University of Auckland. Both courses used the Union/MSF model of OR training. The Pacific course had a focus on TB, NCDs and the interaction between the two, while the Fiji course covered a range of topics related to communicable diseases, NCDs, laboratory issues and health systems.

The Union/MSF's three module approach to OR training was used. This approach is based on the three distinct types of activity involved in the research process: 1) protocol development, 2) data management and analysis, and 3) paper writing. Each 5-day-long module contained content aligned to the sequential activities that the participants need to undertake to make progress with their own research project and to write a paper as first author. The content of each module and the relevant outcomes and milestones are described in Table 1. The maximum number of participants for each course was 12. Each participant was assigned two mentors who worked with them throughout the course, and had access to additional mentors experienced in data management and the software taught during Module Two.

TABLE 1.

Description of The Union/MSF OR course in the Pacific

Mentors on the Pacific operational research courses

The mentors for the courses comprised a team of international and regional experts on TB, diabetes, Pacific health and research from several organisations (Table 2). A team of senior mentors was assisted by junior mentors who had participated in the Fiji OR course. The junior mentors were from the Fiji Ministry of Health, the Fiji NTP and the Fiji National University. SPC technical and project management staff provided technical and organisational support throughout the course. The Fiji National University and the University of Auckland were key academic partners for the project, building on an existing relationship between the two institutions.

TABLE 2.

List of organisations that supported the Fiji and Pacific operational research courses by providing mentors and/or facilities

Over the three courses, we have introduced a number of new mentors to the Union/MSF approach and created a regional pool of mentors who can facilitate on future courses. As not all mentors will be available each year for various courses that may be requested, it will be important to have a pool of mentors who have gone through the course in its entirety. Once mentors have completed an entire course, they have a more in-depth understanding of the ethos, processes, challenges and outputs of the OR course. They are then also able to recommend potential new mentors from among their academic or health services networks. We have opted for a multidisciplinary configuration of mentors; our experience is that it is ideal to have a mix of senior mentors who have extensive experience and highly developed skills in medicine, public health, health care implementation, ethics, OR, policy development, writing, editing and reviewing papers, academic mentorship and adult learning, and junior mentors who are experienced in the delivery of health services in their country and who have some experience in conducting and/or applying research.

INNOVATIONS, ADAPTATIONS AND CHALLENGES IN THE PACIFIC CONTEXT

In the Pacific OR courses, we adapted each course slightly to meet local needs and challenges, while retaining the core course material. The innovations and adaptations are summarised in Table 3 and a selection is more fully described below. The challenges are summarised in Table 4.

TABLE 3.

Innovations and adaptations introduced in the Union/MSF Operational Research training courses in the Pacific, 2011–2013

TABLE 4.

Challenges faced when implementing The Union/MSF operational research courses in the Pacific, 2011–2013

Discussion about research questions from participants before Module One

For the Fiji 2011–2012 course, the two international senior mentors spent 5 hours with the group of participants before Module One to discuss their research questions, including relevance, feasibility and data sources. This enabled various issues to be identified and resolved, and some topics were modified before Module One commenced. This was especially relevant for individuals who were not familiar with research methods. In several cases, participants were unaware about the specifics of the data sources that they were proposing to use, for example, which data are collected and how many years of data are available. For the Fiji 2013–2014 course, the local junior mentors held the same type of meeting before Module One. It was not possible to have a preliminary face-to-face discussion with all the Pacific participants. However, some mentors were able to discuss topics with participants during travel to their countries for other professional missions or via e-mail.

A gap of 2 months between Modules One and Two

Courses being held in other parts of the world have been running Modules One and Two back-to-back, which is a change from the previous practice where they were held approximately 2 months apart.⁴⁴ In the Pacific, we have continued to maintain a gap of approximately 2 months between the first two modules. Reasons for this are specific to the Pacific context. First, given the fact that many health programmes have few staff, it would be hard for programmes to send participants away from their posts for what would be almost 3 weeks. Second, we found that the level of previous exposure to data management and research methods, as well as educational level, was on average lower than in participant cohorts in other courses. Written feedback from participants, collected using a structured questionnaire, showed that the majority (n = 9, 75%) found Module One extremely intense. This finding was confirmed during the verbal feedback session, where all participants mentioned challenges, with some describing how they had experienced a steep, but fulfilling, intellectual and personal learning curve. They appreciated the time given to recover and digest the learning from Module One before embarking on data management and statistics in Module Two. Third, it enabled three points of contact between participants and mentors instead of two, with two before any data collection. This fitted better with the Pacific context, where it is often said that face-to-face verbal communication is preferred over electronic contact.⁴¹ Fourth, the gap allowed most participants to obtain ethics approval before attending Module Two, so that they could begin data collection promptly. Fifth, it allowed participants to check data sources thoroughly before Module Two, so that the variables entered into EpiData (EpiData Association, Odense, Denmark), the analysis software used in Module Two, corresponded to the data sources. Finally, the gap allowed the participants the time to communicate with their colleagues and supervisors about their research protocol and the philosophy of OR. This served to consolidate support, as well as to set up the expectation for participants to transfer information and lessons learned from the course to their colleagues.

Intensified support for the Module Two milestone

Originally, the milestone for Module Two required participants to send evidence of data collection and analysis to the module and course organisers several weeks before Module Three. More recently, participants had to start creating their EpiData files and ‘dummy’ tables and figures (draft versions with no data) during the module and send them 2 weeks after Module Two.⁴⁴ We piloted a new way of running this module: time was allocated for helping participants to finish (or almost finish) their EpiData files and dummy tables and figures during the module. Participants were also required to present them for discussion at the plenary on the last day. This allowed for more intense and personalised technical support during the week, as well as peer review and discussion about analysis for each project individually at the plenary. It also helped participants to start thinking early on about how to prioritise and present their data so that decision makers would pay attention to their findings and recommendations.

Pacific operational research symposium after Module Three

The convention has been for Module Three to finish on day 5 with a plenary session in which participants present their 95% finished papers. For the Pacific course, we asked participants to present their work at a Pacific OR symposium, following course completion. The learning objectives were to distil the essentials from a paper into a Powerpoint presentation using a pre-defined template; practise presenting and responding to questions from a wider audience, including academics, health care workers and decision makers; and prepare for dissemination of their results and policy discussions on their return to post. We also aimed to introduce participants to students, staff and other contacts who work on similar topics in the Pacific and to expand awareness about the course in the region.

To develop the presentations, we followed the same procedure as for paper writing: the participant would sit down with the mentors and edit it side-by-side and/or e-mail it to mentors and receive written feedback. A 2-h practice session in four groups was held on the day before the symposium. Participants familiarised themselves with the technical aspects of presenting, checked the length of their presentations, clarity of the slides and verbal explanations, and practised answering questions from the audience. A senior expert opened each session; three to four participants presented their research findings and policy and practice recommendations, and the session finished with a panel discussion, allowing presenters to respond to audience questions.

PACIFIC OR COURSE OUTPUT

Trained participants

A total of 36 participants enrolled in the three OR courses; of the two courses that have been completed (n = 24), 17 (71%) passed all course milestones. Seven participants did not complete the course: six did not meet the milestones, and/or had to leave the course for family or professional reasons, and one participant died. The participants came from eight PICTs; the majority (n = 24, 67%) were from Fiji, and all had participated in the national Fiji courses (Table 5); 61% of the course participants were female.

TABLE 5.

Country of residence of participants enrolled in operational research courses in the Pacific, 2011–2013

The participants were working in a range of professions; the majority were nurses and physicians (n = 24, 67%; Table 5). A wide range of other professional groups were represented, reflecting the multidisciplinary nature of the training (Table 6). All participants were working in the local health system at the time of commencing the course. The participants undertook research projects on a diverse range of locally relevant topics, including TB, NCDs, the association between TB and diabetes, sexually transmitted infections, including human immunodeficiency virus (HIV), laboratory and diagnostic issues, sepsis, nosocomial infections, congenital rubella and human resources for health (Table 7).

TABLE 6.

Employment profile of participants enrolled in operational research courses in the Pacific, 2011–2013

TABLE 7.

Topics of research projects of those enrolled in operational research courses in the Pacific, 2011–2013

Papers

Participants submitted a total of 19 papers to journals, 10 from the Fiji course and 9 from the Pacific course. All participants were the first author on their papers, reflecting their contribution to their research project and the write-up of their papers. To date, a total of 18 of these are either in press or have been published in a range of peer-reviewed scientific journals.^45–52 The outputs of the two completed courses are summarised in Table 8. Table 9 lists all research projects developed by participants of the three courses.

TABLE 8.

Output from two completed operational research courses run by The Union/MSF and partners in the Pacific, 2011–2013

TABLE 9.

Publication status of participants' projects developed during Pacific operational research courses, 2011–2014

TABLE 9.

(continued)

Trained junior and senior mentors

We endeavoured to train three junior mentors who had been participants in the first Fiji course. Training consists of being paired with a senior mentor and facilitating at least one complete course run by the Union/MSF core team members. Of the three selected, one emigrated before the training opportunity, another has been trained on only one Fiji module so far, and another has been able to train in two Fiji modules but only one of the three Pacific modules. While having local mentors fulfils a very important criterion for local ownership and uptake, this does depend on whether they can be released by employers to complete their training.

Ten new senior mentors have been introduced to the OR course and are now considered part of the pool of senior mentors for any future courses. Although we selected mentors on the basis of their individual experience, skills and approach, we also deliberately created a network that includes a variety of academic, governmental organisations and NGOs, as well as professional disciplines and teaching and mentoring styles.

Increased awareness about OR and expansion of the network

Feedback obtained from participants indicated that the course addresses their individual needs as well as the needs of their institutions and the health programmes in which they are involved. Furthermore, a number of institutional links have been formed and/or reinforced through the courses, with increased awareness among their staff about The Union/MSF approach to OR. In addition, the OR course has been showcased in regional meetings such as the Pacific Stop TB meeting, and is being considered as a component of public health training in the region.

THE FUTURE

Translation of results into policy and practice

Following publication of their research, participants are encouraged to work collaboratively with national Ministries of Health and partner technical organisations to ensure that the results are applied locally and, where appropriate, that they influence the refinement or development of local policies. Dissemination of results is usually the first step. Following the Fiji 2011–2012 course, the Fijian institutions involved organised a research dissemination day in the presence of the Minister of Health and other key policy makers. Several participants from the recent Pacific course have reported that they used their Powerpoint presentations prepared during Module Three to present to colleagues and policy makers on their return. Regarding policy change, the Fiji NTP has recently indicated that several recommendations from the research conducted in the 2011–2012 course have been incorporated into its work plan, including regular cross-checking between laboratory and treatment registers to prevent initial loss to follow-up of smear-positive TB patients, the development of a collaborative TB-diabetes framework and bi-directional screening for TB and diabetes, and improvements in sputum collection and smear examination and related training and reporting. Over the longer term, we aim to document whether changes in policy and practice are being reported in the countries involved.

Ongoing evaluation of output

All participants of the Union/MSF courses are contacted annually to complete a brief questionnaire about their activity related to OR. We are thus able to follow participant output beyond the course. Our aim is to ensure that all participants continue to conduct and publish OR, to pass on understanding about OR and skills to others, and to contribute to the expansion of a regional network of people using OR to improve health outcomes in the Pacific.

Further courses

Requests to hold further national and regional courses have come from PICTs, but the logistics and funding for such courses need to be determined. Potential funding sources include international donors, philanthropic foundations, health research institutes and co-funding between NGOs or other partners that are focused on specific health topics such as HIV, TB and lung health, diabetes, maternal and child health, etc.

Recognition and integration

We are investigating how the OR course might be recognised by academic and governmental institutions such as universities and Ministries of Health. In January 2013, The Union and MSF OR units joined forces with the WHO Special Programme for Research and Training in Tropical Diseases. The three organisations have developed a blueprint for training public health programme staff under the Structured Operational Research Training Initiative (SORT-IT). This recognition of the course should make it even more appropriate for institutions to consider crediting it towards continuing professional education and degrees such as Masters in Public Health. The major organisations currently involved are keen to see how the OR courses can facilitate enhanced collaboration between the academic, non-government and health care implementation sectors.^53,54

Sustainability

It is hoped that the involvement of key individuals and institutions from the region will help to expand OR exponentially. Regionally based mentors can be involved in various ways in promoting OR and providing longer-term support to the graduates of the course. Furthermore, OR may be incorporated into regional policies, strategies, guidelines and funding applications. The involvement of local mentors and experts should facilitate the ongoing development of a cadre of Pacific-based OR experts who can collaborate on future OR projects and continue to apply this knowledge to influence local policy and practice. Holding Module Three and the Pacific OR symposium at the University of Auckland was designed to link participants with relevant university contacts, and motivate them to continue applying their OR skills once they return to their programmes. At the national and regional levels, the OR courses have aimed to promote applied learning and networking that can be further developed and utilised once the course has been completed.

The cost effectiveness of such training as compared to other methods of OR capacity building is an important factor to consider. While we have not carried out a cost-effectiveness analysis, and have not yet been able to monitor long-term outcomes for our participants, we think that our approach to OR capacity building is one of several that can be used. Other approaches might include online learning, one-off training modules that cover distinct pieces of the research process, academic degrees and onsite training in OR during research led by external researchers. All approaches to OR capacity building have cost implications and different outputs, with our approach likely costing more than online learning, but significantly less than an academic degree.

CONCLUSION

Improving OR capacity in the Pacific requires a multipronged approach, with a focus on sustainability and capacity building of local staff who work in health programmes. The Union/MSF OR course, which has a focus on building OR capacity in local staff who conduct and publish OR projects, would seem to complement initiatives that provide Masters and Doctoral scholarships for overseas study.

The Fiji and Pacific OR courses have built capacity for a total of 36 national health programme staff and promoted collaboration between a variety of academic, technical and governmental partners. In addition, junior mentors have been trained and a network of people interested in OR in the region has been developed. Annual evaluations will determine the long-term benefits and impacts of this training. We believe that this model is appropriate for the Pacific context and that it will contribute to the realisation of health programme objectives in the region.