Strengthening anthrax outbreak response and preparedness: simulation and stakeholder education in Namisindwa district, Uganda

Abstract

Background

Anthrax is a zoonotic disease caused by Bacillus anthracis that poses a significant threat to both human health and livestock. Effective preparedness and response to anthrax outbreak at the district level is essential to mitigate the devastating impact of the disease to humans and animals. The current diseaae surveillance in animals and humans uses two different infrastructure systems with online platform supported by established diagnostic facilities. The differences in surveillance systems affect timely outbreak response especially for zoonotic diseases like anthrax. We therefore aimed to assess the feasibility of implementing a simulation exercise for a potential anthrax outbreak in a local government setting and to raise the suspicion index of different district stakeholders for a potential anthrax outbreak in Namisindwa District, Uganda.

Methods

We conducted a field-based simulation exercise and a health education intervention using quantitative data collection methods. The study participants mainly members of the District Taskforce (DTF) were purposively selected given their role(s) in disease surveillance and response at the sub-national level. We combined 26 variables (all dichotomized) assessing knowledge on anthrax and knowledge on appropriate outbreak response measures into an additive composite index. We then dichotomized overall score based on the 80% blooms cutoff i.e. we considered those scoring at least 80% to have high knowledge, otherwise low. We then assessed the factors associated with knowledge using binary logistic regression with time as a proxy for the intervention effect. Odds ratios (ORs) and 95% Confidence intervals (95%CI) have been reported.

Results

The overall district readiness score was 35.0% (24/69) and was deficient in the following domains: coordination and resource mobilization (5/16), surveillance (5/11), laboratory capacity (3/10), case management (4/7), risk communications (4/12), and control measures (4/13). The overall community readiness score was 7 out of 32 (22.0%). We noted poor scores of readiness in all domains except for case management (2/2). The knowledge training did not have an effect on the overall readiness score, but improved specific domains such as control measures. Instead tertiary education was the only independent predictor of higher knowledge on anthrax and how to respond to it (OR = 1.57, 95% CI = 1.07–2.31). Training did not have a significant association with overall knowledge improvement but had an effect on several individual knowledge aspects.

Conclusion

We found that the district’s preparedness to respond to a potential anthrax outbreak was inadequate, especially in coordination and mobilisation, surveillance, case management, risk communication and control measures. The health education training intervention showed increased knowledge levels compared to the pre-test and post-test an indicator that the health education sessions could increase the index of suspicion. The low preparedness underscores the urgency to strengthen anthrax preparedness in the district and could have implications for other districts. We deduce that trainings of a similar nature conducted regularly and extensively would have better effects. This study’s insights are valuable for improving anthrax readiness and safeguarding public and animal health in similar settings.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12917-024-04289-0.

Background

Anthrax is an infectious zoonotic disease caused by the bacterium Bacillus anthracis [1]. It is primarily a disease of animals, especially herbivores such as cattle, sheep, and goats, but it can also affect humans; causing 20,000–100,000 human infections annually worldwide. Approximately 1.1 billion livestock live in areas predicted to be at risk for anthrax globally [2]. Anthrax outbreaks have gross implications on international trade and tourism, requiring an international declaration to the Food Agricultural Organization (FAO) and the World Organization for Animal Health (WOAH) [3].

Since 2017, Uganda has prioritized the enhancement of its syndromic surveillance systems for zoonotic diseases through the implementation of the International Health Regulations (IHR) 2005 [4]. Given the economic implications of such zoonotic diseases, different countries including Uganda set up surveillance infrastructure for tracking and reporting them [5]. For example, Uganda has national one health secretariat with a mandate to coordinate one health activities in the country. There are two strategic committees (Incident Management Team Committee and National Taskforce) established to discuss and make decisions for zoonotic diseases surveillance and control interventions [5]. At ministry level, Ministry of Agriculture, Animal Industries and Fisheries (MAAIF) has animal health epidemiological unit based at National Animal Diseases Diagnostics Centre (NADDEC) aimed at tracking zoonotic diseases across the country [6]. There are reporting platforms for tracking such diseases including mechanisms to report to international levels like to WHO, and WOAH. For Ministry of Health (MoH), they have a fully-fledged department of surveillance, resident fellows under Uganda National Institute of Public Health who support the country to conduct field epidemiology activities from time to time. The fellows undergo a six months’ field epidemiology training programme organized by Makerere University School of Public Health and funded by United States Centres for Disease Control and Prevention (US CDC). There are interactive dashboards at the MoH to display the analytics made for each disease.

There is also a centralized District Health Information Management Systems (DHIS 2) where there is weekly tracking of different disease reports across the country through the M-Trac reporting system [7]. This tracking and reporting is done at sub national level by district biostatisticians supported by records officers employed by government [8]. This tracking is aimed at identifying trends that may be flagged out for early detection of disease outbreaks. To inform the tracking and reporting, there is a list of diseases considered to be priority and indeed followed up from sub-national level to national level of which Anthrax is among these priority diseases [5]. There is also a National Action Plan for Health Security that is always updated after four years as per requirements of International Health Regulations (IHR), 2005, Anthrax is also listed in this plan [5, 9].

Despite Anthrax’s significance, consistent outbreaks, deliberate anthrax surveillance and control programs remain limited [10, 11]. Existent research pinpoints at least 30 Ugandan districts as potential hotspots for anthrax outbreaks [12]. Since this publication, unforeseen outbreaks have occurred in districts of Namisindwa, Kween, and Bududa, further underscoring the existing vulnerabilities in the nation’s surveillance infrastructure. In light of these outbreaks, there’s a palpable need for thorough assessments of the country’s anthrax outbreak readiness. Outbreak simulations are a useful tool for assessing readiness. For instance, simulations in the United States are organized from time to time not only to understand readiness, but also to- find weaknesses and strengthen systems. For example, in 2005, a team of scientists modelled an anthrax outbreak in the United States that was critical in developing Standard Operating procedures (SOPs) for future simulations [13].

Although Uganda has successfully conducted both table-top and field-based simulation exercises for diseases such as Ebola and COVID-19, protocols for these simulations are rarely documented. Yet, these simulations could still be replicated to enhance response mechanisms to diseases. The Africa One Health University Network (AFROHUN), is well-positioned to leverage its community and Ministry ties to conduct such simulations in Uganda to measure anthrax outbreak readiness. We hypothesized that the simulation exercise could guide the assessment of the operational readiness of the district and community to appropriately respond to an anthrax outbreak, which will strengthen field One Health preparedness and enhance zoonotic disease surveillance capabilities. This study aimed at evaluating the preparedness of Namisindwa district to respond to an anthrax outbreak, seeking to illustrate crucial areas necessitating support and intervention.

Methods

Study site

The study was conducted in Namisindwa District, Eastern Uganda. The district is bordered by Bududa District to the north, Republic of Kenya to the east and south, Tororo District to the south-west, and Manafwa District to the west. The district headquarters at Bupoto are located approximately 40 km (25 miles), by road, south-east of Mbale, the largest city of the sub-region. The coordinates of the district are 00°49′N 34°23′E. Most people who stay in this district practice small-scale agriculture and rear different livestock including cattle, goats, sheep, and horses as their livelihood. This district has multiple porous border points with Kenya which promotes illegal movement of cattle from Kenya into Uganda (Fig. 1). In this study area, there are cultures/practices that could increase the risk of anthrax spread like selling and buying of animals, slaughter practices without personal protective equipment, consuming of un-inspected meat or eating of dead carcasses.

Fig. 1.

Map of Uganda showing the location of Namisindwa District [14]

Study design

We conducted a field-based simulation exercise and a health education intervention using quantitative data collection methods. A one-day field-based simulation exercise, and two-day health education training intervention. The outcome measure for the simulation exercise objective was the response readiness for the district. The simulation exercise was co-designed by Ministry of Health, Mbale Regional Emergency Operations Centre (REOC), National One Health Coordination Office, WHO, and MAAIF. Representatives from each body participated in the execution of the study. We used a WHO simulation matrix to score the district in the different aspects of outbreak response including; coordination, risk communication, resource mobilization, laboratory and surveillance, and case management (Appendix I).

Study participants

The study had two fronts and was influenced by the aim. For simulation exercise the particpants were district and community level members whereas for health ediucation intervention, the participants were district healthwortkers and veterinary staff as well as community health workers and community animal health workers. The study participants were purposively selected given their role(s) in the disease surveillance and response at sub-national level. These were mainly district officials who form part of the District Taskforce (DTF). Just as national level where there exist a high-level one health strategic committee from different agencies/ministries/departments, there is also a district-level one. This taskforce is headed by the Head of State (President)’s representative called Resident District Commissioner (RDC). These taskforces were highly pronounced during the COVID-19 pandemic. For zoonotic diseases like Anthrax, the secretary to the committee is the technical person specifically District Health Officer. Part of the members are both technical, political, religious, cultural leaders. These also support in resource mobilization in managing any potential disease outbreak. For this Anthrax, we targeted them to see how they would manage a crisis of Anthrax outbreak. In bid to extend to community-level (lower administrative units called sub counties) where outbreaks occurs, community level local leaders were also considered. These community leaders were selected from selected subcounties. The team specifically went two subcounties nehgbouring/boardering those from Buduuda where the Anthrax outbreak occurred the previous year. The study investigation team monitored and recorded reactions and responses in the different places for about six hours (from 10 a.m. to 4 p.m.). Efforts were made to include surprise elements, whispering the rumor to the local chairpersons of Magale Town Council where slaughtering took place to test adaptability and actions at the community level. The debrief was conducted the following day after the investigation team had met at Mbale RPHEOC to compare notes of what was observed at different places in the field.

Sample size

The sample size for the baseline and endline knowledge assessment and awareness was determined using a Kish-Leslie formula, 1965 [15] The Z-value was 1.96 at 95% confidence interval giving a sample size of 100. The knowledge level awareness estimated to be 50% because no previous studies had measured knowledge on Anthrax preparedness in Uganda. Both baseline and endline knowledge assessment used pre-tested semi-structured questionnaires (Appendix IV) with data from a checklist adopted from an emergency Anthrax and response plan from the 2005 Minosota Anthrax simulation study (Brett Elkin, 2013). Research assistants were trained on the study protocol and administered the questionnaires using the Kobocollect App [16]. The questionnaires and protocols were pretested in Bugema subcounty in Mbale, an area that mimics the study area in terms of culture and economy.

Data collection procedures

The simulation was initiated in August 2023 when the investigation team requested the District Production Office, of Namisindwa District, to bring them to the location of the anthrax outbreak. This outbreak was believed to have originated at the home of a farmer in Mukululu village, Bunamwandu parish, Tsekululu subcounty. In the same month the farmer purchased a male calf from Busulwa market that later presented severe bleeding through the skin and subsequently died two weeks after purchase. The carcass was cut into pieces prior to burial and the ground was saturated with chlorine. The investigation team took a soil sample from the burial site as well as blood samples from the two extant cows on the farm.

The blood samples were immediately sent to the NADDEC laboratories for urgent analysis and the District Veterinary Officer (DVO) was notified. The soil samples were to be shipped to the Ministry of Water and Environment laboratory for analysis but we did not ship them due to logistical reasons. The next morning at 8:45am EAT, Mbale REOC sent a message to the National Public Health Emergency Operations Centre (NPHEOC) through the event-based surveillance system regarding the death of 15 cattle from Namisindwa District. NPHEOC relayed the message via WhatsApp where different national and district officials, including the Namisindwa District Surveillance Focal Person (DSFP), share health-related information/ updates in the country. The investigation team then set for the field at 9:00am, arriving at their various locations at around 10:00am to observe the responses to the rumor.

The investigation team was divided into three teams of four: the first team went to Magale Town Council for community observation, the second team went back to Tsekululu Subcounty for district observation, and the third team went to the district headquarters to observe the DTF’s reaction to the rumor. The investigation team members in the community interacted with randomly selected individuals about the rumor. Efforts were made to include surprise elements, whispering the rumor to the local chairpersons of Magale Town Council where slaughtering took place to test adaptability and actions at community level. The investigation teams monitored and recorded reactions and responses until 4pm, roughly six hours. The main interaction was observed by the District Surveillance Focal person (DSFP) coordinating others. However, a wider discussion was seen during the debrief was conducted with the DTF the following day which we also recorded, and it informed our qualitative paper which is not part of this manuscript.

A workshop-based approach was used in designing the health education training intervention with selected district members. The workshop was organized at the district headquarters after the debrief of the simulation exercise. Baseline and endline knowledge and attitude assessment of one health staff at the district was performed. The baseline survey was conducted just before the training commenced, and the endline assessment was done after the training. Both the pre and post- training assessment was written using the pre-tested tool (See appendix II).

Data management and analysis

The password-protected KoboCollect App hosted the simulation and training interventional survey data collected in the field, while. We used R studio statistical software to analyze the quantitative data we collected. We divided the analysis into two parts: simulation and knowledge assessment.

For the simulation, we computed descriptive statistics mean scores for the six domain observations made at the community and district levels. We also created a composite score for each domain by averaging the scores of the binary questions related to that domain. We then calculated the average score per domain and per location (district or community). We also added up the scores of the six domains to get a total score for each location. The results showed the scores of the district and community levels on various aspects of the response to a disease outbreak. The scores were based on a scale of 0 to 69 for the district level and 0 to 32 for the community level, where higher scores indicate better performance. We combined the total scores of the districts and communities to obtain an overall anthrax preparedness score. We then used this score to classify the overall preparedness (Table 1). To the best of our knowledge, this computation and categorization had not been done and so we innovated it. There was two meetings between STW and AWW to build intellectual consensus before finalization.

Table 1.

Grading of simulation scores

Location	Prepared	Moderately prepared	Insufficiently prepared
District	52–69	35–51	0–34
Community	24–32	16–23	0–15
Overall preparedness (score range)	76–101	50–74	0–49

For the knowledge assessment, the questionnaire had some questions to assess the knowledge using different questions. At the point of baseline/pre-test, we had 100 participants who were trained. These participants attended baseline as well but missed out at the time of pretest, however for regression analysis, only 96 participants were considered (complete case analysis).

The variable for “Experience in years” referred to the duration staff have been practicing and working on clinical management of zoonotic diseases including Anthrax. This was specifically important as it would influence the participants’ knowledge on Anthrax. Using R, we summarized participant demographic characteristics as median (IQR) if continuous, otherwise proportions disaggregated by study period (pre-training and after training. For knowledge and attitudinal variables, we compared pre-intervention estimates (proportions) with post- training evaluation using McNemar tests. We combined 26 variables (all dichotomized) assessing knowledge on anthrax and knowledge on appropriate outbreak response measures into an additive composite index. We then dichotomized overall score based on the 80% blooms cutoff i.e. we considered those scoring at least 80% to have high knowledge, otherwise low. We then assessed the factors associated with knowledge using binary logistic regression with time as a proxy for the intervention effect. All variables judged as epidemiologically relevant in explaining knowledge were considered covariates. Odds ratios (ORs) and 95% Confidence intervals (95%CI) have been reported.

Results

Simulation

The district obtained an overall score of 24 out of 69 (34.8%), performing poorly in each of the six domains: coordination and resource mobilization (5/16), surveillance (5/11), laboratory capacity (3/10), case management (4/7), risk communications (4/12), and control measures (4/13). The community obtained an overall score of 7 out of 32 (21.8%), performing poorly in five of the domains: coordination and resource mobilization (1/7), surveillance (1/6), laboratory capacity (1/6), risk communications (1/4), and control measures (2/7), but received a score of 2 out of 2 in case management. With a combined score of 31, the district and community performed with major challenges. Further exploration of the situation enabled us to develop a strengths, weakness, opportunities, and threats (SWOT) assessment for the district (Appendix III) (Table 2).

Table 2.

Preparedness levels by domain at the community and district level

District level (3 observations)	Scores, n/N (%)
Coordination and resource mobilization	5/16 (31.3)
Surveillance	5/11 (45.5)
Laboratory capacity	3/10 (30.0)
Case management	4/7 (57.1)
Risk communication	4/12 (33.3)
Control measures	4/13 (30.8)
Total score	24/69 (34.8)
Community level (9 observations)
Coordination and resource mobilization	1/7 (14.3)
Surveillance	1/6 (16.7)
Laboratory capacity	1/6 (16.7)
Case management	2/2 (100.0)
Risk communication	1/4 (25.0)
Control measures	2/7 (28.6)
Total score	7/32 (21.8)

Health education/training intervention

Socio-demographic characteristics of participants

The pre and post-assessment respondents were 52.0% female, highly educated (67.9%), and working in health-related fields (37.4%). The most common occupation among the respondents was health worker (37.9%), followed by animal health worker (31.8%) and community health worker (CHW) member (12.8%). Most of the respondents were employed (65.8%), followed by self-employed (26.5%), and the most common experience category was 1 to 5 years (40.3%), followed by 6 to 10 years (30.6%), 11 to 15 years (15.8%) and over 15 years (13.3%) (Table 3).

Table 3.

Socio-demographics of participants

Characteristic	Pre-training, N = 96	Post-training, N = 100
Sex of respondent
Female	49(51.0%)	53(53.0%)
Male	47(49.0%)	47(47.0%)
Median age in years
Median (IQR)	34 (30, 42)	35 (31, 46)
Age in years (categorical)
20–29	20 (20.8%)	17 (17.0%)
30–39	45 (46.9%)	43 (43.0%)
40–49	18 (18.8%)	23 (23.0%)
At least 50	13 (13.5%)	17 (17.0%)
Highest level of education
Primary	7 (7.3%)	12 (12.0%)
Secondary	20 (20.8%)	24 (24.0%)
Tertiary	69 (71.9%)	64 (64.0%)
Occupation
Animal health worker	27 (28.1%)	35 (35.4%)
community leader	10 (10.4%)	5 (5.1%)
Community member	10 (10.4%)	5 (5.1%)
Health worker	43 (44.8%)	31 (31.3%)
Others	2 (2.1%)	2 (2.0%)
VHT member	4 (4.2%)	21 (21.2%)
Employment status
Employed	64 (66.7%)	65 (65.0%)
Self employed	27 (28.1%)	25 (25.0%)
Unemployed	5 (5.2%)	10 (10.0%)
Median years of experience
Median (IQR)	7 (2, 11)	7 (2, 11)
Experience in years*
1–5	38 (39.6%)	41 (41.0%)
6–10	32 (33.3%)	28 (28.0%)
11–15	15 (15.6%)	16 (16.0%)
Over 15	11 (11.5%)	15 (15.0%)

*Experience in years refers to the duration staff have been practicing and working on clinical management of zoonotic diseases including Anthrax

Effect on health education intervention on stakeholders’ knowledge on Anthrax preparedness

The study results revealed that a higher level of education is a significant predictor of knowledge. This means that respondents with tertiary education were more likely to exhibit higher knowledge on anthrax response and how to respond to it than those with primary education (OR 1.57, 95% CI 1.07–2.31, p = 0.023) meaning that the level of education is an effect-modifier of the uptake of the training. Therefore while designing the training there is need to be mindful of the level of education such that relevant content can be packaged for different participants with varying levels of education. The health education intervention did not have a significant association with overall knowledge improvement but had effect on several individual knowledge aspects (Tables 4 and 5).

Table 4.

Knowledge of participants – pre - and post-comparison

Characteristics	Pre-training, N = 96	Post-training, N = 100	p-value
Heard about Anthrax	94 (98%)	100 (100%)	0.239
Sources of Knowledge
Health facility	43 (46%)	62 (62%)
Community members	41 (44%)	48 (48%)
Relatives/Friends	23 (24%)	21 (21%)
Previous infection	2 (2.1%)	1 (1.0%)
Know someone who suffered	5 (5.4%)	6 (6.0%)
School	33 (35%)	35 (35%)
Veterinary Doctor	44 (47%)	57 (57%)
Village health team member	17 (18%)	22 (22%)
Others1	9 (9.6%)	15 (15%)
Cause of Anthrax			0.115
Bacteria	59 (61%)	60 (60%)
Fungi	2 (2.1%)	2 (2.0%)
Don’t know	5 (5.2%)	0 (0%)
Others	5 (5.2%)	8 (8.0%)
Parasites	2 (2.1%)	0 (0%)
Viruses	23 (24%)	30 (30%)
Transmission of Anthrax
Contact with infected animal	67 (70%)	82 (82%)	0.045
Infected animal slaughter	63 (66%)	72 (72%)	0.335
Handling of infected carcass	59 (61%)	74 (74%)	0.060
Soil containing anthrax spores	46 (48%)	70 (70%)	0.002
Eating/ingestion of spore containing meat	69 (72%)	72 (72%)	0.984
Inhalation of anthrax spores	42(44%)	71(71%)	< 0.001
Othertransmission pathways2	6(6.3%)	8(8.0%)	0.634
Signs and symptoms
Skin lesions	73 (76%)	82 (82%)	0.305
Chest discomfort	44 (46%)	71 (71%)	< 0.001
Shortness of breath	48 (50%)	73 (73%)	< 0.001
Fevers and chills	64 (67%)	75 (75%)	0.199
Swelling of the abdomen	51 (53%)	63 (63%)	0.161
Other symmptoms3	10 (10%)	5 (5.0%)	0.154
Preventive and control measures
Burying dead animals	72 (75%)	89 (89%)	0.011
Vaccination	82 (85%)	93 (93%)	0.086
Cooking meat thoroughly	35 (36%)	61 (61%)	< 0.001
Surveillance and inspection of animal slaughter	58 (60%)	75 (75%)	0.029
Centralized animal slaughterhouses	45 (47%)	61 (61%)	0.047
Health education	67 (70%)	70 (70%)	0.975
Wearing PPE during animal slaughter	49 (51%)	68 (68%)	0.016
Hygienic hand washing	41 (43%)	67 (67%)	< 0.001
Other preventive measures4	11 (11%)	6 (6.0%)	0.175
Diagnostic approaches
History of exposure and signs	65 (68%)	74 (74%)	0.332
sampling and laboratory testing	63 (66%)	65 (65%)	0.927
Chest X-ray or a CT scan	32 (33%)	42 (42%)	0.211
All the above	40 (42%)	45 (45%)	0.638
Treatment of anthrax
Antibiotics	55 (57%)	58 (58%)	0.920
Apply herbs to the lesions	10 (10%)	7 (7.0%)	0.396
Painkillers	12 (13%)	17 (17%)	0.375
Medication prescribed by the health worker	61(64%)	72 (72%)	0.205
Disease is self-Healing	3(3.1%)	1(1.0%)	0.361
Don’t know	7(7.3%)	1(1.0%)	0.032
Other treatment options5	5(5.2%)	1(1.0%)	0.113
Knowledge of the anthrax disease	78(81%)	92(92%)	0.027
Burden of anthrax in community	49(51%)	60(60%)	0.207
Risk assessment of area	55 (57%)	65 (65%)	0.268
Surveillance	69 (72%)	72 (72%)	0.984
Laboratory testing	69 (72%)	69 (69%)	0.659
Carcass treatment and disposal	50 (52%)	58 (58%)	0.405
Developing emergency plans	58 (60%)	61 (61%)	0.933
Other prerequisites6	13 (14%)	4 (4.0%)	0.018
Appropriate responses to an anthrax outbreak in humans or animals
Reporting suspected cases to the district RRT	86 (90%)	85 (85%)	0.336
Contact the regional PHEOC As soon as possible	59 (61%)	64 (64%)	0.713
Collecting and transporting samples from suspected cases to designated laboratories	55 (57%)	66 (66%)	0.210
Providing case management and prophylaxis for human cases and contacts	46 (48%)	56 (56%)	0.257
Providing case management and vaccination for animal cases	40 (42%)	32 (32%)	0.161
Implementing IPC measures in facilities and communities	54 (56%)	58 (58%)	0.805
Implementing environmental decontamination of contaminated sites	49 (51%)	61 (61%)	0.160
Communicating accurate and timely information on prevention and control to all stakeholders	58 (60%)	57 (57%)	0.627
Anthrax is a public good disease			0.582
False	50 (52%)	56 (56%)
True	46 (48%)	44 (44%)
Knowledge level			0.823
Low	61 (64%)	62 (62%)
High	35 (36%)	38 (38%)
Anthrax can spread among people through handshake			0.007
False	30 (31%)	15 (15%)
True	66 (69%)	85 (85%)
Anthrax spores can be found in animal skins and hides			0.903
False	15 (16%)	15 (15%)
True	81 (84%)	85 (85%)
Anthrax spores can be found in air			0.020
False	30 (31%)	17 (17%)
True	66 (69%)	83 (83%)
Anthrax spores can survive in the soil for decades			0.118
False	17 (18%)	10 (10%)
True	79 (82%)	90 (90%)
Animal vaccination against anthrax should be annual			0.833
False	6 (6.3%)	7 (7.0%)
True	90 (94%)	93 (93%)
Only gloves are necessary when burying a dead animal of suspected anthrax			0.004
False	45 (47%)	67 (67%)
True	51 (53%)	33 (33%)
Anthrax can spread among people through handshake			0.007
False	30 (31%)	15 (15%)
True	66 (69%)	85 (85%)
Anthrax spores can be found in animal skins and hides			0.903
False	15 (16%)	15 (15%)
True	81 (84%)	85 (85%)
Anthrax spores can be found in air			0.020
False	30 (31%)	17 (17%)
True	66 (69%)	83 (83%)
Anthrax spores can survive in the soil for decades			0.118
False	17 (18%)	10 (10%)
True	79 (82%)	90 (90%)
Animal vaccination against anthrax should be annual			0.833
False	6 (6.3%)	7 (7.0%)
True	90 (94%)	93 (93%)

Note: ¹Training, radios, Farmers, market, political leaders, ²Milk, dogs carrying meat, vets, saliva, sexual intercourse, fomites, microorganisms³Nasal discharge, lymphadenopathy, bleeding, weakness, body aches, mucus from nose/mouth, wounds, appetite loss⁴Quarantine, isolation, calling vets, report to facility, drinking boiled water and wash clothes under the sun, stop taking milk⁵No treatment, vaccines, good diet, call vet, use thermometer⁶Quarantine, reports, funds, Health education, vaccination, vaccines, *26 variables (all binary) were used to calculate knowledge score and based on 80% blooms cutoff, knowledge level was generated as binary variable

Table 5.

Predictors of knowledge on anthrax and response

Characteristic	OR	95% CI1	p-value
Sex of respondent
Female	—	—
Male	0.96	0.84, 1.11	0.624
Age in years (categorical)
20–39	—	—
30–39	0.90	0.74, 1.10	0.306
40–49	0.93	0.71, 1.22	0.614
At least 50	0.93	0.66, 1.30	0.658
Occupation
Animal health worker	0.94	0.67, 1.31	0.702
Community leader	1.04	0.75, 1.45	0.813
Community member	—
Health worker	1.22	0.87, 1.71	0.243
Others	1.30	0.76, 2.20	0.337
VHT member	1.16	0.86, 1.56	0.335
Highest level of formal education
Primary	—	—
Secondary	1.14	0.87, 1.48	0.356
Tertiary	1.57	1.07, 2.31	0.023
Employment status
Employed	—	—
Self employed	1.00	0.80, 1.25	0.983
Unemployed	0.93	0.70, 1.23	0.598
Experience in years (categorical)
1–5	—	—
6–10	1.01	0.83, 1.21	0.948
11–15	1.13	0.90, 1.42	0.308
Over 15	0.93	0.71, 1.21	0.570
Period
Pre-training	—	—
Post-training (estimate of intervention effect)	1.07	0.94, 1.22	0.303

¹CI = Confidence Interval

Regarding the participants’ confidence, comfort, trust, and beliefs regarding anthrax prevention and control in their district, results revealed that the training intervention had a significant positive impact on some aspects of the participant attitudes towards anthrax, while others remained unchanged. For example, the participants became more confident in their knowledge and skills to deal with anthrax cases after the training, as the proportion of those who agreed or strongly agreed increased from 45 to 58% (p = 0.299). They also reported to be more comfortable working with animals or animal products that were infected or suspected of being infected with anthrax, as the proportion of those who agreed or strongly agreed increased from 30 to 49% (p = 0.001). Similarly, they reported being more comfortable working with human cases or contacts that were infected or suspected of being infected with anthrax, as the proportion of those who agreed or strongly agreed increased from 20 to 41% (p < 0.001). On the other hand, the participants reportedly did not show any significant change in their trust in the information and guidance provided by the DRRT, the national authorities or partners, or media on anthrax prevention and control (Table 6) (Appendix II).

Table 6.

Attitudes of participants – pre and post-training comparison

Characteristic	Pre-training, N = 961	Post-training, N = 1001	p-value2
Confident in knowledge and skills to deal with anthrax cases			0.299
Strongly disagree	11 (12%)	13(13%)
Disagree	9 (9.5%)	6 (6.0%)
Neutral	32 (34%)	23(23%)
Agree	24 (25%)	28(28%)
Strongly agree	19 (20%)	30(30%)
Comfortable working with animals or animal products that are infected or suspected of being infected with anthrax			0.001
Strongly disagree	33 (34%)	17(18%)
Disagree	18 (19%)	8 (8.2%)
Neutral	16 (17%)	25(26%)
Agree	11 (11%)	27(28%)
Strongly agree	18 (19%)	20(21%)
Comfortable working with human cases or contacts that are infected or suspected of being infected with anthrax			< 0.001
Strongly disagree	27 (28%)	16(16%)
Disagree	26 (27%)	16(16%)
Neutral	23 (24%)	25(26%)
Agree	7 (7.4%)	29(30%)
Strongly agree	12 (13%)	11(11%)
Trusts Information and guidance provided by the district RRT on anthrax prevention and control			0.096
Strongly disagree	4 (4.3%)	4 (4.2%)
Disagree	12 (13%)	5 (5.2%)
Neutral	24 (26%)	25(26%)
Agree	12 (13%)	25(26%)
Strongly agree	42 (45%)	37(39%)
Trusts information and guidance provided by the national authorities or partners on anthrax prevention and control			0.722
Strongly disagree	3 (3.2%)	2 (2.0%)
Disagree	7 (7.5%)	6 (6.1%)
Neutral	21 (23%)	18(18%)
Agree	19 (20%)	28(29%)
Strongly agree	43 (46%)	44(45%)
Trusts information and guidance provided by the media or social media on anthrax prevention and control			0.149
Strongly disagree	8 (8.6%)	3 (3.1%)
Disagree	17 (18%)	12(13%)
Neutral	25 (27%)	20(21%)
Agree	17 (18%)	24(25%)
Strongly agree	26 (28%)	37(39%)
Believes anthrax is a serious threat to human health and livelihoods in the district			0.107
Strongly disagree	3 (3.2%)	2 (2.0%)
Disagree	10 (11%)	6 (6.1%)
Neutral	21 (22%)	14(14%)
Agree	13 (14%)	27(28%)
Strongly agree	47 (50%)	49(50%)
Believes anthrax can be effectively prevented and controlled in the district			0.364
Strongly disagree	3 (3.3%)	2 (2.1%)
Disagree	8 (8.8%)	3 (3.1%)
Neutral	16 (18%)	16(16%)
Agree	20 (22%)	30(31%)
Strongly agree	44 (48%)	46(47%)

Discussion

We aimed to assess the feasibility of implementing a simulation exercise for a potential anthrax outbreak in a local government setting and to raise the suspicion index of different district stakeholders for a potential anthrax outbreak in Namisindwa District, Uganda. We found the Namisindwa district was not prepared for a potential anthrax outbreak, and they performed their outbreak response with many challenges. Specifically, there were challenges in surveillance, laboratory, risk communication and coordination domains. We also found the two-day health education intervention did not substantially impact the participants’ knowledge levels on anthrax preparedness and response. However, the written test had limitations as it may not exhaustively assess impact on the participants knowledge levels but also, we missed two clinicians who left prematurely before the final day.

The simulation exercise was conducted at the district headquarters and community level to understand how district taskforce members respond to an alert of a potential anthrax outbreak. Their poor performance in response explains how insufficiently prepared these lower local governments are in terms of disease response [7, 17]. There are limited approaches in risk financing, staffing of surveillance officers, coordination and planning and preparedness, making such districts vulnerable to responding to outbreaks [17, 18]. This is concerning because Uganda employs a decentralized governance system where such lower local governments are expected to be self-reliant [19, 20]. Further explanation for this unreadiness could be that the disease response was previously championed and led by ministry of health - level epidemiologists. This means that whenever there was suspected outbreak, the national-level epidemiologists come to investigate and to provide guidance for control of disease outbreaks [21, 22]. Fortunately, there have been deliberate efforts by MoH and implementing partners to build local-level capacities. The current efforts include activating regional emergency operations centres like Mbale, Mbarara, Gulu, Arua, Mubende, Masaka, Lira and Fort Portal [23]. There are also occasional trainings for district surveillance focal persons to build their capacity for managing these threats. Indeed, these investments at the local level could explain the district - level strengths that we identified in Appendix I.

Although we expected robust systems to have been established at the district level given the recent investments through Corona Virus disease 2019 (COVID-19) management and response, we are cognizant that anthrax is not among the diseases for which routine surveillance is conducted by both at MoH and MAAIF. Routine district-level surveillance is conducted for human diseases such as malaria, brucellosis tuberculosis, and respiratory tract infections, and animal diseases such as foot and mouth disease, while rabies surveillance is captured via animal bite cases. There are surveillance gaps in the routine surveillance for some diseases like Malaria or those that present with similar clinical presentations [24–27]. For example, Uganda being a low-income country, malaria surveillance in such lower level is by Rapid Diagnostic Tests (RDTs) which may provide false positives for such diseases but also minimal testing that is conducted due to logistical implications. All these aspects negatively affect early preparedness. Additionally, although anthrax disease is among the seven prioritized diseases under the National Actional Plans for Health Security (NAPHS) and emphasized at the national level, the DTF was not aware of this.

At the community level, specifically in Magale Town Council and Tsekululu Subcounty, we also noted similar vulnerability in terms of response. This clearly shows need for more investment in activating various response domains at different levels. Although different subcounties or town councils had staff responsible for surveillance, there was no coordination for implementing certain pillars at such levels and there was little awareness within the community of DTF. As such, our study results align with a recent simulation exercises conducted in Busia District by Mbale REOC and Baylor Uganda which showed that the district was insufficiently prepared for a potential Ebola outbreak at the Malaba border. Similar findings were observed during the Ebola simulation exercise at Entebbe International Airport in 2019, showing our vulnerability in surveillance and emergency services for any Ebola outbreak [28]. The study findings disagree with Kyobe et al. 2023’s viewpoint that Uganda was prepared for any public health threat. Their viewpoint was piggy backing on the spillover investments from COVID-19, but we found that in the first 8 h after the “purported” alert, the district-level systems were not fully activated to respond, which should worry us as a country [29]. We, however, could agree with Kyobe et al. 2023 that national level alertness is commendable, given routine national taskforce meetings through PHOEC, but this needs to be scaled down to the district and community levels. We also agree that MOH’s surveillance system is more robust than MAAIF’s systems but similar efforts must be made to harness MoH’s strengths through the One Health lens [23, 28, 29]. Through the district interactions, it was clear that the One Health call was still theoretical with minimal implementation during our simulation exercise. This further calls for a deep evaluation of the One Health framework, which is being championed through the National One Health Platform. This could help identify the current strengths and improvement areas to prepare the country for emerging and re-emerging public health threats. The efforts on Joint External Evaluations of the IHR/NAPHS should be extended to implementing the National One Health framework.

These gaps identified calls for further investment in building capacity at different levels by national agencies like the National PHEOC at the MoH and NADDEC under MAAIF [6]. This would make them more robust to respond and manage any potential threats. There is also need to create a vote/budget/funding line for preparedness fund at all levels ranging from the national level to the community level, this would facilitate activities should emergencies occur. Deliberate efforts in activating all domains like coordination and logistics, risk communications, laboratory, and case management at different levels should be highly pronounced.

The health education training intervention showed increased knowledge levels compared to the pre-test and post-test. This is a great indicator and aligns well with our hypothesis, where we articulated that the health education sessions could increase the index of suspicion. Risk communication is critical in disease outbreaks and management, and stakeholders at different levels must invest in this domain [28, 30–32]. Although, in our study, the training did not have a statistically significant impact on knowledge level, we attribute this to (a) the short timeframe between pre and post-assessment would favour recall ability. (b) the short period of intervention as we only conducted in two days for different stakeholders, and (c) the training resources were not translated in the local language. Interestingly, level of participant education was statistically significant in our results analysis. However, it should be noted that different parameters on knowledge greatly improved as it is witnessed between pre-test and post-test. This should inspire more training intervention as a way of increasing awareness for anthrax in such study settings. We used predesigned materials from MOH through Mbale REOC. The health education interventions coupled with stakeholders’ level of education could play a significant role in increasing awareness and raising suspicion index about Anthrax preparedness. Although we notice that the reported knowledge about anthrax being high (ever heard about anthrax) but this could potentially be due to the fact that we had just concluded a simulation exercise in the area and this response could be bias. We suggest that readers should interpret this response with caution.

Our study has several strengths; [1] to the best of our knowledge, this is the first field-based simulation exercise conducted in Africa, particularly in Uganda. We have developed the methodology tools that other scholars could leverage and replicate to promote Anthrax simulations worldwide although we caution that they need to be replicated and adapted to fit local context and language. This could be continuously done in a bid to identify gaps and strengths to inform planning and investment [2]. Our simulation exercise was conducted both at community and district levels, this enabled us to have a 360-degree view of how the lower local government would respond to a potential anthrax outbreak [3]. This simulation exercise was jointly implemented by stakeholders critical in disease preparedness like NADDEC, Mbale REOC, National PHEOC, National One Health Platform, National Public Health Institute-Ministry of Health. This will foster the uptake of observed recommendations for their action. We conducted both quantitative and qualitative evaluations, which enabled us to obtain explanations to what we were observing. The narratives have helped us to appreciate areas for improvement [5]. We alerted the Uganda Police Force -Namisindwa District, Commissioner-Animal Health, MAAIF ahead of this exercise and helped us manage any potential anxiety that could emerge [6]. We also pre-tested our tools at Bugema ward, Mbale City, enabling us to improve our tools before actual data collection [7]. The investigation team was multi-disciplinary with veterinary, public health, epidemiologists, social scientists, medical officers, and nurses. This enabled us to have a wider scope of observation of the strengths and weaknesses of the district. However, we also documented several limitations; [1] anthrax being a notifiable disease, we were cognizant of potential threats to tourism and trade as it could affect the economy grossly. This made us to unblind the district officials within 6 h of execution [2]. We had a network challenge in the field, we had planned three signals during the actual day of the simulation, however, as a team, we did not coordinate fully as our telephone networks went off between the team of Tsekululu and district levels. The only signal that worked was the text message sent on WhatsApp at 7:18 am. We did not execute more than two, which could enable us to probe further [3]. The investigation team were trained virtually about the actual simulation exercise; some did not follow quite well due to internet and network challenges. This could have affected the actual execution of our project in the field [4]. lack of appropriate translation of the training tools into lumasaba (the local language) although we were working with learnt populations who knew English but for purposes of community component, there was need to translate the tools [5]. We are cognizant of the bias caused by conducting a knowledge assessment/intervention just after the simulation exercise and using this measure on the effectiveness of the intervention [6]. Due to logistical challenges, we did not submit the soil samples to ministry of water and environment, it would have been interesting to see the possibility of detecting anthrax spores as they are known to stay longer in the soil. It would have shown the hidden presence of anthrax in the environment.

Conclusion

In conclusion, we found Namisindwa District unprepared for a potential anthrax outbreak, and they performed their response with many challenges, resulting in a score of 31%. Furthermore, the two-day health education intervention did not substantially impact the participants’ knowledge levels.

We recommend that the MoH and MAAIF should continue organizing such simulation exercises since they help to unearth strengths and areas for improvement. A longer period of health education sessions could be designed for improving the knowledge of stakeholders. Specifically, Namisindwa District should; [1] develop an anthrax response plan and share it with the DTF and DRRT members [2], strengthen the district psychosocial support for victims, suspected human cases and farmers that lose their animals [3], recruit or assign a district information officer to support the management of misinformation circulating in the community.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

AFROHUN

Africa One Health University Network

CHW

Community Health Worker

COVID-19

Corona Virus Disease 2019

DTF

District Taskforce

DVO

District Veterinary Officer

FAO

Food Agricultural Organization

IRB

Institutional Review Board

MAAIF

Ministry of Agriculture, Animal Industries and Fisheries

MakSPH

Makerere University School of Public Health

MoH

Ministry of Health

NADDEC

National Animal Diseases Diagnostics Epidemiological Centre

NFT

National Taskforce

REOC

Regional Emmergency Operations Centre

UNCST

Uganda National Council of Science and Technology

WOAH

World Organisation for Animal Health

Author contributions

AWW, EA, WKK, LM conceptualized the day, drafted the proposal and secured funding. LNN; coordinated the data collection process, analysed qualitative data. STW, AWW; developed quantitative data analysis plans, analyzed the quantitative dataset. STW, LNN and AWW have directly accessed and verified the underlying data reported in the manuscript. AWW, AWN, BN, RM, HKI, NMI, NKS; conducted and field data collection. AWW, RM, HK, RS, AA; developed the final protocol for insitutional review board approvals. AWW led the study and developed the first draft manuscript. RS, VN, EA, WKK revised and technically edited the manuscript. LM; supervised the study and all authors reviewed the final version of the manuscript.

Funding

This study was funded by the Africa One Health University Network (AFROHUN) Transition Award (TA), which is a peel-off of the One Health Workforce-Next Generation (OHW-NG) project which aims to develop and strengthen a local One Health (OH) workforce with the capacity to prepare, prevent, detect, and respond to infectious disease outbreaks and complex health challenges in the African region.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

All experiments were performed in accordance with relevant guidelines and regulations (such as the Declaration of Helsinki). We obtained Institutional Review Board (IRB) approval from Makerere University School of Public Health (SPH-2023–463) and the protocol was registered with the Uganda National Council for Science and Technology (HS3136ES). In addition, administrative clearance from Namisindwa district police station was obtained as a process of district entry for the planned activities. We translated the informed consent and data collection tools into the local language. This enabled smooth communication between the research team and all stakeholders. We worked with local leaders and obtained written informed consent and permission for all the participating teams from their organizations/workplaces during all the project activities.

We sought written informed consent from selected respondents before participating in the study. All participants were informed of their voluntary participation in the study and their freedom to withdraw-from the study at any time.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.