Integrating a community-based continuous mass dog vaccination delivery strategy into the veterinary system of Tanzania: A process evaluation using normalization process theory

Abstract

Sustained vaccination coverage of domestic dog populations can interrupt rabies transmission. However, challenges remain including low dog owner participation, high operational costs associated with current (centralized and annually delivered (pulse)) approaches and high dog population turnover. To address these challenges an alternative (community-based continuous mass dog vaccination (CBC-MDV)) approach was designed. We investigated the potential for successful normalization of CBC-MDV into routine practice within the context of local communities and the veterinary system of Tanzania.

Methods

In a process evaluation of a pilot implementation of CBC-MDV, we conducted in-depth interviews with implementers and community leaders (n = 24), focus group discussion with implementers and community members (n = 12), and non-participant observation (n = 157 h) of delivery of the intervention components. We analyzed these data thematically drawing on the normalization process theory, to assess factors affecting implementation and integration.

Main findings

Implementers and community members clearly understood the values and benefits of the CBC-MDV, regarding it as an improvement over the pulse strategy. They had a clear understanding of what was required to enact CBC-MDV and considered their own involvement to be legitimate. The approach fitted well into routine schedules of implementers and the context (infrastructure, skill sets and policy). Implementers and community members positively appraised CBC-MDV in terms of its perceived impact on rabies and recommended its use across the country. Implementers and community members further believed that vaccinating dogs free of charge was critical and made community mobilization easier. However, providing feedback to communities and involving them in evaluating outcomes of vaccination campaigns were reported to have not been done. Local politics was cited as a barrier to collaboration between implementers and community leaders.

Conclusion

This work suggests that CBC-MDV has the potential to be integrated and sustained in the context of Tanzania. Involving communities in design, delivery and monitoring of CBC-MDV activities could contribute to improving and sustaining its outcomes.

Participation in mass dog vaccination campaigns in Tanzania: Benefits of community engagement

Background

Canine rabies causes about 59,000 human deaths each year globally but the disease can be eliminated by sustaining sufficient dog vaccination coverage over several consecutive years. A challenge to achieving high coverage is low participation of dog owners in vaccination campaigns. We explored whether and how previously identified contributory factors to low participation can be addressed through community engagement activities.

Methods

We engaged communities in two wards in Tanzania on dog behavior and handling, safe ways of interacting with dogs, and their perceptions of dog vaccination. We shared and elicited information from them through village meetings, video screenings, posters and leaflets and involved the leadership of one of the wards in planning and implementing a dog vaccination exercise to explore the feasibility of their participation. We assessed the impact of engagement activities with household surveys, meeting reports, observations and focus group discussions. We used a generalized linear mixed-effects model to identify predictors of knowledge and perceptions and compared knowledge amongst respondents before and after engagement activities. Qualitative data was analyzed inductively to explore perceptions of dog handling and vaccination and feasibility, opportunities and barriers to community leadership participation in organizing mass dog vaccination.

Main findings

Knowledge of dog behavior, dog handling, and safe ways of interacting with dogs was positively associated with age (p < 0.0001), dog ownership (p = 0.0203), training (p = 0.0010) and previous experience of a dog bite (p = 0.0002); and was negatively associated with being afraid of dogs (p = 0.0061) and participation in a recent dog vaccination campaign (p = 0.0077). Knowledge was low before and significantly improved after engagement activities. The majority (92%) of respondents believed dog vaccination has no negative effects on dogs. Respondents perceived lack of bonding with their dog as a limitation to the ability to restrain a dog for vaccination. The community performed most roles assigned to them in the dog vaccination exercise, but barriers such as lack of motivation for volunteering exist.

Conclusion

Engaging communities regularly on dog vaccination can improve their knowledge of dog behavior and dog handling techniques, and may help improve owner participation in dog vaccination campaigns.

Development, feasibility and potential effectiveness of community-based continuous mass dog vaccination delivery strategies: Lessons for optimization and replication

Objectives

Dog vaccination can eliminate rabies in dogs, but annual delivery strategies do not sustain vaccination coverage between campaigns. We describe the development of a community-based continuous mass dog vaccination (CBC-MDV) approach designed to improve and maintain vaccination coverage in Tanzania and examine the feasibility of delivering this approach as well as lessons for its optimization.

Methods

We developed three delivery strategies of CBC-MDV and tested them against the current annual vaccination strategy following the UK Medical Research Council’s guidance: i) developing an evidence-based theoretical framework of intervention pathways and ii) piloting to test feasibility and inform optimization. For our process evaluation of CBC-MDV we collected data using non-participant observations, meeting reports and implementation audits and in-depth interviews, as well as household surveys of vaccination coverage to assess potential effectiveness. We analyzed qualitative data thematically and quantitative data descriptively.

Results

The final design included delivery by veterinary teams supported by village-level one health champions. In terms of feasibility, we found that less than half of CBC-MDV’s components were implemented as planned. Fidelity of delivery was influenced by the strategy design, implementer availability and appreciation of value intervention components, and local environmental and socioeconomic events (e.g. elections, funerals, school cycles). CBC-MDV activities decreased sharply after initial campaigns, partly due to lack of supervision. Community engagement and involvement was not strong. Nonetheless, the CBC-MDV approaches achieved vaccination coverage above the critical threshold (40%) all-year-round. CBC-MDV components such as identifying vaccinated dogs, which village members work as one health champions and how provision of continuous vaccination is implemented need further optimization prior to scale up.

Interpretation

CBC-MDV is feasible to deliver and can achieve good vaccination coverage. Community involvement in the development of CBC-MDV, to better tailor components to contextual situations, and improved supervision of activities are likely to improve vaccination coverage in future.

Annual mass dog vaccination campaigns that reach at least 70% of the dog population, should maintain sufficient herd immunity (sustain vaccination coverage above 40%) between campaigns to interrupt rabies transmission. However, it is often challenging to reach 70% of the dog population with annual vaccination campaigns. We hypothesized that a community-based continuous approach to dog vaccination could better maintain high levels of vaccination coverage all-year-round. We describe the development of a community-based continuous approach to dog vaccination in Tanzania, and assessed the feasibility of delivering its components, its potential effectiveness and lessons for its optimization. We found that the approach was well accepted, as its development involved key stakeholders. Although less than half of the components of the community-based continuous approach were delivered exactly as planned, over 70% of dogs were vaccinated and the approach maintained coverage above the critical vaccination threshold throughout the year. We conclude that it is feasible to deliver a community-based continuous approach to dog vaccination, but that some components need further improvement; more supervision and community involvement should lead to better outcomes.

Development of Dog Vaccination Strategies to Maintain Herd Immunity against Rabies

Human rabies can be prevented through mass dog vaccination campaigns; however, in rabies endemic countries, pulsed central point campaigns do not always achieve the recommended coverage of 70%. This study describes the development of a novel approach to sustain high coverage based on decentralized and continuous vaccination delivery. A rabies vaccination campaign was conducted across 12 wards in the Mara region, Tanzania to test this approach. Household surveys were used to obtain data on vaccination coverage as well as factors influencing dog vaccination. A total 17,571 dogs were vaccinated, 2654 using routine central point delivery and 14,917 dogs using one of three strategies of decentralized continuous vaccination. One month after the first vaccination campaign, coverage in areas receiving decentralized vaccinations was higher (64.1, 95% Confidence Intervals (CIs) 62.1-66%) than in areas receiving pulsed vaccinations (35.9%, 95% CIs 32.6-39.5%). Follow-up surveys 10 months later showed that vaccination coverage in areas receiving decentralized vaccinations remained on average over 60% (60.7%, 95% CIs 58.5-62.8%) and much higher than in villages receiving pulsed vaccinations where coverage was on average 32.1% (95% CIs 28.8-35.6%). We conclude that decentralized continuous dog vaccination strategies have the potential to improve vaccination coverage and maintain herd immunity against rabies.

Programmatic factors associated with the limited impact of Community-Directed Treatment with Ivermectin to control Onchocerciasis in three drainage basins of South West Cameroon

Introduction

The CDTI model is known to have enhanced community participation in planning and resource mobilization toward the control of onchocerciasis. These effects were expected to translate into better individual acceptance of the intervention and hence high Treatment Coverage, leading to a sustainable community-led strategy and reduction in the disease burden. A survey revealed that after 10–12 rounds of treatment, prevalence of onchocerciasis was still high in three drainage basins of South West Cameroon and transmission was going on.

Methods

We designed a three (3)-year retrospective (2012, 2013 and 2014), descriptive cross-sectional study to explore the roles of operational challenges in the failure of CDTI to control the disease as expected. We administered 83 semi-structured questionnaires and conducted 12 in-depth interviews with Chiefs of Bureau Health, Chiefs of Centers, CDDs and Community Heads. Descriptive statistics was used to explore indicators of performance which were supported with views from in-depth interviews.

Results

We found that community participation was weak; communities were not deciding time and mode of distributions. Only 6 (15.0%) of 40 Community Drug Distributors reported they were selected at general community meetings as required. The health service was not able to meet and discuss Community-Directed Treatment with Ivermectin activities with individual communities partly due to transportation challenges; this was mostly done through letters. Funding was reported to be inadequate and not timely. Funds were not available to conduct Community-Self Monitoring after the 2014 Mass Drug Administration. There was inadequate health staff at the frontline health facility levels, and some Chiefs of Center reported that Community-Directed Treatment with Ivermectin work was too much for them. The mean operational Community Drug Distributor-population ratio was 1 Community Drug Distributor per 317 populations (range: 194–464, expected is 1:250). Community Drug Distributor attrition rate was 14% (2012), 11% (2013) and 12% (2014) of total Community Drug Distributors trained in the region. Lack of incentive for Community Drug Distributor was primary reason for Community Drug Distributor attrition. Number of Community Drug Distributors trained together by health area ranged from 14 to 127 (mean ± SD = 51 ±32) with duration of training ranging from 4–7 hours (mean ± SD = 5.05 ± 1.09). The trainings were conducted at the health centers. Community Drug Distributors always conducted census during the past three distributions (Mean ± SD = 2.85 ± 0.58). Community-Self Monitoring was facing challenge. Several of the community heads, Chiefs of Bureau Health and Chiefs of Center agreed that Community-Self Monitoring was not being carried out effectively due to lack of incentives for monitors in the communities.

Conclusion

Inadequate human resource, funding issues and transportation challenges during distribution periods reduced the ability of the health service to thoroughly sensitize communities and supervise CDTI activities. This resulted in weak community understanding, acceptance and participation in the process. CDTI in our study area did not achieve sustainable community-led campaign and this may have led to the reduced impact on Onchocerciasis.

River blindness is caused by a very tiny, thread-like worm. The disease is better controlled when affected communities are included in the planning and carrying out of distribution of Ivermectin used to treat the disease. For a community to be able to prevent people from getting this disease, members must take Ivermectin once or twice a year, continuously for about 20 years. Hence, the organization in charge of controlling river blindness (African Programme for Onchocerciasis Control–APOC) decided that when a control programme is started in a community, the community must be involved and assisted to take full charge of the programme so that within 12 years the community can sustain the distribution of Ivermectin for as long as necessary. This community directed strategy prevented river blindness in many communities. However, after 10–12 years of implementation, studies found that river blindness largely persists in communities in three drainage basins in South West Region of Cameroon. This paper discussed the operational challenges that the programme may have faced in these areas.