Collective participation is crucial for creating public goods, particularly in under-resourced communities where disease prevention is paramount. Vaccination, while offering individual protection, requires reaching a threshold level of coverage to achieve herd immunity and disease elimination. This study uses rabies control through dog vaccination in Tanzania as a case study to investigate the collective action problem. Rabies, causing approximately 59,000 human deaths annually, disproportionately affects Africa and Asia, mainly through dog bites. While post-exposure prophylaxis is an option, mass dog vaccination is more cost-effective and offers long-term benefits. Successful campaigns have eliminated rabies in many areas; however, achieving the necessary vaccination coverage remains a challenge due to both demand-side (community participation) and supply-side (vaccine availability and delivery) factors. Demand-side factors include poverty, competing priorities, perceptions of rabies risk, awareness of vaccination campaigns, and community ability to solve collective action problems. Supply-side barriers include infrastructural challenges affecting vaccine access. This research focuses on two communication strategies (mobile phone text messaging and promotion by community leaders) and operational changes to increase rabies risk salience and reduce participation costs, thereby addressing the collective action problem.

The introduction adequately reviews existing literature on collective action problems related to public goods provision, focusing on the challenges in achieving herd immunity through vaccination campaigns. The authors cite relevant works highlighting the importance of community participation in disease eradication efforts and the factors influencing vaccination uptake, including poverty, competing health priorities, perceptions of risk, and access to services. The literature review establishes the context for the study by showing the need for effective strategies to overcome barriers to collective action in rabies vaccination campaigns.

The study employed a randomized controlled trial involving 56 villages in Tanzania’s Morogoro Rural district. Villages were randomly assigned to four groups: routine advertising only, routine advertising plus community leader advertising, routine advertising plus text messaging, and routine advertising plus both community leader and text messaging. Prior to interventions, workshops and focus group discussions were held to identify potential issues and refine the study design and messaging. Text messages emphasized a real-life rabies death in a nearby village, encouraging vaccination, and providing campaign details. Community leaders received letters inviting them to promote the campaign using a sample message or at their discretion. Routine advertising included posters and loudspeakers. Vaccination campaigns were held, and vaccination coverage was estimated using post-vaccination transects. Data was analyzed using generalized linear mixed models (GLMMs) to assess the effects of advertising interventions, controlling for factors such as pastoralist presence and livestock field officer presence. Qualitative data from interviews and questionnaires were used to explore the mechanisms underlying the findings and understanding factors influencing participation. A small experiment using cash incentives was conducted to create an indicator of community propensity for collective action. This involved giving participants money with instructions to either keep it or return it, with a collective reward if a threshold of returns was met. This was integrated into household questionnaires along with questions on how participants learned about vaccination campaigns.

Neither text messaging nor community leader advertising alone significantly impacted vaccination coverage. However, when used together, they substantially increased participation (approximately 70% coverage compared to around 30% for text messaging only and around 60-65% for other groups). Operational changes, such as increasing the number of vaccination points and extending the operating hours, significantly improved coverage, even surpassing the impact of the combined advertising interventions (Figure 3). Qualitative data suggests that the anecdote about a rabies death in the text messages increased rabies risk salience. Community leaders were seen as vital for mobilization but not as a primary information source; their role was in encouraging participation rather than solely informing people. The effectiveness of the interventions was shown to be context-dependent. A model including operational changes, year, and other covariates further supported the substantial positive impact of operational changes on vaccination coverage. The collective action indicator (derived from a cash-return experiment) wasn’t found to be statistically significant, possibly due to limited between-village variation or the perception of rabies as an individual rather than a collective problem.

The findings challenge the assumption that simply providing information about the benefits of vaccination is sufficient to achieve high participation. While information dissemination was important, operational changes—making vaccination more accessible and convenient—proved crucial for increasing uptake. The combined effect of the two advertising interventions suggests that targeting both information and motivation is necessary but alone insufficient for high participation in collective action. The success of the interventions depended heavily on the context, highlighting the need for culturally tailored and locally sensitive approaches. The study's results have implications for public good provision and interventions needing community participation, particularly those requiring threshold coverage for herd immunity. The study emphasizes the need for policymakers to understand local contexts, combining information dissemination strategies with improvements in service delivery to achieve desired outcomes.

The study demonstrates that while advertising interventions can raise awareness, operational changes are crucial for significantly improving participation in rabies vaccination campaigns. The combined use of advertising and operational changes offers a more effective approach. Future research could investigate other operational factors influencing participation, further explore the interplay between information dissemination and convenience, and conduct similar studies in other contexts to test the generalizability of the findings.

The study couldn't fully isolate the effects of advertising from operational changes. The collective action indicator might not have fully captured the community's propensity for collective action or the framing of rabies as a collective issue. Data on other village-level changes that might affect participation were limited in the analysis of operational changes. The cash-return experiment, while providing a proxy for collective action propensity, might have raised ethical considerations. The sample size of 56 villages, while reasonably sized for the power analysis, may not fully capture the diversity of contexts that affect rabies vaccination uptake in a country like Tanzania.