Chasing rabies herd immunity: Evaluating dog vaccination strategies and post-vaccination survey reliability in urban and peri-urban Bangladesh

Catherine Swedberg^1,2*Umme Ruman Siddiqi³Ravikiran Keshavamurthy¹Md Sohel Rana⁴Kamrul Islam³Erin Kennedy¹Yasmeen Ross¹Sarah Bonaparte¹Frederic Lohr²Hasan Sayedul Mursalin³Md Kamruzzaman³Luke Gamble²Andrew D Gibson²Jesse D Blanton¹Ryan M Wallace¹

• ¹Centers for Disease Control and Prevention (CDC), Atlanta, United States
• ²Worldwide Veterinary Service, Cranborne, United Kingdom
• ³Directorate General of Health Services, Ministry of Health and Family Welfare, Dhaka, Bangladesh
• ⁴Department of Livestock Services, Ministry of Fisheries and Livestock, Dhaka, Bangladesh

Introduction: Mass dog vaccination (MDV) is essential for eliminating dog-mediated rabies, responsible for over 95% of the estimated 74,000 annual human rabies deaths. Achieving ≥70% vaccination coverage necessary for herd immunity remains challenging, underscoring the need for effective vaccination strategies and reliable post-vaccination evaluation methods. Methods: MDV campaigns were conducted in four site in Bangladesh (two urban, two peri-urban) using three strategies: standard capture-vaccinate-release (CVR), enhanced CVR (eCVR), and roaming static point (RSP). Post-vaccination sight re-sight (SRS) and household surveys were used to characterize dog populations and estimate coverage. Three enumeration methods were compared to assess reliability and utility for campaign evaluation. Results: Over 12 working days, 9,195 dogs were vaccinated. eCVR achieved the highest operational efficiency (40.2 dogs/vaccinator/day), followed by standard CVR (36.6) and RSP (18.2). Post-vaccination surveys indicated that over 93% of dogs were free-roaming, with ~30% unowned, highlighting limitations of static-point strategies. Human-to-dog ratios (HDRs) were lower than the 100:1 planning estimate and varied widely across sites (mean: 67.8; range: 21.1–129.6), with no significant difference between urban and peri-urban areas (p = 0.479). Coverage estimates differed by enumeration method: 19% (dog density), 32% (HDR), and 47% (SRS), with comparable SRS and HDR estimates (p = 0.920), and dog-density estimates significantly lower (p = 0.014). Averaging across methods, eCVR achieved the highest coverage (45%), followed by standard CVR (37%) and RSP (16%), with RSP significantly lower than eCVR (p = 0.028). Discussion: The wide heterogeneity in HDRs highlights the limitations of using a single ratio for national dog population extrapolation. Effective rabies control requires locally tailored vaccination strategies, real-time monitoring, and robust enumeration techniques to guide planning and ensure reliable evaluation of campaign impact.