Fig. 2. The influence of vaccine type and vaccination coverage on rabies transmission.

The relationship between mean vaccination coverage (%) and the effective reproductive number, R_e. Vaccination coverage was projected using data from vaccination campaigns and assumed to wane with demographic turnover of the dog population and vaccine longevity (11). R_e was estimated as the average number of secondary cases generated by each primary case, from the construction of 1,000 bootstrapped transmission trees, following previously described methods (12). Individual estimates of R_e and local vaccination coverage (both at the site of each rabies case) were averaged across six-month time windows, with symbols indicating the year of each six-month estimate, and scaled by the number of cases contributing to the estimate (17 data points in total, from the second half of 2008 until the end of 2016). There was a strong negative relationship between transmission and vaccination coverage, based on a weighted linear regression after removal of the outlier from the first half of 2015, following use of the ineffective vaccine (regression coefficient = −0.005, p value = 0.005, R-squared = 0.39)