Abstract:
Non-live rabies vaccines have been associated with both beneficial and detrimental effects on host population morbidity and mortality rates to unrelated infections in people and animals, and these non-specific effects may differ by sex. Previous animal studies may have been affected by bias, including selection bias due to loss to follow up in randomized controlled trials (RCTs). We previously reported results of an RCT in dogs on the effect of primary rabies vaccine administered at 6 weeks of age on all-cause mortality over a 7-week follow-up period, in a high-mortality population of owned dogs. Here, we report the results from the same trial of a second vaccination at 13 weeks of age, compared to a primary vaccination. Because a relatively high proportion of study subjects (30%) were lost to follow-up in the RCT, we also conducted an analysis to control for possible selection bias over both periods (6 to 13 weeks and 13 to 20 weeks of age). We found that primary rabies vaccination at 6 weeks of age substantially increased the hazard of death from all causes over the next 7 weeks among females (hazard ratio [HR] 2.69, 95% confidence intervals [CI] 1.27–5.69), but not among males (HR 0.91, 95% CI 0.32–2.59). Among survivors, administration of a second dose of rabies vaccine at 13 weeks of age was associated with a decreased hazard of death among males (HR 0.33, 95% CI 0.10–1.02) but not females (HR 1.64, 95% CI 0.59–4.58), when compared to the group receiving their first dose at this age. Based on our causal assumptions, we show that these results were not affected by selection bias. In this high-mortality dog population, receipt of a non-live rabies vaccine substantially affected all-cause mortality rates, with this effect being strongly modified by sex.
