Million Death Study in India estimates rabies deaths

  • Community News

The Million Death Study (MDS) seeks to assign causes to all deaths in 6,671 randomly selected areas of India (each with a population of around 1,000 people) between 2001 and 2014. Investigators use an enhanced verbal autopsy technique to attribute a cause to each identified death.

A new study used MDS data from years 2001-2003 (122,429 surveyed deaths, of which only 8.7% were of unknown cause), and found 95 recorded due to rabies, with a further 45 deaths due to dog bites being confirmed as rabies cases. Averaged across the 3 years of data, this corresponds to a rate of 1.3 rabies fatalities per 1,000 deaths. These deaths were not distributed evenly across the country, they almost all occurred in rural areas, and outside hospitals. 65% of patients did not seek treatment and only 1 received a full course of PEP vaccines (no RIG).

Combining this proportion of deaths due to rabies with United Nations Population Division estimates of deaths in India, the authors estimated that 12,700 deaths in India were due to symptomatically identifiable furious rabies in 2005. This figure is the first attempt to estimate rabies deaths based on a representative sample of actual deaths from across India, and is considerably lower than previous estimates (17,137 and 19,713) for rabies deaths in India. However, the verbal autopsy method would not detect deaths due to paralytic or abnormal rabies. The proportion of rabies cases presenting with paralytic or abnormal symptoms is unknown, though figures of less than 20%, or one third have been suggested.

The authors conclude that although canine rabies elimination in India is not likely to be achieved soon, the fact that rabies cases are clustered suggest that significant reductions in human deaths could be possible with targeted control measures.

Summarised by Louise Taylor from the paper: Suraweera et al (2012) Deaths from Symptomatically Identifiable Furious Rabies in India: A Nationally Representative Mortality Survey. Negl Trop Dis 6(10): e1847