Malaria-attributed death rates in India.

摘要

Neeraj Dhingra and colleagues use verbal autopsy procedures to estimate that 205 000 deaths attributable to malaria occur annually in India before 70 years of age (with plausible lower and upper bounds of 125 000-277000 deaths annually). We believe that Dhingra and colleagues might have substantially overestimated the number of malaria deaths in India for the following reasons. First, previous validation studies show that verbal autopsy procedures tend to overestimate malaria deaths in areas where malaria transmission is low or absent. Verbal autopsy overestimated malaria deaths in children by 47% in a low-transmission setting in Uganda (Kampala), and overestimated malaria deaths in adults by 200% in a high-transmission setting in Ghana where the proportional mortality of malaria among adults was very low (table). In Kisoro, Uganda, where no malaria deaths occurred in children younger than 5 years, verbal autopsy attributed 12% of deaths to malaria (unpublished data).