The effect of extreme temperature and precipitation on cause-specific deaths in rural Burkina Faso: a longitudinal study

BACKGROUND: Extreme weather is becoming more common due to climate change and threatens human health through climate-sensitive diseases, with very uneven effects around the globe. Low-income, rural populations in the Sahel region of west Africa are projected to be severely affected by climate change. Climate-sensitive disease burdens have been linked to weather conditions in areas of the Sahel, although comprehensive, disease-specific empirical evidence on these relationships is scarce. In this study, we aim to provide an analysis of the associations between weather conditions and cause-specific deaths over a 16-year period in Nouna, Burkina Faso. METHODS: In this longitudinal study, we used de-identified, daily cause-of-death data from the Health and Demographic Surveillance System led by the Centre de Recherche en Santé de Nouna (CRSN) in the National Institute of Public Health of Burkina Faso, to assess temporal associations between daily and weekly weather conditions (maximum temperature and total precipitation) and deaths attributed to specific climate-sensitive diseases. We implemented distributed-lag zero-inflated Poisson models for 13 disease-age groups at daily and weekly time lags. We included all deaths from climate-sensitive diseases in the CRSN demographic surveillance area from Jan 1, 2000 to Dec 31, 2015 in the analysis. We report the exposure-response relationships at percentiles representative of the exposure distributions of temperature and precipitation in the study area. FINDINGS: Of 8256 total deaths in the CRSN demographic surveillance area over the observation period, 6185 (74·9%) were caused by climate-sensitive diseases. Deaths from communicable diseases were most common. Heightened risk of death from all climate-sensitive communicable diseases, and malaria (both across all ages and in children younger than 5 years), was associated with 14-day lagged daily maximum temperatures at or above 41·1°C, the 90th percentile of daily maximum temperatures, compared with 36·4°C, the median (all communicable diseases: 41·9°C relative risk RR 1·38 95% CI 1·08-1·77, 42·8°C 1·57 1·13-2·18; malaria all ages: 41·1°C 1·47 1·05-2·05, 41·9°C 1·78 1·21-2·61, 42·8°C 2·35 1·37-4·03; malaria younger than 5 years: 41·9°C 1·67 1·02-2·73). Heightened risk of death from communicable diseases was also associated with 14-day lagged total daily precipitation at or below 0·1 cm, the 49th percentile of total daily precipitation, compared with 1·4 cm, the median (all communicable diseases: 0·0 cm 1·04 1·02-1·07, 0·1 cm 1·01 1·006-1·02; malaria all ages: 0·0 cm 1·04 1·01-1·08, 0·1 cm 1·02 1·00-1·03; malaria younger than 5 years: 0·0 cm 1·05 1·01-1·10, 0·1 cm 1·02 1·00-1·04). The only significant association with a non-communicable disease outcome was a heightened risk of death from climate-sensitive cardiovascular diseases in individuals aged 65 years and older associated with 7-day lagged daily maximum temperatures at or above 41·9°C (41·9°C 2·25 1·06-4·81, 42·8°C 3·68 1·46-9·25). Over 8 cumulative weeks, we found that the risk of death from communicable diseases was heightened at all ages from temperatures at or above 41·1°C (41·1°C 1·23 1·05-1·43, 41·9°C 1·30 1·08-1·56, 42·8°C 1·35 1·09-1·66) and risk of death from malaria was heightened by precipitation at or above 45·3 cm (all ages: 45·3 cm 1·68 1·31-2·14, 61·6 cm 1·72 1·27-2·31, 87·7 cm 1·72 1·16-2·55; children younger than 5 years: 45·3 cm 1·81 1·36-2·41, 61·6 cm 1·82 1·29-2·56, 87·7 cm 1·93 1·24-3·00). INTERPRETATION: Our results indicate a high burden of death related to extreme weather in the Sahel region of west Africa. This burden is likely to increase with climate change. Climate preparedness programmes-such as extreme weather alerts, passive cooling architecture, and rainwater drainage-should be tested and implemented to prevent deaths from climate-sensitive diseases in vulnerable communities in Burkina Faso and the wider Sahel region. FUNDING: Deutsche Forschungsgemeinschaft and the Alexander von Humboldt Foundation.