Lagged association between climate variables and hospital admissions for pneumonia in South Africa

Abstract

Pneumonia is a leading cause of hospitalization in South Africa. Climate change could potentially affect its incidence via changes in meteorological conditions. We investigated the delayed effects of temperature and relative humidity on pneumonia hospital admissions at two large public hospitals in Limpopo province, South Africa. Using 4062 pneumonia hospital admission records from 2007 to 2015, a time-varying distributed lag non-linear model was used to estimate temperature-lag and relative humidity-lag pneumonia relationships. Mean temperature, relative humidity and diurnal temperature range were all significantly associated with pneumonia admissions. Cumulatively across the 21-day period, higher mean daily temperature (30 C relative to 21 C) was most strongly associated with a decreased rate of hospital admissions (relative rate ratios (RR): 0.34, 95% confidence interval (CI): 0.14–0.82), whereas results were suggestive of lower mean daily temperature (12 C relative to 21 C) being associated with an increased rate of admissions (RR: 1.27, 95%CI: 0.75–2.16). Higher relative humidity (>80%) was associated with fewer hospital admissions while low relative humidity (<30%) was associated with increased admissions. A proportion of pneumonia admissions were attributable to changes in meteorological variables, and our results indicate that even small shifts in their distributions (e.g., due to climate change) could lead to substantial changes in their burden. These findings can inform a better understanding of the health implications of climate change and the burden of hospital admissions for pneumonia now and in the future.