Estimating the Effects of Mean, Inter-, and Intra-day temperature variations on mortality among 7 Tropical and Subtropical Cities of Southeast Asian Countries

Abstract

Extensive research has been done in probing the effects of temperature on health, with cascades of studies supporting the non-linear relationship. However, there are ample evidence which have explored the effects of inter- and intra-day variations in relation to mortality, let alone among cause-specific ones. This study focuses in determining the risks associated with the mean, inter- and intra-day temperatures in all-cause and cause-specific mortality. We used a Quasi-Poisson regression coupled with a distributed lag non-linear model in estimating the relative risks (RR) associated with (1) mean temperature, (2) inter-day variations, and (3) intra-day variations in relation to all-cause mortality among the cities of the Philippines, Vietnam and Malaysia. We further explored the variations in the risks in both cardiovascular- and respiratory-related mortality. In all-cause mortality, at mean temperature, most cities exhibited increase in the risk with Manila having the highest RR at 1.35 (95% CI: 1.21-1.51) per 1°C increase. On the contrary, risks observed in both inter- and intra-day variations were non-definitive, with some tend to be protective. In cause-specific analysis, there were also irreconcilable directions of the risks in both inter- and intra-day variations among the seven cities, with slight increase in inter-day risks observed in Klang Valley with an RR of 1.15 (95% CI: 0.90-1.46) and intra-day risks in Cebu (RR=1.15, 95% CI: 0.98-1.34) per 1°C increase; both observed in cardiovascular-related mortality. In summary, results of this study showed that mean temperature accounts for higher risks in both all-cause and cause-specific mortality, with inconsistent risks levels in both inter- and intra-day variations. Particular focus on the effects of mean temperature on mortality should be considered for policy and planning strategies. While caution should be taken with regard to the interpretation of risks associated with inter- and intra-day variations.