Ambient temperatures (heat and cold) are associated with mortality, but limited research is available about groups most vulnerable to these effects in rural populations. We estimated the effects of heat and cold on daily mortality among different sociodemographic groups in the Vadu HDSS area, western India. We studied all deaths in the Vadu HDSS area during 2004-2013. A conditional logistic regression model in a case-crossover design was used. Separate analyses were carried out for summer and winter season. Odds ratios (OR) and 95% confidence intervals (CI) were estimated for total mortality and population subgroups. Temperature above a threshold of 31 °C was associated with total mortality (OR 1.48, CI = 1.05-2.09) per 1 °C increase in daily mean temperature. Odds ratios were higher among females (OR 1.93; CI = 1.07-3.47), those with low education (OR 1.65; CI = 1.00-2.75), those owing larger agricultural land (OR 2.18; CI = 0.99-4.79), and farmers (OR 1.70; CI = 1.02-2.81). In winter, per 1 °C decrease in mean temperature, OR for total mortality was 1.06 (CI = 1.00-1.12) in lag 0-13 days. High risk of cold-related mortality was observed among people occupied in housework (OR = 1.09; CI = 1.00-1.19). Our study suggests that both heat and cold have an impact on mortality particularly heat, but also, to a smaller degree, cold have an impact. The effects may differ partly by sex, education, and occupation. These findings might have important policy implications in preventing heat and cold effects on particularly vulnerable groups of the rural populations in low and middle-income countries with hot semi-arid climate.
Keywords: Cold; Heat; India; Mortality; Socioeconomic factors; Temperature.