BACKGROUND: Although blood pressure decreases in response to high ambient temperature, little is known about whether the ambient temperature can induce clinical hypotension events. Therefore, we conducted a time-series analysis to evaluate the association between hypotension hospital visits and ambient temperature in seven metropolitan cities of Korea.

METHODS: We used the National Health Insurance Database, which contains the complete hospital visit data of the entire Korean population. We collected hospital visit data of seven metropolitan cities and linked the number of daily hypotension hospital visits to city-level ambient temperature, relative humidity, and air pollution levels from 2011 to 2015. Time-series analysis using the Poisson generalized additive model was conducted for each metropolitan city and we meta-analyzed the time-series results using the random effect model.

RESULTS: There were 132,097 hospital visits for hypotension during our study period. A 1 degree Celsius (°C) increase in ambient temperature was associated with 1.1% increase in hospital visits for hypotension on lag day 0. Effects of ambient temperature lasted for 7 days, showing greater effects in shorter lag days. Subgroup analysis by sex and income groups showed similar results, but effects of ambient temperature on hypotension hospital visits was higher in the younger age group compared to older age group (aged over 65 years old). The results were unchanged when we applied cumulative lags, different case definitions, degrees of freedom per year, and multi-pollutant model adjusting for air pollutants.

CONCLUSIONS: Hospital visits for hypotension were positively associated with ambient temperature. Increased hypotension events in response to increased ambient temperature might explain the high cardiovascular mortality on hot days.