Association of winter indoor-temperature variability with 5-year changes in home blood pressure: the Smart Wellness Housing Cohort Survey in Japan

Cold indoor temperatures acutely raise blood pressure, but their long-term effects remain uncertain. We investigated whether winter indoor-temperature levels and temporal variability were associated with 5-year changes in home blood pressure. This nationwide cohort included 700 adults from 483 households with complete data. Living-room temperature was measured at 10-min intervals, and home blood pressure was measured twice each morning and evening for 2 weeks in winter at baseline and at the 5-year follow-up. Linear mixed-effects models with household random intercepts were adjusted for baseline blood pressure, individual covariates, and mean indoor temperature in analyses of variability. Mean and minimum indoor temperatures were not associated with 5-year changes in systolic or diastolic home blood pressure. In contrast, a 10 °C/day higher cumulative indoor-temperature variation was associated with 0.42 mmHg (95% confidence interval, 0.04–0.80) and 0.27 mmHg (0.01–0.54) greater increases in systolic and diastolic home blood pressure, respectively. Each additional daily indoor-temperature swing ≥3 °C and ≥5 °C was associated with 0.32 mmHg (0.004–0.636) and 0.49 mmHg (0.04–0.93) greater increases in systolic blood pressure, respectively. A model-based comparison of cumulative indoor-temperature variation values 1 standard deviation above versus below the mean yielded an estimated difference of 1.47 mmHg (0.13–2.81) in the 5-year change in home systolic blood pressure. The sensitivity analysis yielded broadly consistent results. Greater winter indoor-temperature variability was associated with larger 5-year increases in home blood pressure independently of mean indoor temperature. Clinical Trial Registration URL: http://www.umin.ac.jp/ctr/index.htm . UMIN Clinical Trials Registry, Trial No. UMIN000042196.