The effects of ambient temperature variation on respiratory hospitalizations in summer, New York State

Abstract

A time-series study was used to assess the effect of temperature variation during summer on respiratory disease in New York State. Daily respiratory admissions were linked with various meteorological indicators including daily and weekly temperature variation from June-August, 1991-2004. Two-stage Bayesian hierarchical models were used to first compute percent excess risks at the region level while controlling for air pollutants and time-varying variables using Poisson generalized additive models, and then to pool statewide estimates together after controlling for regional confounders. This study found that the daily temperature range between maximum and minimum temperature was associated with a 0·27-0·38% increased risk of admission. Minimum temperature (TMIN) above the previous 6-day average was associated with a 0·93% higher risk of respiratory morbidity. Multiday temperature ranges within 5 and 7 days were associated with 0·49 and 0·73% increases in admissions, respectively. We concluded that daily and multiday temperature variation may increase respiratory hospitalizations with a larger risk associated with TMIN.