Objectives The objective of the study was to analyse the association between the presence of air conditioning in hospital wards and the intrahospital mortality during the 2003 heatwave, in mainland Portugal.

Methods Historical cohort study design including all patients aged 45 or more who were hospitalised in the 7 days before the heatwave. The outcome was survival during the 18 days the heatwave lasted and during the 2 days after the end of the heatwave. A comparison group was also selected in four analogous periods without any heatwave event during January to May 2003. Data were obtained from the 2003 hospital discharges database. Air conditioning presence in hospital wards was determined using a survey sent to hospital administrations. A Cox-regression model was used to estimate the confounder-adjusted HR of death, during the heatwave and the comparison period, in patients in wards with air conditioning (AC+) versus patients in wards without air conditioning (AC−).

Results 41 hospitals of mainland Portugal (49% of all hospitals in mainland Portugal) participated, and 2093 patients were enrolled. The overall confounder-adjusted HR of death in AC+ patients versus AC− patients was 0.60 (95% CI 0.37 to 0.97) for the heatwave period and 1.05 (95% CI 0.84 to 1.32) for the comparison group.

Conclusions The study found strong evidence that, during the August 2003 heatwave, the presence of air conditioning in hospital wards was associated with an increased survival of patients admitted before the beginning of the climate event. The reduction of the risk of dying is estimated to be 40% (95% CI 3% to 63%).