Heat Waves, Drugs with Anticholinergic Effects, and Outcomes in Older Hospitalized Adults

Objectives: To determine whether heat waves (HWs) affect exposure to drugs with anticholinergic effects (DACEs) on admission and the effect of such on length of stay (LOS) and in-hospital mortality in older adults. Design: Retrospective. Setting: Metropolitan teaching hospital. Participants: Individuals aged 65 and older with at least one chronic condition (n = 307) admitted to a medical ward during five consecutive HWs (≥5 consecutive days of a maximum temperature of ≥35 °C or ≥3 consecutive days of a maximum temperature of ≥40 °C) recorded between 2007 and 2009. Individuals admitted before or after each HW, matched for HW period, age, and day of the week admitted, served as controls (n = 1,114). Measurements: Data on clinical and demographic characteristics, Charlson Comorbidity Index (CCI), total medication and DACE exposure on admission (any DACE, number of DACEs, or Anticholinergic Risk Scale (ARS) score), LOS, and in-hospital mortality were collected using electronic medical data. Results: No significant differences in total medications, DACE exposure, CCI, number of daily admissions, LOS, or in-hospital mortality were observed between the HW and control groups. DACE exposure (ARS) predicted longer LOS during non-HW (β = 0.70, standard error (SE) = 0.27, P =.01) but not HW (β = 0.02, SE = 0.46, P =.96) periods. ARS did not predict mortality during non-HW (odds ratio (OR) = 1.02, SE = 0.12, P =.87) or HW (OR = 1.21, SE = 0.23, P =.31) periods. Conclusion: The effect of DACE exposure on LOS, but not mortality, is different between HW and non-HW periods in older hospitalized adults. This should be considered when assessing the effect of DACEs in pharmacoepidemiological studies.