Introduction: Medications with anticholinergic effects are associated with increased mortality. Previous analyses could not differentiate between specific anticholinergic effects or the associated occurrence of comorbidity and reduced functional status. We, therefore, investigated whether the anticholinergic risk score predicted death in older patients discharged from rehabilitation, in whom data on comorbid disease and functional status were available.
Methods: Data on sex, age, admission, discharge, length of stay, statin use, antiplatelet therapy (a surrogate marker of cardiovascular disease), measures of functional ability including the 20-point Barthel score on admission and discharge and time to death after discharge were recorded from a single inpatient rehabilitation service between 1999 and 2008.
The total anticholinergic risk score for each patient was calculated based on previously published weightings for strength of the anticholinergic effect. Cox regression analysis was used to test the effect of quartiles of anticholinergic risk score on survival, with adjustment for the above variables.
Results: A total of 3,355 patients were included in the analysis. The mean age was 81.6 (SD 7.6) years and 1,316 (39%) were male. The mean follow-up time was 5.2 years; 1,300/3,355 (39%) died during the follow-up. Anticholinergic risk scores for each quartile were: 0 to 1, 2–3, 4–5 and 6 or more. Cox regression analysis demonstrated an increased risk of death for those in the highest quartile compared with the lowest quartile (hazard ratio 1.20, 95% CI: 1.02–1.41); the association remained significant after adjustment for other baseline variables (hazard ratio 1.31, 95% CI: 1.09–1.58, P for trend <0.001).
Conclusion: The anticholinergic risk score independently predicted mortality after discharge from rehabilitation even after adjustment for function and comorbidity.