Dementia and Risk of Adverse Warfarin-Related Events in the Nursing Home Setting

Background

Little attention has been focused on the safety of medications administered to treat non illnesses in nursing home residents with dementia. It is unclear whether this population is at increased risk of adverse drug events.

Objectives

To test the hypotheses that in nursing home residents with dementia prescribed warfarin have less time in therapeutic range and a higher incidence of nonpreventable and preventable adverse warfarin events compared to nursing home residents without dementia after controlling for facility and patient characteristics.

Methods

A prospective cohort embedded in a clinical trial of nursing home residents prescribed warfarin in 26 nursing homes in Connecticut was observed for up to 12 months. The primary outcome measures included adverse warfarin events (AWEs) (injuries resulting from warfarin use), potential AWEs (INR [international normalized ratio] >4.5 and management error), and AWE preventability based on physician reviews of medical record abstractions. Potential confounders included nursing home structural characteristics (eg, number of beds and for-profit status), nursing staff time, and nursing home regulatory deficiencies (pharmacy, administrative, quality of care, and all other deficiencies). Multivariable Poisson regression analysis was used to determine the independent association of dementia with potential and preventable AWEs using generalized estimating equations to account for clustering within nursing homes.

Results

Residents with dementia had no difference in the number of INR monitoring tests or percentage of days in the therapeutic range, but did have an increased risk of AWEs (adjusted incidence rate ratio [IRR], 1.47; 95% confidence interval [CI], 1.20–1.82), and preventable or potential AWEs (adjusted IRR, 1.36; 95% CI, 1.06–1.76) after adjustment for patient characteristics, nursing home quality, and case mix. Greater nursing staff time was protective for preventable and potential AWEs (adjusted IRR, 0.66; 95% CI, 0.48–0.90) but not for nonpreventable AWEs.

Conclusion

A diagnosis of dementia was associated with increased risk of nonpreventable and preventable or potential AWEs. Greater nursing staff time was associated with lower risk of preventable AWEs. These findings have implications for quality-of-care reporting and patient safety.