(P-049) Prevalence and Factors Associated with Oral Anticoagulant Use in the General Nursing Home Population

Background: Anticoagulants are indicated for treatment and prevention of several clinical conditions. Prior studies have examined anticoagulant utilization for specific indications and in community-dwelling populations. Decision-making regarding anticoagulant prescribing in the nursing home setting is particularly challenging because advanced age and clinical complexity places most residents at increased risk for adverse drug events.

Objectives: To estimate the prevalence of oral anticoagulant use (overall, warfarin, and direct-oral anticoagulants (DOACs)) and identify factors associated with oral anticoagulant use among the general population of residents living in nursing homes.

Methods: This point prevalence study was conducted among 506,482 residents in US nursing homes on October 31, 2016 who were enrolled in Medicare fee-for-service. Covariates including demographics, clinical conditions, medications, cognitive impairment, and functional status were obtained from Minimum Data Set 3.0 assessments and Medicare Part A and D claims. Oral anticoagulant use was identified using dispensing dates and days supply information from Medicare Part D claims. Adjusted prevalence ratios (aPR) for associations between covariates and any anticoagulant use were estimated with robust Poisson models. For the three-level outcome (warfarin user, DOAC user, non-user), multinomial logistic models estimated adjusted odds ratios (aOR).

Results: Overall, 11.8% of residents used oral anticoagulants. Among users, 44.3% used DOACs. Residents with body mass index >40 kg/m2 (aPR: 1.66; 95% Confidence Interval (CI): 1.61 -1.71), with functional dependency in activities of daily living, polypharmacy, and higher CHA2DS2-VASc stroke risk scores had a higher prevalence of oral anticoagulant use. A lower prevalence of anticoagulant use was observed for women, residents with limited life expectancy (aPR 0.80; 95% CI 0.76-0.83), those with moderate-to-severe cognitive impairment (aPR: 0.67; 95% CI 0.65-0.68), those using NSAIDs or antiplatelets, and non-white racial/ethnic groups. Factors associated with warfarin use were similar to associations observed with DOACs.

Conclusions: Approximately one in eight nursing home residents use oral anticoagulants and among oral anticoagulant users, only slightly more residents used warfarin than DOACs. Multiple geriatric comorbidities and concomitant drugs were associated with oral anticoagulant utilization. Further research is needed to inform resident-centered shared decision-making that explicitly considers individual specific risks and benefits of anticoagulation at all stages of the medication use continuum.