Retrospective Evaluation of a Pragmatic Deprescribing Initiative in a Skilled Nursing Facility System

Context: Polypharmacy, referring to multiple medications or the use of more drugs than are medically necessary, may lead to undesirable consequences, increasing the risk of adverse drug effects and leading to increases in falls and hospitalizations among older persons. Our objective was to retrospectively determine whether a pragmatic deprescribing protocol reduced 8 common classes of medications in two skilled nursing facilities (SNFs) in a single system.

Objective: To determine whether a deprescribing effort reduced several key classes of medications, and the overall number of medication classes per patient, among long-term residents of SNFs.

Study Design and Analysis: Retrospective, longitudinal pre/post evaluation. Data from before and during the implementation of the deprescribing effort (2018-2019) were compared with post-intervention years (2020-2021).

Setting and Populations Studied: Long-term resident data reported through annual comprehensive minimum data set (MDS) reviews conducted at two SNFs located in central New York State between 2018 and 2021 (N=11,862).

Intervention/Instrument: Multi-faceted, pragmatic interdisciplinary deprescribing effort to reduce medications in SNF residence including clinician education, guideline development, and individual chart reviews began in 2019.

Outcome Measures: Overall percentage of patients on each of eight classes of medications, compare between pre- and post- intervention periods via cross-tabulation with chi-square. Mean total medication classes for each patient compared pre/post via one-way Analysis of Variance (ANOVA).

Results: Mean number of medications were lower, post-intervention, at both facilities (mean=1.88 pre vs. 1.85 post, NS). Significant drops were observed in 4 of 8 categories, including Diuretics (-3.7%, p=.001), Opioids (-4.9%, p=.001), Antipsychotics (-2.0%, p=.002), Antibiotics (-1.6%, p=.045). Antianxiety medications were non-significantly reduced by 0.5%. Antidepressant usage increased (4.6%, p<.001), as did Anticoagulants (4.4%, p<.001). Hypnotic usage went up slightly, but was rare in both periods (0.9% pre vs. 1.4% post, p=.004), and represents a real difference of 3 patients in total.

Conclusions: A combined, comprehensive approach to deprescribing was associated with a reduction in overall number of medication classes per resident and in several key classes of medications. Antidepressant usage likely increased as a safer offset to reductions in higher risk medications.