International Deprescribing Guidelines Did Not Impact Actual Practice in Deprescribing of Potentially Inappropriate Medications for Nursing Home Residents: An Interrupted Time Series Analysis.

Abstract

Background: Deprescribing guidelines reduce the use of potentially inappropriate medications (PIMs) in trial settings; however, their real-world impact remains unclear. Therefore, this study assesses deprescribing trends and the impact of guideline publications (STOPPFrail, proton pump inhibitors [PPIs], and antipsychotics) on these trends among nursing home residents with limited life expectancy in Belgium.

Methods: Deprescribing was assessed using linked healthcare reimbursement data for all residents aged 65 years and older who died between 2014 and 2019. In total, 15 PIMs from STOPPFrail version 1 were selected. Deprescribing was operationalized as discontinuing at least one PIM in the last 4 months of life among those who had been prescribed these medications chronically between 6-12 months prior to death. To identify changes in the trend of deprescribing, we employed a joinpoint linear regression model. We calculated the average quarterly percent change (AQPC) and 95% confidence intervals (CIs). In addition, we used autoregressive integrated moving average (ARIMA) modeling to explore the impact of publication of these guidelines on four commonly used PIMs: PPIs, antipsychotics, lipid modifying agents, and calcium.

Results: Among 244,865 residents, 169,782 (69.3%) were chronically prescribed at least one PIM and 50,487 (29.7%) had at least one discontinued. The prevalence of deprescribing declined from 31.7 to 27.66% between the first quarter of 2014 and the fourth quarter of 2019, with an average quarterly percent change decline of - 0.47% (95% CI - 0.85, - 0.10). No joinpoints were identified, indicating a consistent linear trend with no interruptions or statistically significant shifts in the rate of change in deprescribing prevalence. ARIMA modeling found that the publication of deprescribing guidelines had no impact on deprescribing trends.

Conclusions: Despite the high use of PIMs, and the publication of the STOPPFrail, PPI, and antipsychotic deprescribing guidelines, deprescribing rates remained low and even decreased. These findings emphasize the importance of implementation efforts that go well beyond guideline publications to effectively change deprescribing practices.