Electronic health record interventions to reduce postoperative pregabalin prescribing: a quality improvement initiative.

Abstract

Purpose: Perioperative gabapentinoids may not provide meaningful analgesia and can have significant adverse events. Our objective was to estimate the association of two electronic health record (EHR) interventions with pregabalin prescribing by the acute pain service (APS) at a multi-site academic health sciences network.

Methods: We conducted a quality improvement initiative using a retrospective observational cohort and a quasi-experimental interrupted time series design. Following a baseline period (19 January 2021-19 January 2022), we introduced a Best Practice Advisory that warned of pregabalin's risks for sedation or respiratory depression. On 19 June 2022, pregabalin was removed as a standard checkbox in the APS orders. The primary outcome was the proportion of patients receiving pregabalin during their APS admission. The balancing measure was the highest postoperative day one pain score. Analysis used segmented regression in an interrupted time series design to estimate the immediate (level) change, trend (slope), and total counterfactual differences, controlling for the preintervention trend.

Results: We included 10,667 patients (5,559 preintervention, 2,271 postintervention 1, and 2,837 postintervention 2). Preintervention, 1,284 APS admissions had a pregabalin order (23.1%) compared with 379 (16.7%) after intervention 1 and 490 (17.3%) after intervention 2. Our interrupted time series analysis did not identify significant immediate, trend, or total counterfactual differences associated with the interventions (intervention 1, total counterfactual P = 0.76; intervention 2, total counterfactual P = 0.11). Only the preintervention trend (-0.2% per week, 95% confidence interval, -0.5 to -0.1) was significantly different (P < 0.001). No changes in pain intensity scores occurred despite decreased pregabalin use over time.

Conclusion: We did not identify a significant association of EHR interventions with pregabalin prescribing. Nevertheless, a continued downtrend in pregabalin prescribing was not associated with worsening acute pain.