2019 HSR&D/QUERI National Conference

1096 — Engaging patients to promote deprescribing of potentially inappropriate medications

Lead/Presenter: Amy Linsky,  COIN - Bedford/Boston
All Authors: Linsky AM (Center for Healthcare Organization and Implementation Research, Boston), Kressin N (Center for Healthcare Organization and Implementation Research, Boston), Bokhour BG (Center for Healthcare Organization and Implementation Research, Bedford) Rosen AK (Center for Healthcare Organization and Implementation Research, Boston) Stolzmann K (Center for Healthcare Organization and Implementation Research, Boston) Simon SR (Center for Healthcare Organization and Implementation Research, Boston)

Few initiatives target medication dose reduction or discontinuation of potentially inappropriate medications. While some believe providers have primary responsibility, patients can also facilitate deprescribing. We sought to understand how medication-specific brochures activate Veterans to discuss medications with their providers and affect deprescribing.

We identified Veterans at one VA Medical Center with a primary care visit over a 3-month period eligible for one of two cohorts: 1) chronically prescribed a proton pump inhibitor (PPI), or 2) diabetes (DM) at risk of hypoglycemia and prescribed either insulin or a sulfonylurea. Each subject was mailed a VA-tailored EMPOWER ("Eliminating Medications through Patient Ownership of End Results") brochure two weeks prior to their visit. Chart review identified two outcomes: 1) documented discussion about the target medication (possible vs. no), and 2) target medication changes (dose increase or no change vs. dose decrease or discontinuation ["deprescribing"]). Control subjects were patients seen the month prior to the intervention who would have been eligible. Descriptive statistics examined patient characteristics (demographics and select comorbidities and medications) and prevalence of the two outcomes. Chi-square tests examined the association of receiving brochures with discussions and deprescribing. Within each cohort, chi-square tests assessed relationships between patients' characteristics with each outcome.

There were 254 Veterans in the intervention group (217 PPI, 37 DM) and 95 in the control group (80 PPI,15 DM). Subjects were predominantly male (94%), age > = 65 years (80%), and white (82%). There was a possible discussion for 32/254 (12.6%) intervention vs 1/95 (1.1%) control subjects (p = 0.001). Target medications were deprescribed for 37 (14.6%) intervention vs. 4 (4.2%) control subjects (p = 0.008). No medication or comorbidity was statistically associated with deprescribing except within the PPI cohort, where 24 (20%) of those with gastroesophageal reflux disease experienced deprescribing compared to 11 (6.25%) of those without reflux (p = 0.0004).

Targeted distribution of medication-specific brochures in advance of a primary care appointment increased both discussions about and deprescribing of potentially inappropriate medications. Engaging patients can lead to changes in clinicians' prescribing practice, specifically for deprescribing.

Stopping or reducing inappropriate medication has the potential to improve quality, patient safety, and efficiency.