randomized Japan Automobile Manufacturers Association This trial shows that simple behaviorally designed electronic health record prompts can change prescribing habits in primary care without the need for new staff, additional time, or complex programs.

study: Electronic medical record intervention and deprescribing for older adults.. Image credit: Yelena Temirgaliyeva/Shutterstock.com

In a recent study published in Japan Automobile Manufacturers AssociationResearchers evaluated whether two interventions, precommitment and boosting, could increase the rate at which physicians stop prescribing drugs that should be restricted to older adults, such as benzodiazepines and sedative-hypnotics.

Compared to usual care, both behavioral science-based interventions increased deprescribing rates, with precommitment strategies having a greater effect, although overall deprescribing rates remained modest.

Why do elderly people continue to take dangerous drugs?

Potentially inappropriate medications, such as non-benzodiazepine sedative-hypnotics, benzodiazepines, and strong anticholinergic drugs, are commonly prescribed to older adults despite well-documented risks, including increased falls and hospitalization. Although clinical guidelines state that long-term intake of these drugs should be limited, discontinuing their prescription remains uncommon in daily practice. Barriers include limited clinician time, patient resistance, inertia, and inadequate practice tools.

Traditional deprescribing interventions, such as pharmacist-led reviews, expert involvement, and patient education, have had mixed effectiveness and are often resource-intensive. In contrast, electronic medical records (EHR)-based clinical decision support tools have been successful in improving evidence-based prescribing, but their effectiveness in reducing inappropriate long-term drug use in primary care is less clear.

Behavioral science provides strategies for overcoming decision-making barriers, such as precommitment and reinforcement. This may be particularly suitable for integration into society. EHR Improve workflow by avoiding time-sensitive clinical decision-making.

Physician-level randomization targets high-risk prescribing patterns

The researchers tested: EHR Interventions based on behavioral science principles may increase the rate at which physicians stop prescribing drugs that may have harmful consequences in older adults, compared with usual care.

They conducted a three-group, cluster-randomized clinical trial at a medical center within a large health system in Massachusetts. A total of 201 primary care physicians (PCP) were randomly assigned in a 1:1:1 ratio to either usual care, precommitment intervention, or add-on intervention.

Patients are eligible if they are 65 or older and have an appointment that includes them. PCP During the study period, participants received prescriptions for high-dose benzodiazepines, non-benzodiazepine sedative-hypnotics, or at least two strong anticholinergic drugs.

Randomization was done at the physician level to reduce contamination. The precommitment intervention included a sequential intervention EHR This notice urges doctors to first discuss the drug’s risks with patients and then stop prescribing the drug. In the booster intervention, a deprescription prompt was presented first, followed by an optional prompt. EHR Rather than being sent automatically, if your clinician requests it, you will receive a reminder in your inbox approximately 4 weeks later. Physicians administering conventional treatment received no visible instructions.

The primary outcome was deprescribing of at least one targeted drug, defined using: EHR Data on discontinuations, dose reductions, or non-updates, including both active and passive discontinuations. The researchers accounted for clustering in their analysis and applied statistical corrections for multiple comparisons.

Failure to reduce dose increases drug discontinuation

Of 1,146 eligible patients (approximately 70% female) with a mean age of 73.6 years, 32.5% experienced at least one potentially inappropriate medication discontinuation during an average follow-up of approximately 290 days.

Deprescription occurred most frequently in the precommitment group (36.8%), followed by the add-on group (34.3%), and least frequently in usual care (26.8%). This means that even with intervention, approximately one in three patients experienced deprescribing.

Compared to usual care, patients whose physicians received pre-intervention interventions were 40% more likely to have their medications discontinued, corresponding to an absolute increase of 10.4%.

The booster intervention also significantly increased deprescribing, with a relative increase of 26% and an absolute difference of 6.5% compared with usual care. These effects remained consistent after adjustment for patient demographics and most prespecified subgroups.

The intervention was particularly effective for patients taking only one class of targeted drugs, but the effect was smaller and not statistically significant for patients taking multiple classes of drugs.

No significant differences were observed between groups for secondary outcomes measuring pill quantity or cumulative dose of prescribed drugs, indicating that overall drug exposure was not reduced despite high discontinuation rates. No serious adverse events related to prescription discontinuation were reported, and mortality was low in all groups, although mortality was numerically higher in the booster group.

Behavioral EHR nudges change clinician prescribing behavior

This test demonstrated that EHR Behavioral science-based interventions have the potential to significantly increase discontinuation or reduction of potentially inappropriate medications in older adults receiving primary care, absent short-term safety concerns.

Both pre-commitment and booster strategies were effective, with the pre-commitment strategy producing the greatest improvement. These findings extend previous research by showing that: EHR– Embedded behavioral nudges do not require additional staff or patient care programs and can change clinician behavior at scale, but cannot eliminate inappropriate drug use.

Key strengths include randomized design, integration into routine clinical workflow, broad inclusion criteria, and intended use. EHR-Based results. The pragmatic nature of this study increases its generalizability to real-world primary care settings.

Some limitations should be noted. Prescription discontinuations that occur outside the health care system may not be captured, and passive discontinuations do not necessarily reflect intentional actions by clinicians. This study did not report downstream clinical outcomes such as falls or hospitalization, although they were prespecified. This is because claims-based data linkage was not available at the time of publication and outcomes based on secondary doses showed wide confidence intervals. Furthermore, the findings reflect a single academic health care system.

Overall, this study supports the use of behaviorally based information. EHR We offer the tool as a scalable and effective approach to reducing potentially inappropriate drug use in older adults, while highlighting the need for complementary strategies to achieve greater clinical efficacy.