Targeting both patients and physicians with information about deprescribing proton pump inhibitors (PPIs) may be more effective than communicating with physicians alone, recent research suggests.

There is a “great overuse” of PPIs in the French healthcare system, said lead study author Jean-Pascal Fournier, MD, PhD, family physician in the Department of General Medicine at Nantes University in Nantes, France. About 40% of PPI prescriptions in the country could be considered inappropriate for indication or duration, he said in an interview.

In a study of 34,409 patients from 683 general practices (representing 1498 practitioners), an intervention targeted the physicians, or both the patients and physicians, to reduce PPI use. The combination intervention was found to significantly reduce PPI use at 1 year — in 14.9% of patients — compared with targeting physicians only (7.7%; adjusted absolute difference, 6.7%; 95% CI, 5.4%-8.2%; P < .001) or with usual care (7.0%; adjusted absolute difference, 6.9%; 95% CI, 5.7%-8.3%; P < .001). Results were published online in JAMA Internal Medicine.

The combined intervention showed efficacy, with double the reduction rate of PPIs compared with the physician-facing intervention, said Fournier. “Our conclusion is that the component that works in our intervention is mainly the component directed to patients.”

Fournier said he expected to see more of a difference in the physician-facing intervention. “Maybe because the intervention was not sufficiently tailored to the GPs’ [general practitioners’] needs; maybe it was a bit passive to just send them materials.”

The DEPRESCRIPP trial was conducted from November 2020 to November 2021 in primary care practices in two regions of western France (the Loire-Atlantique and Vendée). GPs were eligible if they had more than 100 patients on the practice list at baseline.

Patients were eligible if they were older than 18, affiliated with the French National Health Insurance System, and had been dispensed more than 300 defined daily doses (DDDs) of PPIs in the year preceding baseline. Investigators excluded patients considered at risk for gastroduodenal lesions and older than 65 and those treated with corticosteroids, anticoagulants, or platelet aggregation inhibitors.

Practices were randomized 1:1:1 to one of three groups. The patient- and GP-facing intervention consisted of educational materials mailed to both patients and their GPs. Patients received a brochure about PPIs accompanied by a motivational cover letter. The brochure was designed to increase patient knowledge about PPIs, how they work, risks of long-term use, and alternative approaches such as lifestyle changes, and to encourage discussion with a health care professional about PPI dose reduction or discontinuation. GPs received a letter and details of a PPI decision-support deprescribing algorithm. Participants in the usual care group did not receive any intervention.

The primary outcome was PPI dose reduction at 1 year, measured as a 50% or more reduction in annual PPI use and proxied by reimbursement claims during the intervention year compared with the year preceding mailing of the educational materials. Secondary outcomes were gastroesophageal reflux disease (GERD) symptoms, assessed using the GERD Impact Scale (GIS); and patient attitudes toward deprescribing, assessed using the French version of the Revised Patient Attitudes to Deprescribing questionnaire. Both surveys were sent to a 10% random patient sample in each group 1 month before and 1 year after baseline. Data were analyzed from January to July 2025.

Mean patient age was 68.6 years; 56.7% were women. Mean baseline annual PPI use was 413.7 DDDs. Outcomes were analyzed using a modified intention-to-treat approach. All randomized patients and physicians were included except for those under legal guardianship, physicians and patients who left the practice between randomization and intervention, and those who moved or died during the study period. PPI dose reduction rate and GIS scores at 1 year were compared across study groups using a generalized linear mixed model.

At baseline, 1826 of 3455 patients (52.9%), corresponding to the 10% patient sample, responded to the GIS. Patients reported high GERD symptom frequencies with mean scores from 3.03 to 3.28 in the three groups. At 1 year, 1532 patients (44.3%) responded to the GIS. Mean scores did not differ significantly in the groups for upper gastrointestinal (GI) symptoms, other acid-related GI symptoms, and impact of symptoms on daily activities.

“It’s definitely a good study” with wide sample sizes that was cluster-randomized, said Sander Veldhuyzen van Zanten, MD, MSc, MPH, PhD, emeritus professor of medicine, Division of Gastroenterology at the University of Alberta in Edmonton, Alberta, Canada. But what was not clear from the study is how many patients reduced from two DDDs of PPIs to one or who discontinued the medication, said van Zanten, who co-authored a recent article about Canada’s new Choosing Wisely toolkit for deprescribing PPIs. He also would have liked to know about other symptoms beyond those related to GERD measured on the GIS scale.

The study also mentioned some of the associated risks of PPIs, such as bone fractures and Clostridioides difficile infection, but those are very rare, van Zanten said. “It’s an incredibly safe drug,” he said, but it makes sense to deprescribe it if it isn’t needed. Initial recommended treatment duration for PPIs is typically up to 8 weeks, he said. Options for deprescribing include tapering, taking them on demand as needed until symptoms resolve, or lowering the dose or frequency to a level that prevents symptoms from recurring.

The study was supported by a grant from the French Ministry of Health. Fournier and van Zanten reported having no relevant financial disclosures. 

Karen Blum is a freelance medical/science writer in the Baltimore area.