Overview
Proton pump inhibitors, often called PPIs, are among the most commonly used medicines for acid-related stomach problems such as gastroesophageal reflux disease, peptic ulcer disease, and prevention of stomach injury from certain medications. While these drugs can be very effective and life-changing when used for a clear reason, they are also frequently continued longer than necessary. Long-term or unnecessary use may expose patients to avoidable harms and add costs without improving outcomes.
This cluster randomized clinical trial examined whether a deprescribing intervention could safely reduce potentially inappropriate PPI use in primary care. Deprescribing means carefully reducing or stopping a medication when the expected benefits no longer outweigh the risks. The study found that a combined intervention directed at both patients and general practitioners was more effective than usual care, and more effective than a GP-only approach, in lowering PPI use without worsening long-term reflux-related symptoms.
Why PPI deprescribing matters
PPIs are powerful acid-suppressing medicines. Common examples include omeprazole, esomeprazole, pantoprazole, lansoprazole, and rabeprazole. They are often appropriate for short-term treatment of reflux symptoms, healing of ulcers, or protection in patients at high risk of bleeding. However, many people remain on them for years after the original reason has resolved.
Inappropriate long-term PPI use has been linked in observational studies to several concerns, including nutrient deficiencies, kidney problems, certain infections, bone-related complications, and drug interactions. Not every patient on a PPI is harmed, and many do need ongoing therapy. The key point is to identify those who no longer have a strong indication for treatment and to offer a safe step-down plan.
This issue is especially relevant in primary care, where prescribing is long term and often managed across multiple years. Patients may continue a medication because it was never reviewed, because they fear symptoms will return, or because there is uncertainty about when and how to stop. Clinicians may also be cautious about discontinuation if they worry about rebound acid hypersecretion, a temporary increase in acid-related symptoms after stopping a PPI.
Study design
This was a pragmatic, cluster-randomized, open-label clinical trial conducted in primary care practices in two regions of western France between November 12, 2020, and November 11, 2021. The study included adults aged 18 years or older who had been taking a PPI for at least one year, along with their general practitioners.
The trial used cluster randomization at the level of the practice, rather than individual patient randomization. This approach helps avoid contamination, meaning it reduces the chance that clinicians would apply the same intervention to patients in different study arms. A total of 683 primary care practices and 1,498 GPs participated, representing 34,409 patients.
Practices were assigned in a 1:1:1 ratio to one of three groups:
1. A patient- and GP-facing intervention
2. A GP-facing intervention only
3. Usual care
The patient- and GP-facing intervention combined two elements. Patients received a brochure explaining why and how PPIs might be deprescribed, while their GPs received a letter with a deprescribing algorithm. The algorithm outlined how to identify potentially inappropriate use and how to reduce or stop treatment in a stepwise way.
In the GP-only group, clinicians received the letter and algorithm, but patients did not receive the brochure. The usual care group received no targeted deprescribing materials.
Main outcome and how it was measured
The primary outcome was PPI dose reduction, defined as at least a 50% reduction in annual PPI use. This was measured in defined daily doses, or DDDs, and estimated using reimbursement claims data. In practical terms, the researchers looked at whether the amount of PPI dispensed or reimbursed over time dropped substantially after the intervention.
A secondary outcome was symptom burden, measured with the Gastroesophageal Reflux Disease Impact Scale, or GIS, in a 10% sample of participants. This scale helps assess whether reflux symptoms worsened in a clinically meaningful way after reducing PPI use.
Key results
The participants were older on average, with a mean age of 68.6 years, and a little more than half were women. Mean baseline annual PPI use was 413.7 DDDs, indicating prolonged treatment in many patients.
The intervention that targeted both patients and GPs produced the best results. PPI dose reduction occurred in 1,710 of 11,442 patients, or 14.9%, in the patient- and GP-facing group. In comparison, dose reduction occurred in 825 of 11,732 patients, or 7.0%, in the usual care group. After adjustment, this was an absolute difference of 6.9 percentage points.
The same combined intervention also outperformed the GP-only strategy. Dose reduction occurred in 14.9% of patients in the combined intervention group versus 7.7% in the GP-only group, an adjusted absolute difference of 6.7 percentage points.
Importantly, GIS scores did not differ significantly between groups. This suggests that reducing PPI use did not lead to a meaningful worsening of long-term gastroesophageal reflux symptoms in the sampled patients.
What the findings mean in clinical practice
These results support a simple but important principle: deprescribing is more successful when both the clinician and the patient are involved. A letter to the GP with an algorithm may improve prescribing awareness, but pairing that with direct patient education appears to strengthen the effect.
Why might this work better? Patients who understand why a medication is being reviewed are more likely to accept gradual dose reduction. They may also feel reassured that stopping or lowering the dose is planned, monitored, and reversible if symptoms return. Meanwhile, clinicians are given a structured pathway that reduces uncertainty and supports consistent decision-making.
The study also provides reassurance that a targeted deprescribing approach can reduce medication use without obvious harm to reflux-related quality of life, at least over the follow-up captured in this trial. That is particularly important because fear of symptom relapse is one of the biggest barriers to stopping PPIs.
Practical deprescribing approach
In routine practice, deprescribing should always begin with a review of the original indication. Some patients should continue PPIs long term, including those with severe erosive esophagitis, Barrett esophagus in selected cases, Zollinger-Ellison syndrome, or those requiring gastroprotection because of substantial bleeding risk.
For patients without a clear ongoing indication, common approaches include:
– Reducing the dose stepwise
– Switching to on-demand use
– Changing from daily to every-other-day dosing temporarily
– Stopping the drug and using antacids or an H2-receptor blocker for short-term symptom control if needed
It is also helpful to counsel patients about rebound symptoms, which may occur temporarily after dose reduction or stopping. These symptoms do not necessarily mean the disease is worsening; they may reflect the stomach adjusting after acid suppression is removed. A gradual taper and a clear follow-up plan can make discontinuation more acceptable.
Strengths and limitations of the study
This trial has several strengths. It was large, pragmatic, and based in real-world primary care settings, which makes the findings highly relevant to everyday practice. The cluster-randomized design reduced contamination between groups, and the use of claims data allowed objective measurement of prescribing behavior.
There are also limitations. The study measured prescription or reimbursement changes rather than direct pill ingestion, so actual patient adherence may differ from dispensed medication. Symptom outcomes were assessed in only a 10% sample, so the reflux-related safety assessment was less comprehensive than the prescribing data. In addition, the trial evaluated a specific intervention in a particular health system, so results may vary depending on local practice patterns, access to follow-up, and patient expectations.
Even so, the overall message is robust: educational and algorithm-based deprescribing strategies can meaningfully reduce potentially inappropriate PPI use.
Bottom line
This randomized clinical trial showed that a combined patient- and GP-facing deprescribing intervention reduced PPI use more effectively than usual care or a GP-only approach, without a significant negative impact on reflux symptoms. The findings support routine medication review in primary care and show that patient engagement is a key part of successful deprescribing.
For clinicians, the study reinforces the value of checking whether a PPI is still needed, especially in patients who have remained on therapy for years. For patients, it highlights that stopping or lowering a PPI can often be done safely when guided by a thoughtful plan.
Clinical takeaway
When PPIs are continued without reassessment, they may become a “default” medication rather than a necessary one. This study suggests that a brief, structured conversation supported by written information and a clear deprescribing algorithm can help correct that pattern. In primary care, where long-term medication stewardship is essential, involving both doctor and patient appears to be the most effective route to safer prescribing.
