Reducing unnecessary PPI use may help save lives

Follow guidelines and implement electronic systems to improve use.

Proton-pump inhibitors (PPIs) are often given prophylactically to hospitalized patients even though they are known to increase the risk of serious conditions such as Clostridium difficile infection, osteoporosis, pneumonia, and more. Many patients are also on unneeded PPIs before they enter and after they leave the hospital, putting them at even greater risk of complications. Hospitalists may be able to help.

They are in a unique position to make a real difference on the issue of PPI overuse, by both altering their own prescription practices and addressing other physicians', said Charles F. Seifert, PharmD, a professor of pharmacy practice at Texas Tech University Health Sciences Center in Lubbock.

Photo by Thinkstock
Photo by Thinkstock

“I get sick and tired of physicians who know this isn't right, but they are not going to change it because they don't want to infringe on another physician's turf. I completely understand, but there comes a time when we need to act. We need to get patients off PPIs and try to curb this epidemic because it's rampant,” Dr. Seifert said.

While growing awareness of the dangers is helping reduce PPI overuse in some areas, there is much variation in practice, according to Bradley A. Sharpe, MD, FACP, a professor of clinical medicine and a hospitalist at the University of California San Francisco (UCSF).

“It's very much a local practice culture issue that hospitals need to examine and grapple with,” Dr. Sharpe said. “It is clear that overprescribing or misprescribing by hospitalists is an issue in many places.”

Hospitalists can take the lead in resolving this issue by bringing their own prescriptions into accordance with evidence and guidelines, as well by helping to improve others' practices by working on quality improvement projects, experts said.


One of the challenges of reducing PPI overuse is that the drugs have very few acute short-term side effects, so it can feel benign to prescribe them for the duration of a hospital visit, according to Dr. Sharpe. This is particularly common in patients the hospitalist is worried about in general, such as people with multiple comorbidities and/or advanced age, he said.

However, clear guidelines around appropriate PPI prescribing that caution against such overuse have been around since 1999. Physicians who overprescribe the drugs either don't know the guidelines exist or don't know the data, speculated Dr. Sharpe.

“The reasons to prescribe PPIs in the hospital are quite narrow,” he said. “They are appropriate for patients who we think are having a gastrointestinal bleeding episode but not for prevention of such an episode.”

Reducing prophylactic use is a key way that hospitalists can improve PPI prescribing, agreed Matthew Pappas, MD, FACP, a hospitalist at the Ann Arbor Veterans Administration Medical Center and the University of Michigan.

“With very rare exceptions, hospitalists should not start PPIs as a preventive measure,” Dr. Pappas said. “Physicians should not start PPIs in the hospital unless their patient is having an active GI bleed.”

A challenge to implementing the experts' advice is that electronic medical records often include PPIs on the checklist of things that should be considered upon admission. “Often there is an area for prophylaxis of conditions, such as venous thromboembolism prophylaxis, so it is easy to just check the box for PPIs at that point too,” said Dr. Sharpe. “We have made it too easy to prescribe them.”

Successful efforts

To make it easier for prescribers to do the right thing, UCSF conducted a quality improvement project to reduce unnecessary PPI use in the hospital, Dr. Sharpe said. First, a team created a consensus-based guideline on the topic, then had an educational intervention for key prescribers. A pharmacist-led review and intervention followed. The result was a statistically significant reduction in inappropriate use of stress ulcer prophylaxis in the ICU.

“We focused on making sure PPIs were prescribed appropriately in the ICU and then stopped when patients leave the ICU,” Dr. Sharpe said.

One of the keys was active pharmacist engagement; pharmacists reached out to prescribers when they saw use that they thought might be inappropriate and suggesting either discontinuation or alternative medications. “The combination of education and intervention can be a recipe for decreasing inappropriate use,” Dr. Sharpe said.

At Texas Tech, nurses, physicians, and pharmacists have worked to improve PPI use with a medication reconciliation process at discharge. “We generate a list of all the medications patients are on, and physicians have to check off if they want the patient to continue on the medication. Having them make this conscious decision has been very effective in reducing things that would have slipped by in the past,” said Dr. Seifert.

Texas Tech has also experimented with having an educational prompt about C. diff appear in the electronic medical record when PPIs are prescribed. It has brought interesting results, with clinicians overprescribing another drug instead, Dr. Seifert reported.

“It stated if PPIs are not indicated and gave them the options to prescribe H2 blockers instead,” he said. “Instead of not giving acid suppression, almost all of our patients are now on famotidine. We are trying to change prescribing habits by some of these computer prescriber order entry initiatives, but we didn't have the result we thought we were going to have.”

Consequences of not optimizing inpatient use

The consequences of not optimizing PPI use in the hospital are significant, according to a study Dr. Pappas and colleagues published in the November 2015 Journal of General Internal Medicine.

On the basis of existing data about the risks of PPIs (pneumonia and C. diff) and the benefits (reduced gastrointestinal bleeding), they simulated the risk of PPI use in non-ICU hospitalized patients—both prescriptions begun in the hospital and continued from outpatient care—and found that PPI use is exposing many inpatients to a higher risk of death than they would otherwise have.

“This is a small effect—1 additional death per 800 to 1,000 patients,” Dr. Pappas said. “It's a small effect, but it's a consistent effect. With over 35 million hospitalizations a year, that effect would lead to potentially several thousand unnecessary deaths.”

There are some risks in the other direction, he noted. Although his simulation predicted overall harm from continuing unneeded outpatient PPIs on admission, Dr. Pappas warned that patients who have been on PPIs for a long time may have rebound symptoms once the drugs are discontinued. “For example, it is conceivable that it could lead to chest pain that leads to other workups in the hospital and overwhelm the negative effect of continuing the PPI,” he said.

Because of these considerations, he suggested physicians lean toward trying to remove patients from PPIs while hospitalized but also counsel them about the risks and benefits of doing so.

Although for most patients neither the risks nor benefits are likely to be enormous, reducing PPI overuse is an important consideration in improving health care overall, according to Dr. Pappas. “While [the risk] is small relative to, say, the mortality of stroke, this is one of the details that can elevate our care from good to great,” he said.