Health-care providers in the United States have been prescribing too many opioids for too many people, leading to a public health epidemic of overuse and accidental deaths. Studies estimate that in 2015, more than one in three American adults took a prescription opioid—and 22,000 people died because of opioid overdose.

Now, new research provides yet another reason for restraint in opioid prescribing. A study by Andrew Wiese and colleagues at Vanderbilt University called “Opioid Analgesic Use and Risk for Invasive Pneumococcal Diseases” shows opioid use may raise the risk of serious infections.

We interviewed Sascha Dublin, MD, PhD, a drug safety researcher at Kaiser Permanente Washington Health Research Institute and a primary care physician (internist) with Washington Permanente Medical Group. She and her colleague Michael Von Korff, ScD, recently co-wrote an editorial called “Prescription Opioids and Infection Risk: Research and Caution Needed,” which was published along with the Vanderbilt study, in the Annals of Internal Medicine.

Is it true that opioids make serious infections more likely?

They may. While we don’t know for sure, there is now a growing body of evidence that these drugs raise the risk for serious infections. The evidence comes from studies in the test tube and in animals—and now from three large studies in humans as well.

How does this latest study add to our understanding of the infection risk from opioids?

This is the third study done in a large population of people, and all three studies are remarkably consistent:

•The first, by our team in 2011, showed a substantial increase in pneumonia risk, particularly with new use of opioids.
•The second, by the same group that published the current work, focused on patients with rheumatoid arthritis.
•The current study is large and focuses on patients age five and older receiving Medicaid.

It’s true that in all these studies people were not randomly assigned to receive different treatments. But all three used sophisticated research methods to make up for the fact that people who are using opioids tend to have poorer health and more diseases than nonusers.

The fact that all three studies, each using different populations, have found very similar findings should make us sit up and take notice. This really cements the idea that opioids raise infection risk.

How much do opioids raise the risk of infections?

Use of opioids is linked to a 40% to 60% greater risk for pneumonia and other serious infections. The risk is highest with new use, long-acting opioids, and opioids that have been found to suppress the immune system. The risk with new use seems to be very high: In the study we did at Kaiser Permanente in 2011, people who had just started taking opioids had more than three times the risk of pneumonia in the first 2 weeks, compared to people not taking opioids.

What impact could opioids have on the immune system?

For many years, strong work in immunology and basic science has suggested that many opioids may suppress the immune system and increase infection risk. But until 2011, no large studies had been done in people.

This new study is really important because it suggests that even with the opioids that don’t suppress the immune system, infection risk is still increased. This could be happening through another way, such as making people sleepier, which can make them breathe in small amounts of saliva, carrying bacteria down into their lungs.

What research needs to be done next?

We urgently need research to compare various opioid medications head to head, such as comparing morphine with hydrocodone or tramadol. If a care provider believes that a patient needs an opioid, not much evidence is available right now to guide which one to prescribe. Evidence from the test tube suggests some opioids might harm the immune system but others might not—and studying this question in large human populations would be a major step forward.

We also urgently need more information about whether the higher infection risk persists with long-term use—or whether it’s elevated only for the first week or two. Some animal studies suggest the immune suppression might wear off after a few weeks, as the body gets used to the opioid.

We should do more and larger epidemiologic studies in older people with chronic disease and those with impaired immune function. We can use new techniques to make our research more efficient.

Do people with chronic pain need opioids?

Probably some people get some benefit from these drugs. But beyond the safety concerns, there is a real lack of evidence that opioids are effective for decreasing pain and improving function in people with chronic pain—particularly when used long term.

We need to start thinking about all the other options that people have for pain control and making sure patients know about the range of things they might be able to try—such as exercise, massage, and safer medications such as acetaminophen and ibuprofen—that are probably considerably safer and may work quite well for them.

What message should providers and patients take from this latest research?

Everyone is at risk for infections like pneumonia. But some of the patients being prescribed long-term opioids are especially vulnerable—such as patients over 65 or those with diseases like rheumatoid arthritis. I hope this research will help empower patients to ask their care providers whether they really need an opioid, how much it is likely to help them, and what they could choose to do instead.

And I hope it will reinforce the message that providers must be consistently cautious and closely monitor opioid prescribing in all patients—not only those who are seen as being at higher risk for drug addiction or overdose. In my own practice, I am going to start thinking about infection risk when I consider an opioid, and I will discuss this new evidence with my patients. Even when we aren’t 100% certain, our obligation as doctors is to protect our patients from harm.