Rules to control opioid prescribing don’t always work as intended, studies say

An opioid prescription.Associated Press

Faced with a soaring death toll from opioid-related overdoses, federal and state policymakers in recent years have enacted measures intended to keep doctors from prescribing too many opioid painkillers.

But two studies published Wednesday in the journal JAMA Surgery suggest that such well-intentioned efforts sometimes don’t have the desired effect.

One found that after new rules made it harder to refill prescriptions for the painkiller hydrocodone, surgeons prescribed even more of the drug immediately after surgery.

The other study found that a requirement to consult a database that would reveal patients at risk of opioid misuse took up a lot of surgeons’ time but had no effect on prescribing practices — at least at one New Hampshire hospital during a six-month period.

The research might hold lessons for Massachusetts policymakers, who have sought to limit opioid prescribing as overdose deaths have risen to about four or five a day.

Both studies were fairly narrow, based in limited geographic areas and examining prescriptions by one type of doctor — surgeons. But they point to a broader issue, said Dr. Michael Barnett, a Harvard health-services researcher who studies opioid prescribing but was not involved in this research.

“Clinician behavior is harder to predict, when you put these kinds of limits on it, than we’d like to think,” said Barnett, a professor at the Harvard T.H. Chan School of Public Health. “Regardless of the law you put in place, physicians are going to respond to what patients need. . . . We need to ask a harder question: How do we influence health care decisions?”

Excessive prescribing played a major role in the genesis of the opioid crisis, although today street drugs cause the vast majority of overdoses. Controlling prescribing has thus become a focus of government efforts to combat opioid addiction, particularly in Massachusetts, where laws and policies have sought to prevent excessive opioid use and opioid prescribing has decreased by 30 percent.

In 2014, the US Drug Enforcement Administration reclassified hydrocodone — the main ingredient in Vicodin — from “Schedule III” to the more restrictive “Schedule II,” which meant that patients could no longer get prescriptions refilled over the phone.

To measure the effect on post-surgical prescribing, researchers at the University of Michigan examined prescriptions for 21,955 patients who underwent elective surgery in 75 Michigan hospitals from 2012 to 2015.

They found that hydrocodone refills did go down after the drug was reclassified — but patients left the hospital with prescriptions for more pills.

“Well-intended policy changes to curb opioid prescribing can have perverse effects,” said Dr. Michael Englesbe, a University of Michigan surgery professor and one of the study authors. He speculated that surgeons, aware that refills would be difficult, wrote heavy prescriptions to make sure that patients would have adequate relief and would not call the office in desperate pain on weekends or other times when it would be difficult to make appointments.

The study has implications for other types of prescribing limits, the authors assert. For example, a 2016 Massachusetts law requires doctors to limit first-time opioid prescriptions to seven days. That requirement could also lead to overprescribing if doctors are worried about their patients coming up short. A seven-day prescription for two tablets every four hours would hand a patient 84 tablets, when typically only 15 or fewer are needed for elective surgery, Englesbe said.

Ironically, the surgeons who boosted their initial prescriptions were acting on a faulty premise, according to Englesbe.

Previous research shows, he said, “the number of pills you give someone has no relationship to their likelihood of calling for a refill. The more pills you give a patient, the more they take, and they don’t rate their pain care any better. It’s counterintuitive.”

It’s up to the physician community to solve the problem through education about what’s best for patients, Englesbe said. Additionally, physician behavior often changes when insurers create financial incentives to follow best practices.

Although the study looked only at Michigan, Englesbe said that similar practices are probably happening elsewhere.

Dr. Melissa A. Covington, an anesthesiology professor at the University of Vermont’s Larner College of Medicine, said the Michigan study highlights the importance of gathering data on the effects of policies and guidelines.

“It’s important to look at the real-world outcome and how that changes prescribing,” she said.

In the other study, researchers at Dartmouth-Hitchcock Medical Center in New Hampshire examined the effects of a state requirement that surgeons consult a database that lists every prescription for controlled substances, called the prescription drug monitoring program, before prescribing opioids. Massachusetts has a similar requirement. The database can reveal whether a patient is obtaining drugs from multiple doctors or pharmacies, a sign of misuse.

The Dartmouth study compared surgeons’ opioid-prescribing practices at one hospital in the six months before and after the requirement went into effect, and found no change in how much they prescribed. Additionally there was no instance in which checking the database led to the identification of even one high-risk patient who was then denied an opioid prescription.

But Covington said checking the database brings other benefits beyond reducing the quantity of painkillers prescribed: It can inform prescribing decisions by revealing other medications patients are taking.