Clinicians were told their patient had died of an overdose. Then opioid prescribing dropped

The letters arrived from the San Diego County medical examiner’s office, informing clinicians that one of their patients had died from a prescription drug overdose.

These letters appear to have had an impact — prescriptions of addictive painkillers dropped.

In a small, randomized trial, researchers showed that this intervention — aimed at making the abstract issue of safe prescribing individually tangible — led to a slight reduction in the amount of opioids these clinicians prescribed. What’s more, prescribers who received the letters doled out fewer of the most powerful doses and appeared to start fewer patients on opioids compared with doctors who did not receive the letters.

Authors of the study published Thursday and experts not involved with the research urged other communities around the country to adopt the same strategy.

“Hearing about one person’s death can be really impactful,” said Jason Doctor, the lead author of the study and an expert in behavioral science and policy at the University of Southern California. “People often don’t change their behavior unless they have a really salient, personal experience.”

The approach described in the new paper, which was published in the journal Science, was easy to implement, inexpensive, and relied on existing tools, the authors said. Medical examiners’ offices already track overdose deaths for health authorities, and prescribing histories can be looked up through prescription drug monitoring programs, which almost all states have. (Missouri, the last holdout state, has attempted to start one, though its status is in flux.)

In effect, the letters simply served to close the information gap between medical examiners and prescribers.

“I think it’s a no-brainer,” said Dr. Andrew Kolodny, co-director of Brandeis University’s Opioid Policy Research Collaborative, who was not involved with the study. “If a prescriber’s patient ultimately loses their life to a drug overdose, that prescriber should be notified — there should be that feedback.”

Still, such an effort wouldn’t be feasible everywhere, said Dr. Kim Collins, a forensic pathologist in South Carolina and 2018 president of the National Association of Medical Examiners. Depending on a jurisdiction’s policies and coroner and medical examiner system, pathologists can’t always disclose causes of death of individual patients.

“If there was a way to do that, it would be great,” said Collins, who was not involved with the study. “But by law you just can’t do it in every location.”

Freewheeling prescribing of painkillers helped fuel the opioid epidemic that now kills tens of thousands of Americans a year. Many became addicted by using prescribed drugs — either their own or those diverted from someone else — before turning to street drugs such as heroin.

Although prescribing rates have fallen in recent years, efforts such as sending general letters to frequent prescribers haven’t always been shown to be effective. Top-down recommendations sometimes meet resistance from patients as well as prescribers, who bristle at attempts they think threaten their autonomy or relationships with patients.

For the study, the researchers took a more subtle tack.

They focused on 170 people whose primary or contributory cause of death was a prescription drug overdose (many had both prescription and illicit drugs in their systems) and who had received a prescription for a controlled substance in the year before their deaths. Then, using California’s prescription monitoring program, called the Controlled Substance Utilization Review and Evaluation System, or CURES, they identified hundreds of clinicians — nurse practitioners, dentists, doctors, and others — who had prescribed the controlled substances.

More than 380 of those prescribers received a letter signed by Dr. Jonathan Lucas, San Diego County’s chief deputy medical examiner.

“This is a courtesy communication to inform you that your patient [Name, date of birth] died on [date],” the letter said. “Prescription drug overdose was either the primary cause of death or contributed to the death.”

The letter included recommendations for safer prescribing, including using CURES, avoiding providing both opioids and benzodiazepines, and reducing the length and strength of opioid prescriptions.

“We did not want it to be a punitive letter,” Doctor said. “We just wanted to make them aware of the death.”

The awareness seems to have changed behavior, though marginally. While the 438 prescribers in a control group — those who didn’t get a letter — kept prescribing roughly the same amount of opioids each day, those who received the letter reduced their opioid prescriptions within a few months from an average of 72.5 morphine milligram equivalents a day to 65.7 MMEs. (A standard Vicodin pill is 5 MMEs.)

“It would have been nice to see more of an impact, though it’s hard to get patients’ prescriptions down,” Kolodny said.

The reduction was modest, Doctor acknowledged. But both he and Kolodny said a small initial reduction might be responsible on the prescribers’ part. Some advocates argue that slashing prescriptions too rapidly can hurt patients who legitimately need painkillers and even leave them vulnerable to seeking illicit drugs (though experts debate just how pervasive a problem this actually is).

“On the plus side, you could say this means that the doctors didn’t overreact and cut off all their prescriptions,” Kolodny said.

Letter recipients also reduced their prescriptions of high-dose opioids (50 MMEs and 90 MMEs a day) more than practitioners in the control group. And they seemed to be slightly less likely to start a new patient on opioids than other doctors, though limitations in what the researchers could track in CURES made it difficult to know whether these patients were truly getting their first opioid prescription.

The authors identified several possible reasons why prescribing declined among those sent a letter: There was a sense that they were being monitored; they were no longer just exposed to patients who returned to appointments and were uneventfully using the medicines; and “people rely upon knowledge that is impactful, recent, and easy to retrieve from memory,” they wrote.

Doctor and his colleagues are continuing to track if the letters have a long-term effect on prescribing, and moving ahead, he said he hoped medical examiners would consider starting such a campaign. He proposed adding information about medication-assisted therapy — which is the gold standard for treating opioid addiction — to the letters.

“It’s not going to be a silver bullet,” Doctor said about the approach. But, he added, “I’m hoping that counties will start using this, maybe even health systems will start asking for this type of data. Even states could apply this.”