Study Finds Low Risk of Overdose From Prescription Opioids
/By Pat Anson, PNN Editor
A large new study that identifies the top risk factors for an overdose involving prescription opioids has inadvertently shown just how low the risk is in the first place.
In a systematic review published in the Canadian Medical Association Journal, researchers at McMaster University looked at 28 studies involving nearly 24 million patients in the United States, Canada and the United Kingdom who were prescribed opioids for chronic pain.
They found 10 “predictors” associated with a higher risk of a fatal or nonfatal overdose, such as a patient taking high doses, having a substance abuse problem, and having multiple prescribers or pharmacies.
Many of these risk predictors are already well-known. What’s different about this study is that the researchers calculated the odds of an overdose happening in different situations and doses. That’s where things get interesting:
“High-certainty evidence from 14 studies involving 1,315,173 patients showed a linear dose–response relationship with opioid overdose. The association was small at a 50-mg morphine equivalent dose/day (OR 1.69, 95% CI 1.50–1.90) and large at 90 mg (OR 2.57, 95% CI 2.08–3.18), with an absolute risk 2.6 per 1000 for fatal overdose and 5.1 per 1000 for nonfatal overdose at a 90-mg morphine equivalent dose/day.”
In plain English, the risk of a fatal overdose at 90 morphine milligram equivalents (MME) is not “large.” It’s only 0.26% at 90 MME, which is considered a fairly high dose. The risk of a fatal overdose is even lower at 50 MME – just 0.16% -- a level that researchers say is “small-to-trivial.”
I’m not a statistician, but 0.26% and 0.16% seem like pretty small odds – similar to the lifetime risk of dying in a car accident, fire or drowning. Unlike opioids, there is no talk of a ban on swimming pools, motor vehicles or matches.
“I think that most people living with chronic pain would agree with your interpretation,” says co-author Jason Busse, DC, a clinical epidemiologist and professor of anesthesia at McMaster University. “I do think, however, that a minority of patients will place greater value on the possibility of overdose and death even though the absolute risk is small.”
Busse’s involvement in the study is notable, because he was the lead investigator in Canada’s 2017 opioid guideline, which suggest that initial doses of opioids be kept under 50 MME and strongly recommend that they never exceed 90 MME. The CDC opioid guideline in the United States makes similar recommendations.
Given the small risk of an overdose actually happening at 50 or 90 MME, the new study would seem to debunk both guidelines. Busse sees it a bit differently, telling me by email that the overdose calculations will help patients understand the risks associated with prescription opioids.
“Our work in this area has suggested that most people living with chronic pain, who have not found sufficient relief with non-opioid therapy, would be interested in a trial of opioids. Specifically, when provided with the evidence for benefits and harms, including the risk of overdose, that most patients in whom non-opioid therapy has proven insufficient would elect to pursue a trial of opioid therapy,” Busse said.
“By presenting the evidence to patients, and helping them to understand the anticipated benefits and harms, we can help ensure that the decisions they make are the right ones for themselves.”
Unfortunately, pain patients in the U.S. and Canada don’t get to make decisions for themselves. Decisions are made for them by doctors, pharmacists, regulators, and law enforcement. Patients increasingly have trouble finding a provider willing to treat them or getting a prescription filled at a pharmacy.
And because the “voluntary” opioid guidelines are usually treated as mandatory, patients who are prescribed opioids are often kept at ineffective low doses that are well below 90 or even 50 MME.
‘Opiods Kill and Opioids Are Bad’
Experts say the low risk of overdose from prescription opioids was established in previous studies, but people got caught up in opioid hysteria and ignored the evidence. The new study, they say, is no different.
“This paper examines well plowed ground and provides no new insight. Quite the contrary, it obfuscates through oversimplification of the problem,” says Stephen Nadeau, MD, a Professor of Neurology at the University of Florida College of Medicine. “The essential message is that opioids kill and opioids are bad.”
Nadeau says Busse and his co-authors ignored factors like genetic differences in opioid metabolism and put too much emphasis on the risk of a patient having multiple prescribers or pharmacies. The latter could simply be a sign that they were abandoned by a doctor or turned away by a pharmacy, not doctor shopping. Statistics mined from databases don’t tell you that.
“This paper takes a rigorous statistical approach to explain what is happening in a highly heterogeneous population in which there is a high probability of misconstruing the sources of variance. I think it would have been OK if published in 2015, but we have learned a thing or two since then and now the paper serves only to obfuscate and mislead,” said Nadeau.
Two findings in the study worth highlighting are that researchers found little risk of an overdose when a patient is co-prescribed opioids with benzodiazepines, sedatives or muscle relaxants – the so-called “Holy Trinity.” The overdose risk is also “small-to-trivial” when a patient is given long-acting opioids instead of short-acting ones. Those findings contradict the recommendations made in the U.S. and Canadian guidelines.
“The opioid crisis has generated interest in identifying patients at higher risk of addiction or overdose and has led to the development of several screening tools; however, these instruments have either not been validated or shown poor psychometric properties,” wrote lead author Li Wang, PhD, a researcher at McMaster University. “Our findings suggest that awareness of, and attention to, several patient and prescription characteristics, may help reduce the risk of opioid overdose among people living with chronic pain.”
One of the co-authors of the study is David Juurlink, MD, a member of Physicians for Responsible Opioid Prescribing (PROP), an anti-opioid activist group. Like several other PROP members, Juurlink disclosed that he has been a paid expert witness in legal cases involving opioids.
Previous studies have also found that the risk of overdose is small — 0.3% — in Medicaid and Medicare patients prescribed opioids for the first time. Another study of Medicare patients found that over 90% had little to no risk of overdosing. Low risk patients had only 0.006% chance of an overdose.
