番茄社区

ORLANDO -- Opioid overdoses fell after the FDA implemented a class-wide Risk Evaluation and Mitigation Strategy (REMS) for extended release/long-acting (ER/LA) opioids, but deaths did not, researchers reported here.

In an analysis of claims data, there was a small decrease in the rate of opioid poisoning and overdose deaths among all ER/LA opioid users in a propensity score-adjusted model after the program was implemented in 2013 (IRR 0.81, 95% CI 0.69 to 0.95), according to , a senior scientist at HealthCore, an Anthem company, and colleagues.

However, there was a numerical increase in the rate of opioid overdose deaths after implementation, so the analysis couldn't determine a cause-and-effect pattern, Stephenson said in a presentation at the American Academy of Pain Medicine (AAPM) meeting.

"We have to be careful about interpreting these things because there are so many REMS-type activities going on, so we can't say how much of the change was due to implementation," Stephenson noted.

, of Swedish Health in Seattle, the newly appointed AAPM president, said that the study did not evaluate differences among providers who had gone through the REMS training versus those who have not.

"Some studies have shown that those who did the education felt more comfortable [prescribing] opioids and were doing urine screens more frequently. But the education isn't mandatory," said Stanos, who was not involved in the study.

The ER/LA opioid REMS was approved in July 2012 with the goal of reducing serious adverse outcomes from the inappropriate prescription, misuse, and abuse of these powerful opioids. Last year, a 2-day FDA hearing concluded that it's been impossible to tell whether the program has had an effect on opioid overdose and death. Panelists at that meeting voted to overhaul the REMS, which they felt was ineffective.

Stephenson's group looked at data from the HealthCore Integrated Research Database, which collects claims information from parent company Anthem, as well as available de-identified data from Medicaid from three states. She noted that mortality assessments couldn't be made for Medicaid patients.

They included patients who'd had at least one ER/LA opioid prescription, and compared the 2 years before REMS implementation (July 1, 2010 to June 30, 2012) with the 2 years after it became effective on July 1, 2013. Overall, in the commercially insured population, they studied 81,958 opioid analgesic users before the REMS and 53,135 after it was implemented. In the Medicaid population, there were 4,589 ER/LA opioid users before implementation and 6,518 users after.

The researchers conducted several analyses of overdose events in commercially-insured patients, and most were non-significant, but risk was lower in the active period in the propensity-score adjusted versions for both current exposure and all person-time, they reported.

Those findings were similar for Medicaid patients as well, they said.

With regard to opioid overdose deaths, the rate was weakly and imprecisely elevated after the program was implemented for commercially insured patients, they reported (IRR 1.14, 95% CI 0.73 to 1.77). Also, the rate of all-cause mortality was slightly increased during this time (IRR 1.04, 95%CI 1.00 to 1.08).

Stephenson pointed out that among all ER/LA opioid users, the incidence of opioid overdose or poisoning was substantially higher in the Medicaid population, but adjusted estimates were similar to those in the commercially insured (IRR 0.75, 95% CI 0.55 to 1.02).

She warned again that it was not clear the extent to which this represents an increase in opioid deaths due to inappropriate use, appropriate use in sicker patients, or a more complete ascertainment of deaths through the wider recognition of opioid overdose deaths and changes in coding.

Since the REMS is one of many efforts to address opioid safety, she reiterated that it's not possible to conclusively determine the extent to which the REMS specifically has impacted changes in opioid overdose rates.

"All we can say is, looking at the 2 years before REMS and the 2 years after, this is what we found in our database and in Medicaid data from three states," she stated.

Comment