番茄社区

"Prescription Opioid Use Among Adults With Chronic Pain: United States, 2019," the government report's title reads. Inside, based on responses from the National Health Interview Survey (NHIS) that year, it says, "22.1% of U.S. adults with chronic pain used a prescription opioid in the past 3 months."

Given that there might be no hotter topic around treatment for chronic pain than the utility of opioids, that sounded like important news.

But a closer look is likely to leave the field scratching its collective head as to whether these data mean anything at all -- and raises questions about the extent to which broad, poly-topic surveys like the NHIS can inform practice and policy.

Certainly the issue of opioids for chronic pain needs clarity. Having long endorsed opioid therapy, the medical community and policymakers reversed course over the past 10 years or so, such that patients who'd been maintained for years found it difficult or impossible to renew their prescriptions. Many physicians, fearing repercussions, stopped writing opioid scripts altogether.

So, it would be interesting to see just how commonly opioids are prescribed nowadays for chronic pain, given the many discouragements faced by patients and physicians.

Certainly the NHIS would seem to be a reasonable data source to get at the question. It's big (more than 7,000 people for the current analysis) and representative of the population at large.

The new , by researchers at the National Center for Health Statistics (NCHS), is based on responses to two of the NHIS's questions:

• "In the past 3 months, how often did you have pain? Would you say never, some days, most days, or every day?"

• "During the past 3 months, have you taken any opioid pain relievers prescribed by a doctor, dentist, or other health professional?"

An of 2019 NHIS data by the same researchers had classed those answering the first question with "most days" or "every day" as those with chronic pain, and they did the same in the new report.

Of that group -- which, in the earlier analysis, worked out to 20.4% of the whole adult sample -- 22.1% said "yes" to the second question.

And that's where the trouble lies.

It's not that these data were self-reported by respondents, open to varying interpretation, with no cross-check against medical records. Nor is it the researchers' arbitrary definition of chronic pain, although one could easily argue with it. It's the intersection with the use of opioids.

That is, could some of those reporting prescription opioid use previously had "chronic pain" that, thanks to treatment, is now reduced to pain some days or never? Wouldn't those individuals legitimately be said to have chronic pain? HIV patients don't stop being considered as such when their viral loads become undetectable with treatment.

One might wonder whether the 22.1% were merely those patients for whom opioids were not working. But one can't conclude that either, because the NCHS report didn't get into detail about patients' pain -- its intensity either at present or over time. Perhaps those respondents who reported pain most days or every day despite taking opioids still had it less severely. (In fairness, the report's authors hinted at a future analysis looking at "high impact" vs less severe chronic pain.)

Nor did the NCHS report distinguish between cancer and non-cancer chronic pain -- an important distinction for policy-making, since there is no controversy over opioids' appropriateness for cancer pain.

So in sum, the report can't really tell us how many people with chronic pain were using prescription opioids in 2019. And despite the authors' repeated allusions to the purported dangers of prescription opioids, as context for the analysis, the report says nothing at all about the appropriateness of the usage it did identify.

Comment