Far and away, the most common initial exposure to narcotics is through a healthcare encounter. Heroin, opium, and other preparations are far less common than the ubiquitous prescription narcotics inundating our population. As opiate overdose-related morbidity and mortality climbs, increasing focus is rightly turned to the physicians supplying these medications.
This most recent article is from the New England Journal of Medicine, and is focused on the prescriptions provided in the Emergency Department. The Emergency Department is not one of the major prescription sources of narcotics, but may be an important source of exposure, regardless. Through a retrospective analysis of a 3-year cohort of Medicare beneficiaries, these authors defined two treatment groups: patients treated by a lowest-quartile of physician opiate prescribing rates, and those treated by a highest-quartile of physician opiate prescribing rates. The lowest quartile provided narcotics to approximately 7% of ED visits, while the highest to approximately 24%. In the subsequent 12 month period, those who received treatment by the highest-quartile of physician prescribing were more likely to fill at least an additional 6-month supply of another opiate. This adjusted odds ratio of 1.30 compared with the lowest-quartile includes a dose-response relationship with the two middle quartiles, as well.
The authors note this, essentially, means for every 48 patients prescribed an opiate above the lowest prescribing baseline, one additional patient then receives a long-term prescription they otherwise would not. Their calculation is a little odd – factoring both the additional likelihood of a prescription and the absolute increase in subsequent prescription rates. The true value likely lies between that and the NNH calculated from the absolute percentage difference – 0.35%, or ~280. No reliable or specific harms were detected with regards to these patients – additional Emergency Department visits, deaths by overdose, or subsequent encounters for potential side effects were similar between the groups. It is reasonable, however, to expect these additional prescriptions have some small number of downstream harms.
There are many indirect effects measured here, including pinning the entire primary outcome observation on clinical “inertia” resulting from the initial Emergency Department prescription. They also could not, by their methods, specifically attribute a prescription for opiates to any individual physician – they used the date of an index visit matched to a filled prescription to do so.
That said, the net effect here probably relates to less-restrictive prescribing resulting in prescriptions dispensed to patients for whom dependency is more likely. The effect size is small, but across the entire healthcare system, even small effect sizes result in potentially large absolute magnitudes of effect. The takeaway is not terribly profound – physicians should be judicious as possible with regard both their prescribing rate and the number of morphine equivalents prescribed.
Finally, the article concludes with a pleasing close-up photograph of a tiger.
“Opioid-Prescribing Patterns of Emergency Physicians and Risk of Long-Term Use”
http://www.nejm.org/doi/full/10.1056/NEJMsa1610524