"First, we confirmed that the rate of diagnosed OUD has increased steadily among commercially insured adults, and we documented how the age distribution of OUD has changed. In 2008 diagnosed OUD among the youngest age group (ages 18–24) was more than double that among the oldest group (ages 55–64). However, in 2017 diagnosis rates exhibited a hump-shaped pattern in age, with the highest rate (4.75 per 1,000 enrollees) among the middle-aged (people ages 35–44) and the greatest increase among the near-elderly (ages 55–64).

"Second, even as diagnosed OUD rates have increased and overdose deaths have soared, the likelihood that patients would receive any treatment has plunged. This decrease was driven by a reduction in MAT rates, coupled with relatively steady rates of medication-free treatment (except for the youngest age group, which experienced a surge in this treatment modality). In 2008, 60 percent of those diagnosed with OUD in our primary sample had at least one claim for OUD medication, but that rate had declined to 50 percent by 2017. Notably, there has been a course reversal: The nadir of 43 percent occurred in 2016, and the subsequent surge in MAT among patients with OUD has coincided with a sharp increase (27 percent) in the number of practitioners who completed the training required to prescribe buprenorphine.5,6 However, the aggregate time-series trend in treatment rates is deeply concerning in light of the high and increasing reported prevalence of OUD and its life-threatening sequelae.

"Third, our analysis of medical spending for commercially insured patients with OUD revealed several new facts and points to areas for future research. We documented high average spending for patients with OUD, both before and after diagnosis (around $30,000 per person during the twelve-month period spanning a diagnosis). We found that receiving an OUD diagnosis was followed by a sizable increase in medical spending, but that increase was much larger for patients who chose medication-free treatment than for those who chose medication or no treatment. However, we also found significant differences in prediagnosis spending across these groups, which implies that selection effects must be addressed before a causal link between treatment modality and spending is inferred."


Karen Shen, Eric Barrette, and Leemore S. Dafny. Treatment Of Opioid Use Disorder Among Commercially Insured US Adults, 2008–17. Health Affairs 2020 39:6, 993-1001.