Spillover effects in healthcare refer to the unintended consequences – both positive and negative – of programs and policies that affect populations beyond those targeted. For example, an initiative by the Centers for Medicare and Medicaid Services (CMS) tested sending warning letters to doctors prescribing a high volume of antipsychotic drugs in the Medicare population. That approach led a decline in those prescriptions, even to commercially insured patients. Spillover effects are fascinating, are found all over healthcare, and can amplify or diminish the effects of any given program.
Many readers are aware of CMS’s Oncology Care Model (OCM): a value-based payment model that allowed oncology practices to voluntarily share clinical and financial risk data about their patients with Medicare. OCM ran from 2016 to 2022 and is rightly heralded as a first-of-its-kind approach, in that it was the first alternative payment model in specialty medicine. It garnered lots of attention and practice participation, but also dealt a heavy financial loss to CMS, which had to write off more than $300 million in net losses, as reported by the official analysis. In response, CMS sharply reduced the size of upfront payments and mandated downside risk in its successor, the Enhancing Oncology Model (EOM), which began last year.
Despite the absence of a “primary effect,” we wondered whether the OCM may have produced a spillover effect. If we did find one, this would suggest that a traditional evaluation limited to the savings experienced only by the model sponsor would represent an underestimate of the total impact.