Vertical Integration and Advanced Practice Provider Employment in Physician Practices

Vertical integration has been a focal point of recent healthcare industry behavior and regulatory concern. Much is known about pricing and spending changes following hospitals’ purchases of physician practices; however, little is known about how the configuration of providers changes, namely the employment of nurse practitioners (NPs) and physician assistants (PAs), after integration. Using SK&A data, we examined a unique balanced panel (2008-2015) of 144,289 physician practices to estimate the effect of vertical integration on practices’ employment of NPs and PAs. We used a generalized difference-in-differences model with two-way fixed effects, which we then complemented with an event study analysis. During the analytic period, we found that vertical integration increased the likelihood that practices employed advanced practice providers. For NPs, there was a 1.2-percentage point increase in the likelihood of employing at least one NP following vertical integration across all practices. Relative to the sample mean, this was a 6% effect due to health system ownership, and it was unaffected by the inclusion of time-varying covariates. We stratified our sample by practice specialty (i.e., primary care, non-primary care, and multispecialty), and the strongest effect was found among non-primary care practices, which demonstrated a 9% increase in the probability of employing NPs post-integration. Further, in the event study analysis, there was an immediate increase in the likelihood of NP employment in non-primary care practices at the time of vertical integration, which then grew by roughly 40% in the following year and persisted thereafter. Similar patterns in NP employment did not emerge for primary care or multispecialty practices. Conversely, there was no effect of vertical integration on PA employment in physician practices, regardless of specialty type. Supplementary analyses suggest improved productivity as an underlying motivation for greater NP employment. We combined non-primary care physician data from our SK&A panel with physician-level inpatient and ambulatory procedure output data in the state of Florida from the Agency for Health Care Administration. Vertical integration was positively correlated with total procedure output at the physician level. On average, a physician produced 18 more procedures per year after vertical integration, which was a 7% increase relative to the analytic sample mean. Yet, the greater number of procedures came almost entirely from ambulatory procedures. Vertical integration was associated with 15 additional ambulatory procedures per year or an 8% increase relative to the sample mean. The effect of vertical integration on inpatient procedures was non-significant. Our results show that vertical integration leads to increased blending of high-skill provider types within physician practices, and these changes appear to coincide with enhanced productivity. The stronger response of NP employment among non-primary care practices indicates that NP clinicians are unlikely to be physician substitutes, rather, as our complementary data suggest, physicians are able to devote more time to performing procedures post-integration. These observed changes suggest improved efficiency in care delivery and are broadly appealing, especially since “team-based” care remains challenging and perhaps aspirational for many healthcare providers.