Determinants of referral network size for screening colonoscopies in the Veterans Health Administration after the implementation of the MISSION Act.

Abstract

To measure key characteristics of the Veterans Health Administration's (VHA) Community Care (CC) referral network for screening colonoscopy and identify market and institutional factors associated with network size. VHA electronic health records, CC claim data, and National Plan and Provider Enumeration System. In this retrospective cross-sectional study, we measure the size of the VHA's CC referral networks over time and by VHA parent facility (n = 137). We used a multivariable linear regression to identify factors associated with network size at the market-year level. Network size was measured as the number of physicians who performed at least one VHA-purchased screening colonoscopy per 1000 enrollees at baseline. Data were extracted for all Veterans (n = 102,119) who underwent a screening colonoscopy purchased by the VHA from a non-VHA physician from 2018 to 2021. From 2018 to 2021, median network volume of screening colonoscopies per 1000 enrollees grew from 1.6 (IQR: 0.6, 4.6) to 3.6 (IQR: 1.6, 6.6). The median network size grew from 0.63 (IQR: 0.30, 1.26) to 0.92 (IQR: 0.57, 1.63). Finally, the median procedures per physician increased from 2.5 (IQR: 1.6, 4.2) to 3.2 (IQR: 2.4, 4.7). After adjusting for baseline market characteristics, volume of screening colonoscopies was positively related to network size (β = 0.15, 95% CI: [0.10, 0.20]), negatively related to procedures per physician (β = -0.12, 95% CI: [-0.18, -0.05]), and positively associated with the percent of rural enrollees (β = 0.01, 95% CI: [0.00, 0.01]). VHA facilities with a higher volume of VHA-purchased screening colonoscopies and more rural enrollees had more non-VHA physicians providing care. Geographic variation in referral networks may also explain differences in the effects of the MISSION Act on access to care and patient outcomes.