Trends in Primary Care Access at Community Health Centers 2012-2016: Evidence from an Experimental Study

Background: Community Health Centers (CHCs) have historically provided better primary care access for medically underserved populations than private physician offices. Prior to the Affordable Care Act (ACA), Medicaid-insured patients seeking a primary care appointment had greater success at CHCs than at non-CHC offices. Additionally, CHCs were much more likely to offer flexible financing to uninsured patients (Richards et al., 2014). However, many provisions of the ACA, notably including major insurance expansions and significantly expanded funding for CHCs, have affected meaningfully the demand for primary care services—including at CHCs—and CHCs’ capacity to deliver them (Cole et al., 2017; Polsky et al., 2017). How key measures of access to care at CHCs have changed following the ACA’s implementation remains unknown.

Methods: We conducted an experimental audit study in of primary care access in ten states (Arkansas, Georgia, Iowa, Illinois, Massachusetts, Montana, New Jersey, Oregon, Pennsylvania, Texas) in 2012/2013 and 2016. Trained auditors posing as new patients called CHCs and non-CHC primary care offices requesting the earliest available appointment. Auditors, whose insurance status (employer-sponsored insurance, Medicaid, or uninsured) and clinical scenario were randomly assigned, recorded whether they could get an appointment and if so, the wait time (days). Uninsured callers asked how much they needed to bring to the appointment and whether they could make a payment arrangement; in particular, we analyzed the dichotomous outcome whether the uninsured caller could bring $75 or less to the appointment. We computed difference-in-differences estimators to examine whether the changes in appointment availability and wait times (all insurance types) and financial flexibility (uninsured patients only) were different for CHCs versus non-CHCs in 2016 versus 2012/13. In sensitivity analyses, we also estimated difference-in-differences regressions adjusting for caller, office, and county characteristics—results were similar.

Findings: Appointment availability at CHCs increased for all insured patients between 2012/2013 and 2016. Among individuals with employer-sponsored insurance, appointment availability increased from 82.8% to 91.8% (p=.006) at CHCs and decreased at non-CHCs from 84.7% to 82.7% (p=.032) (difference-in-differences: 11.0 percentage points, p=.002). Among individuals with Medicaid, appointment availability increased from 80.4% to 93.8% (p=.001), and increased at non-CHCs from 56.1% to 60.9% (p=.004). The difference-in-differences for Medicaid patients of 8.6 percentage points (p=.030) indicates a significantly greater increase at CHCs than at non-CHCs. Appointment availability for uninsured callers remained unchanged at both CHCs (p=.595) and non-CHCs (p=.511). The share of uninsured callers who could bring $75 or less to the visit decreased at both CHCs (from 36.4% to 23.2%, p=.002) and non-CHCs (from 14.2% to 10.5%, p<.001), and the difference-in-differences was not significant. Mean days to an appointment – generally less than 2 weeks for all insurance types – increased slightly at non-CHCs, but stayed unchanged at CHCs during the study period.

Conclusion: CHCs remain an important entry point to primary care for undeserved patients. Appointment availability has improved for patients with employer-sponsored insurance or Medicaid at CHCs versus at non-CHCs under the ACA. However, these gains may not be sustained if CHCs’ expanded funding is not renewed.