Access to Primary Care for New Medicaid Patients and Preventive Care Utilization: Evidence from an Audit Study

Access to primary care is a longstanding concern for adults in the United States, particularly for adults with Medicaid coverage. Poor access to a regular provider is a potential barrier to utilization of preventive care, including screening for chronic diseases and lifestyle management. The Affordable Care Act is projected to dramatically increase the number of Medicaid-insured adults, but many of these adults reside in areas where there is currently low supply of primary care providers.

This study examines the relationship between primary care supply and preventive care use in this study. Our measures of appointment availability were collected as part of a large-scale audit study that used trained interviewers posing as patients. These measures uniquely enable us to assess whether preventive care use for different groups may be more closely associated to the appointment availability for adults with private insurance versus Medicaid. We hypothesized that both privately insured and Medicaid-insured adults would be more likely to use preventive care in markets with higher appointment availability, and that the Medicaid appointment rate would matter most for adults likely to participate in Medicaid.

The audit study was conducted in ten diverse states, and provided information on appointment availability for Medicaid patients in 525 counties. We linked these measures to data on preventive care use (check-ups, flu shots, pap smears, PSA tests, and mammograms) among 52,672 non-elderly adults residing in the counties from the 2012 Behavioral Risk Factor Surveillance System. Although we could not directly observe whether an adult in the BRFSS sample was enrolled in Medicaid versus private insurance, we could assign all adults in our sample a probability of Medicaid enrollment based on data from the Current Population Survey (CPS). Using a set of variables common to the CPS and the BRFSS we estimated a logistic regression model for the probability of being enrolled in Medicaid among non-elderly adults with any current insurance coverage. The simulated Medicaid probability demonstrated a high concordance with actual Medicaid participation.

We found that a higher appointment availability rate for Medicaid in the county was associated with elevated, but not statistically significant, use of virtually all measures of preventive care among adults likely to be enrolled in Medicaid. By contrast, we found that a higher appointment rate for privately insured callers was associated with significantly lower use of most types of preventive care among adults likely to be privately insured. These associations were eliminated after adjusting for county poverty and uninsurance rates, indicating that the relationship for privately-insured populations is likely to be mediated by other features of the county environment related to demand for care.

The fact that the association between appointment rates and preventive care use are different for privately insured and Medicaid-insured populations underscores that access for these two populations is likely to be affected by very different market-level features. Future work to be presented at ASHE, will further explore whether these different relationships are mediated by local differences in practice patterns for these populations or in market-level segregation.