RESUMO
There are concerns that physicians serving low-income, Medicaid patients, in the United States are less likely to adopt electronic health records and, if so, that Medicaid patients will be denied the benefits from electronic health record use. This study seeks to determine whether physicians treating Medicaid patients were less likely to have adopted electronic health records. Physician surveys completed during physicians' license renewal process in Arizona were merged with the physician licensing data and Medicaid administrative claims data. Survey responses were received from 50.7 percent (6,780 out of 13,380) of all physicians practicing in Arizona. Physician survey responses were used to identify whether the physician used electronic health records and the degree to which the physician exchanged electronic health records with other health-care providers. Medicaid claims data were used to identify which physicians provided health care to Medicaid beneficiaries. The primary outcome of interest was whether Medicaid providers were more or less likely to have adopted electronic health records. Logistic regression analysis was used to estimate average marginal effects. In multivariate analysis, physicians with 20 or more Medicaid patients during the survey cycle were 4.1 percent more likely to use an electronic health record and 5.2 percent more likely to be able to transmit electronic health records to at least one health-care provider outside of their practice. These effects increase in magnitude when the analysis is restricted to solo practice physicians This is the first study to find a pro-Medicaid gap in electronic health record adoption suggesting that the low income patients served by Arizona's Health Care Cost Containment System are not at a disadvantage with regard to electronic health record access and that Arizona's model of promoting electronic health record adoption merits further study.