Understanding Sexual Activity and Chlamydia Testing Rate Based on Linked National Survey and Medicaid Claims Data

BACKGROUND: Monitoring adherence to national recommendations for annual chlamydia screening of female adolescents and young adult women is important for targeting quality improvement interventions to improve low screening rates. However, accurate measurement of rates may vary depending on the data source used to determine eligible sexually-active women. METHODS: The 2001–2004 NHANES data linked with Medicaid administrative data by respondent’s unique identifier, the 2011–2012 NHANES data, and the 2004 and 2010 Medicaid data were used in this cross-sectional analysis. We defined self-reported sexual activity by self-reported sexual behaviors, claim-identified sexual activity by reproductive-related claims among women who had ≥ one healthcare claim, HEDIS-defined sexual activity by reproductive-related claims among women who were enrolled in Medicaid for ≥330 days and had ≥ one healthcare claim, and chlamydia tests by claims submitted in the 12 months prior to the survey interview. RESULTS: Of Medicaid women aged 18–25 years, 91.5% self-reported to be sexually-active. Of self-reported sexually-active women aged 18–25 years, 92.0% had ≥ one healthcare claim in the 12 months prior to the survey interview; of this subpopulation, only 58.8% were enrolled in Medicaid for ≥ 330 days in the 12 months prior to the survey interview; of this further subpopulation, 74.1% had healthcare claims identifying them as sexually-active in the 12 months prior to the survey interview. Of HEDIS-defined sexually-active women, 42.4% had chlamydia testing. CONCLUSION: Our study suggests that the number of sexually-active women aged 18–25 years used as the denominator in the chlamydia testing measure could be significantly different, depending upon the definition applied and the data used. Our data highlight the limited representativeness of Medicaid population in the current HEDIS measure on chlamydia testing when a high proportion of women who were enrolled in Medicaid for <330 days had been excluded from the measure. The interventions that can improve the proportion of women who were enrolled in Medicaid for ≥ 330 days among all young Medicaid women are needed not only for improving health care services, but also for measuring quality of healthcare.