Methodological issues in evaluating expanded Medicaid coverage for pregnant women

Area of Science:

• Public Health
• Perinatal Epidemiology
• Health Services Research

Background:

• US Congress expanded Medicaid coverage for pregnant women to reduce infant mortality.
• Evaluating the impact of these Medicaid expansions on perinatal outcomes is crucial.
• Accurate data linkage is essential for assessing the effectiveness of health policy.

Purpose of the Study:

• To assess the effectiveness of expanded Medicaid coverage on pregnancy outcomes.
• To evaluate the feasibility of linking Medicaid enrollment data with vital records.
• To identify potential biases in outcome assessment due to data completeness.

Main Methods:

• Linked Medicaid enrollment files for mothers and infants to birth, death, and fetal death certificates in Tennessee (1984-1987).
• Assessed data completeness for mother, delivery, and child information among Medicaid-reimbursed births.
• Compared characteristics and mortality rates between births with complete and incomplete data.

Main Results:

• Complete data on mother, delivery, and child were available for only 75% of Medicaid-reimbursed births.
• Medicaid births with complete data exhibited different characteristics and lower mortality rates than those with missing elements.
• Incomplete data significantly underascertained fetal, perinatal, and neonatal mortality rates.

Conclusions:

• Evaluating expanded Medicaid coverage requires awareness of data completeness issues.
• Studies relying on incomplete data may seriously underestimate mortality rates.
• Improved data collection and linkage are necessary for accurate assessment of Medicaid's impact on maternal and infant health.