Identifying Cohabiting Couples in Administrative Data: Evidence from Medicare Address Data

Area of Science:

• Social Epidemiology
• Health Services Research
• Data Science

Background:

• Marital status is a key social determinant of health, income, and social support.
• Administrative data often lacks marital status information, limiting its use in research.
• Identifying cohabiting couples is crucial for comprehensive health and socioeconomic analyses.

Purpose of the Study:

• To assess the feasibility of identifying cohabiting couples using exact address and ZIP code history in Medicare administrative data.
• To determine if the identified couples exhibit characteristics of known married couples.

Main Methods:

• Utilized the 2018 Medicare Vital Status file and 2011-2014 Master Beneficiary Summary Files.
• Developed an algorithm using exact address data and ZIP code history to identify potential cohabiting couples.
• Compared characteristics of identified beneficiaries with known married couples from the Health and Retirement Study linked to Medicare claims.

Main Results:

• The algorithm successfully identified Medicare beneficiaries consistent with cohabiting couples.
• Identified beneficiaries exhibited characteristics aligned with assortative mating.
• The identified couples closely resembled married couples in a linked dataset.

Conclusions:

• Exact address and ZIP code history data offer a feasible method for identifying cohabiting couples within administrative datasets.
• This approach holds promise for incorporating relationship status into analyses of healthcare claims and other administrative data.
• Improved identification of cohabiting couples can enhance research on social determinants of health.