Comparison of middle aged and older cancer survivors in the US Health and Retirement Study (HRS) and the Surveillance, Epidemiology, and End Results (SEER).

Abstract

Introduction: We evaluated the effectiveness of the US Health and Retirement Study (HRS) in representing middle-aged and older cancer survivors by comparing individual- and county-level characteristics with those of a comparable cohort in Surveillance, Epidemiology, and End Results (SEER).

Methods: We identified incident cancer survivors aged ≥ 50 years in the HRS and SEER biennially from 2000 to 2020. We calculated proportions of individual- level and county-level sociodemographic attributes for the sampling-weighted HRS and SEER. We calculated the standardized differences (SD) between the HRS and SEER, with an SD of ≥ 0.1 indicating a meaningful difference.

Results: Cancer survivors in the HRS and SEER had similar sociodemographic characteristics, with some exceptions. Across most years, the HRS had a lower proportion of cancer survivors in the younger baseline age group (e.g., in 2020, 1.3% in HRS vs. 7.4% in SEER for ages 50-54), but a higher proportion of non-Hispanic White (e.g., in 2020, 75.7% in HRS, 68.3% in SEER), and married (e.g., in 2020, 59.5% in HRS, 53.2% in SEER), all with SD ≥ 0.1. The general populations of their data collection areas were similar, while the HRS over-represented counties with a higher proportion of Hispanic residents.

Conclusions: The sociodemographic profiles of middle-aged and older cancer survivors in the HRS and SEER were similar, with some minor exceptions, reflecting their distinct objectives and data collection methodologies. Understanding the comparability between HRS and SEER is crucial for ensuring that HRS data can reliably inform cancer survivorship research across the US population while providing additional longitudinal aging and covariates data.