Comparison of Young Adult Female Cancer Survivors Recruited from a Population-Based Cancer Registry to Eligible Survivors

Abstract

Background: Population-based cancer registries provide a resource to recruit young adult cancer survivors who may not be easily identified otherwise. Methods:We compared demographic and cancer-related characteristics of participants in a cohort of female young adult cancer survivors to those of eligible survivors in the Georgia Cancer Registry, a population-based registry in the United States. We examined associations between survivor characteristics and non-participation using logistic regression and associations between survivor characteristics and different types of non-participation (refusal, unable to contact, or unresolved vs. interviewed) using polytomous regression. Results:The Georgia Cancer Registry was able to contact 60% of eligible women (3,061/5,137). Of those, 78% agreed to study contact (n=2,378), and of those, 56% were interviewed (n=1,342). Participation was similar across age at contact and at diagnosis but varied across cancer type from 17% for cervical cancer to 32% for breast cancer. White women were slightly more likely to be interviewed (28%) than African-American women (23%), which was mostly attributable to greater difficulty in contacting African-American women (odds ratio 1.7, 95% confidence interval: 1.5, 2.1). Conclusions:The greatest challenge to recruiting women was contacting them, which differed across some but not all demographic and cancer-related characteristics. When successfully contacted, most survivors agreed to participate. Impact: Population-based cancer registries can serve as an invaluable resource to recruit representative samples of young adult cancer survivors, who are otherwise difficult to identify.