Longitudinal associations between psychosocial morbidity, chronic health conditions, quality of life, and mortality in adult survivors of childhood cancer.

Abstract

e24147 Background: Psychosocial late effects of childhood cancer are often studied in isolation despite high rates of co-morbidity. We examined patterns of psychosocial morbidity and their associations with subsequent chronic health conditions, quality of life, and mortality in adult survivors of childhood cancer. Methods: St. Jude Lifetime Cohort participants (n = 4,009; median[range] 30.6[18.0-64.8] years of age; 20.4[5.5-52.3] years from diagnosis; 51% female) completed baseline questionnaires. Latent class analysis characterized psychosocial morbidity using 18 individual indicators across four domains: functional independence (e.g., independent living), emotional health (e.g., depressive symptoms), health behaviors (e.g., alcohol use), and socioeconomic status (e.g., employment). Generalized linear models, adjusted for age, sex, and race/ethnicity, examined longitudinal associations between class membership at cohort entry and persistent/new onset Grade 2-4 chronic health conditions from baseline to follow-up (n = 2,355, median f/u = 5.8 years). In separate adjusted models, associations with persistently poor/worsening quality of life at follow-up [SF-36 t-score <40] (n = 2,276, median f/u = 5.7 years) were examined. Cox Proportional models were used to examine associations between class membership and all-cause mortality adjusting for age, sex, and race/ethnicity. Results: Four latent classes were identified: (1) functional non-independence (21%); (2) risky health behaviors (15%); (3) poor emotional health (14%); and (4) low psychosocial morbidity (49%). Survivors in the functional non-independence and poor emotional health classes had increased risk of persistent/new onset Grade 2-4 cardiac (RR = 1.23, 95% CI 1.09-1.40; RR = 1.16, 95% CI 1.04-1.31), pulmonary (RR = 1.28, 95% CI 1.14-1.42, RR = 1.15; 95% CI 1.02-1.29), neurologic (RR = 2.22, 95% CI 1.93-2.56; RR = 1.97, 95% CI 1.70-2.28), and musculoskeletal conditions (RR = 1.85, 95% CI 1.58-2.15; RR = 1.46, 95% CI 1.21-1.75) compared to low psychosocial morbidity. Each class of psychosocial morbidity was significantly associated with persistently poor/worsening quality of life from baseline to follow-up compared to low psychosocial morbidity. Compared to the low psychosocial morbidity class, functional non-independence (HR = 3.37, 95% CI 2.43-4.67), risky health behaviors (HR = 1.77, 95% CI 1.21-2.59), and poor emotional health (HR = 2.70, 95% CI 1.89-3.84) classes were associated with increased risk of all-cause mortality. Conclusions: Baseline psychosocial morbidities experienced by survivors of childhood cancer are associated with persistent and new onset chronic health conditions, poor/worsening quality of life, and mortality. Interventions targeting psychosocial late effects may mitigate adverse health outcomes in survivors.