Unemployment among adult survivors of childhood cancer: a report from the childhood cancer survivor study

Abstract

Background: Adult childhood cancer survivors report high levels of unemployment, although it is unknown whether this is because of health or employability limitations.

Objectives: We examined 2 employment outcomes from 2003 in the Childhood Cancer Survivor Study (CCSS): (1) health-related unemployment and (2) unemployed but seeking work. We compared survivors with a nearest-age CCSS sibling cohort and examined demographic and treatment-related risk groups for each outcome.

Methods: We studied 6339 survivors and 1967 siblings ≥25 years of age excluding those unemployed by choice. Multivariable generalized linear models evaluated whether survivors were more likely to be unemployed than siblings and whether certain survivors were at a higher risk for unemployment.

Results: Survivors (10.4%) reported health-related unemployment more often than siblings (1.8%; Relative Risk [RR], 6.07; 95% Confidence Interval [CI], 4.32-8.53). Survivors (5.7%) were more likely to report being unemployed but seeking work than siblings (2.7%; RR, 1.90; 95% CI, 1.43-2.54). Health-related unemployment was more common in female survivors than males (Odds Ratio [OR], 1.73; 95% CI, 1.43-2.08). Cranial radiotherapy doses ≥25 Gy were associated with higher odds of unemployment (health-related: OR, 3.47; 95% CI, 2.54-4.74; seeking work: OR, 1.77; 95% CI, 1.15-2.71). Unemployed survivors reported higher levels of poor physical functioning than employed survivors, and had lower education and income and were more likely to be publicly insured than unemployed siblings.

Conclusions: Childhood cancer survivors have higher levels of unemployment because of health or being between jobs. High-risk survivors may need vocational assistance.

Figure 1. Relative Risk and 95% Confidence Interval (CI) of health-related unemployment and unemployed but seeking work for survivors of specific cancers type compared to siblings*

*Adjusted for current age, sex and race. Regressions for all survivors vs. siblings and for specific cancer diagnoses were performed using generalized linear models clustered by the survivor-siblings pairs to generate relative risk estimates because of the high proportion of central nervous system tumor patients and bone tumor patients reporting health-related unemployment (25% and 13%, respectively).

Figure 2. Percent reporting selected socioeconomic characteristics by survivors and siblings and percent with low SF-36 physical (PCS) and mental (MCS) component scores for survivors by employment*

*Because a random sample of siblings was given the SF-36 questions, we lacked enough responses for SF-36 calculation for siblings.