ideally have results from studies that had the following six characteristics (Steiner et al., 2004):
Inclusion of cancer survivors that represented the entire population of U.S. cancer survivors. Many studies are based on survivors followed at one cancer center, or who are from particular geographic areas. Their employment experience may not reflect that of the nation. Ideally, survivors would be selected for study from population-based cancer registries.
Designed to provide a prospective and longitudinal look at work outcomes so that both short-term and long-term work outcomes could be assessed and the dynamic nature of employment could be understood.
Include assessments of work, including information on the type, amount, content, physical demands, cognitive demands, and attitudes about work.
Include assessments of the impact of cancer on the economic status of the individual and the family.
Identify moderators of work return and work function, particularly those that are susceptible to intervention (e.g., availability of health insurance and disability benefits to offset lost income).
Include a cohort of survivors that is sufficiently large to allow multivariate statistical analysis and that provides information on important groups (e.g., minority groups, cancer types).
The committee reviewed the literature published in the past 10 years on the employment experience of U.S. cancer survivors who were studied in 1992, the year the ADA took effect, or later.2 Most of the studies reviewed had some, but rarely all, of the ideal attributes just described. There are few prospective studies of cancer’s effects on employment, but those that are available provide important insights into how interventions could be designed to assist cancer survivors.
In one prospective study, women with invasive breast cancer were less likely to work 6 months following diagnosis relative to a control sample of women. Breast cancer survivors who remained working worked fewer hours than women in the control group (Bradley et al., 2005a). At 12 months, however, many women who had stopped working had returned to work (Bradley, 2004). The nonemployment effect of breast cancer diagnosis and treatment at 6 months was twice as large for African-American women. Similar findings were evident among men with prostate cancer. Here, 28