Duke 2008 | ASHE | Equity and Efficiency in Health and Healthcare

Cancer represents a substantial hardship for patients and their families in multiple domains beyond health and survival. Earlier cancer diagnosis and better treatment in recent decades have combined to expand the number of cancer survivors. A small body of research has examined the long-term labor market outcomes of childhood cancer survivors, as well as immediate impacts of adult cancer on disability and absence from work. Much less is known about the short and longer term economic impacts of a cancer diagnosis on individuals and families. Using nationally representative longitudinal data from the Panel Study of Income Dynamics (PSID), 1968-2005, we will estimate the income and wealth changes of families in which a household member has reported having had cancer. We ask the following questions: 1.) How do individual and spousal incomes change during several years following a cancer diagnosis? It is possible that spouses of cancer patients may compensate for the decline in the patient's income by working or earning more, or they may reduce their labor market participation to care for their spouse. 2.) How does household wealth change in the years following a cancer diagnoses? Life-threatening illness has been reported as the most common reasons for bankruptcy. There may be substantial heterogeneity in the economic impact of cancer on families depending on the type of cancer, type of treatment, age of the patient, as well as the economic and social resources of the individuals at baseline. We will test whether the impact of cancer on family income and wealth differs by these factors.

Questions about whether you have ever had cancer have been asked in each PSID wave since 1999. These questions also ask how long ago the cancer occurred, allowing us to reconstruct the approximate year of diagnosis for survey respondents in 1999 and onward. We will use multiple analytic approaches. Drawing on the job displacement literature, we will estimate the past, current, and future income of individual i as a function of a series of dummy variables that indicate whether a cancer "event" is reported for each period for an individual (or their spouse), including individual level fixed effects, period effects, and other fixed and time-varying covariates. We will also use growth mixture models to identify distinctive patterns in the income trajectories for cancer families, and examine the association of the patterns with the family demographic and socioeconomic correlates. Since wealth is captured less frequently in the PSID (every 5 years from 1984 until 1999, every two years thereafter), we will examine the change in family wealth before and after the cancer "event."