July 21, 2016

Employment and Income Losses among Cancer Survivors

By Chichun Fang

Cancer diagnosis. Stamp, stethoscope, syringe, blood test and pi

A cancer diagnosis can have significant impacts on individual well-being beyond its negative consequences on health longevity. A worker who is diagnosed with cancer may have to either reduce hours of work or quit working altogether. Hence, one important concern is the economic cost of cancer diagnoses, measured by the amount of lost earnings among cancer survivors. In a paper recently published in Cancer, Anna Zajacova, Jenn Dowd, SRC Professor Bob Schoeni, and Health and Retirement Study Co-Investigator Bob Wallace studied how cancer diagnosis affects employment outcomes.

One of the major challenges in studying the causal effect of a cancer diagnosis on employment outcomes is that cancer morbidity is not randomly assigned. If we only have a one-time “snapshot” of the population and observe that, say, people who have been diagnosed with cancer tend to have lower earnings, we cannot really tell whether lower earnings are “caused” by cancer diagnosis or by other factors that are correlated with both cancer diagnosis and lower earnings. For example, those with low income on average are more likely to smoke cigarettes, and tobacco use is associated with lung cancer.

To overcome this challenge, Professors Zajacova, Dowd, Schoeni, and Wallace relied on a research design in which individuals were observed multiple times, for up to five years both before and after the cancer diagnosis. By comparing the employment outcomes of the same individual over time (that is, before and after a cancer diagnosis), they were able to differentiate the effect of a cancer diagnosis from other factors that influence employment and income. They used data from the Panel Study of Income Dynamics (PSID), which is a longitudinal household survey conducted by the SRC since 1968 and is the longest-run survey of its kind in the world.  PSID respondents are surveyed every other year, and the researchers would know when the diagnosis occurred and what the employment outcomes were in the years preceding and following the diagnosis.

The researchers examined the effect of a cancer diagnosis on four outcomes: employment, hours worked, individual income, and total family income. They found significant effects across all four dimensions. The probability of being employed dropped by 10 percentage points, hours of work declined by 200 hours in the first year after diagnosis, and annual labor earnings dropped 40% within two years and stayed low for at least five years. Total family income also reduced by 20% initially but eventually recovered within four years. Such effects were largely driven by male cancer survivors, and one possible explanation for the sex difference is lower labor-force participation among women overall.

With advancements in medical technology, cancer is increasingly becoming a chronic disease (rather than a terminal illness). This study shows that the negative impact of cancer on the economic well-being of cancer survivors and their family members is not only large but also long-lasting, suggesting that workplace and employment policies should be considered to buffer such health challenges.