The economic consequences of hospital admission for individuals with health insurance

Conference: Topics in Applied Microeconomics

Client: The Becker Friedman Institute for Research in Economics at the University of Chicago

After a hospital stay, how well do patients recover from the financial consequences of their care? Matthew J. Notowidigdo’s work seeks to address this essential health care impact question. He was especially interested in determining whether the consequences for those who have medical insurance differed from those patients who did not. Notowidigdo, who is from Northwestern University, coauthored the paper with Carlos Dobkin from UC Santa Cruz, and Amy Finkelstein and Raymond Kluender from Massachusetts Institute of Technology.

Notowidigdo and his colleagues analyzed hospital discharge data for one million, non-pregnant adult patients in California and matched it to consumer credit data. They found that among insured and uninsured patients, the economic impact of hospitalization was marked and surprisingly similar. The study focused solely on patients who had not been hospitalized for at least three years prior to their admittance, so that it would be less likely that chronic conditions were at play in the results.

Among the insured within in four years following the hospital stay, the researchers showed a 25% increase in unpaid bills, a 33 percent increase in consumer bankruptcy filings, a six percent decrease in borrowing limits, a nine percent decrease in credit card balances, and a seven percent decrease in auto installment loans. For the uninsured, the impact was even greater on unpaid bills and bankruptcies, but they endured similar consequences on borrowing limits and access to credit. The researchers also estimated that both the insured and uninsured experienced a one to two percent annual drop in household consumption for at least four years following the hospital stay. However, the elderly fared on the whole much better. They experienced far fewer negative financial impacts across all parameters studied, except for a small uptick in unpaid bills sent to collectors.

Perhaps most significantly, the researchers documented that hospital stays were pivotal in five percent of all bankruptcies among the uninsured and in three percent of bankruptcies among the insured following hospitalization. Again, among the elderly, there was little effect.

Notowidigdo theorizes that insured patients may have been affected so significantly because of income lost during and after their hospital stay rather than due to the medical expenses they stacked up. The study estimates that only about 20 percent of the decline in credit and borrowing access experienced by the insured was due to medical costs not covered by insurance. The study also shows that approximately half of the insured’s decline in credit access can be attributed to unemployment. In other words, Notowidigdo says, the health insurance is solely insurance against medical expenses. Notowidigdo suggested a need to learn more about the direct impact on income after hospital stays and the potential role of income insurance as a next step for deeper inquiry.

—Jennifer Roche

More work samples