Research Highlights Featured Chart
February 4, 2026
A Danish study shows that an adverse health shock increases the chances of criminal behavior.
A cancer diagnosis can be devastating, often placing heavy medical, emotional, and financial burdens on patients and their families.
Beyond their personal costs, serious health shocks have broader consequences, as shown by authors Steffen Andersen, Elin Colmsjö, Gianpaolo Parise, and Kim Peijnenburg in a paper in the American Economic Journal: Applied Economics. The authors find that cancer diagnoses lead to measurable increases in criminal behavior, representing a previously overlooked negative externality on society.
The study draws on comprehensive Danish administrative data linking health with criminal records for 368,317 individuals diagnosed with cancer between 1980 and 2018. Figure 1 from the authors’ paper shows their main result through an event-study that tracks criminal activity in the years surrounding a cancer diagnosis.
The horizontal axis represents the time relative to a diagnosis, from six years before to ten years after the event. The vertical axis measures changes in crime propensity in percentage points, with the year immediately preceding diagnosis serving as the baseline. The vertical bars represent 95 percent confidence intervals.
The chart shows that in the years leading up to a diagnosis, the estimates hovered near zero and remained statistically insignificant. In the year of a diagnosis, criminal activity declined slightly, likely reflecting the physical demands of treatment and the availability of savings accumulated before the health shock.
However, roughly two years after a diagnosis, crime propensity rose above the pre-cancer baseline and became statistically significant. The effect continued growing through year five before stabilizing at elevated levels that persisted through year ten.
Overall, the average treatment effect of a cancer diagnosis amounted to a 0.10 percentage point higher likelihood of committing a crime, representing a 14 percent increase relative to the baseline annual crime rate of 0.69 percent.
The authors also demonstrate that welfare policies can moderate these effects. Exploiting variation created by a 2007 Danish municipal reform, they show that reductions in social support amplified the crime-inducing effects of cancer diagnoses. This finding suggests that adequately funded safety net programs may generate benefits beyond their direct recipients by reducing the negative spillovers that health shocks impose on broader society.