This paper examines whether health shocks influence labour supply. Estimates rely on detailed health measures that not only are more homogeneous than commonly-used self-reported health measures, but they also can be interpreted as plausibly exogenous with respect to labour supply. Further, this paper investigates the unique role that certain fringe benefits, namely health insurance and paid sick leave, play in the link between health shocks and subsequent employment activity. Results show that, compared to commonly-used self-reported measures of health, health problems defined by the US government as ‘priority’ conditions correlate with smaller labour supply shifts, but non-work-related injuries lead to larger shifts. Second, the arrival of a health shock appears to reduce the probability of remaining employed full time, while also increasing the likelihood of quitting work. Relatively few full-time workers who acquire health problems switch to part-time employment. Third, in the event of a health shock, sick leave appears to facilitate reductions in employment activities, while employer-provided insurance appears to hinder such adjustments.