Using a matched insurant–general practitioner panel data set, we estimate the effect of a general health-screening program on individuals' health status and health-care cost. To account for selection into treatment, we use regional variation in the intensity of exposure to supply-determined screening recommendations as an instrumental variable. We find that screening participation increases inpatient and outpatient health-care costs up to 2 years after treatment substantially. In the medium run, we find cost savings in the outpatient sector, whereas in the long run, no statistically significant effects of screening on either health-care cost component can be discerned. In sum, screening participation increases health-care cost. Given that we do not find any statistically significant effect of screening participation on insurants' health status (at any point in time), we do not recommend a general health-screening program. However, given that we find some evidence for cost-saving potential for the sub-sample of younger insurants, we suggest more targeted screening programs. Copyright © 2014 John Wiley & Sons, Ltd.