Objectives: After the firework disaster in Enschede, The Netherlands, on 13 May 2000, a longitudinal health study was carried out. Study questions were: (1) did the health status change over this period; and (2) how is the health status 18months after the disaster compared with controls? Study design: A longitudinal comparative study with two surveys at 3 weeks and 18 months after the disaster. Methods: A control group for the affected residents was included in the second survey. Respondents filled in a set of validated questionnaires measuring their physical and mental health problems. Results: The prevalence of physical and emotional role limitations, severe sleeping problems, feelings of depression and anxiety, as well as intrusion and avoidance decreased from 3 weeks to 18 months after the disaster for the affected residents. Independent of background characteristics and other life events, residents had 1.5 to three times more health problems than the control group; for example, physical role limitations (odds ratio [OR] ¼ 1.5, 95% confidence interval [CI] 1.2–2.0) and anxiety (OR ¼ 3.1, 95% CI 2.4–4.2).Conclusions: Although health problems decreased compared with 3 weeks after the disaster, 18 months after the disaster, the affected residents had more health problems than the people from the control group.