Exercise-induced bronchospasm (EIB) is widely used in epidemiological studies to investigate the prevalence of asthma. We aimed to determine seasonal variations in the prevalence of EIB in Ghanaian school children from urban-rich (UR), urban-poor (UP), and rural (R) schools. We have previously reported the prevalence of EIB in 9-16-year-old children to be 3.1% in the dry season, with UR children having a significantly higher prevalence of both EIB and atopy compared to UP or R children. In the current study, the prevalence of EIB was assessed in the same 1,095 children in the wet season (5 months following the initial study) using the same methodology. Exercise provocation consisted of free running outdoors for 6 min. In the wet season, 17/1,095 children [1.55%, 95% confidence interval (CI): 0.91-2.47] had a positive response to exercise, compared to our previous report of 34 children (3.1%, 95% CI: 2.15-4.32) with EIB in the dry season (dry vs. wet season, difference 1.55, 95% CI: 0.41-2.69). The proportion of children with a positive response to exercise in the UR school fell from 4.2% (25/599) to 1.3% (8/599) (difference 2.9, 95% CI: 1.2-4.5). In the wet season, there was no difference in the prevalence of EIB among the UR, UP, and R children. Only five of 1,095 subjects (0.5%, 95% CI: 0.15-1.07) demonstrated EIB in both seasons. In conclusion, although exercise challenge remains a useful tool for determining asthma prevalence in epidemiological studies, seasonal variations in the pattern of responses may occur and the results should be interpreted with caution. © 2002 Blacwell Munksgaard.