The purpose of this study is to assess the concentration-response relations between conjunctivitis and exposure to ambient ozone. This retrospective study includes emergency department (ED) visits for conjunctivitis in Edmonton, Canada, for the period April 1992 – March 2002. Daily average levels of ozone, of temperature and relative humidity were estimated and used for the period of the study. For each of the considered exposure lags, from 0 to 9 days, six various models were fitted to estimate the concentration-response function. The goodness of fit was assessed using Akaike information criterion. During the period of the study, 17,211 ED visits for conjunctivitis were recorded and used. For all subjects together, a positive statistically significant association was obtained for the exposure lagged by 5 days. For female subjects, lags 1, 3, and 9 had positive statistically significant associations (lag 2 had negative associations). For male subjects only lag 5 had a positive statistically significant association. The estimated non-linear concentration-response functions for the considered groups (all, males, females) and lags, revealed the associations along the exposure levels. The fitted shapes are described by the parameters and may have various forms. The estimated function are useful to determine the risk associated with exposure to ground-level ozone.