Numerous epidemiological studies have reported that ozone (O(3)) and temperature are independently associated with health outcomes, but modification of the effects of O(3) on health outcomes by temperature, and vice versa, has not been fully described. This study aimed to investigate effect modification by temperature on the association between O(3) and emergency ambulance dispatches (EADs) in Japan. Data on daily air pollutants, ambient temperature, and EADs were obtained from eight Japanese cities from 2007 to 2015. A distributed lag non-linear model combined with Poisson regression was performed with temperature as a confounding factor and effect modifier to estimate the effects of O(3) on EADs at low (<25th percentile), moderate (25th-75th percentile), and high (>75th percentile) temperature for each city. The estimates obtained from each city were pooled by random-effects meta-analysis. When temperature was entered as a confounder, the estimated effects of O(3) on EADs for all acute, cardiovascular, and respiratory illnesses were largest at lag 0 (current-day lag). Therefore, this lag was used to further estimate the effects of O(3) on EADs in each temperature category. The estimated effects of O(3) on EADs for all acute, cardiovascular, and respiratory illnesses in all eight Japanese cities increased with increasing temperature. Specifically, a 10 ppb increase in O(3) was associated with 0.80 % (95 % CI: 0.25 to 1.35), 0.19 % (95 % CI: -0.85 to 1.25), and 1.14 % (95 % CI: -0.01 to 2.31) increases in the risk of EADs for all acute, cardiovascular, and respiratory illnesses, respectively, when city-specific daily temperature exceeded the 75th percentile. Our findings suggest that the association between O(3) and EADs for all acute, cardiovascular, and respiratory illnesses is the highest during high temperature. Finding of this study can be used to develop potential mitigation measures against O(3) exposure in high temperature environment to reduce its associated adverse health effects.