This study examined the spatiotemporal variations in extreme values of relative humidity (RH), temperature, temperature-humidity index (THI) and the effect of climatic variables on THI in the northern parts of Iran from 1986 to 2016. The THI was generated for each of the 30 meteorological stations. Previously, trend analysis was calculated in different seasons using quantile regression. For the most part, the results indicate a declining and rising trend in RH and temperature, respectively. The highest percentage of decreasing trend slopes was for maximum RH (≤-50 %), but for maximum temperature, the highest percentage of increasing slopes in summer and winter (≥30 %) was identified on the southern coast of the Caspian Sea. However, a decreasing trend (%≤-15) was detected for the maximum temperature in autumn. The highest increasing (decreasing) trend slope of THI (30 % and −30 %) was for low values of THI in spring and winter (autumn and winter). The minimum and maximum temperatures positively affected THI in all seasons, and the value of the effect coefficient was higher in the coastal areas. Also, high values of minimum RH in the spring, summer, autumn, and winter seasons had a positive (negative) effect in 73, 100, 33, and 0 % (26, 0, 66, and 100 %) of the stations, respectively. But the maximum RH had a positive (negative) effect on 80, 100, 33, and 6 % (20, 0, 66, and 93 %) of stations, respectively. The findings of this study are generally helpful in identifying changes in weather extremes and their effects on human health.