Although several studies have documented that latitude might be an effect modifier of the association between temperature and mortality, little is known about how much latitude modifies the temperature-mortality relationship. In this study, we examined this research question using a distributed lag non-linear model and meta-regression analysis based on data from 13 large cities of eastern US from the US National Morbidity, Mortality, and Air Pollution Study. We found that cold effects lasted about 1 month while hot effects were acute and short-term. Meta-regression analysis showed that latitude modified both the cold and hot effects with statistical significance. The cold effect decreased with the latitude increment, with -0.11 % change of mortality effect for 1° increment, while the hot effect increased with the latitude increment, with 0.18 % change of mortality effect for 1° increment. This finding indicates the importance of latitude on temperature-related mortality risk, which is helpful for city to develop localized effective adaptation strategy in the context of climate change.