Abstract: Objective To explore the effects of extreme low temperatures on respiratory hospital admissions in Guangdong Province and effect modification of different regions and age groups. Methods Based on the time-series data of respiratory hospital admissions and meteorological variables in Panyu District (developed region) and Xingning County (underdeveloped region) of Guangdong Province during 2006-2011, the relationship between extreme low temperatures (below the 5th percentile of daily average temperature distribution) and respiratory hospital admissions was analyzed with distributed lag non-linear model, and stratified analysis by sex and age group was also conducted.Results The mean of daily average temperature was 22.7 ℃ in Panyu, and the total respiratory hospital admissions were 24 216 in Panyu Central Hospital with a daily average of 11.1 during 2006-2011. The mean of daily average temperature was 21.9 ℃ in Xingning, and the total respiratory hospital admissions were 19 934 in Xingning People’s Hospital with a daily average of 9.1 in the same period. The effect of daily average temperature on respiratory hospital admissions was non-linear, and low temperatures had significant lagged effects for the 2 regions, with the RRs reaching the maximum of 1.04 (95% CI:1.02-1.06) at 6.4 ℃ on lagged 13 days for Xingning and 1.03 (95% CI:1.02-1.04) at 9.9 ℃ on lagged 10 days for Panyu, respectively. For the cumulative lagged effects during the 25 lagged days, the RR s were 1.48 (95% CI:1.13-1.94), 1.45 (95% CI:1.12-1.82) and 1.29 (95% CI:1.10-1.51) respectively at the 5th percentile (10.9 ℃), 10th percentile (14.3 ℃) and 25th percentile (18.2 ℃) of daily average temperature distribution for Panyu, and were 1.61 (95% CI:1.17-2.21), 1.45（95% CI:1.08-1.96）and 1.11 (95% CI:0.94-1.35) for Xingning. The change of effects (RRs) associated with extreme low temperatures along the lagged days showed a similar trend of converse “U” shape for various sub-groups, with the effects emerging on 2-4 days after exposure, growing up later, and decreasing after reaching the peak. The effect of extreme low temperatures had the greatest impact on ≥65 age group in Xingning with an RR of 1.86（95% CI:1.04-3.31）. Conclusion Extreme low temperatures can significantly impact on respiratory hospital admissions for the elderly, especially in underdeveloped regions.