Abstract: Objective To explore the application of segmented linear regression of interrupted time series design in the evaluation of policy intervention and SAS implementation, and to provide reference for public health policy researchers in methodology application. Methods Taking the influence of the two-child policy on the cesarean section （CS） rates as an example, SAS software was used for the segmented linear regression model in the CS rates at Ⅳ-class maternity hospitals in Guangzhou between January 2013 to December 2016, including principle of the model, SAS program, controlling for autocorrelation and adjustment of covariates. Results The segmented linear regression analysis showed that there was no significant change in the level and trend of CS rate at Ⅳ-class maternity hospitals in Guangzhou both during the period of the selective two-child policy and the universal two-child policy after adjusting maternal age and the history of CS. Conclusion The segmented linear regression model is a powerful tool for evaluating the effects of policy by estimating the changes in the level and trend of observed indicators before and after the intervention, but attention should be paid to the applicable conditions and the interpretation of the results for its limitations.

Key words: Interrupted time series, Segmented linear regression, Effects of policy, SAS software