Statistical Analysis

We used a comparison-population, interrupted time series design (an ecologic study design) to test the hypothesis that the log odds of LBW among monthly cohorts born to parous people after the CTC payment (ie, July-December 2021) would differ from their expected values. We derived expected values from preterm births among parous people during CTC payments, LBW among births to nulliparous people (the comparison group) during CTC payments, and autocorrelation in LBW among births to parous people prior to CTC payments. We further adjusted for shocks to (ie, outliers in) LBW that occurred prior to the CTC payments, such as those that could have arisen from the COVID-19 pandemic. We implemented this design using the following steps (further detail is provided in the eMethods in Supplement 1):

The data management was performed using SAS, version 9.4 statistical software (SAS Institute Inc), and the time series analyses were performed using SCA Statistical System, version 5.2 (Scientific Computing Associates). A 2-sided P < .05 was considered significant.