Statistics analysis system software

We used Statistics Analysis System software (version 9.3, SAS Institute, Cary, NC, USA) for statistical analysis. All study variables were categorical and the differences between the two study hospitals were compared using the χ² test. Using all women in the two study hospitals as the standard population, we calculated the weighted average annual CS rate standardized by age, education level, parity, CS history, and Robson classification over the 10 years based on the weights from combined population, to compare the trends between the two study hospitals. The CS rates were tested for changes over time using χ² tests for trend. Crude odds ratio (cOR) and 95% CI were calculated for the association between year and the proportion of nulliparous women, with truncation analysis at year 2014. To understand the change in 10 years in composition of women in type of delivery by parity and CS history, we plotted the data annually. Similarly, we plotted the annual composition of CS cases by parity and CS history to understand the 10 years change, which indicated the change of each group's contribution to the overall CS rate. A p < 0.05 was considered to be statistically significant.

Hardy–Weinberg equilibrium of the genotype distribution for each SNP among the control group was examined by a goodness-of-fit χ2test. Chi-square (χ2) test was used to evaluate the distribution differences of selected demographic characteristics as well as each allele and genotypes of rs3217927 between the cases and controls. Unconditional univariate and multivariate logistic regression analyses were performed to obtain crude and adjusted odds ratios (ORs) for estimating risk of childhood ALL and their 95% confidence intervals (CIs) with adjustment for diagnosis age, gender, parental alcohol use, tobacco-smoking and housing painting status. Independent-sample t-test was used for analyzing the results of CCND2 mRNA expression. Two-sided P values were selected and P<0.05 was considered statistically significant. All the statistical analyses were performed using Statistics Analysis System software (version 9.1; SAS Institute, Cary, NC).

Hardy-Weinberg equilibrium of the genotype distribution among the controls was evaluated by a goodness-of-fit χ² test to identify possible selection genotyping errors and bias. Nonparametric Mann-Whitney test was used for analyzing the distribution differences of age between the cases and controls. Chi-square (χ²) test was used to estimate the distribution differences of gender as well as each genotype of rs1800469 and rs2317130 between the two groups. Unconditional multivariate logistic regression analyses were performed to obtain crude and adjusted odds ratios (ORs) for risk of acquired AA and their 95% confidence intervals (CIs) with adjustment for diagnosis age and gender. Two-sided P values were selected, and P < 0.05 was set as the threshold for statistical significance. Haploview version 4.0 was used to calculate the D’ value and r2 value among the two SNPs. All statistical analyses were performed using Statistics Analysis System software (version 9.2; SAS Institute, Cary, NC).