Association between the number of live births and hazard of all-cause premature mortality stratied by potential risk factors
We also conducted stratified analyses according to the potential risk factors including age (born year), race, BMI, Townsend deprivation index, smoking status, moderate drinking, diabetes, hypertension, asthma, emphysema and chronic bronchitis in Supplementary Table 2. The association of parity with hazard of all-cause premature mortality appeared was not significant in Other ethnic group participants (P-trend = 0.447), where the protective effect of mothers with double children on all-cause premature death had disappeared (HR2=1.02, 95% CI: 0.31-3.31). We also found the harmful effects of the nulliparous women on all-cause premature death had become not significant in obese participants (BMI≥ 30 kg/m2) (HR0=1.12, 95% CI: 0.99-1.27). Notably, the observed significant association of the number of live births with hazard of all-cause premature mortality appeared to be abolished in current-smokers (P-trend = 0.003) and overdrinker (P-trend = 0.114). Meanwhile, the chronic diseases such as diabetes, asthma, emphysema and chronic bronchitis also had an impact on the relationship between parity and all-cause premature mortality. We did not find significant interactions between other potential confounders and the number of live births on hazard of all-cause premature mortality.