Discussion
In this prospective study of 272494 participants from UK Biobank, we found that compared with those who had only one child, childless women had the higher all-cause premature mortality and the lower years of life expectancy, whereas those with two, three, or four children had the lower all-cause premature mortality and the higher years of life expectancy. And these associations were independent of biological factors, lifestyle, socioeconomic level, and pre-existing diseases. Similarly, this study found that the relationship between parity and all-cause premature mortality was also observed in sensitivity analyses, demonstrating the robustness of the results. Moreover, comparied with women who had one child, the estimated values of FI, ΔKDM-biological age and HD adjusted for multiple covariates in women with two or three children were lower, and the correspondings in females with five or more chilldren were higher.
Our study found that the childless women had the higher all-cause mortality, and women who had two, three, or four children had the lower all-cause mortality, compared to women with one child, which are similar with the results reported in a Sweden population[13]. Differently, the relationship between five or more live births and all-cause premature mortality was significant in the Sweden population. Although the correspondings relationship in our study was not significant, there is also a clear upward trend in the RCS curve, which may be attributed to the more covariates we have corrected. The finding in subgroup analyses suggested that association between the number of live births and hazard of all-cause premature mortality appeared to be different among different races, which may be attributed to the genetic reasons of race being a potential factor affecting this relationship. In addition, obesity, current-smoking, and excessive alcohol consumption mask the relationship between parity and all-cause premature mortality, possibly due to the fact that obesity[33,34], smoking[35], and alcohol consumption[36] themselves can cause a series of physiological changes, damage health, and even lead to premature death. For example, the obese persons are more likely to experience metabolic changes during pregnancy than their normal weight counterparts[37,38]. And in obese individuals, lipids, oxidized LDL particles, and free fatty acids activate the inflammatory process and trigger the atherosclerosis and CVD[39-40]; and obesity leads to changes in the structure and function of the heart[41], leading to heart failure[42], atrial fibrillation[43,44] and sudden cardiac death[45]. Similarly,the chronic diseases also had an impact on the relationship between parity and all-cause premature mortality.
In addition, our study provides novel evidence to show the relationship between parity and mortality. Our results on the cause-specific premature mortality indicate that the hazard of all-cause premature mortality associated with more number of live births could be partly attributed to CVD-specific and cancer-specific premature mortality. Such observations are consistent with previous evidence linking parity with various conditions including cardiovascular disease[46] and cancer[47]. For the subtypes of CVD-specific premature mortality, we found that higher number of live births was significantly associated with lower hazard of CHD-specific and stroke-specific premature mortality, compared with women with one child. These observations were supported by the results from the parity and stroke study, and the parity and coronary heart diseases study, in which parity may confer a moderate long-term protective effect on the risk of subarachnoid hemorrhage (SAH)[48,49] and CHD[50]. Future investigations are warranted to explore the association of the number of live births with various cardiovascular disease subtypes. In addition, evidence from experimental and epidemiological studies has shown that parity was related to breast cancer[51-53], ovarian cancer[53,54], cervical cancer[55]. We found that nulliparous women had the highest hazard of all-cause premature mortality, which be partly attributed to the continuous stimulation of ovarian hormones to the body of nulliparous women may increases the cancer-specific premature mortality, especially breast cancer and ovarian cancer[5,6], and further increases the all-cause premature mortality of nulliparous women.
For the first time, we reported that childless women was associated with the lower life expectancy at the age of 40 years by 1.41 (95% CI: 0.78–2.01) years for women, and women with two, or three or four children had an average 1.10 (95% CI: 0.63–1.74), 1.00 (95% CI: 0.54–1.74), 1.00 (95% CI: 0.05–1.84) higher years of life expectancy, respectively, compared with participants who had one child. Meanwhile, we found that accelerated biological aging and functional decline were potential influencing factors, as the estimated values of FI, ΔKDM-biological age and HD adjusted for multiple covariates have significant changes. Our research is similar with the changes in HD caused by parity in previous articles, giving rise to the U-shape for the overall relationship between parity and biological aging[56]. Moreover, we also found that women who gave two or three births were the least likely to age, while women with five or more children are the most likely to age, which can be attributed the fact that reproduction can consume a large part of resources, reducing the efforts invested in the maintenance and repairment of somatic tissues, and then leading to aging and death in women[57-60]. In addition, studies have shown that the social and psychological pressure caused by women’s by high parity may be another potential factor accelerating biological aging and functional decline[61].
The present findings may have several public health implications. First, the evidence is complementary to those on the association between the number of live births and all-cause mortality and fills the gap in the relationship between parity and all-cause premature mortality, life expectancy, and aging. Second, the evidence may inform the recommendations on behavioral changes regarding reproduce. The parity is easily assessed in clinical and public settings, and may be useful for future reproduce interventions.
Although our study have some advantages including the large sample size, the extensive covariate correction, and the consistent results in several sensitivity and subgroup analyses, several potential limitations should be carefully considered in this study. Firstly, we could not exclude the possibility that high number of live births is a marker for a lower socioeconomic level. However, subgroup analyses indicated that the positive association between the number of live births and hazard of mortality was consistent across the subgroups of socioeconomic level. Secondly, in this study we obtained maternal live births by questioning “How many children have you given birth to? (Please include live births only) ” Because this question did not include the number of stillbirths, spontaneous miscarriages or termination that women may have had previously, we could not obtain specific date associated with these situations. As described earlier, if the number of births affects long-term health conditions, the number of stillbirths, spontaneous miscarriages or termination will also be an important source of information. We need a further study including these basic facts in order to investigate the effects of number of stillbirths, spontaneous miscarriages or termination on long-term health. Thirdly, although we controlled for many socioeconomic factors and lifestyle factors, we lacked explicit information on marital status, on age at first birth , and on attitudes and values that may be associated with both fertility and health. These unmet covariates may affect the relationship between parity and all-cause premature death. Fourthly, an important limitation of this study is that the UK Biobank is not representative of the general population due to the voluntary participation[24]. Further studies are needed to conirm our findings, especially in populations that are more representative of the UK population.
In conclusion, our study indicates that the relationships between the number of live births and hazard of all-cause premature mortality, life expectancy, and aging. obesity, current-smoking, and excessive alcohol consumption may affect the association between the parity and all-cause premature mortality. Further clinical trials are warranted to validate these findings.