Parity and osteoporotic fracture risk in postmenopausal women: a dose-response meta-analysis of prospective studies

The present dose-response meta-analysis shows linearly decreased hip fracture (HF) risk and nonlinearly decreased osteoporotic fracture (OF) risk associated with increasing number of parity of up to five live births among postmenopausal women.

Epidemiological reports suggest that parity is associated with reduced OF risk among women. However, these findings are controversial. Here, we present a meta-analysis of prospective studies of parity in relation to OF risk.

We performed systematic searches using Medline and Embase from January 1, 1966, to December 31, 2014, with limits of language in English and prospective study design. Relative risks (RRs) and confidence intervals (CIs) were derived mainly using random-effects models. Categorical, dose-response, heterogeneity, publication bias, and subgroup analyses were conducted.

We analyzed 10 articles of 19 independent reports from 1966 to 2014, comprising a total of 217,295 participants and 26,525 cases of OF. Compared to nulliparous women, the OF and HF risks of parous women with at least one live birth were reduced by 11 % (95 % confidence interval (CI) 3-19 %; I (2) = 77.1 %, p < 0.001) and 26 % (95 % CI 17-35 %; I (2) = 19.5 %, p = 0.287), respectively. Representative nonlinearly and linearly inverse dose-response associations were found between parity (range of 0-6) and OF risk (p nonlinearity = 0.0163; I (2) = 79.7 %, p < 0.001), and between parity (range of 0-5) and HF risk (p nonlinearity = 0.054; I (2) = 76.5 %, p < 0.001), respectively. The lowest risk reduction for OF of 25 % (95 % CI 16-33 %) was observed for five live births. And, the summary risk reduction for HF was 12 % (95 % CI 9-15 %) for each one increased live birth.

We found that increasing number of parity is associated with linearly reduced HF risks among women. The association between parity of six or more live births and HF risks should be studied further in future.