Dose-response association between adult height and all-cause mortality: a systematic review and meta-analysis of cohort studies

Background: We conducted a systematic review and meta-analysis from published cohort studies to examine the association of adult height and all-cause mortality and to further explore the dose-response association. Methods: PubMed, The Cochrane Library, The Ovid, CNKI, CQVIP and Wanfang databases were searched for articles published from database inception to 6 February 2018. We used the DerSimonian-Laird random-effects model to estimate the quantitative association between adult height and all-cause mortality and the restricted cubic splines to model the dose-response association. Results: We included 15 articles, with 1 533 438 death events and 2 854 543 study participants. For each 5-cm height increase below the average, the risk of all-cause mortality was reduced by 7% [relative risk (RR) = 0.93, 95% confidence interval (CI), 0.89-0.97] for men and 5% (RR = 0.95, 95% CI, 0.90-0.99) for women. All-cause mortality had a U-shaped association with adult height, the lowest risk occurring at 174cm for men and 158cm for women (both P-nonlinearity< 0.001). Relative to the shortest adult height (147 cm for men and 137cm for women), men at 174cm had a 47% lower likelihood of allcause mortality and women at 158 cm a 33% lower risk of all-cause mortality. Conclusions: Our study suggests that the relation between adult height and all-cause mortality is approximately U-shaped in both men and women.