Age at menarche. The reproducibility of self-reported menarcheal age and the association between age at menarche and total- and cardiovascular mortality - The Tromsø Study

Abstract

Background: The possible relationships between age at menarche and total as well as cardiovascular mortality has been investigated, but has not been established. Aim: The first aim was to investigate the reproducibility of self-reported menarcheal age from 1986-1987 (Tromsø 3) to 1994-1995 (Tromsø 4). The second aim was to examine the association between age at menarche and cardiovascular- or all-cause mortality. Methods: This is a prospective cohort based on data from the Tromsø study. Follow-up for reproducibility purpose was 7 years. Pearson’s correlation analysis and Bland Altman plot was used in the reproducibility study. In the investigation of all-cause and cardiovascular disease mortality follow-up was from date of attending Tromsø 4 until 30.06.2015 (mean follow up on 18.7 years) and until 31.12.2012 (mean follow-up on 16.7 years), respectively. Cox survival analysis was used to investigate the association between menarcheal age and mortality. Results: Reported menarcheal age in Tromsø 4 was significantly related to that reported in Tromsø 3 (r = 0.84, p <0.001). Mean difference in reported menarcheal age was negligible, 0.01 years. A total of 2203 women died during the follow up in the investigation of all-cause mortality. During the follow up of cardiovascular mortality, 654 women died from cardiovascular disease (184 from stroke, 250 from ischemic heart disease and 220 from other cardiovascular related causes). There was no association between age at menarche and total mortality after adjusting for confounding factors. For total cardiovascular mortality there was an indication of a weak positive linear relationship after adjustments. One year increase in age at menarche was associated with 7 % increased cardiovascular mortality (HR: 1.07, 96% CI: 1.01 – 1.14, p =0.03). Conclusion: Self-reported age at menarche in Tromsø 4 was strongly correlated with that reported in Tromsø 3, both when combined and in stratified age-groups. A positive linear trend was observed between menarcheal age and cardiovascular disease mortality after adjusting for confounding factors.