3885 Factors shared by monozygotic twins explain unexpected associations between frailty and menopausal hormone replacement therapy

Abstract Introduction Menopausal hormone replacement therapy (HRT) is first-line treatment for distressing vasomotor symptoms, and increasingly popular. However, data on the association of HRT with ageing-related conditions including frailty is lacking. Method We analysed women in the Danish population registry (n = 471,206), Danish Twin Registry and TwinsUK cohort (n = 1547). In Denmark, we assessed frailty age 65, 70 and 75, using a modified Hospital Frailty Risk Score. This linked to national prescribing data, to ascertain HRT use by age 55, adjusting for birth year, education and income. In TwinsUK, we assessed frailty after age 65 using a frailty index (TwinFI). HRT use, and covariates (BMI, smoking, alcohol use, education and deprivation) were ascertained from retrospective questionnaires. Analysis used conditional/fixed-effects generalised estimating equations, examining within twin-pair frailty associations. Results HRT associated with higher frailty scores among both Danish women (beta = 0.21, P < 0.001) and TwinsUK (beta = 0.035, p = 0.01) equivalent to ageing 5 years. However, in TwinsUK, HRT only associated with higher frailty for the within-pair analyses in dizygotic twins (Dizygotic: beta = 0.061, p = 0.01, Monozygotic: beta = 0.013, p = 0.36). This was robust to multiple sensitivity analyses, including HRT subtypes, duration and imputation. Conclusions We found an unexpected relationship between HRT use and higher frailty across both Denmark and the UK. However, twin analysis suggests this relationship is not causal, but confounded by factors shared within monozygotic twins, including genetics. Further investigation is needed to determine whether individuals prescribed HRT are already on the path to frailty, and whether this would be an apt opportunity to intervene.