Background Early identification of modifiable risk factors is essential for the prevention of frailty. This study aimed to explore the causal relationships between a spectrum of genetically predicted risk factors and frailty.

Methods Univariable and multivariable Mendelian randomisation (MR) analyses were performed to explore the relationships between 22 potential risk factors and frailty, using summary genome-wide association statistics. Frailty was accessed by the frailty index.

Results Genetic liability to coronary artery disease (CAD), type 2 diabetes mellitus (T2DM), ischaemic stroke, atrial fibrillation and regular smoking history, as well as genetically predicted 1-SD increase in body mass index, systolic blood pressure, diastolic blood pressure, low-density lipoprotein cholesterol, triglycerides, alcohol intake frequency and sleeplessness were significantly associated with increased risk of frailty (all p<0.001). In addition, there was a significant inverse association between genetically predicted college or university degree with risk of frailty (beta −0.474; 95% CI (−0.561 to –0.388); p<0.001), and a suggestive inverse association between high-density lipoprotein cholesterol level with risk of frailty (beta −0.032; 95% CI (−0.055 to –0.010); p=0.004). However, no significant causal associations were observed between coffee consumption, tea consumption, serum level of total testosterone, oestradiol, 25-hydroxyvitamin D, C reactive protein or moderate to vigorous physical activity level with frailty (all p>0.05). Results of the reverse directional MR suggested bidirectional causal associations between T2DM and CAD with frailty.

Conclusions This study provided genetic evidence for the causal associations between several modifiable risk factors with lifetime frailty risk. A multidimensional approach targeting these factors may hold a promising prospect for prevention frailty.