Objective: Frailty and chronic pain are closely related, but the causal relationship between them needs to be further explored with high-level evidence. Mendelian randomization (MR) was used to analyse the bidirectional causal relationship between nine chronic pain disorders and two frailty indicators in this study. Methods: We used Causal Analysis Using Summary Effect Estimates (CAUSE) as the primary method of analysis. The inverse-variance-weighted, simple model, weighted model, penalized weighted median and MR‒Egger regression methods were used to evaluate the robustness of the results. Results: The frailty index (FI) was significantly associated with the number of chronic pain sites (multisite chronic pain, MCP, padjust < .001) and the risk of chronic widespread pain (CWP, padjust < .001). The Fried frailty score (FFS) was significantly associated with MCP (padjust < .001), the risk of CWP (padjust < .001) and chronic back pain (padjust = .031). In the reverse analysis, both MCP and CWP were significantly positively associated with the FI (padjust < .001; padjust = .003) and FFS (padjust < .001; padjust = .009). Conclusions: This study revealed a bidirectional causal relationship between frailty and chronic pain, with the number of pain sites playing a key role. This finding has significant implications for effectively managing frailty and chronic pain in older adults.