Background: the association between pain characteristics and frailty risk is uncertain. Objective: to investigate the separate impact of the frequency, intensity and location of pain on frailty risk and its possible mechanisms. Methods: prospective cohort of 1505 individuals ≥63 years followed between 2012 and 2015 in Spain. In 2012, pain was classified into: lowest pain (Score 0), middle pain (Score 1-4) and highest pain (Score 5-6). Incident frailty was assessed in 2015 as having ≥3 Fried criteria or a Frailty Index (FI) ≥0.30. Results: in multivariate analyses, the risk of frailty (measured with the Fried criteria or the FI) increased progressively with the frequency of pain, its intensity and the number of pain locations. Compared with those having the lowest pain score, the odds ratio (95% confidence interval) of Fried-based frailty was 1.24 (0.56-2.75) in the middle score and 2.39 (1.34-4.27; P-trend <0.01) in the highest score. Corresponding values for frailty as FI ≥0.30 were 1.39 (0.80-2.42) and 2.77 (1.81-4.24; P-trend <0.01). Odds ratios did not change after adjustment for alcohol intake, Mediterranean diet adherence or sedentary time, but were reduced with adjustment for pain-associated chronic diseases (cardiovascular disease, diabetes, chronic lung disease, osteomuscular disease and depression). A higher pain score was linked to higher risk of exhaustion and low physical activity (two out of five Fried criteria) and to a worse score in all FI domains. Conclusion: frequency, intensity and location of pain were associated with higher risk of frailty. Study associations were partly explained by pain-associated morbidity.