Background: The gender-specific role of marital status for the development of frailty has not been clarified. This study evaluates the gender differences in the association between marital status and frailty development, and the possible modifying effect by age cohort in such a relationship. Methods: The sample included 2179 community-dwelling older adults involved in the Swedish National Study on Aging and Care in Kungsholmen, followed up for 6 years. Participants stable in marital status over time were categorized as partnered, widowed, single, and divorced. Changes were classified as losing one's partner and gaining a partner. Frailty was defined as the presence of three or more criteria among: weight loss, low physical activity, slow walking speed, weakness, and exhaustion. The association between marital status and frailty, with death as an alternative outcome and controlling for confounders, was estimated with multinomial logistic regressions. Results: Men who remained single (odds ratio [OR] = 2.50, 95% confidence interval [95% CI] 1.05 - 5.98) and those who lost their partner (OR = 2.59, 95% CI 1.16 - 5.77) had higher odds of frailty than those with a partner. The OR differed between younger (60-80 years) and older (>= 81 years) women (p-(interaction) = 0.04). The youngest women who remained divorced had a higher risk of frailty (OR = 2.75, 95% CI 1.24 - 6.08) than those who still had a partner. Conversely, older women who lost their partner had 80% (95% CI 0.05-0.86) lower odds of frailty than those with a partner. Conclusions: Marital status can influence frailty development differently for women and men. This gender-specific influence may vary by age cohort, perhaps in response to sociocultural factors.