Previous studies suggest social support is associated with musculoskeletal health in later life. We explored this relationship further in community-dwelling older adults, by considering associations between different aspects of social support and musculoskeletal health in community-dwelling adults. Participants from the Hertfordshire Cohort Study reported level of confiding/emotional, practical, and negative support using the Close Persons Questionnaire. Muscle strength was measured by grip strength dynamometry, and physical capability by timed up-and-go, chair rises, and walking speed tests. Bone mineral density (BMD) was assessed using dual-energy x-ray absorptiometry. Linear regression, adjusted for age, sex, body mass index (BMI), alcohol, smoking, physical activity, social class, and diet, was used for analysis. 1842 men and women (mean age 65.7 years) participated. Low emotional support correlated with weaker grip strength and poorer physical capability tests, although estimates were not robust to adjustment for confounders. Low practical support was linked to shorter timed up-and-go (β - 0.171, 95%CI - 0.319, - 0.024) and walking speed times (β - 0.157, 95%CI - 0.306, - 0.007), following adjustment for confounders. Negative support (i.e. the perceived inadequacy of the support received) was associated with lower grip strength (β - 0.145, 95%CI - 0.223, - 0.067) and slower walking speeds (β 0.159, 95%CI 0.004, 0.314). No social support exposures were associated with BMD. Different types of social support are linked to various measures of musculoskeletal health in older adults. Limited requirement for practical support correlated with better physical capability, while negative support correlated with poorer outcomes. No social support measure was associated with BMD.