The purpose of this study is to examine correlates of self-perceived fracture risk (SPR) and relationships between SPR and subsequent bone density and microarchitecture in the UK arm of the Global Longitudinal Study of Osteoporosis in Women. 3912 women completed baseline questionnaires detailing medical history and SPR;​ 492 underwent HRpQCT scans of the radius and tibia and DXA scans of total body, hip, femoral neck and lumbar spine a median of 7.5 years later. Correlates of SPR were examined and a cluster analysis of potential predictors of SPR performed. SPR in relation to HRpQCT and aBMD parameters was examined using linear regression with and without adjustment for anthropometric, demographic and lifestyle covariates. Mean (SD) baseline age was 69.0 (9.0) years;​ 56.6% reported a similar SPR;​ 28.6% lower SPR;​ 14.9% higher SPR compared to women of similar age. In mutually-adjusted analysis, higher SPR was associated (p < 0.05) with:​ lower physical activity and educational attainment;​ use of anti–osteoporosis medications (AOM) and calcium supplements;​ greater number of falls in the previous year;​ history of fracture since aged 45;​ family history of hip fracture;​ and increased comorbidity. Higher SPR, history of fracture, and use of AOM, calcium and vitamin D clustered together. Even after adjustments that included AOM use, higher SPR was associated with:​ lower radial trabecular volumetric density and number, and higher trabecular separation;​ lower tibial cortical area and trabecular volumetric density;​ and lower aBMD at the femoral neck. Despite greater AOM use, women with higher baseline SPR had poorer subsequent bone health.