We may still only assess fracture risk with moderate accuracy. This is in spite of several new visualisation techniques, biochemical markers and risk-factor based calculation tools. There are thus probably additional factors that affect fracture risk, to be found. Bad luck is probably not the only explanation. FRAX is the fracture prediction tool that is most widespread worldwide. FRAX includes BMD of the femoral neck along with eleven other risk factors for fracture. The aims of this study was to see if self-estimated health measured with a visual analog scale (VAS) was associated with hip-fracture risk. In 1999, 351 Swedish women aged 69-79 years at inclusion, were included in a population-based prospective cohort study. They estimated their global health by putting an x on a 100 millimeter line drawn between worst imaginable health and best imaginable health. At the same visit they were also assessed with FRAX including bone mineral density (BMD). Ten years later, Swedish medical records were used for follow-up regarding fractures and mortality. The main outcome was a hip fracture. No participant was lost to follow up. During a median follow up time of 9.8 years, 40 participants (11%) had a hip fracture. The age adjusted hazard ratio (HR) of a hip fracture was 3.70 for the lowest quartile of self-estimated health compared to the highest quartile of self-estimated health. This relation was unaffected by adjustment for BMD as well as for any other single risk factor of the risk factors included in FRAX. Not even adjustment for all risk-factors in FRAX together, altered this relation between self-estimated health and hip fracture risk. However, adjustment for maximum one-leg standing time, made the relation become non-significant! Self-estimated health thus seems to say something about hip-fracture risk not covered by any of the clinically most commonly considered risk-factors.
Disclosure: The authors declared no competing interests. This study was funded by the Stockholm County Council in Sweden.