Age and Low Heel Bone Mineral Density predicted Hip and Pelvic fracture-risk in 245 women aged 72-98: A 20-Year Population based Primary Care Study in Rural Sweden

Nils Larsson, Hans Thulesius, Ulrica Mölstad, Anna Lindgren, Anna Segernäs, Robert Eggertsen, Håkan Johansson, Pär Wanby, Ferdinando Petrazzuoli, Märit Wallander, Daniel Albertsson

Keywords: Hip fracture, Pelvic Fracture, 20-year follow-up, Bone Mineral Density (BMD), Clinical Risk Factors, Fragility Fracture, Primary Care, Fracture Risk, Hazard ratios (HR), Dual X-ray and Laser (DXL)

Background:

Background: Hip and pelvic fractures have high mortality rates and incidence at advanced age. We wanted to evaluate possible long-term clinical risk factors in order to improve fracture prediction and thereby prevention.

Research questions:

Research question: Could clinical risk factors and heel bone mineral density (BMD) assessment in primary care predict long-term risk for hip and pelvic fractures?

Method:

Methods: From a population-based fracture prevention intervention study in rural Sweden we included 285 women aged 72-98 years who provided data in 2003, from a questionnaire with 15 clinical risk factors and heel BMD measurements using Dual X-ray and Laser (DXL). With fracture and mortality data from population registries and medical records, we calculated risk of first hip or pelvic fracture up to 20 years for their remaining lifetime measured with hazard ratios (HR) using Cox regression analyses.

Results:

Results: Data from 245 deceased women was analyzed. Median age at death was 90 years. Age and heel BMD were independent risk factors for suffering a first hip or pelvic fracture (n=60); HR for continuous age was 1.09 (1.04-1.15, p<0.001) indicating that fracture risk increased with 4-15% for each life year; HR for age ≥80 years was 2.81 (1.66-4.78, p<0.001) indicating that fracture risk increased with 66-378% for women aged ≥80 years as compared to women <80 years. HR for heel BMD was 1.58 (1.23-2.03, p<0.001) per standard deviation decreased t-score, i.e. fracture risk increased by 23-103% for each standard deviation decrease in t-score.

Conclusions:

Conclusion: In a population-based study of women aged 72-98 years from rural Sweden, both continuous age, age ≥80 years, and heel BMD were independent of each other and associated with hip and pelvic fractures during a 20-year follow-up. No other clinical risk factors did significantly predict fractures. Thus, especially age, but also BMD appeared suitable as long-term fracture risk prediction.

Points for discussion:

Could age over 80 years be used in consultations for a reminder of high risk of fragility fractures?

Is long term BMD screening valuable in primary care?

#143