Send to

Choose Destination
Osteoporos Int. 2013 Apr;24(4):1225-33. doi: 10.1007/s00198-012-2063-1. Epub 2012 Jul 10.

Hip fracture and other predictors of anti-osteoporosis drug use in Norway.

Author information

Department of Chronic Diseases, Norwegian Institute of Public Health, Oslo, Norway.


This study aims to find predictors of anti-osteoporosis drug (AOD) use. Known risk factors of osteoporosis, i.e., age, hip fracture, and corticosteroid use were found to be predictors of AOD use, in addition to a number of other drugs used. Higher socioeconomic position did not favor the use of AOD.


This study deals with studying predictors of anti-osteoporosis drug treatment in Norwegian women and men.


All Norwegian women and men≥50 years were included (n=1,407,392). Data were taken from different data sources, (1) the Norwegian Prescription Database (drug use in 2004-2005); (2) the Nationwide Census 2001 (marital status, education and resident county); (3) the National Hip Fracture Database (hip fractures 2003-2005); and (4) the National Population Register (date of death/emigration). We estimated the hazard ratios (HR) for incident treatment by Cox proportional hazard regression.


In 2005, 10,332 women (1.5%) and 1,387 men (0.2%) were new users of anti-osteoporosis drugs (incident treatment). Age was a statistically significant predictor of incident treatment in both women and men, with HR ranging from 1.7 to 3.2 (per 10 years). A middle educational level in men strongly predicted incident treatment [HR 2.0 (CI 1.1-3.8)], but not in women after full adjustment. A previous hip fracture, increasing number of drugs used and use of corticosteroids were all predictors of incident treatment in both genders after adjustments. Corticosteroid use [HRwomen=4.0 (CI 3.8-4.2)] had a higher HR for incident treatment than hip fracture [HRwomen=2.0 (CI 1.8-2.3)]. Marital status and area of residency were not predictors of incident treatment in either gender, after adjustments. The predictors of prevalent treatment were only slightly different from incident treatment in 2005.


Age, previous hip fracture, number of drugs used, and use of corticosteroids were positively related to treatment in both genders. In men, a middle educational level predicted treatment.

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Springer
Loading ...
Support Center