Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

When to measure bone density


Women with low bone density have an increased propensity for fractures when older. Knowing that bone density is low is helpful in decisions about treatment. But in which women do we measure it? There are a number of clinical decision rules, but which is best? Testing them in a population with known bone density tells us that some are better than others [1].


An age and sex stratified sample of Canadian women who were aged 25 years or older were selected randomly in 1996 and 1997. For the study of bone density menopausal women aged 45 years or older were eligible. Women taking HRT for less than five years were excluded, but those taking HRT for more than five years were included, and there were sensible reasons for this.

Bone mineral density was measured at the femoral neck using dual-energy X-ray absorptiometry. The results were divided between those with normal bone mineral density (T score less than -1 SD units), osteoporosis (T score <-2.5 SD units), and various intermediate values of moderate osteopenia (T score -2 to -2.5) and mild osteopenia (-1.0 to -2.0).


There were 3,288 women, and after exclusions for those taking bone-sparing medicines, with diagnosed osteoporosis, or with missing data not allowing calculation of scores, 2,365 remained. They had a mean age of 66 and a mean weight of 69 kg; 97% were white.

Clinical decision systems tested are shown in Table 1. They came from a systematic review of the literature. SCORE and ORAI had the best discriminatory performance at each cut point. The performance of the scores at detecting women with osteoporosis and normal bone mineral density in 45-65 year old women is shown in Figure 1.

Table 1: Clinical decision systems tested

Factor Score
National Osteoporosis Foundation (NOF), test if score ≥1
Age ≥65 years 1
Weight <57.6 kg 1
History of fracture 1
Family history of fracture 1
Current cigarette smoker 1
Simple Calculated Osteoporosis Risk Estimation (SCORE), test if score ≥6
Race not black 5
Rheumatoid arthritis present 4
History of fracture at wrist, hip of rib 4 each
Age ≥65 years 3 times first digit of age
Oestrogen therapy never used 1
Weight (-1 times weight in lb/10)
Osteoporosis Risk Assessment Instrument (ORAI), test if score ≥9
Age 55-64 5
Age 65-74 10
Age 75+ 15
Weight 60-70 kg 3
Weight <60 kg 9
Not currently taking oestrogen 2
Age, Body Size, No Estrogen (ABONE), test if score ≥2
Age >65 1
Weight <63.5 kg 1
Never used OC or oestrogen therapy for at least 6 months 1
Body weight criterion, test if weight <70kg

Figure 1: Results for women with osteoporosis and normal bone density


It is rare to find an examples where a single clinical decision aid has been tested in an independent sample. To find a study that systematically compares all the available decision aids is remarkable. There are several ways in which this information might be used. Many women have bone density measured whatever decision system was used, and some will argue for mass screening. This study is the essential aid to decision making.


  1. SM Cadarette et al. Evaluation of decision rules for referring women for bone densitometry by dual-energy X-ray absorptiometry. JAMA 2001 286: 57-63.
previous or next story in this issue