Background Peak bone mass is determined by the genetic potential as modified by nutrition, exercise, hormonal status and disease. Osteoporosis may be prevented by identifying and modifying risk factors in younger age groups. This paper reports on the distribution of bone density (BMD) and prevalence of risk factors in 225, of a target 300.

Objectives The aims of this study were to identify the frequency of osteoporosis and identify the key associates of low bone density in young women in 2 regions of England.

Methods Women (excluding pregnant) aged 20–29 were selected at random from 9 GP practices (6 in Cornwall [C]and 3 in Surrey [S]) in a 2 stage random sampling process, and invited by letter to participate. Demographic, lifestyle and anthropometric data were collected by questionnaire, 7-day food diary and examination. BMD was measured at the lumbar spine (LS) and femoral neck (FN) using dual-energy X-ray absorptiometry (DXA).

Results The main demographic and anthropometric data are given in Table 1. No differences were found between the 2 groups. However, a lower BMD was found in C at the FN (t = 2.13, p = 0.03). Although the meaning of t-score is uncertain in women of this age, osteopenia (t-score <-1.0 and >-2.5) was found in 21% women at LS and 11% at FN. Osteoporosis (t-score <-2.5) was found in 2(1%) cases at the LS only, both in C. Various risk factors were identified in the total group. Amenorrhoea >6 months duration in 29(13%), BMI <19 in 13(6%), history of smoking in 134(59%). 74(33%) women had a history of fracture however none fulfilled the standard criteria for low trauma fracture. 18(8%) women reported a history of possible or definite eating disorder. In analysing the association between risk factors and BMD weight had the predominant effect at FN (r2 = 0.181 B = 0.425 p < 0.001) and LS (r2 = 0.112 B = 0.334 p < 0.001). Risk factors significantly associated with BMD were history of smoking (t = -2.31 p = 0.02) and amenorrhoea (t = -2.09 p = 0.04). Associations between self-reported milk intake and BMD were analysed using the GL Model. Those with a current consumption of 1 pint/day or more had BMD.094 (SD.043 p = 0.03) greater than those who drank milk less than once/week. Both adolescent and current low consumption of milk was associated with a lower BMD for weight than expected.

Conclusion The prevalence of osteoporosis was low in young UK women but 25% were osteopenic. BMD measurement may be of limited use at this age for predicting future osteoporotic fracture in the normal population. The overall prevalence of identified risk factors was low but smoking and body weight are both modifiable thus health promotion messages should aim to target these risk factors.