ORIGINAL ARTICLE Clinical risk indices, prediction of osteoporosis, and prevention of fractures: diagnostic consequences and costs Chatlert Pongchaiyakul Æ Nguyen D. Nguyen John A. Eisman Æ Tuan V. Nguyen Received: 10 July 2004 / Accepted: 2 October 2004 / Published online: 17 September 2005 Ó International Osteoporosis Foundation and National Osteoporosis Foundation 2005 Abstract The primary aim of this study was to validate the Osteoporosis Self-Assessment Tools for Asians (OSTA) instrument in Thai women, and to evaluate its utility in terms of economic costs and fracture predic- tion. Femoral neck and lumbar spine BMD was mea- sured by dual energy X-ray absorptiometry in 322 Thai women, aged 60±10 years (mean±SD; range: 45– 84 years). The women were classified as having osteo- porosis if their BMD T -scores were £ )2.5. The OSTA score was calculated for each woman using her age and weight according to the formula: 0.2(weight)age). Wo- men with OSTA scores £ )1 and >)1 were classified as ‘‘high risk’’ and ‘‘low risk,’’ respectively. The prevalence of osteoporosis was 33% by femoral neck or lumbar spine BMD. Using the OSTA score, 165 (51.2%) women were classified as high risk. The sensitivity, specificity and positive predictive value of OSTA was 82% 64% and 53%, respectively. If the OSTA score is used to identify women with high risk of fracture, and assuming that the incidence of fracture among osteoporotic and non-osteoporotic women are 2% and 1% per year, respectively, the OSTA score can identify 59% of frac- ture cases correctly, and 41% are expected to be missed. Furthermore, if the high-risk subjects identified by OSTA are to be treated, and if the treatment reduces fracture incidence by 50%, and assuming that the treatment cost is $1 per day, then the cost to prevent one fracture is estimated to be $48,530. Results of this study suggest that, in the Thai population, the OSTA score had high sensitivity but low specificity and low positive predictive value in the identification of osteoporotic women. Its use in the general population can result in a high false-positive rate and incur significant cost to the community. Keywords Asia Æ Clinical risk index Æ Fracture Æ Osteoporosis Introduction Osteoporosis and its ultimate consequence of low-trau- ma fracture represent one of the major public health problems, not just in Western countries [1] but also in Asian countries [2]. It is projected that by the end of this century, 50% of all hip fractures in the world will occur in Asia [3]. Prevention of osteoporotic fractures in Asia is therefore of paramount importance. Bone mineral density (BMD) measured by dual-en- ergy X-ray absorptiometry (DXA) is a primary predictor of osteoporotic fractures and a surrogate measure of osteoporosis [1]. In some developing countries, DXA is not widely available, and the cost of BMD measurement is high. Bone mineral density is highly related to age and body weight. Indeed, the two factors collectively account for 40–60% variance of BMD in the population [4–8]. It is reasonable, therefore, that clinical risk factors (such as age and weight) could be used for identifying subjects with low BMD and, hence, high-risk fracture individuals in developing countries. The Osteoporosis Self-Assessment Tools for Asians (OSTA) instrument, largely derived from age and body weight, has been found to be a good and simple tool for the identification of women with osteoporosis risk [9]. However, the sample in which OSTA was devel- oped mainly came from the Chinese population, among whom lifestyles and behavioral factors are likely different from other developing populations, such as Thai. The ultimate aim of identifying high-risk Osteoporos Int (2005) 16: 1444–1450 DOI 10.1007/s00198-005-1996-z C. Pongchaiyakul Æ N.D. Nguyen Æ J.A. Eisman T.V. Nguyen (&) Bone and Mineral Research Program, Garvan Institute of Medical Research, 384 Victoria Street, Darlinghurst Sydney, NSW 2010, Australia E-mail: t.nguyen@garvan.org.au Tel.: +61-2-92958277 Fax: +61-2-92958241 C. Pongchaiyakul Division of Endocrinology, Department of Medicine, Faculty of Medicine, Khon Kaen University, Khon Kaen, Thailand