Development and assessment of a personal evaluation system for risk factors of osteoporosis

在日本和歌山县的一个山村进行了一项为期10年的队列研究，测量了骨密度(BMD)，以提供有关成年和老年人口骨质疏松症累积发病率的信息，并开发一种评估系统，利用基线问卷中的潜在危险因素预测骨质疏松症的发病率。受试者包括从1913-1952年出生的居民中按性别和年龄随机选择的400名受试者，并在基线、3、7和10年使用双能X射线骨密度仪(DXA)测量腰椎和股骨近端的骨密度。在40岁至70岁的男性和女性中，10年来骨质疏松症的累积发病率在腰椎分别为2.0%和15.0%，在股骨颈分别为10.5%和18.7%。对既往病史、家族史、药物、锻炼、饮食习惯、饮酒、吸烟等生活方式因素对累积发病率的影响进行Logistic回归分析，但未发现显著相关性。预测10年后腰椎骨质疏松发病率的回归公式为：Y=EXP(1.41X1-0.02X2-0.65X3-18.88X4+17.94)(X1：性别，男0，女1；X2：年龄(岁)，X3：体重(公斤)，X4：骨密度(g/cm2))R2=43.0%。预测10年以上股骨颈骨质疏松发生率的回归公式为：Y=exp(-1.94X1+0.02X2-0.04X3-19.22X4+13.63)(X1：性别，男0，女1；X2：年龄(岁)；X3：体重(公斤)；X4：骨密度(g/cm~2))R2=30.1%。我们的研究表明，对生活方式因素的评估可能不足以预测骨质疏松的发生率。

Bone mineral density (BMD) was measured over a 10-year period in a cohort study in a mountain village in Wakayama Prefecture, Japan, to provide information on cumulative incidence of osteoporosis in the mature and elderly population and to develop an assessment system to predict incidence of osteoporosis using potential risk factors in a baseline questionnaire. Subjects comprised 400 subjects selected randomly by sex and age from a full list of residents born in 1913-1952, and BMD of the lumbar spine and proximal femur was measured using dual energy X-ray absorptiometry (DXA) at baseline and 3, 7 and 10 years. Cumulative incidence of osteoporosis over 10 years in men and women aged in their 40s to 70s was 2.0% and 15.0% at the lumbar spine, respectively, and 10.5% and 18.7% at the femoral neck, respectively. We analyzed the influence of lifestyle factors such as past history, family history, medicines, exercise, dietary habits, alcohol and smoking on cumulative incidence by logistic regression analysis after adjustment for age and gender, but did not find any significant associations. Regression formula for predicting the incidence of osteoporosis at the lumbar spine over 10 years~ was as follows : Y=exp(1.41X1-0.02X2-0.65X3-18.88X4+17.94) (X1: sex, male 0, female 1; X2: age (years), X3: weight (kg), X4: BMD (g/cm2)) R2=43.0%. Conversely, regression formula for predicting the incidence of osteoporosis at the femoral neck over 10 years as follows : Y=exp(-1.94X1+0.02X2-0.04X3-19.22X4+13.63)(X1: sex, male 0, female 1; X2: age (years); X3:weight (kg); X4: BMD (g/cm2)) R2=30.1%. Our study demonstrates that the assessment of lifestyle factors might not be sufficient to predict incidence of osteoporosis.