The 130th Annual Meeting of APHA |
Kathleen Bainbridge, MPH and MaryFran Sowers, PhD. Department of Epidemiology, University of Michigan, 339 East Liberty, Ann Arbor, MI 48104-2205, (734) 615-9570, kbbridge@umich.edu
As part of Michigan’s osteoporosis prevention efforts, a six-item self-administered risk instrument was developed for pre- and perimenopausal women in order to classify women between the ages of 20 and 50 into risk groups for low and high bone mineral density (BMD) according to easily assessed risk factors. Repeated BMD measurements of the lumbar spine and the femoral neck were collected for 614 women over six years. Information on body size, diet, physical activity, alcohol use, smoking behavior, reproductive factors, menopausal status, history of chronic illness, medication use, and family history of osteoporosis were also assessed. Linear mixed models were developed to identify risk factors for BMD and risk factors for BMD change. Using data for half of the sample, generalized estimating equations were developed, and six risk factors (body weight, alcohol use, family history of osteoporosis, irregular menses, bilateral oophorectomy, and reproductive cancer) were identified for either high (femoral neck z-score > 2 and < 3.5) or low (femoral neck z-score < -2) BMD. The risk instrument assessed these six risk factors with items weighted according to their strength of association, and a risk score was calculated using the other half of the sample. ROC curves were plotted to identify cutpoints for classifying women into risk groups. The instrument classifies women as low BMD with moderate specificity (64-66%). This instrument may be a useful tool to identify pre- and perimenopausal women who are not likely to be of low BMD.
Learning Objectives:
Keywords: Primary Prevention, Risk Factors
Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.