Primary Hypothesis: Risk factors such as smoking, alcohol, and caffeine are associated with osteoporosis in men.

Intervention: None

Primary Outcomes: Percent reduction in cross-sectional bone mineral density over time.

Study Abstract: The goals of this project are to establish a new cohort of male veterans and describe associations between potential risk factors and baseline bone mineral density (BMD) as well as rates of BMD loss.

The study is a two-year observational prospective study of 1000 male veterans aged 50 and older. The subjects are drawn from two existing VA cohorts, the Normative Aging Study (NAS) and the Veterans Health Study (VHS). Examinations to be performed on each subject (baseline and two follow-up exams) include BMD measurement at the hip, forearm and total body, anthropometry, hand grip strength, and questionnaires on current diet, physical activity, tobacco, alcohol use, sunlight exposure, health related quality of life (Medical Outcomes Study Short Form), depression (Brief Depression Scale), fracture history, medications and psychological factors (PrimeMD, MiniMental). Additional data on historical exposure to risk factors, previously obtained in the NAS or VHS studies, will also be correlated with BMD and BMD loss. Multivariate linear and logistic models with control for potential confounders will be used in the main analyses. Examinations take place at the Boston VA Outpatient Clinic.

Examinations on each subject include BMD measurement at the hip, forearm and total body, height and weight, hand grip strength, leg extensor strength, and mobility. Questionnaires on current diet, physical activity, tobacco, alcohol use, sunlight exposure, health related quality of life (Medical Outcomes Study Short Form), depression (Brief Depression Scale), fracture history, medications and psychological factors (PrimeMD, MiniMental) are administered. Each subject also contributes a nonfasting blood sample and a 24-hour urine sample.

Beginning in April, 1999, two additional measures are being obtained as a result of a 4-year supplemental VA Epidemiology grant awarded to Dr. Miller.

These are the heel ultrasound measure and a 4-day physical activity level determined by ambulatory monitor.

Preliminary cross-sectional analysis of mean BMD for femoral neck, total body and forearm BMD suggest the percent reduction in cross-sectional BMD between ages 60-69 and 80-89 is approximately 9% per decade at the femoral neck, 5%/decade at the total body, and 8%/decade at the forearm. These age-related patterns agree well with other studies of BMD in older men.

The x-ray densitometer was installed at the Boston VA Outpatient Clinic in April, 1998 and the study technician was certified for bone density measurements during the three-day training session that followed. A precision study was conducted in the following month. In vivo reproducibility was determined by performing 5 repeat scans of each of the three sites on 5 volunteers ranging in age from 21 to 50 years. Subjects were repositioned between each scan. The coefficient of variation (CV) of femoral neck BMD in our laboratory is 0.92%; for whole body BMD, 0.72%; and for distal radius BMD, 1.04%. An aluminum spine phantom with known density (1.254 gm/cm2) is scanned every two weeks on the DPX-IQ. The mean ? standard deviation of 15 measurements is 1.250 ? 0.0058, coefficient of variation = 0.46%. There has been no detectable drift in phantom values during the study start-up phase.

When completed, this study will be one of the largest in the US to study risk factors for osteoporosis in men, historically an understudied group in relation to osteoporosis. Several of the hypothesized risk factors (smoking, alcohol, caffeine, for example) are amenable to change and represent potential areas of intervention. Thus, not only can the VA make a valuable contribution to furthering our understanding of osteoporosis in men in general, but strategies may be identified to reduce morbidity as well as reduce health care costs associated with fracture in the elderly.