Prospective association of low serum total testosterone levels with health care utilization and costs in a population-based cohort of men.

Despite the burgeoning interest in the field of andrology, no studies have specifically addressed the impact of serum testosterone levels on healthcare utilization and costs. We analysed data from the population-based cohort Study of Health in Pomerania (SHIP), Germany, to assess the association of serum testosterone levels with self-reported health care utilization and costs at baseline and at 5 years follow up. Study sample comprised 2023 men at baseline, of whom 1530 men were repeatedly examined. Low and high serum testosterone levels, defined according to the age-specific 10th and 90th percentile, were compared with reference subjects with serum testosterone levels ≥10th -≤90th percentile. Two-part econometric models were applied adjusting for socio-economic and medical confounders. Cross-sectional models revealed higher numbers of outpatient visits and higher costs for both, men with low (+19.1 and +19.9%, respectively) and high serum testosterone levels (+25.3 and +30.2%, respectively), whereas number of inpatient days and costs were not associated with serum testosterone levels. Adjustment for age, educational level, income, waist circumference, smoking status, physical activity and alcohol consumption did not considerably alter the results. Longitudinal models revealed a significant association of low serum testosterone levels with increased number of follow-up outpatient visits (age-adjusted: +28.6%) and costs (+38.0%) only. Low and high serum testosterone levels were associated with increased short-term outpatient health care costs, whereas low serum testosterone levels appear to be predictive of long-term outpatient health care costs. Cost-effectiveness studies of available treatments are necessary to identify benefits for physicians, patients and health care system as a whole.