Over the past decade, testosterone (T) treatment has increased markedly in older hypogonadal men, yet the long-term benefits and risks of T treatment are unknown because prior studies have been inadequately powered. A significant public health challenge is to assess the long-term risks of T treatment in older men with low T levels in the absence of a long-term randomized controlled trial. We propose to conduct a pharmaco-epidemiologic study to ascertain long-term risks of T treatment in a national VA database of ~480,000 older, hypogonadal men that includes ~160,000 men who initiated T treatment between January 1, 2001 and July 1, 2012, and have a maximal follow-up time of 10.5 years. Our primary outcomes will be combined cardiovascular (CV) events (recurrent and incident myocardial infarction, ischemic stroke, and venous thrombosis) and incident aggressive prostate cancer (Gleason grade >8 or PSA >20 ng/dl or T3 stage or greater or nodal or metastatic disease). Secondary outcomes will be specific CV events, overall (aggressive and non-aggressive) incident prostate cancer, and total mortality. We will examine outcomes by different T formulations (intramuscular, patch, and gel) and whether risks are modified by age, ethnicity, and prevalent diabetes. A unique strength of the VA database is its linkage to Medicare data and a VA Prostate Cancer registry that will enhance ascertainment of outcomes. Our research team has relevant experience in: 1) the use of a VA database to assess effects of T treatment in hypogonadal men;2) outcome assessment of CV events, prostate cancer, and mortality;and 3) pharmaco-epidemiology of hormone treatment. We will employ several analytic techniques to strengthen the results of our study. Specifically, we will use a vetted propensity score analysis to adjust for the non-randomization to T-treatment and will adjust for potential bias related to increased prostate cancer screening in the T-treated group, by adjusting for screening intensity with PSA levels, frequency of PSA levels, and prostate biopsy. Strengths of this study include 1) a large cohort of older, ethnically diverse, hypogonadal men with high medical morbidity and event rates and a long follow-up period, 2) primary aims focused on clinically relevant outcomes and 3) innovative analytic approach and methods. Our team has successfully collaborated in the past and has the clinical, pharmaco-epidemiologic, and biostatistical expertise to complete a high-quality study on the long-term risks of T- treatment. Given the high prevalence of hypogonadism and increasing number of T-treated men, this study will provide crucial data to clarify risks of T treatment and provide impetus for a large, randomized, controlled, clinical trial.

Public Health Relevance

Testosterone treatment is increasing markedly in older men, even though the long-term risks of testosterone treatment are unknown. We will use a clinical database of 500,000 men with low testosterone levels to examine if testosterone treatment is associated with serious health problems such as aggressive prostate cancer, heart attacks, strokes, blood clots, and death. These results will be highly relevant for public health, due to th large and growing numbers of older men treated with testosterone, who may be at risk for serious health problems from long-term testosterone treatment.

National Institute of Health (NIH)
National Institute on Aging (NIA)
Research Project (R01)
Project #
Application #
Study Section
Special Emphasis Panel (ZAG1-ZIJ-8 (M1))
Program Officer
Joseph, Lyndon
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Seattle Institute for Biomedical/Clinical Research
United States
Zip Code
Walsh, Thomas J; Shores, Molly M; Krakauer, Chloe A et al. (2018) Testosterone treatment and the risk of aggressive prostate cancer in men with low testosterone levels. PLoS One 13:e0199194
Shores, Molly M (2018) Testosterone treatment and cardiovascular events in prescription database studies. Asian J Androl 20:138-144
Shores, Molly M; Matsumoto, Alvin M (2014) Testosterone, aging and survival: biomarker or deficiency. Curr Opin Endocrinol Diabetes Obes 21:209-16
Shores, Molly M (2014) The implications of low testosterone on mortality in men. Curr Sex Health Rep 6:235-243