| |
Testosterone Therapy is Associated With Depression, Suicidality,
and Intentional Self-Harm: Analysis of a National Federated
Database
|
| |
| |
Download the PDF here
Our analysis not only failed to demonstrate a reduction in the development of MDD and suicidality after the initiation of testosterone in testosterone deficient men, but even demonstrated a modest increase in these psychiatric comorbidities.
CONCLUSION
Testosterone use is independently associated with mental health comorbidities. Typically, when patients are prescribed testosterone, they are counseled regarding the risks of polycythemia, gynecomastia, infertility, and physiologic dependence, but our study reveals that men who use testosterone should also be screened for and counseled about risks of depression and suicidality. As this study could only establish an association rather than true causation, future studies are necessary to investigate potential dose-response relationships, whether withdrawal from testosterone plays a role, and whether patient selection or preventative measures can mitigate the risk.
Sirpi Nackeeran, BA,1 Mehul S Patel, MD,1 Devi T Nallakumar, BS,1 Jesse Ory, MD,2 Taylor Kohn, MD,3
Christopher M Deibert, MD, MPH,4 Chase Carto, BSE,1 and Ranjith Ramasamy, MD1
ABSTRACT
Background: Long-term use of testosterone can be associated with mood destabilizing effects. Most studies investigating psychiatric complications of anabolic steroids have used small samples, but a comprehensive assessment of the risk of developing mental health disorders after testosterone use has not been performed at the population level.
Aim: To determine whether testosterone therapy is associated with major depressive disorder or suicide attempts in men.
Methods: We conducted a retrospective cohort study of 70.3 million electronic health records collected from 46 healthcare organizations encompassing flagship hospitals, satellite hospitals, and outpatient clinics since 2008 to determine whether testosterone use is associated with major depressive disorder and suicide attempts in a large population. We included men 18 or older who either used testosterone or did not, defined by reported use, insurance claim, or prescription use of testosterone documented in the electronic health record. We propensity-score matched by age, race, ethnicity, obesity, and alcohol-related disorder. Additionally, a sub-group analysis was performed in testosterone deficient (<300 ng/dL) men comparing those with TD on testosterone therapy to a control group of men with TD who are not using testosterone.
Outcomes: We determined measures of association with a new diagnosis of major depressive disorder and suicide attempt or intentional self-harm following testosterone use within 5 years.
Results: A total of 263,579 men who used testosterone and 17,838,316 men who did not were included in the analysis. Testosterone use was independently associated with both Major Depressive Disorder (OR 1.99, 95% CI 1.94-2.04, P < .0001) and Suicide Attempt/Intentional Self-Harm (OR 1.52, 95% CI 1.40-1.65, P < .0001). Results remained significant in testosterone deficient sub-group analysis.
Clinical Implications: Men who use testosterone should be screened for and counseled about risks of depression and suicidality.
Strengths and Limitations: Strengths of this study include a large sample size, the ability to account for chronology of diagnoses, the use of propensity score matching to control for potentially confounding variables, and the consistency of results with sub-group analyses. Limitations include the potential for incorrect coding within the electronic health record, a lack of granular information regarding testosterone therapy adherence, the possibility that unrecorded testosterone or anabolic steroid use were prevalent but not captured within the control group, and a lack of data regarding testosterone withdrawal.
Conclusion: Testosterone use is independently associated with new-onset mental health disorders. Future studies are necessary to elucidate the role that androgen withdrawal plays and whether a causal relationship exists.
|
|
| |
| |
|
|
|
