TRT and Mental Health: Anxiety, Depression & the First 90 Days

You started TRT because you were tired, flat, and struggling to motivate yourself. A few months later, a lot of men report feeling genuinely better. But the path there is not a straight line. There is a week-three anxiety spike that catches people off guard. There is a period where you feel worse before you feel better. There is the question of whether your antidepressant is still doing anything now that your hormones are optimised. And there is the persistent confusion between clinical depression and the neurological state that untreated low testosterone creates.

This article maps the first 90 days of TRT for mood, explains the physiology behind each phase, gives you the seven blood markers that matter most for mental health monitoring, and addresses the SSRI question directly.

How testosterone affects your brain

Testosterone is not just a muscle hormone. Androgen receptors are expressed throughout the brain, with particularly high density in the amygdala (around 65% of cells are androgen receptor-positive), the prefrontal cortex, and the hippocampus. These are regions that govern emotional regulation, executive function, and memory formation. The receptor distribution alone tells you that testosterone is deeply integrated into brain function.

The serotonin pathway is one of the clearest mechanisms. Sex steroids modulate serotonin synthesis and transport in the dorsal raphe nucleus. Research in male primates shows that reproductive hormones regulate tryptophan hydroxylase-2 (TPH2), the rate-limiting enzyme in serotonin synthesis, and the serotonin transporter (SERT) through both androgen and estrogen receptor pathways (Bethea et al., 2015). This is the same pathway targeted by SSRIs, which block SERT reuptake. When testosterone is low, serotonin turnover is impaired at a fundamental level.

The dopamine pathway is equally important. Testosterone's motivational and reward-seeking effects run through the mesolimbic dopamine system, the same circuit involved in goal-directed behaviour, anticipation, and pleasure. Studies blocking D2 receptors eliminate testosterone's rewarding effects, confirming this dependency (Packard et al., 1998). This is directly relevant to the overlap between hypogonadism and anhedonia, the symptom where nothing feels rewarding or worth pursuing.

The relationship between testosterone levels and depression risk follows a U-shaped curve. Both hypogonadal and hypergonadal men have elevated depression risk (odds ratio 1.871), with the lowest risk in the mid-range (Kratzik et al., 2007). Supraphysiological testosterone is not the answer to mood problems, and neither is chronically low T. The goal is normalisation.

Low testosterone mimics clinical depression so closely that it has its own diagnostic category: PADAM (partial androgen deficiency of the ageing male). Fatigue, anhedonia, insomnia, poor concentration, reduced motivation, and depressed affect are all shared features with major depressive disorder (MDD) (Amore, 2005). The practical consequence of this overlap is discussed in the misdiagnosis section below.

Beyond monoamines, testosterone has antidepressant-like effects through HPA-axis modulation. Testosterone suppresses cortisol secretion via negative feedback at both the pituitary and adrenal levels, reducing the chronic cortisol burden that is one of the most consistent biological findings in major depression (Wainwright et al., 2016). When testosterone is normalised, the HPA axis tends to recalibrate, and morning cortisol often falls to a more appropriate level.

The first 90 days: a week-by-week timeline

The most detailed longitudinal dataset on TRT and mood comes from Saad et al. (2011), a 30-week registry study of 407 hypogonadal men on long-acting testosterone undecanoate. It remains the best-characterised timeline for mood outcomes on TRT (Saad et al., 2011).

Weeks 1-2: the honeymoon phase

The first two weeks are usually positive. Testosterone rises into range, placebo effects contribute, and men typically report notable improvements in energy and mood. Wang et al. (1996) documented this in a 60-day randomised controlled trial of 51 hypogonadal men: significant decreases in anger, sadness, tiredness, and nervousness were measurable early in the treatment course (Wang et al., 1996).

One important nuance from the Wang data: before treatment, mood correlated significantly with testosterone levels, as expected. After testosterone was normalised, that correlation disappeared. This suggests a threshold effect: once testosterone is in range, further increases do not linearly improve mood. Normalisation is the goal, not maximisation.

Weeks 3-6: the adjustment valley

This is where the timeline gets complicated, and where most men are not adequately warned.

As testosterone rises, aromatisation to estradiol increases. A large retrospective analysis of over 34,000 men on injectable testosterone found that elevated estradiol (above 42.6 pg/mL) was present in 20.2% of the cohort at some point during treatment (Tan et al., 2015). If you are prone to heavier aromatisation (higher body fat, larger infrequent doses), your estradiol rise can be pronounced in weeks 3-4 specifically.

Elevated estradiol in men has been independently associated with depressive symptoms in men under 60 (Stanikova et al., 2018). More broadly, estradiol-driven mood shifts in this window, particularly anxiety and emotional lability, can feel like the TRT is making things worse. It is not. It is a transitional hormonal state.

On the depression side, the Saad data shows detectable improvement in depression scores beginning at the 3-6 week mark. The 52.9% remission rate for subthreshold depression versus 18.8% placebo at 12 weeks comes from Shores et al. (2009), a double-blind RCT in 33 older men with low testosterone and minor or subsyndromal depression (Shores et al., 2009). The direction is clearly positive in this window, but the noise from transient E2 elevations can obscure the signal.

Injection frequency matters here. Large infrequent injections create sharp testosterone peaks that drive disproportionate aromatisation in the first 48-72 hours post-injection, followed by a relative trough before the next dose. This peak-to-trough cycling is the mechanism behind the mood swings that are common with biweekly or monthly injection schedules and far less common with more frequent dosing.

Weeks 6-12: stabilisation

By week 6, most men are past the adjustment valley. Depression continues improving. Quality of life improvements accumulate. The hormonal environment is more stable as the body adapts to the new testosterone baseline.

The Saad timeline shows that maximum mood benefit takes 18-30 weeks to fully develop. This matters for managing expectations. If you are at week 8 and still not feeling as good as you hoped, you are not a TRT failure. You are on schedule. The nervous system changes underlying mood improvement are not instantaneous.

Why anxiety gets worse before it gets better

The paradox of early TRT anxiety is best understood through the dual role of estradiol in the brain.

Estradiol is not just an aromatisation byproduct. It is required for testosterone's anxiolytic effects. Research by Carrier and colleagues demonstrated that blocking aromatisation specifically in the hippocampus eliminates testosterone's mood and anxiety benefits in animal models (Carrier et al., 2015). The brain needs to convert some testosterone into estradiol locally to produce the full psychological effect. Systemic estradiol and locally-aromatised estradiol have different effects: this is why crashing your E2 with an aggressive aromatase inhibitor produces flat affect and depression rather than improved mood. You are eliminating the very conversion that testosterone's brain benefits depend on.

At the same time, peripheral estradiol elevation above approximately 40 pg/mL is associated with irritability, emotional lability, and anxiety symptoms. The same hormone that is essential for mood stabilisation at physiological levels can drive instability when it rises disproportionately.

The peak-to-trough dynamics of injection protocols amplify this. Standard intramuscular injections given weekly or biweekly produce large swings in serum testosterone, with peaks in the first 48-72 hours followed by a gradual decline. More frequent subcutaneous dosing (daily or EOD) flattens this curve significantly. The sharp peaks from large infrequent injections drive proportionally higher aromatisation in the days immediately post-injection. This is when anxiety is most likely to be elevated.

The solution is not an aromatase inhibitor. The solution is more frequent, smaller injections: EOD or daily subcutaneous dosing produces the flattest testosterone curve and the most stable estradiol levels. For a full discussion of E2 management, see Estradiol on TRT.

The bloodwork panel for mood monitoring

Most TRT monitoring focuses on total testosterone and haematocrit. For mental health, you need a broader panel. These are the seven markers worth tracking.

For a full discussion of testing timing, see Blood Test Timing on TRT. For mood monitoring, always draw at trough: the morning before your next injection. Run the full panel at weeks 4-6 after starting or adjusting your protocol, then again at week 12. The sensitive estradiol assay (LC-MS/MS) is essential; standard immunoassays are unreliable in the male range.

When LH and FSH are relevant: if you have not yet confirmed the diagnosis of hypogonadism, low LH and FSH alongside low testosterone confirms secondary hypogonadism (the hypothalamic-pituitary axis is not driving the testes). This has different causes and implications than primary hypogonadism. Most men already on TRT have suppressed LH and FSH from exogenous testosterone, which is expected.

Is it low T or depression? The misdiagnosis problem

Hypogonadal men are 4.2 times more likely to be diagnosed with depression than eugonadal men (Shores et al., 2004). The symptom overlap between hypogonadism and major depressive disorder is almost complete: fatigue, anhedonia, sleep disruption, poor concentration, reduced motivation, low libido, and depressed mood appear in both diagnostic frameworks (Amore, 2005).

The problem is that routine testosterone screening is almost never performed in psychiatric settings. A literature review found that testosterone measurement is rarely incorporated into the initial workup for depression, despite the evidence base for the hormone-mood relationship (Smith et al., 2018). The result is a significant number of men who receive antidepressant prescriptions for a condition that is partly or entirely driven by untreated hypogonadism.

A 2026 systematic review found that TRT significantly improves mild-to-moderate depression in hypogonadal men, with a clinically meaningful effect size across multiple validated depression instruments (Canal de Velasco et al., 2026). For men with concurrent low testosterone and depression, treating the hypogonadism is a necessary component of any treatment plan.

The diagnostic bloodwork to request if you are experiencing depression before or without TRT: morning fasted total testosterone, free testosterone (calculated or direct), LH, FSH, prolactin, SHBG, and sensitive estradiol. This panel takes one blood draw and can differentiate hormonal from non-hormonal causes of your mood state. It is not standard psychiatric practice to order this, which means you may need to request it explicitly.

If total testosterone comes back below 280-300 ng/dL with symptoms, optimising hormones before or alongside starting an antidepressant is a defensible clinical approach supported by the evidence. Discuss this explicitly with your doctor.

TRT and SSRIs: can you take both?

The short answer is yes, and the combination is sometimes better than either alone.

First, the concern about SSRIs suppressing testosterone. The data does not support this. Giltay et al. (2012) compared testosterone levels across over 2,100 participants, including those on SSRIs versus those not taking antidepressants. Men on SSRIs had similar or slightly higher testosterone levels than untreated controls (Giltay et al., 2012). The fear that starting an SSRI will tank your testosterone is not supported by population-level data.

The more interesting finding goes the other direction: testosterone may augment SSRI response. Seidman and Rabkin (1998) reported that in a series of hypogonadal men who had failed to respond to SSRIs, adding testosterone supplementation produced dramatic improvement: Hamilton Depression Rating Scale scores dropped from 19.2 to 4.0 on average (Seidman & Rabkin, 1998). The hypothesis is straightforward. If the neurochemical substrate that SSRIs act on (serotonin synthesis and transport) is impaired by low testosterone, blocking reuptake with an SSRI cannot fully compensate.

The practical decision tree: if you are being evaluated for depression and your testosterone has not been checked, check it first. If it is low, optimise hormones before or alongside starting an SSRI. If you are already on an SSRI that was partially effective and you suspect hypogonadism, adding TRT may produce the improvement the SSRI alone could not achieve.

On SSRI choice, sexual side effects are worth discussing with your prescribing doctor if libido and function matter to you. The spectrum runs from paroxetine (worst sexual side effects) through most SSRIs, to bupropion and mirtazapine (lowest rates of sexual dysfunction). For the overlap between libido, prolactin, and dopamine, see Low Libido on TRT.

Compounds that make mental health worse

19-nors: nandrolone and trenbolone

If you are using 19-nor compounds and experiencing depression, anhedonia, or blunted motivation, prolactin is the first thing to check.

Nandrolone and trenbolone activate progesterone receptors, which stimulate prolactin release from the pituitary. Elevated prolactin suppresses dopaminergic tone in the mesolimbic system, directly impairing the reward circuit that drives motivation and pleasure. Beyond prolactin, nandrolone has been shown to alter sigma-1 receptor neurosteroid binding, with potential independent effects on mood (Elfverson et al., 2011).

A 2026 analysis of psychiatric outcomes in AAS users found that past AAS users had significantly higher psychiatric diagnosis rates than non-users, and that psychiatric comorbidity persisted even after biochemical testosterone recovery (Grant et al., 2026). The 19-nor class carries distinct neurological risks via the prolactin-dopamine pathway that testosterone-only cycles do not.

If you are running 19-nors and your mood is suffering, test prolactin. If it is above 20 ng/mL, cabergoline is the standard intervention. See On-Cycle Support Protocol for dosing guidance. Do not add more TRT or adjust your estradiol management first: prolactin is the variable to fix.

Crashed estradiol from AI overuse

Estradiol suppression below 20 pg/mL produces a distinctive and debilitating mood state: flat affect, anhedonia, emotional blunting, and depression that does not respond to SSRIs because it is not a serotonin-mediated state. It is a consequence of removing a hormone that the brain depends on.

Castanho et al. (2014) found that lower estradiol in older men was associated with poorer cognitive function and higher rates of depressive symptoms, independent of testosterone levels (Castanho et al., 2014). The brain requires estradiol for normal function, and aromatase inhibitor overuse removes it.

The scenario plays out predictably. A man on TRT develops water retention or nipple sensitivity at week 4. He reads online that he needs to control his estradiol and starts 0.5 mg anastrozole three times weekly. Four weeks later his E2 is in the single digits, his joints are grinding, his mood is flat, and he feels worse than before starting TRT. He assumes TRT does not work for him. It does work, but the anastrozole is counteracting it.

If you are on an AI and experiencing flat mood or depression, reduce or stop the AI before making any other protocol changes. Allow four to six weeks for estradiol to recover before evaluating your mood again.

Post-finasteride syndrome

This deserves specific mention because the mental health consequences can be severe.

5-alpha reductase inhibitors (5-ARIs) like finasteride block the conversion of testosterone to dihydrotestosterone, but they also block the synthesis of allopregnanolone, a neurosteroid that is a positive allosteric modulator of GABA-A receptors. Allopregnanolone is one of the neurosteroids involved in anxiety regulation. Blocking its synthesis can produce significant psychiatric consequences in susceptible individuals.

Melcangi et al. (2017) found that 50% of men reporting post-finasteride syndrome (PFS) met criteria for major depressive disorder (Melcangi et al., 2017). A broader mechanistic review identified at least six potential pathways through which finasteride could cause depression: neurosteroid alteration, dopamine dysfunction, reduced neurogenesis, neuroinflammation, HPA axis activation, and epigenetic changes (Saengmearnuparp et al., 2021).

A 2025 Mendelian randomisation study found no causal link between genetically-predicted 5-ARI exposure and depression at the population level (Wang et al., 2025). The emerging picture is that PFS affects a neurologically susceptible subset rather than the majority of users. For a full discussion of finasteride on TRT, see TRT and Hair Loss.

The clinical takeaway: if you are using finasteride and your mood has worsened significantly, this is a known adverse effect in susceptible individuals. Stopping finasteride may not produce immediate resolution; neurosteroid normalisation can take months.

Practical protocol for mood stability

The recommendations that follow from the evidence above:

Injection frequency. Daily or EOD subcutaneous injections produce the flattest testosterone curve and the most stable estradiol levels. If you are experiencing mood swings, anxiety peaks post-injection, or a prolonged adjustment valley, moving to more frequent smaller doses is the first protocol adjustment to make.

Estradiol management. Do not use an AI pre-emptively or aggressively. Let estradiol find its level on your protocol and only intervene if you have symptoms. The single biggest cause of TRT-era depression that I hear about in the community is crashed E2 from unnecessary AI use.

Sleep. Hormone fluctuations worsen mood through sleep disruption, and the relationship is bidirectional. AAS use and significant testosterone changes affect sleep architecture and circadian rhythm regulation (Klonteig et al., 2024). If sleep quality deteriorates during protocol changes, address it directly rather than attributing all mood symptoms to hormones.

Timeline management. If mood has not improved after 3-6 months of optimised TRT with normal bloodwork, psychiatric evaluation is warranted. Hormones are one input into mood. Psychiatric comorbidity is a stronger predictor of persistent depression than testosterone level alone (Grant et al., 2026). TRT is not a treatment for clinical depression independent of hypogonadism.

When to escalate. See a psychiatrist if suicidal ideation is present at any point; if mood does not improve after 3-6 months of optimised protocol; if you have a pre-existing psychiatric diagnosis and are starting TRT; or if you are using 19-nor compounds and experiencing anhedonia that persists after normalising prolactin.

For men considering TRT for the first time and wondering what the process looks like, see Starting TRT in Australia for a practical walkthrough of the clinical process.

Post-cycle depression is a separate but related phenomenon. After stopping AAS, the hypothalamic-pituitary axis takes time to recover endogenous production, and the crash in testosterone during this window can precipitate significant mood deterioration. See PCT Bloodwork for how to monitor and manage this transition.

Key takeaways

• Testosterone has direct effects on serotonin synthesis, dopaminergic reward circuits, and HPA axis regulation. Low T mimics clinical depression at a neurobiological level.
• The first 90 days of TRT have a predictable mood timeline: early improvement in weeks 1-2, a potential adjustment valley in weeks 3-6 from rising estradiol, stabilisation from weeks 6-12, and maximum mood benefit at 18-30 weeks.
• Anxiety in weeks 3-4 is often transient and related to peak-to-trough E2 fluctuations. More frequent smaller injections resolve this more reliably than aromatase inhibitors.
• Seven markers matter for mood monitoring: estradiol (sensitive assay), free testosterone, SHBG, cortisol, prolactin, TSH, and free T3. Test at trough, weeks 4-6 and week 12.
• Hypogonadal men are 4.2 times more likely to receive a depression diagnosis. Testosterone screening is almost never performed in psychiatry. Request it yourself if you are being evaluated for depression.
• TRT and SSRIs can be used together. Testosterone may augment SSRI response in hypogonadal non-responders. SSRIs do not suppress testosterone.
• 19-nor compounds elevate prolactin and suppress dopamine, producing anhedonia distinct from estradiol-related mood effects. Crashed E2 from AI overuse produces flat affect that does not respond to antidepressants.
• Psychiatric comorbidity predicts persistent mood problems more strongly than testosterone level alone. TRT is not a standalone treatment for clinical depression.

---

References

1. Saad, F., Aversa, A., Isidori, A. M., Zafalon, L., Zitzmann, M., & Gooren, L. (2011). Onset of effects of testosterone treatment and time span until maximum effects are achieved. European Journal of Endocrinology, 165(5), 675-685. PubMed

2. Bethea, C. L., Phu, K., Belikova, Y., & Bethea, S. C. (2015). Localization and regulation of reproductive steroid receptors in the raphe serotonin system of male macaques. Journal of Chemical Neuroanatomy, 66-67, 19-27. PubMed

3. Packard, M. G., Schroeder, J. P., & Alexander, G. M. (1998). Expression of testosterone conditioned place preference is blocked by peripheral or intra-accumbens injection of alpha-flupenthixol. Hormones and Behavior, 34(1), 39-47. PubMed

4. Kratzik, C. W., Schatzl, G., Lackner, J. E., Lunglmayr, G., Brandstatter, N., Rucklinger, E., & Huber, J. (2007). Mood changes, body mass index and bioavailable testosterone in healthy men: results of the Androx Vienna Municipality Study. BJU International, 100(3), 614-618. PubMed

5. Amore, M. (2005). Partial androgen deficiency and neuropsychiatric symptoms in aging men. Journal of Endocrinological Investigation, 28(11 Suppl Proceedings), 49-54. PubMed

6. Wainwright, S. R., Workman, J. L., Tehrani, A., Hamson, D. K., Chow, C., Lieblich, S. E., & Galea, L. A. (2016). Testosterone has antidepressant-like efficacy and facilitates imipramine-induced neuroplasticity in male rats exposed to chronic unpredictable stress. Hormones and Behavior, 79, 58-69. PubMed

7. Wang, C., Alexander, G., Berman, N., Salehian, B., Davidson, T., McDonald, V., Steiner, B., Hull, L., Callegari, C., & Swerdloff, R. S. (1996). Testosterone replacement therapy improves mood in hypogonadal men: a clinical research center study. Journal of Clinical Endocrinology & Metabolism, 81(10), 3578-3583. PubMed

8. Tan, R. S., Cook, K. R., & Reilly, W. G. (2015). High estrogen in men after injectable testosterone therapy: the low T experience. American Journal of Men's Health, 9(3), 229-234. PubMed

9. Stanikova, D., Luck, T., Bae, Y. J., Thiery, J., Ceglarek, U., Engel, C., Enzenbach, C., Wirkner, K., Stanik, J., Kratzsch, J., & Riedel-Heller, S. G. (2018). Increased estrogen level can be associated with depression in males. Psychoneuroendocrinology, 87, 196-203. PubMed

10. Shores, M. M., Sloan, K. L., Matsumoto, A. M., Moceri, V. M., Felker, B., & Kivlahan, D. R. (2004). Increased incidence of diagnosed depressive illness in hypogonadal older men. Archives of General Psychiatry, 61(2), 162-167. PubMed

11. Shores, M. M., Kivlahan, D. R., Sadak, T. I., Li, E. J., & Matsumoto, A. M. (2009). A randomized, double-blind, placebo-controlled study of testosterone treatment in hypogonadal older men with subthreshold depression. Journal of Clinical Psychiatry, 70(7), 1009-1016. PubMed

12. Carrier, N., Saland, S. K., Duclot, F., He, H., Mercer, R., & Kabbaj, M. (2015). The anxiolytic and antidepressant-like effects of testosterone and estrogen in gonadectomized male rats. Biological Psychiatry, 78(4), 259-269. PubMed

13. Smith, J. B., Rosen, J., & Colbert, A. (2018). Low serum testosterone in outpatient psychiatry clinics: addressing challenges to the screening and treatment of hypogonadism. Sexual Medicine Reviews, 6(1), 69-76. PubMed

14. Canal de Velasco, L. M., Gonzalez Flores, J. E., Vargas, L., & Morales Arteaga, J. L. (2026). Psychiatric and cognitive effects of testosterone therapy in adult men: a systematic review of clinical evidence and mechanistic insights. Cureus, 18(2), e102894. PubMed

15. Giltay, E. J., Enter, D., Zitman, F. G., Penninx, B. W., van Pelt, J., Spinhoven, P., & Roelofs, K. (2012). Salivary testosterone: associations with depression, anxiety disorders, and antidepressant use in a large cohort study. Journal of Psychosomatic Research, 72(3), 205-213. PubMed

16. Seidman, S. N., & Rabkin, J. G. (1998). Testosterone replacement therapy for hypogonadal men with SSRI-refractory depression. Journal of Affective Disorders, 48(2-3), 157-161. PubMed

17. Elfverson, M., Johansson, T., Zhou, Q., Le Greves, P., & Nyberg, F. (2011). Chronic administration of the anabolic androgenic steroid nandrolone alters neurosteroid action at the sigma-1 receptor but not at the sigma-2 or NMDA receptors. Neuropharmacology, 61(7), 1172-1181. PubMed

18. Grant, B., Kean, J., de Silva, N. L., Aguilera, R., Quinton, O. C. B., Gumssani, M., Bassett, P., Dhillo, W. S., Lingford-Hughes, A., Wolff, K., & Jayasena, C. N. (2026). Clinical features of androgen abuse withdrawal in men during the first year of cessation: a community dwelling study. Journal of Clinical Endocrinology & Metabolism, advance online publication. PubMed

19. Castanho, T. C., Moreira, P. S., Portugal-Nunes, C., Novais, A., Costa, P. S., Palha, J. A., Sousa, N., & Santos, N. C. (2014). The role of sex and sex-related hormones in cognition, mood and well-being in older men and women. Biological Psychology, 103, 158-166. PubMed

20. Melcangi, R. C., Santi, D., Spezzano, R., Grimoldi, M., Tabacchi, T., Fusco, M. L., Diviccaro, S., Giatti, S., Carra, G., Caruso, D., Simoni, M., & Cavaletti, G. (2017). Neuroactive steroid levels and psychiatric and andrological features in post-finasteride patients. Journal of Steroid Biochemistry and Molecular Biology, 171, 229-235. PubMed

21. Saengmearnuparp, T., Lojanapiwat, B., Chattipakorn, N., & Chattipakorn, S. (2021). The connection of 5-alpha reductase inhibitors to the development of depression. Biomedicine & Pharmacotherapy, 143, 112100. PubMed

22. Wang, J., Lin, D., Zhao, J., Wei, S., Wang, H., & Li, W. (2025). Finasteride use does not lead to depression or suicide: insights from a large-scale cohort study and Mendelian randomization analysis. Journal of Cosmetic Dermatology, 24(12), e70579. PubMed

23. Dai, W., Liu, J., Xie, H., Teng, Z., Luo, W., Yuan, H., & Zhang, X. (2023). Association between subclinical hypothyroidism and psychotic features in Chinese young adults with first-episode and untreated major depressive disorder. Journal of Affective Disorders, 333, 209-215. PubMed

24. Klonteig, S., Scarth, M., & Bjornebekk, A. (2024). Sleep pathology and use of anabolic androgen steroids among male weightlifters in Norway. BMC Psychiatry, 24(1), 62. PubMed