Siobán D. Harlow, Mardge Cohen, Suzanne E. Ohmit, Paula Schuman, Susan Cu-Uvin, Xihong Lin, Ruth Greenblatt, Alejandra Gurtman, Ann Khalsa, Howard Minkoff, Mary A. Young, Robert S. Klein
4 min
Women living with HIV often report menstrual irregularities, which are frequently attributed to the virus itself or antiretroviral treatments. This study sought to determine whether other common factors—specifically substance use and the use of psychotherapeutic medications—contribute to these menstrual disruptions in women who are HIV-seropositive and seronegative.
Researchers analyzed prospective menstrual calendar data from 1,075 women enrolled in the Women’s Interagency HIV Study (WIHS) and the HIV Epidemiology Research Study (HERS). Participants recorded daily bleeding or spotting patterns over six months. The researchers adjusted for demographic factors, body mass index (BMI), and depressive symptoms to isolate the effects of substance use (alcohol, tobacco, marijuana, crack cocaine, injection drugs, and methadone maintenance) and psychotherapeutic medications (such as antidepressants and antipsychotics) on cycle length and variability.
The study found that substance use and psychotherapeutic medications are strong predictors of menstrual dysfunction. Specifically, women on methadone maintenance therapy and those who used injection drugs faced significantly higher odds of experiencing cycles of 90 days or longer. Similarly, the use of psychotherapeutic medications was associated with a nearly doubled risk of amenorrhea and an increased likelihood of very short menstrual cycles (less than 18 days). These associations remained significant even after adjusting for depressive symptoms, suggesting that the medications themselves, rather than the underlying conditions they treat, may be influencing the neuroendocrine axis.
These findings suggest that clinicians should look beyond HIV status and antiretroviral therapy when evaluating menstrual complaints in women. Because substance use and the use of psychotropic drugs are prevalent in this population, they represent important, potentially modifiable, or manageable contributors to menstrual health. Recognizing these associations allows for more accurate clinical counseling and may help patients understand that their menstrual symptoms could be a side effect of their medications or substance use.
Alex: [pressing] And did they account for the psychological burden of living with a chronic, stigmatized illness? Stress alone can suppress GnRH pulsatility.
Sam: [steady] They did. They adjusted for depressive symptoms using the CES-D, and also controlled for BMI and demographic factors. The association between these medications and menstrual dysfunction persisted after those adjustments. That the effect is independent of depressive symptoms themselves is an important piece of evidence—it is not simply that sicker or more distressed women are more likely to be on these medications and also more likely to have irregular cycles. [[RP_SECTION:clinical-implications-for-practice|Clinical implications for practice]]
Alex: [deliberate] So the clinical implication is not just to check viral load. It is to take a granular look at the full medication list.
Sam: [precise] Exactly. If a patient presents with amenorrhea, the default clinical instinct is to look for disease progression. This study argues that the more likely driver is the pharmacology of their daily regimen. The diagnostic blind spot is not ignorance of the biology—it is the salience of the primary diagnosis crowding out everything else.
Alex: [reflective] Where does the research need to go from here? The binary use-versus-no-use framing seems like it leaves a lot on the table. [[RP_SECTION:future-research-directions|Future research directions]]
Sam: [broader perspective] It does. The obvious next step is dose-response work—mapping how specific psychotropics at specific doses shift cycle length, ideally with concurrent hormone monitoring. That would move the field from identifying a correlation to defining a clinical threshold: at what point does the neuroendocrine disruption become significant enough to require management? That question is still open.
Alex: [quiet conviction] It is a useful reminder that in complex clinical populations, the most visible explanation is not always the correct one. The treatment burden can rival the disease itself.
Sam: [measured, concluding] That is the core of it. And it is a clear call for clinicians to broaden their diagnostic lens—to think about the neuroendocrinologic consequences of the entire regimen, not just the index condition. Thanks for listening to ResearchPod.