Schizophrenia and schizoaffective disorders (collectively referred to as schizophrenia spectrum disorders, SSD) show distinct patterns between men and women, including age of diagnosis, incidence rates, clinical presentation, and treatment response (Ochoa et al., 2012). For example, we often observe a “female-specific and time-dependent deterioration” (Brand et al., 2024, pg. 893) in women over 45 who also experience higher incidence rates and more psychosis relapses than age-matched men, and younger women with SSD (Sommer et al., 2023).
One explanation for these sex differences is the oestrogen protection hypothesis (review, Sharpe, 2003). Menopause, typically occurring between ages 45-55, is marked by fluctuating and declining oestrogen levels (Burger et al., 2007), leading to stable and low levels after the final menstrual period (Harlow et al., 2012). Oestrogen has neuroprotective effects and modulates neurotransmitters systems, including those which are integral to the neurobiology of schizophrenia (Ross et al., 2006). Menopause may elicit greater vulnerability to psychosis and other psychiatric symptoms (Marwick, 2024) due to reduced hormonal modulation of critical neurotransmitter systems.
This decline in oestrogen may contribute to the increase in SSD symptoms. This has prompted research into how menopause hormone treatment (MHT) could help mitigate psychosis relapse, further supported by reports suggesting that higher doses of antipsychotic did not prevent this deterioration in women over 45 (Sommer et al., 2023). MHT, otherwise known as Hormone Replacement Treatment (HRT), uses synthetic progestogen and/or oestrogen to alleviate menopausal symptoms like hot flushes. There are mixed findings for the beneficial impact of MHT on SSD in menopausal women, but not in a systematic and real-world context – until now. The authors (Brand et al., 2024) studied the real-world effectiveness of MHT in preventing psychosis relapse in women of menopausal age with SSD.
Does the decline in oestrogen during menopause increase risk of schizophrenia diagnoses and psychosis relapse in women over 45?
Methods
This cohort study identified women (biological sex) with an SSD diagnosis who were hospitalised in Finland between January 1972 to December 2014, as recorded in a hospital discharge register. The cohort was restricted to women who initiated MHT from the ages of 40-62 across 1995 to 2017, with MHT exposure derived from the prescription register.
Follow-up extended from MHT initiation to 2017 (or patient death). Psychosis relapse was measured as:
- Hospitalisation due to psychosis (primary outcome)
- Hospitalisation due to psychiatric reason (secondary outcome)
Outcomes were recorded on the hospital discharge register. Women who changed MHT prescription were included multiple times in the follow-up, with each prescription examined separately.
Statistical analysis
Analyses compared MHT periods of use to non-user periods. During these time periods, primary and secondary outcomes were analysed using stratified Cox models (by age groups: 40-49, 50-55, 56-62). Each patient was their own control when they had changes to MHT and a hospitalisation event. Hazard ratios were then conducted to measure risk of relapse for each outcome with MHT use.
Results
Cohort
An initial cohort of 30,785 women was identified, and this was reduced to 3,488 women on MHT. The follow up varied between 3-15 years, depending on when MHT initiation occurred and when a patient exited the study. Women with shorter follow-up periods were either recruited towards the end of the study or began a new MHT regimen.
From the cohort, 52.70% had a least one hospitalisation due to psychosis and 63.20% had a least one hospitalisation due to psychiatric reason. Moreover, 70.90% of women were on MHT for more than a year, and 54.50% of the cohort had just one period of MHT use.
Importantly, 95.50% of the cohort reported at least one non-use period, allowing for a within-subject comparison of the impact of MHT on hospitalisation outcomes.
Real-world effectiveness of MHT on reducing risk of hospitalisation due to psychosis
Overall, MHT use was associated with a 16% reduction in risk of relapse, particularly among women aged 40-55. In contrast, women who initiated MHT from the ages of 56-62 did not experience a reduced risk of relapse.
When comparing MHT formulations, it was found that oestrogen-only or oestrogen with progestogen showed similar effects, reducing risk of relapse by 14 to 21%. However, administrative route was important, as transdermal administration did not significantly reduce risk of relapse, whereas oral administration reduced risk of relapse by 13-18% for both oestrogen-only and oestrogen with progestogen combinations.
Different oestrogenic and progestogenic compounds, varying in chemical composition and potency, produced differences in effectiveness. Oestradiol-only and oestrogen with levonorgestrel, MPA, and norethisterone led to a 15 to 25% lower risk of relapse. In contrast, oestriol-only or oestrogen with dydrogesterone did not reduce relapse risk.
Real-world effectiveness of MHT on reducing risk of hospitalisation due to psychiatric condition
The authors also found that risk of hospitalisation due to psychiatric reason decreased with MHT use, and the findings for different age groups, MHT formulations and administrative routes mirror the primary outcome results.
Hormone treatment was associated with a 16% reduced risk of psychosis relapse, and this relationship was strongest when hormone treatment was initiated earlier (e.g. from the ages of 40-55).
Conclusions
In a cohort of 3,488 women with SSD at menopausal age, MHT was linked to a 16% reduction in relapse risk. This effect was particularly notable among women who began MHT between the ages of 40-55, highlighting a time-sensitive benefit of intervention during menopause. Similar trends were observed with risk of hospitalisation due to psychiatric condition. Effectiveness of reducing risk in women on MHT was dependent on method of administration and formulations. These findings emphasise the need to tailor MHT regimens to individual profiles to maximise benefits for relapse prevention in women with SSD.
Personalising hormone treatments during menopause – considering age and treatment type and formulation – may optimise mental health outcomes for women with schizophrenia spectrum disorders.
Strengths and limitations
Brand et al. (2024) highlight a promising avenue of research for treating SSD using MHT, which corroborates previous findings on SSD symptoms and MHT use (Lindamer et al., 2001). The findings emphasise the potential to reduce hospitalisation due to psychosis and psychiatric reason, which are clinically-significant outcomes. While the consistency across MHT formulations strengthens these findings, the differences between oral and transdermal administration raises further questions about correct administration or adherence to regimen.
The naturalistic, observational data allows for better generalisation to real-world settings. Objective measures of relapse (hospitalisation) provide practical insight into severe SSD cases, but the authors note that this is less applicable to milder SSD cases where symptom worsening might manifest in functional impairments or non-hospitalised psychotic episodes. Future research may wish to establish the effects of MHT on SSD through objective scales that explore worsening SSD in day-to-day instances.
A key strength of this study is the longitudinal tracking, allowing participants to serve as their own control over time. This allows for control over time-dependent factors, like baseline illness severity or age. Antipsychotic treatment was controlled to isolate the relationship between MHT and psychosis relapse. However, important confounding factors, such as ethnicity, family history of psychosis and lifestyle were not reported or controlled. These factors influence SSD diagnosis, progression, and outcomes, and the absence of ethnicity data limits generalisation of these findings across diverse populations where SSD presentation and access to treatment may vary.
Another methodological factor is that menopause was determined by age rather than menstrual cycle information, which is typically used to characterise menopause (Harlow et al., 2012). This approach may obscure the exact phase of menopause, as different stages, such as perimenopause, can involve more pronounced menopausal symptoms that could exacerbate SSD. Better characterisation of menopause in this cohort would clarify the timing and effects of MHT initiation on relapse prevention.
Addressing methodological gaps and generalisability to the cohort sample would provide deeper insight and encourage personalised treatment.
Implications for practice
This paper highlights the importance of prioritising women’s health through appropriate systems, services, and support for women at risk of psychosis relapse, particularly women with prior hospitalisations. Tailored interventions, including improving access to MHT in this vulnerable group, could play a significant role in reducing relapse risks and enhancing mental health outcomes. While there remain some negative connotations surrounding MHT use, which are often centred around its safety, the authors emphasise that a well-managed, individualised dose, which is tailored to factors such as age, can make MHT a beneficial treatment for women.
While this study demonstrates the impact of MHT on objective measures of relapse, it does not explore subjective patient-reported outcomes, such as patient perceptions of wellbeing, symptom management, or daily functioning. Incorporating patient-reported outcomes in future research would provide a more holistic understanding of the impact of MHT, ensuring that interventions not only alter clinical markers but also meaningfully enhance patients’ lived experiences.
While this paper focuses on the clinical implications for women with existing diagnoses, it also raises important questions about the effects of menopause on mental health in women without prior psychiatric conditions, in line with the oestrogen hypothesis. By raising awareness, healthcare providers can help women better understand the relationship between menopause and mental health, enabling earlier intervention and support for those who may be at risk of developing psychosis or other mental health challenges during this transition.
Further exploration is needed to ensure the safety and compatibility of MHT when used alongside prescribed antipsychotic medications. This will draw better conclusions about treatment efficacy in women over 45, where antipsychotic efficacy plateaus (Sommer et al., 2023). Additionally, research into the neural mechanisms underlying the relationship between SSD and sex steroid hormones is also critical. This will allow for more effective, targeted therapies for women at risk of relapse during menopause to ensure improved mental health outcomes.
More research is needed on MHT safety, patient-reported outcomes, and the biological mechanisms linking sex hormones to mental health
Statement of interests
No conflict of interest to declare.
Links
Primary paper
Brand, B. A., Sommer, I. E., Gangadin, S. S., Tanskanen, A., Tiihonen, J., & Taipale, H. (2024). Real-world effectiveness of menopausal hormone therapy in preventing relapse in women with schizophrenia or schizoaffective disorder. American Journal of Psychiatry, 181(10), 893-900. DOI: 10.1176/appi.ajp.20230850
Other references
Burger, H. G., Hale, G. E., Robertson, D. M., & Dennerstein, L. (2007). A review of hormonal changes during the menopausal transition: focus on findings from the Melbourne Women’s Midlife Health Project. Human reproduction update, 13(6), 559-565.
Harlow, S. D., Gass, M., Hall, J. E., Lobo, R., Maki, P., Rebar, R. W., … & STRAW+ 10 Collaborative Group. (2012). Executive summary of the Stages of Reproductive Aging Workshop+ 10: addressing the unfinished agenda of staging reproductive aging. The Journal of Clinical Endocrinology & Metabolism, 97(4), 1159-1168.
Lindamer, L. A., Buse, D. C., Lohr, J. B., & Jeste, D. V. (2001). Hormone replacement therapy in postmenopausal women with schizophrenia: positive effect on negative symptoms?. Biological psychiatry, 49(1), 47-51.
Marwick K. The influence of the menopause in first onset of mental illness. The Mental Elf, 3 Oct 2024.
Ochoa, S., Usall, J., Cobo, J., Labad, X., & Kulkarni, J. (2012). Gender differences in schizophrenia and first‐episode psychosis: A comprehensive literature review. Schizophrenia research and treatment, 2012(1), 916198.
Ross, C. A., Margolis, R. L., Reading, S. A., Pletnikov, M., & Coyle, J. T. (2006). Neurobiology of schizophrenia. Neuron, 52(1), 139-153.
Sharpe, R. M. (2003). The ‘oestrogen hypothesis’–where do we stand now? 1. International journal of andrology, 26(1), 2-15.
Sommer, I. E., Brand, B. A., Gangadin, S., Tanskanen, A., Tiihonen, J., & Taipale, H. (2023). Women with schizophrenia-spectrum disorders after menopause: a vulnerable group for relapse. Schizophrenia Bulletin, 49(1), 136-143.
