“Male dominance – and with it the superiority of the male body – was cemented into medicine’s very foundations… Women were marked by their anatomical difference from men, and medically defined as faulty, defective, deficient.”
— Elinor Cleghorn (2021, pp. 6-7)
The history of medicine is marked by an entrenched misogyny that has shaped what is studied, how it is studied, and what is taken to count as legitimate knowledge. Biological models have overwhelmingly centred male physiology, male life courses, and male assumptions about risk. Women’s experiences have often been treated as deviations from this norm, rather than as phenomena worthy of study in their own right. This bias is particularly stark in relation to women in midlife. Cultural narratives frequently portray menopausal women as losing beauty, value, and sexuality, while healthcare disparities persist through medical bias and a relative absence of focused research (Gibbons, 2025a).
Against this background, this new UK study of self-harm and suicide in women aged 40–59 is very welcome. It addresses an under-examined group using robust, long-term data.
Women, particularly women in midlife, often are not the focus of health research. This study seeks to explore self-harm and suicide in women aged 40-59 years.
Methods
The study draws on routinely collected hospital data from three English cities and includes more than 14,000 hospital presentations for self-harm by women aged 40–59 between 2003 and 2016, with mortality follow-up through national Office for National Statistics records to the end of 2019. The authors examine rates of hospital-presenting self-harm, methods used, clinician-recorded precipitants, clinical responses including psychosocial assessment and psychiatric admission, repetition of self-harm, and suicide mortality. A comparison group of younger women aged 25–39 is included, and differences within midlife are explored using five-year age bands.
Results
The authors report several key findings. Rates of hospital-presenting self-harm in women aged 40–59 were high but lower than in younger women, and self-harm rates declined steadily across midlife, with the lowest rates in women aged 55–59, while remaining relatively stable over time overall. Compared with younger women, women in midlife were more likely to report financial difficulties, alcohol problems, physical health problems, mental health problems, and bereavement as precipitants of self-harm, with physical health problems and bereavement increasing with age within midlife and alcohol involvement common across all age bands. Despite declining self-harm presentations and reduced longer-term repetition in older midlife women, clinical responses became more intensive with age, with higher rates of psychosocial assessment and psychiatric inpatient admission. All-cause mortality increased markedly across midlife, and suicide mortality was more than twice as high in the oldest midlife women compared with the youngest, although the absolute number of suicide deaths was small.
Suicide mortality was more than twice as high in the oldest midlife women compared with the youngest women.
Conclusions
The authors conclude that women in midlife are not a homogeneous group, that vulnerabilities and care needs change across this period, and that greater attention to alcohol use, financial stressors, physical and mental health problems, and age-related social transitions may be important for clinical formulation and safety planning.
Women in midlife are not a homogenous group; we need to explore what factors contribute to the increased rates of suicide in this population.
Strengths and limitations
To understand what this study really tells us, we need to do something that suicide research still struggles to do, and that is not done clearly in the authors’ interpretation: we need to separate self-harm from suicide, and examine each carefully rather than assuming they are simply different expressions of the same problem. One of the enduring difficulties in suicide research is the tendency to collapse self-harm and suicide into a single continuum, despite decades of evidence that they are related but distinct phenomena (Gibbons 2024a, 2024b; Gibbons, 2025a). Self-harm, which is common, and suicide, which is rare, are often treated as interchangeable, an assumption that obscures their different functions, motivations, and dynamics, and can confound both research and clinical practice (Gibbons, 2025b).
Self-harm often functions as a way of coping with distress: regulating overwhelming affect, communicating pain, or maintaining connection (Gibbons, 2024a). Suicide, by contrast, more often involves withdrawal from relationship and a move toward psychic finality, an attempt to escape unbearable mental pain rather than to signal it (Shneidman, 1993). Although self-harm is statistically associated with later suicide (Geulayov et al., 2019; Hawton et al., 2015; Tidemalm et al., 2015), this does not mean the two are psychologically equivalent. They may share vulnerabilities, such as loss of mentalisation or collapse of symbolic processing, while remaining clinically and phenomenologically different acts (Gibbons, 2024).
When self-harm is examined in its own right in this study, a clear age-related pattern emerges. Hospital-presenting self-harm is more common in younger women aged 25–39 than in women aged 40–59, and within midlife it declines steadily across successive five-year age bands. Women aged 55–59 present far less often than women aged 40–44. The consistency of this pattern across both between-group comparisons and within-group comparisons suggests that self-harm, as a hospital-visible behaviour, becomes less common as women move from young adulthood into and through midlife. Across all midlife age groups, self-poisoning was the dominant method of self-harm. With increasing age, overdoses involving benzodiazepines became more common, while patterns of repetition also changed. There was no difference in 12-month repetition of self-harm between age groups, but longer-term repetition decreased with age, indicating that older midlife women were less likely to re-present repeatedly after an initial episode. These findings about self- harm stand independently of any conclusions about suicide, and are consistent with previous research in this area.
One possible way of understanding this age-related decline in self-harm, is developmental. In my own work on the psychodynamics of self-harm (Gibbons, 2024a), I have described self-harm as an acting-out behaviour that occurs when emotional experience cannot be adequately symbolised, such that distress must be communicated across the body boundary rather than through words. From this perspective, declining self-harm with age may reflect an increased capacity to mentalise and articulate distress verbally rather than enact it physically. This does not imply that distress diminishes with age, but that its mode of expression may change, with the body becoming less necessary as a site of communication when symbolic thought is more available. Importantly, this interpretation concerns self-harm only and tells us nothing about suicide.
What, then, does the study tell us about suicide? Because the cohort is restricted to women who had already self-harmed and presented to hospital, the suicide findings apply only to a particular subgroup of those who die by suicide. This is an important strength, as it allows long-term outcomes to be examined in a clearly defined population, but it also highlights a persistent limitation in suicide research: the tendency to focus on what is most visible to services rather than where most suicide deaths actually occur, namely among people not in contact with mental health services, many of whom have no recorded history of mental illness.
The length of follow-up in this study is crucial. Index self-harm presentations occurred between 2003 and 2016, with mortality outcomes tracked until the end of 2019. Participants were therefore followed for up to approximately 16 years, with a minimum follow-up of around three years. Suicide analyses were conducted using person-years-at-risk, confirming that this was a long-term outcome study rather than an examination of short-term suicide risk following self-harm. Over this period, the risk of dying by suicide was low, around 1 in 100. Of the 6,147 women who could be traced for mortality follow-up, 77 died by suicide, corresponding to approximately 1.2–1.3% of the self-harm–experienced midlife cohort. This finding is consistent with other contemporary cohort studies of suicide following hospital-presenting self-harm Hawton et al., 2015).
Suicide mortality was not evenly distributed across midlife. Within the cohort, suicide mortality increased with age, such that women aged 55–59 experienced more than twice the suicide mortality of women aged 40–44, although the absolute number of deaths remained small in all age bands. The study does not show that suicide follows escalating or repeated self-harm, that it can be predicted from the frequency of self-harm, or that increased clinical contact prevents it.
These findings are interesting, but they can be interpreted in more than one way. One question the study does not address directly is how this pattern compares with men in the same age groups. Evidence from large hospital-presenting self-harm cohorts, suicide mortality in men increases with age; in register-linked Scandinavian data, men in the 45–59 age band have higher first-year suicide incidence than men aged 30–44, and large English multicentre studies also show a positive age–risk relationship (Geulayov G 2019, Tildeman 2015). This raises the possibility that what we are seeing in midlife women is not simply a biological story, but a relational and developmental one, shaped by gendered pathways of distress, visibility, and help-seeking.
What this study ultimately shows is that self-harm and suicide are not interchangeable outcomes, but different expressions of distress that unfold across midlife in different ways.
Implications for practice
There is no current evidence that suicide can be predicted, and self-harm and suicide are not interchangeable phenomena. Suicidal ideation has a low predictive value, and although self-harm is associated with increased risk at a population level, it does not narrow down a group at risk of suicide in a way that is clinically useful.
Our role in mental health is not to predict suicide; that is a distortion of our reality-based task. When someone expresses suicidal thoughts or self-harms, they are communicating distress. These acts may signal a loss of capacity to mentalise, a collapse in the ability to put feelings into words. Our task is not to forecast death, but to turn towards the person and ask what these communications mean. If we can help individuals restore symbolic thinking and name their distress, we are more likely to help them re-engage with life. As in other areas of healthcare, our aim is not solely to prevent death, but to help people live.
For women, midlife is often a period of profound transition. It can involve biological change, but also loss, reorientation and the renegotiation of identity. It may be a time of mourning of youth, fertility, roles and relationships, but also a time of reawakening and consolidation, a reclaiming of authority and experience. The stigma directed at women as they age needs to be challenged. Cultural narratives portray decline and invisibility, yet many women experience midlife as a period of integration and strength. Personally and clinically, I have never felt more visible or powerful than at this stage of life.
In summary, the implications for practice are not about refining prediction, but about deepening formulation. Each woman presenting in midlife requires an individual, biopsychosocial formulation that takes seriously her history, her relational world, her physical health, her social and economic context, and the developmental transitions of this stage of life.
Expressions of suicidality or self-harm should be understood as communications of distress, often emerging when emotions cannot yet be symbolised. Our role is to help restore that capacity, to think, to name, to mourn, and to make meaning, rather than to focus narrowly on risk prediction. Good practice is thoughtful, relational and contextual, not algorithmic.
Our aim is not solely to prevent death, but to help people live. Each woman presenting in midlife requires an individual, biopsychosocial formulation that takes seriously her history, relational world, physical health and social context.
Statement of interests
Rachel Gibbons is a consultant psychiatrist, psychoanalyst and group analyst. She has written previously about the menopause transition and the risks of over-biological framings of women’s mental health. AI was used as part of the drafting process for this blog. Rachel Gibbons reviewed and edited all content for accuracy and appropriateness.
Edited by
Laura Hemming.
Links
Primary paper
Caroline Clements, Harriet Bickley, Keith Hawton, Galit Geulayov, Keith Waters, Jennifer Ness, Samantha Kelly, Ellen Townsend, Louis Appleby, Nav Kapur. (2025). Self-harm in women in midlife: rates, precipitating problems and outcomes following hospital presentations in the multicentre study of self-harm in England. The British Journal of Psychiatry, 227(1), 456-462.
Other references
Cleghorn, E. (2021). Unwell Women: A Journey Through Medicine and Myth in a Man-Made World.
Geulayov G, Hawton K, Casey D, et al. (2019). Suicide following presentation to hospital for non-fatal self-harm in the Multicentre Study of Self-harm in England. The Lancet Psychiatry, 6(12), 1022–1030.
Gibbons, R. (2024a). The psychodynamics of self-harm. BJPsych Advances, 31(3), 164-172.
Gibbons, R., (2024b). Understanding the psychodynamics of the pathway to suicide. International Review of Psychiatry, pp.1-9.
Gibbons, R. (2025a). The menopause transition: a call for a holistic approach. BJPsych Bulletin, 1-3.
Gibbons, R. (2025b). Rethinking suicide prevention: from prediction to understanding. BJPsych International, 22(4), 131-134.
Hawton K, Bergen H, Cooper J, et al. (2015). Suicide following self-harm: findings from the Multicentre Study of Self-harm in England, 2000–2012. Journal of Affective Disorders, 175, 147–151.
Tidemalm D, Beckman K, Dahlin M, Vaez M, Lichtenstein P, Långström N, Runeson B. (2015). Age-specific suicide mortality following non-fatal self-harm: national cohort study in Sweden. Psychological Medicine, 45(8), 1699–1707.
Shneidman, E. S. (1993). Commentary: Suicide as psychache. Journal of Nervous & Mental Disease, 181(3), 145-147.