A new study of almost 8,000 mothers with postpartum depression shows that it has nine subtypes, each with its own pattern of symptoms, trauma history and pain and with distinct links to genetic vulnerability. The findings give clinicians and researchers a more precise starting-point for preventing and treating mental health problems after birth – a field in which early recognition can change the course of a mother’s life and her child’s well-being.
Three months after giving birth, Susan sits on the sofa with her baby asleep on her chest and the TV on mute. She cries in short bursts, sleeps poorly and feels a hard knot in her stomach every time the baby begins to whimper. During pregnancy, she had severe pelvic pain and several hospitalisations. Only at her most recent appointment, months after the birth, did her general practitioner ask her to complete a questionnaire about how she was doing mentally.
Her score is clearly above the threshold for postpartum depression – which can remain hidden for months unless someone asks the right questions at the right time.
The Norwegian Mother, Father and Child Cohort Study (MoBa) has thousands of mothers like Susan – a reminder that postpartum depression affects not just individuals but whole families and communities who move into a depressive state after birth.
Nevertheless, the paths into illness look very different when women are followed through pregnancy, delivery and the first months with their child. These are the trajectories that psychiatrist and geneticist Anna E. Bauer, affiliated with the University of North Carolina, analysed together with colleagues in Norway, Sweden and the United Kingdom. She wanted to know whether the patterns recognised from clinical work also appear in the questionnaire responses and registry data.
The study shows that they do: using an unsupervised machine-learning method that groups people by shared patterns without predefining what the subtypes should look like, the team identified nine statistically distinct subtypes – patterns within the data that may correspond to clinically meaningful groups, ranging from lives marked by abuse and stress to everyday circumstances of harsh living conditions.
The researchers behind the study want to give clinicians clearer landmarks when they meet women like Susan. Leading that effort is Helga Ask, senior author and Research Professor and Head of the PsychGen Centre for Genetic Epidemiology and Mental Health at the Norwegian Institute of Public Health in Oslo.
“We need to spot the most vulnerable mothers while they are still coming in for antenatal visits with the midwife,” says Helga Ask, adding: “Mothers with a heavy trauma history should be met differently than mothers whose main burden is pain and exhaustion – so that conversations, medication, pain management and social support reach the right woman at the right time.”
What the MoBa data reveal about mothers
More than 95,000 pregnant women in MoBa completed questionnaires during pregnancy and again when the baby was six months old. From this cohort, Bauer and colleagues selected 7,859 mothers who, six months after birth, scored above the established cut-off on the Edinburgh Postnatal Depression Scale. For these mothers, the team had detailed information on psychiatric history, traumatic experiences, symptom patterns before and after birth and obstetric and physical health factors during pregnancy and delivery.
“From the outset, we decided that the main analysis should rely on information that can, in principle, be gathered in routine practice,” says Helga Ask. She adds that the subtypes will only make a real difference if clinicians can identify them using the data that midwives, general practitioners and psychiatrists already work with every day.
To explore biological influences, the researchers also used polygenic risk scores – a way of summarising many small genetic risk variants into a single estimate of inherited vulnerability. These scores do not diagnose conditions but indicate population-level tendencies into one overall measure – for depression, post-traumatic stress disorder (PTSD), attention-deficit/hyperactivity disorder (ADHD) and other mental disorders. The analysis pointed to higher genetic risk for PTSD in the trauma-related subtypes and higher genetic liability for ADHD in some of the others.
Nine patterns emerge – and most hold up
To make sure the patterns were reliable, the researchers first analysed one half of the material (discovery sample) and then checked whether the same patterns appeared in the other half (test sample).
The first step was to simplify the large amount of information, so the algorithm could focus on the most important patterns. In the discovery sample, an algorithm grouped together mothers who resembled one another across symptoms, medication use, psychiatric and trauma history, pain, physical health and lifestyle. Nine subtypes emerged.
When the researchers repeated the analysis in the second half of the material, seven of the subtypes reappeared with similar profiles, indicating moderate-to-strong stability. Two showed lower reproducibility and will require confirmation in larger and more diverse cohorts.
“Here we examine whether the two patterns consolidate as independent subtypes or blend into neighbouring trajectories,” says Helga Ask.
In addition, statisticians, geneticists and clinicians reviewed the patterns together.
Three forces that shape the illness pathways
Across the nine subtypes, three main themes – or axes – emerge, offering a clearer map of why postpartum depression consistently differs so much from one mother to the next across the clustering dimensions in the mothers’ stories. Trauma is the first: some mothers carry experiences of violence, abuse or other severe events; they cluster together and show higher genetic risk for PTSD and related conditions.
The body under strain is the second: for many women, pregnancy and birth are physically demanding, with severe pain, disrupted sleep and complications; that burden colours subsequent depressive symptoms and calls for practical, everyday measures – targeted pain management, support for sleep and recovery and help to get breastfeeding and routines on track.
Everyday circumstances form the third axis: education, relationship stability, finances and practical support weigh more heavily in the more severe subtypes and can prolong the course of illness and make it more serious.
In Helga Ask’s view, the conclusion is clear: postpartum depression is not a single reaction to becoming a mother but a heterogeneous set of conditions shaped by inherited vulnerability, past experiences and present circumstances. The subtypes represent statistical patterns and are not yet clinical categories, but they offer a structured way to understand this diversity.
How the findings can change clinical care
The team will now examine whether the subtypes appear similarly in other countries and in more diverse populations. In parallel, they plan targeted trials that test interventions in selected groups – trauma-focused psychotherapy, strengthened social support and regular follow-up for the most burdened women and more ambitious pain management and closer monitoring after birth for women with severe pain and physical complications.
Helga Ask also notes that genetic liability for ADHD was higher for some subtypes, and that ADHD-related traits are often overlooked in prenatal and postnatal care.
“Here the subtypes can serve as a map that helps us to spot these mothers and arrange support for structure, sleep and everyday life with an infant,” she says.
The aim is for mothers like Susan to meet a healthcare system that recognises their subtype – so support and treatment can be organised in the combinations they actually need – and mothers are met early, accurately and compassionately before the illness becomes overwhelming.
