Postpartum Depression: Signs, Causes, and What Actually Helps

The baby is finally asleep. This is the window everyone told you to sleep in, and you’re lying there wide awake running a list. Whether the latch was right. Whether you should’ve gone back in when she fussed. Whether you love her enough, which wasn’t a question you knew was available until about a week ago and now turns up most nights somewhere around three.

At the six-week check someone asked how you were doing and you said fine, because the alternative was explaining a feeling you don’t have words for and can’t defend. It isn’t sadness, exactly. It’s more like standing slightly outside your own life, watching a woman hold a baby and finding her difficult to identify with.

That gap between what postpartum depression is assumed to look like and what it actually feels like is why so many people go months without naming it. Here’s what the research describes, including the parts that rarely make it into the leaflet.

It often doesn’t look like sadness

The cultural image is a woman crying and unable to get out of bed. That happens. It’s also one presentation among several, and probably not the most common one.

What people describe more often is irritability that feels chemical, a short fuse with a partner who’s done nothing in particular. Or anxiety that attaches itself to the baby’s breathing, the car seat, the stairs. Or a flatness where the feeling was supposed to be, which frightens people more than sadness would, because sadness at least resembles caring about something.

Many people also report a specific and lonely version of guilt: everyone keeps saying this is the happiest time of your life, and you can’t work out what’s wrong with you that it isn’t. If you’ve been through our piece on the signs of depression in adults, much of this will look familiar, because it’s the same condition presenting in circumstances that disguise it.

The performance is its own exhausting job. Answering “fine” at the six-week check, posting the photo, holding it together for visitors, then putting it down the moment the door closes. That pattern has a name and a research literature, and we’ve written about what it costs to keep functioning while depressed. New parenthood supplies an unusually good cover story for it, because everyone already expects you to look wrecked.

And the circumstances do disguise it. Exhaustion, appetite changes, disrupted sleep and low libido are all diagnostic features of depression and all completely ordinary after a birth. That overlap is a real problem for detection, and it’s part of why so much depends on what a clinician asks rather than what a checklist catches.

How common it is depends on what you’re counting

You’ll see “one in seven” almost everywhere. It’s worth knowing where that comes from, because two different numbers get used interchangeably and they measure different things.

Screening figures come from questionnaires. In 2018 the CDC found that 13.2% of mothers across 31 US sites reported postpartum depressive symptoms, though the range between states was wide: 9.7% in Illinois, 23.5% in Mississippi (Centers for Disease Control and Prevention, 2020).

Diagnostic figures come from clinical interviews, and they’re lower. A meta-analysis restricted to studies using diagnostic interviews found 12.1% for any depressive condition within the first postpartum year, with major depression at roughly one in 15 (Bai et al., 2023).

Neither number is wrong. A screening questionnaire is designed to over-identify, because the cost of missing someone is higher than the cost of a follow-up conversation. But it means “one in seven mothers have postpartum depression” is a slightly stronger claim than the evidence makes. If you’ve screened positive and are now reading the internet at 4am, that gap is worth holding onto: a positive screen is a reason to talk to someone, not a diagnosis you’ve already received.

The four-week window is a filing rule, not a description

Here’s something that surprises people. The DSM-5-TR doesn’t contain a diagnosis called postpartum depression. What it has is a specifier, “with peripartum onset”, applied to a major depressive episode that begins during pregnancy or within four weeks of delivery (American Psychiatric Association, 2022).

Two things follow, and both matter more than they sound.

The first is that roughly half of episodes labelled postpartum actually began before the birth. The manual uses “peripartum” rather than “postpartum” for exactly this reason.

The second is that the four-week boundary is contested within psychiatry itself. Plenty of people develop symptoms at four months or eight months, and researchers have argued the window should be extended, with proposals commonly landing around six months. Most epidemiological research ignores the four weeks entirely and uses the first postpartum year.

So if your symptoms started at five months and you’ve been quietly telling yourself it doesn’t count, that reasoning traces back to a filing convention that specialists are actively arguing about. Diagnostic thresholds exist to keep clinicians consistent with each other. They were never meant as a bar you have to clear before your difficulty counts as real.

The thoughts nobody warns you about

A very large proportion of new parents experience sudden, unwanted, vivid images of something terrible happening to the baby, and for many the thought involves harm they might do themselves. These are unwanted intrusive thoughts, and prospective research finds them normative rather than exceptional, peaking in the first eight weeks and mostly resolving by six months (Collardeau et al., 2024).

The finding that matters most: there is no association between having these thoughts and going on to harm your baby. The horror you feel is evidence that the thought runs against what you want, not a preview of it.

What is different about postpartum psychosis

Everything above describes thoughts you find horrifying. Postpartum psychosis is a different condition, and the distinction is the single most useful thing to understand here.

It affects roughly one to two women per 1,000 births. Onset is usually fast, typically within the first two weeks, and it involves a genuine break with reality: beliefs that are not true and cannot be reasoned with, hearing or seeing things that are not there, dramatic shifts in mood or energy, confusion, not sleeping for days without feeling tired.

The critical difference is not how frightening the thought is. It is whether the person recognises it as a thought. Intrusive thoughts feel alien and abhorrent to the person having them. In psychosis, an idea about the baby may feel true, reasonable, or even necessary, and the person may not be distressed by it at all.

Postpartum psychosis is a medical emergency, in the same category as chest pain rather than something to raise at the next appointment. It also responds well to prompt treatment, which is why speed matters so much. If someone you love has recently given birth and is not sleeping, seems confused, or is saying things that do not track, that is an emergency department today.

Why it happens, and the part nobody can tell you

Honest answer: there isn’t a single cause, and anyone offering you one is simplifying.

The hormonal story is the popular one. Oestrogen and progesterone fall sharply after delivery, and that shift is real. But it happens to everyone who gives birth, and most people don’t develop depression, so the drop alone doesn’t explain who does. Current thinking points towards differences in sensitivity to hormonal change rather than the size of the change itself.

Sleep deprivation is more than a background complaint. Fragmented sleep independently affects mood regulation, and new parents are experiencing it at an intensity that would be considered an intervention in a laboratory study.

Then there’s everything the biology framing leaves out. A previous episode of depression is among the strongest predictors. So are a difficult or frightening birth, feeding that isn’t working, financial strain, an absent or unhelpful partner, isolation, and having no one to hand the baby to.

That last cluster gets discussed less because it’s harder to fix than a chemical explanation. But it’s also the more hopeful half, because circumstances can change in ways that hormone receptors can’t.

What the evidence actually supports

The strongest evidence here is for talking therapies, and it’s stronger than most people expect.

The US Preventive Services Task Force reviewed the evidence and issued a grade B recommendation that clinicians provide or refer pregnant and postpartum people at increased risk to counselling interventions (US Preventive Services Task Force, 2019). Two approaches carried that recommendation: cognitive behavioural therapy and interpersonal therapy.

Note what’s being claimed. This is a recommendation about prevention. The evidence was strong enough to justify offering therapy to people at elevated risk before they became unwell, which is an unusually assertive position for a body that’s conservative by design.

Interpersonal therapy deserves a mention for fitting the situation so neatly. It focuses on role transitions, relationship disputes and social support, which is a fair description of what most new parents are dealing with underneath the symptoms.

Medication is often part of treatment and is a conversation for you and a qualified prescriber, including questions about feeding. That’s outside what this article can usefully cover, and anyone who tells you the answer without knowing your history is guessing.

Practical support belongs in the treatment list too. Someone taking the baby for four hours so you can sleep properly isn’t an alternative to treatment; for a lot of people it’s the thing that makes treatment possible.

Screening is a doorway, not a verdict

The questionnaire you were handed was probably the Edinburgh Postnatal Depression Scale, ten items, developed in 1987 and still the most widely used tool in the field (Cox, Holden, & Sagovsky, 1987). It’s deliberately built to avoid physical symptoms like fatigue and appetite change, precisely because those are unreliable indicators in the postpartum period.

Worth knowing about it: the last item asks about thoughts of harming yourself. Any endorsement of that item warrants a proper conversation regardless of the total score, including the passive version, the “everyone would be better off without me” version. It isn’t scored as one point among ten.

People also under-report on these forms, and not irrationally. If you believe an honest answer might trigger a referral you don’t understand, understating is a reasonable-feeling decision. But screening exists to open a door, and the person reading it is far more used to these answers than you’d guess.

When to get help sooner rather than at the next appointment

Some of this can wait for a scheduled appointment. Some can’t.

Today, not later: any thought of ending your life, however vague. Any belief about your baby that feels true rather than intrusive. Confusion, or not sleeping at all across several days. Being unable to care for yourself or the baby.

This week: symptoms that have persisted beyond two weeks and aren’t lifting. Anxiety that’s interfering with sleep even when the baby’s asleep. Avoiding being alone with the baby. Feeling detached from her.

Worth raising regardless: any of this at all. The bar for asking is much lower than people set it, and “I don’t think I’m bad enough to bother anyone” is one of the most common things clinicians in this field hear. If you’re weighing it up, our piece on the signs that therapy is worth considering is a reasonable place to start, though the postpartum year lowers the threshold rather than raising it.

The takeaway

The most useful thing research offers here isn’t a symptom list. It’s permission to stop treating your own experience as evidence of a character problem.

The thoughts that terrify you at 3am are, in the overwhelming majority of cases, a well-documented and self-resolving feature of a nervous system on high alert around something it has decided is precious. The flatness isn’t a verdict on whether you love your baby. The fact that your symptoms started at five months rather than three weeks says something about a classification system, not about you.

None of that means waiting it out. It means the thing you’d be describing if you told someone is ordinary to them, treatable, and far more common than the silence around it suggests.

Frequently asked questions

Sources

1. American Psychiatric Association. (2022). Diagnostic and Statistical Manual of Mental Disorders (5th ed., text revision). Peripartum onset specifier.
2. Bai, Y., Li, Q., Cheng, K. K., Caine, E. D., Tong, Y., Wu, X., & Gong, W. (2023). Prevalence of postpartum depression based on diagnostic interviews: A systematic review and meta-analysis. doi:10.1155/2023/8403222
3. Centers for Disease Control and Prevention. (2020). Vital Signs: Postpartum depressive symptoms and provider discussions about perinatal depression, United States, 2018. Morbidity and Mortality Weekly Report, 69(19).
4. Collardeau, F., et al. (2024). Prevalence and course of unwanted, intrusive thoughts of infant-related harm. Journal of Clinical Psychiatry, 85(3), 23m15145.
5. Cox, J. L., Holden, J. M., & Sagovsky, R. (1987). Detection of postnatal depression: Development of the 10-item Edinburgh Postnatal Depression Scale. British Journal of Psychiatry, 150, 782–786.
6. Raza, S. K., & Raza, S. Postpartum psychosis. StatPearls. StatPearls Publishing.
7. US Preventive Services Task Force. (2019). Interventions to prevent perinatal depression: US Preventive Services Task Force recommendation statement. JAMA, 321(6), 580–587.