Prescribing Sleep: An Overlooked Treatment for Postpartum Depression

In 1892, Freud described a case of a woman who presented with psychiatric symptoms only after giving birth, a “hysterique d’occasion.” Upon delivery, she developed loss of appetite, vomiting, pain with nursing, agitation, and insomnia. Ultimately, she had to send her child to a wet nurse. Two years later, she became pregnant again. Remarkably, the same symptoms recurred after delivery, threatening a similar separation from her baby. Freud’s prescription? She should be more assertive and “break out against her family with some acrimony.” Under hypnosis, he told her to ask, “What had happened to her dinner? Did they mean to let her starve? How could she feed the baby if she had nothing to eat herself?” She complied, “and…remonstrated with her mother in a way quite unlike herself.” The symptoms resolved and Freud claimed victory. It is a testament to his influence that, for the next hundred years, postpartum depression (PPD) was seen as a conflict arising over a mother’s unmet needs. Yet little progress was made in treatment, as clinicians focused not on meeting mothers’ needs but on persuading them to go away.

It is only recently, over the last 20 to 30 years, that biological explanations of PPD have gained traction. Decades of careful studies highlighted some women’s sensitivity to neurohormonal fluctuations, culminating in 2019, when the FDA approved the first drug specifically for PPD. Brexanolone, an allosteric modulator of the GABA_A receptor, was a seeming miracle cure. Its clinical trials enrolled postpartum women¹ with moderate to severe depression within 6 months of delivery. They spent 3 days and nights receiving an inpatient IV infusion and stopped breastfeeding for 7 days. After 60 hours, women’s scores on the Hamilton Depression Rating Scale (HAM-D) dropped an average of 17 points (with most of the improvement in the first 12 hours), bringing their depression to the level of mild or none (1). Their illness had melted away.

While the results were impressive, there may be more to the story. Women in the placebo arm also experienced a marked decrease in their HAM-D scores, by an average of 13 points, bringing those with moderate to severe depression into the mild range—again, almost dissolving depression in a mere 2.5 days. Moreover, of those women considered placebo responders, 91% had not relapsed 30 days later (1). What could account for this rapid and durable response in a disease that typically inflicts months of suffering with conventional treatments? Could there be another biological system beyond GABA, targeted by brexanolone and placebo, causing both groups to improve dramatically? One ready hypothesis is that there was a second active intervention taking place: a complex biopsychosocial approach to restoring women’s sleep.

There is a robust literature on the toxic impact of acute sleep deprivation. Keeping someone awake for 19 hours impairs their speed, accuracy, and hand-eye coordination on tasks to the same level as a blood alcohol concentration of 0.1 (0.08 being enough to get you a DUI [driving under the influence] in most states) (2). Not sleeping can dramatically lower resilience and affect decision-making capacity, a trick used by those in power across history—from the Spanish Inquisition of the 1400s to modern-day policing—to break down psychological defenses and extract false confessions. This effect is neatly illustrated by a 2016 experiment, in which the number of college students who signed a false confession after being kept awake just 8 hours overnight jumped from 18% to 50% (3).

The data are even more dire for chronic sleep deprivation. A review of 64 studies summarizing the effects of shift work and sleep disruption found that night shifts and short sleep (defined as 4-7 hours a night) are particularly noxious and linked to increases in cardiovascular disease, diabetes, stroke, and cancer (4). Frighteningly, a 2019 study of medical residents found that long work hours during their intern year advanced cellular aging in a dose-dependent relationship through increased telomere length loss. Those who averaged 75 hours per week had their telomeres shorten by 758 base pairs, a loss equivalent to 31 years of life (5).

Shift work and intern year, however, are no comparison for the disruption experienced by new mothers. Postpartum women, waking frequently to care for their infants, suffer a mind-bending cocktail that combines acute and chronic sleep deprivation during the exact window of time when vulnerability to hormonal fluctuations puts women at the greatest risk of depression. So why haven’t we explored restoring maternal sleep as a potential treatment?

Most studies that have tried to move the dial on maternal sleep have done so indirectly: by getting infants to fall asleep more quickly and independently, and to wake less frequently. In other words, let’s change the baby. The results of this approach have been strikingly unimpressive, with one review baldly entitled “Behavioral sleep interventions in the first six months of life do not improve outcomes for mothers or infants” (6). Study after study has replicated the same findings: prior to 6 months, babies are reliably resistant to changing their sleep.

The other obvious approach, rather than changing the baby, would be to increase maternal sleep directly. But just telling a mother to sleep is as ridiculous as telling her to fly. Protecting her sleep requires challenging deep cultural and structural factors, both within families and within the medical establishment. Few have looked at this. One recent study looking at paternal involvement noted its groundbreaking status—in 2015—as “the first to investigate the triadic links between infant sleep, maternal sleep, and paternal involvement in infant care” (7). Enticingly, they found that greater paternal involvement in infant care predicted greater sleep consolidation for both mothers and infants at 6 months.²

To overcome these structural factors requires the buy-in of multiple stakeholders and takes time. There is no shortcut. There are at least 4 core principles to prescribing sleep (Figure 1): 1) Change the message so that women see self-care rather than self-sacrifice as the hallmark of a good mother; 2) Consolidate sleep and educate about the consequences of sleep deprivation; 3) Recruit help (to the extent possible) because newborn night feedings are best understood as a job for more than 1 person; and 4) Encourage a flexible approach to breastfeeding.

Figure 1.

Sleep deprivation is bad for the brain. Prescribing sleep for new moms is a powerful intervention that involves 4 core principles. 1) Change the message. Help mothers view protecting their own sleep as crucial to supporting the health of the whole family. 2) Consolidate sleep. Larger blocks of sleep have a more powerful impact than multiple smaller ones. 3) Recruit help (to the extent possible). Taking shifts may involve flexible sleeping arrangements, e.g., one person with the baby and one person in a quiet separate room, swapping places in the middle of the night. The person sleeping first should go to bed as early as possible (forget the dishes!). 4) Flex the breast. Women can add additional pumping sessions during the day to allow them to gradually space out night-time pumping and involve a second person in bottle feeding at night.

What might widespread adoption of these changes look like? The closest example comes from Australia and the research of pediatrician Harriet Hiscock, who has been studying infant sleep patterns and their relation to maternal depression for decades. She has built an infrastructure around supporting postpartum sleep that is now widely available across the country from the moment of delivery. The education is extensive: nurses who train families on sleep strategies first undergo training themselves. The focus is reading infant signals (rather than changing infants) and on the family as a system. The public resources invested are significant. When outpatient services are not enough, additional options are covered by insurance such as home visiting, day programs, and even residential sleep schools where entire families can stay for up to 4 nights while infant sleep experts observe and offer suggestions and support.

Early data are impressive. Hiscock’s approach shows that targeting the whole family is cost-effective, is scalable, and reduces maternal depression symptoms up to 2 years later (8). Additionally, providing parental guidance around sleep before delivery may have the power to prevent PPD or reduce symptoms.

What would it take to implement a comprehensive, integrative approach like this in the U.S.? For some families, a simple conversation could spur them to reallocate resources to protect maternal sleep. But given the vast socioeconomic and health care disparities, for most families this will not be an option. Yet the societal cost of not treating is high: 14.5% of women experience a new episode of depression in the first 3 months after delivery, at a price of $32,000 per untreated mother–child pair, with total costs reaching into the billions (9,10). Brexanolone, even at $34,000 a treatment, presents significant savings when compared with an inpatient psychiatric admission or maternal suicide; psychosocial interventions such as night doulas or paid family leave are likely still cheaper. Foundational research on the cost and efficacy of prescribing sleep can set the stage for political action.

A hundred years ago, Freud was on to something—not with hypnosis, but with how he framed the issue. The family’s failure to respond to a new mother’s needs was a serious problem. The effective treatment of PPD requires us to call in the cavalry—fathers, partners, family members, anyone who can help protect moms’ sleep—and prioritize women’s health as a family and societal necessity.