The nights are hard

Diana Chirinos is a psychologist at Northwestern University Feinberg School of Medicine who studies insomnia in widowed people. Her findings show that those who have the greatest rumination – the thoughts that go round and round in one's head – also have the greatest difficulty sleeping.² When we have had significant loss, we often dread going to bed, because with nothing to distract us, our thoughts turn to grief. For widowed people, the bed is usually a strong reminder of loss. One of Chirinos's participants commented, "I don't like sleeping in my bed . . . It's funny how you form habits – I'm still on the edge of the bed, the whole bed is empty. I could spread out, but I don't."

Chirinos is also studying how to treat insomnia in bereaved people, and part of her motivation is that research shows sleep medications are not effective. For example, in a randomized study in London, bereaved people were prescribed twenty pills of either low-dose diazepam or a placebo, provided within two weeks of the death of their partner and to be taken as needed across the following six weeks. This mirrors the way sleep medications are often given to bereaved people by general practitioners. From baseline to the end of the six weeks, more people in the diazepam group had trouble getting to sleep in thirty minutes or less after going to bed, but fewer people in the placebo group had trouble getting to sleep.³ This seems counterintuitive. After all, why would sleep medication not help improve their sleep? The answer may be related to what is called the 3P model of insomnia.

This 3P model is a general explanation of how sleep problems can arise, and not just during bereavement. It focuses on three conditions that must come together for us to develop insomnia: predisposing, precipitating, and perpetuating factors.⁴ The first p is for predisposing factors that increase the risk of developing insomnia, and in the case of bereavement, these are the factors present before the death of a loved one. They include genetically determined factors, personality traits, and environmental situations that are already in place when a loss happens. For example, a long period of caregiving prior to a loss is a predisposing factor that can in- crease insomnia during bereavement. The second p is for a precipitating event, in this case the death event. Loss often triggers initial sleep difficulty that initiates short-term insomnia. It is well documented that most of us experience initial sleep difficulties in bereavement. In a systematic review, eighty-five different published studies described these grief-related sleep problems.⁵ Most important is the third p, because these are the perpetuating factors. These are the things that maintain insomnia once we have begun having difficulty sleeping because of our loss.

Perpetuating factors are also things that we may have some control over, unlike the things that happened before the loss or the fact that our loved one has died. Essentially, some of the things we do to cope with common, expected sleep problems during bereavement may perpetuate the insomnia, turning it into a long-term problem. For most people who have sleep difficulties during bereavement, normal sleep patterns will return. Only a couple of studies examine the natural course of sleep difficulty in bereavement, but one Dutch study followed recent widows over time. The widows showed significantly more sleep difficulties at four months than a nonbereaved group of women. Overall, the widows returned to their baseline sleep patterns after a year, although their sleep was also significantly better by seven and ten months after their loss.⁶

So, what coping strategies for insomnia backfire, perpetuating our short-term sleep difficulties and turning them into long-term insomnia? Now we can unravel the mystery of why prescribed diazepam did not help insomnia in the study of bereaved people I described earlier.

The physiological co-regulation we share with our bonded loved one means that when our loved one dies, we must learn to regulate our system without this relationship, previously a part of our own functioning. Learning to live with their absence includes adjusting our sleep system to go to sleep, to stay asleep, and to wake up without them there, or even simply without the comforting knowledge they are somewhere in the world. Grieving is partly learning to regulate our physiology without the input of their warmth, their smell, their touch, or their texting us good night. Human beings are resilient, and our sleep system will adjust, even though it takes time and is frustrating and we are often tired a lot. But if we introduce medication as a part of that adjustment, our system is adjusting to the loss by adopting those chemicals as a part of our sleep system. This means that we are regulating with the input of those medications. We will still need to adjust to functioning without them, eventually. If we are taking the pills intermittently, as in the London study, our sleep system has difficulty knowing what to expect and how to adjust on the nights we do not take them. This is why sleep medication can ironically be a factor perpetuating insomnia.

Introducing medication or other substances to help us get to sleep, or to keep us alert during the day, are not the only perpetuating factors that can turn insomnia into a long-term problem. Particular beliefs about sleep and specific behaviors have been proven to maintain insomnia. These include using the bed for anything other than sleep or sex, like watching TV or using a smartphone. We want to condition ourselves to the idea that getting into bed means sleep will follow. Similarly, spending a long time in bed unsuccessfully trying to get to sleep or to return to sleep weakens the bed-means-sleep conditioning. If you have insomnia, and you are taking more than fifteen minutes to get to sleep (or back to sleep), the recommendation is to get up and do something quiet in another place until you feel sleepy again (and preferably without a lit screen in front of you). Have you ever lain in bed, so annoyed by the fact that you cannot sleep that it is impossible to relax? Some people become more upset and anxious about their insomnia over time and develop unrealistic expectations about sleep, such as the idea that unbroken sleep throughout the night is required to feel rested, or they underestimate the amount of time they are asleep, or they believe they will not be able to work the next day if they are tired. Having these thoughts during the night paradoxically make it harder to fall asleep, as we get more frustrated because of the outcome we anticipate the following day.

Most important as we learn to reregulate our sleep system is to keep a consistent schedule. Getting up in the morning at the same time, even when we are tired or had a poor night, is key. Getting exercise or movement in your day (but not right before bed) and refraining from daytime naps (including falling asleep in front of the TV in the evening) are vital to our sleep system. Although I usually take a nap every day, if I am experiencing a bout of insomnia or when I have jet lag, I skip my nap to help consolidate my sleep to nighttime hours.

But the most important thing I can tell you is that cognitive behavioral therapy for insomnia (CBT-I) is a proven effective treatment. In head-to-head comparisons, CBT-I is effective and does not interfere with long-term sleep function like sleep medication can. CBT-I is available (even as teletherapy) from board-certified specialists.⁷ You may read a list of sleep problems like the one in the previous paragraph and believe you already know these things, but it is very different to have a specialist identify your particular perpetuating factors and explain which techniques will ameliorate them. There are mathematical formulas for how long to spend in bed given your specific sleep patterns, and the best half hour of the day to nap if you feel it is necessary. These formulas, techniques, and tailored interventions can be gained only through seeing a professionally trained clinician. I cannot encourage you enough to get behavioral treatment for insomnia if you are suffering from it – it can be a turning point in healing after loss.

Benefits of Treating Insomnia Beyond Grief Therapy 

There is an additional reason to seek treatment for insomnia, even if you already have a grief support group or are in therapy related to your grief. First, in a large study I did with Dutch colleagues Maud de Feijter and Annemarie Luik, and Mayo Clinic's Brian Arizmendi, we discovered that having insomnia prior to the death of a loved one increased the chances of developing prolonged grief disorder.⁸ So not only can grief cause insomnia but already having insomnia can also make grieving harder. Second, several studies have now documented that even after people with prolonged grief disorder have successfully completed psychotherapeutic treatment, poor sleep may persist.⁹ That is, even though feelings of yearning have decreased, and new skills for coping with waves of grief have developed, insomnia can remain an issue. The good news is that following CBT-I treatment, 100 percent of bereaved people said they would recommend this brief intervention for guidance in how to restructure their sleep–wake schedule.¹⁰

Sleep abnormalities have health risks of their own, even unrelated to medical consequences of bereavement. Surprisingly, although sleeping too little on average (less than five hours) is associated with poor health, sleeping too much (eight and a half hours or more) is also associated with poor health. One way researchers assess good health is to consider the combined health of many physiological systems together, including cardiovascular, immune, lipid-metabolic, glucose-metabolic, sympathetic, parasympathetic, and hypothalamic-pituitary-adrenal systems. This is sometimes called the multisystem biological risk index. Colleagues from UCLA, including Michael Irwin, Jude Carroll, and Teresa Seeman, have shown both sleeping too few hours on average and sleeping too many hours on average are associated with poorer multisystem functioning of the body.¹¹ Data for this finding comes from a longitudinal study of over one thousand participants, called the Midlife Development in the United States (MIDUS) study.

While too many and too few hours of sleep reported in the study were both associated with poorer physiological regulation, the number of hours we sleep is distinct from sleep quality. Sleep quality is how well we feel we have slept, a subjective sense of how restorative one's sleep was. Compared to a normal sleep duration (between six and a half and eight and a half hours), all short sleepers reported poor sleep quality. But in long sleepers, it was only those who reported poor sleep quality who showed multisystem biological risk. For example, I would be considered a long sleeper – and I have been a long sleeper all my life. I do best when I get about nine hours of sleep. This is not necessarily a bad thing for my health, because subjectively I feel all my hours of sleep are of good quality. However, if you are in bed for eight and a half or more hours but are not getting good sleep during this time, this may be a health issue and worth your seeking treatment.

• Extracted from The Grieving Body by Mary-Frances O'Connor, PhD and reprinted with permission from HarperOne, an imprint of HarperCollins Publishers. Copyright 2025.

References

2. D. A. Chirinos (January 29, 2022). "Sleep Disturbances in Spousal Bereavement: Developing a Targeted Intervention," Neurobiology of Grief International Network (NOGIN) Conference, Tucson, AZ.

3. J. Warner, C. Metcalfe, and M. King, "Evaluating the Use of Benzodiazepines Following Recent Bereavement," British Journal of Psychiatry 178, no. 1 (2001): 36–41, doi.org/10.1192/bjp.178.1.36.

4. A. J. Spielman, L. S. Caruso, and P. B. Glovinsky, "A Behavioral Perspective on Insomnia Treatment," Psychiatric Clinics of North America 10, no. 4 (1987): 541–53, doi.org/10.1016/S0193–953X(18)30532-X.

5. M. Lancel, M. Stroebe, and M. C. Eisma, "Sleep Disturbances in Bereavement: A Systematic Review," Sleep Medicine Reviews 53 (2020): 101331, doi.org/10.1016/j.smrv.2020.101331.

6. E. E. Beem et al., "Psychological Functioning of Recently Bereaved, Middle-Aged Women: The First 13 Months," Psychological Reports 87, no. 1 (2000): 243–54, doi.org/10.2466/pr0.2000.87.1.243.

7. Society of Behavioral Sleep Medicine, "United States" (member map), accessed July 15, 2024.

8. M. de Feijter et al., "The Longitudinal Association of Actigraphy- Estimated Sleep with Grief in Middle-Aged and Elderly Persons," Journal of Psychiatric Research 137 (2021): 66–72, doi.org/10.1016/j.jpsychires.2021.02.042.

9. A. Germain et al., "Treating Complicated Grief: Effects on Sleep Quality," Behavioral Sleep Medicine 4, no. 3 (2006): 152–63, doi.org/10.1207/s15402010bsm0403_2; and P. A. Boelen and J. Lancee, "Sleep Difficulties Are Correlated with Emotional Problems Following Loss and Residual Symptoms of Effective Prolonged Grief Disorder Treatment," Depression Research and Treatment (2013): 739804, doi.org/10.1155/2013/739804.

10. P. A. Carter, S. Q. Mikan, and C. Simpson, "A Feasibility Study of a Two-Session Home-Based Cognitive Behavioral Therapy–Insomnia Intervention for Bereaved Family Caregivers," Palliative & Supportive Care 7, no. 2 (2009): 197–206, doi.org/10.1017/S147895150900025X.

11. J. E. Carroll et al., "Sleep and Multisystem Biological Risk: A Population-based Study," PLoS One 10, no. 2 (2015): e0118467, doi.org/10.1371/journal.pone.0118467.