Sleeping When You Are On Call: Why the Chance to Sleep Does Not Fix It

Being on call is a state rather than a task. That is why the chance to sleep does not repair it: the vigilance runs whether or not anything happens, so a carer handed a free night will frequently lie awake through the whole of it, waiting for a sound that is not coming. Almost every piece of advice written about carers and sleep treats this as insomnia and offers insomnia’s answers. It is not insomnia. It is a job that has no end of shift, and it responds to a different set of moves.

The scale of it is measurable at both ends. Among community-dwelling US adults aged 45 and over, 36.7 percent of carers got insufficient sleep, defined as under 7 hours in 24, using BRFSS data from 2015 to 2017 across 44 states plus the District of Columbia and Puerto Rico (CDC with the National Association of Chronic Disease Directors, published 2018)¹. Among carers responding to Carers UK’s survey run between June and August 2025, 65 percent found it hard to get a good night’s sleep (State of Caring 2025: The cost of caring, published October 2025, 10,539 respondents)². That survey is self-selecting and Carers UK says so plainly: the sample was 82 percent female, 90 percent White British and 61 percent caring 50 or more hours a week, so read it as carers who responded rather than as carers in general².

I have four years of this. My mother came home twelve days after coronary bypass surgery in April 2022, and for the first fortnight I did not sleep so much as monitor. The bedroom door stayed open, the landing light stayed on, and I learned the exact noise the airing cupboard makes when the heating comes on, because I sat up for it three times a night. In February 2023 she went down in the kitchen at six in the morning, and after that the listening got worse rather than better, because now I had evidence. When two paid carers finally started coming four mornings a week in January 2025, about a year later than I should have accepted them, the mornings improved immediately and the nights did not change at all. That gap is the subject of this page.

This article sits under caring for a parent at home and belongs to the same cluster as carer burnout and your own health, which sets out what sustained caring does and, just as importantly, what the evidence does not support.

Why does the chance to sleep not fix it?

Because you are not tired in the way a short night makes you tired. You are running a monitoring process, and the process has no off switch that somebody else can flip for you.

Three things keep it running. The listening is trained. Months of waking to a specific noise turns the noise into a category your brain now checks for continuously, and one covered night does not untrain it. The responsibility does not transfer with the room. Even when another person is physically in the house, you are usually still the one who knows what a bad night looks like for this particular person, which means you are still adjudicating. And the anticipation is worse than the event. Most carers I have listened to describe waking before anything happens, at the hour it usually happens, on nights when it does not happen at all.

This is why the first night of respite so often disappoints, and why carers conclude from it that they are beyond help. They are not. The state takes several consecutive nights to stand down, not one, and a single covered night is a test of the wrong thing. If you are planning a break around this, plan more than one night: respite care explained covers the forms it takes and how it is funded, and it is honest about what the trial evidence does and does not show.

Intensity is the variable that decides how bad this gets. Census 2021 counted 5.0 million unpaid carers in England and Wales, of whom 1.5 million were caring 50 or more hours a week, with 260,000 more people providing 20 to 49 hours a week than in 2011 (ONS, published January 2023)³. In the United States the average is 27 hours a week, and 24 percent of carers provide 40 or more hours (National Alliance for Caregiving and AARP, Caregiving in the US 2025, published July 2025)⁴. Nobody counts night hours separately, which is precisely why they get absorbed.

What is actually waking you?

Six things, and it is worth knowing which one you are dealing with, because they have completely different solutions.

• The listening itself. Nothing has happened. You woke to check. This is the largest category and the one no equipment solves.
• The continence round. A planned or unplanned change, once or several times a night. Practical, and often improvable: see continence and keeping dignity.
• The person getting up. Walking to the bathroom, walking for no clear reason, or trying to and not managing. This is a falls question as much as a sleep question, and it is covered in wandering and keeping someone safe.
• Pain, or distress that reads as agitation. Frequently the real cause of a bad night in somebody who cannot say what hurts.
• Calls, and the person calling out. Sometimes for a reason, sometimes as part of a late-day pattern covered in sundowning and difficult nights.
• Your own mind, at four in the morning, running the list. Appointments, forms, the thing you did not chase.

Write down which of the six woke you, for a fortnight, with times. It takes ten seconds a night and it is the single most useful thing you can hand a professional, because “she is up a lot” and “she was up at 1.10, 3.40 and 5.15, and on three of the fourteen nights she was confused about where she was” produce entirely different responses. Keeping a care file is where that record lives.

What should be ruled out before you treat this as a sleep problem?

A new or worsening night disturbance in the person you look after is a change to report, not a fact of the household. This is the part of the page that matters most, because the households where the nights suddenly got worse are usually the households where something is wrong.

New confusion is the first thing to act on. NICE describes delirium as a clinical syndrome characterised by disturbed consciousness, cognitive function or perception, with an acute onset and fluctuating course, which usually develops over 1 to 2 days, and states plainly that “it can be prevented and treated if dealt with urgently” (guideline CG103, published July 2010, substantively updated 18 January 2023)⁵. Fluctuation is the giveaway, and it very often looks worse in the late afternoon and at night, which is exactly when a carer is least equipped to interpret it. The quiet form is the one that gets missed, and NICE tells clinicians to be “particularly vigilant for changes that may indicate hypoactive delirium, which are often missed, such as withdrawal, slow responses, reduced mobility and movement, worsened concentration and reduced appetite”⁵. Delirium or dementia sets out the distinction in full.

Then the ordinary causes of a broken night that are treatable and get mistaken for old age:

• Pain, including pain that shows only as restlessness, resistance to being moved, or calling out.
• A sudden change in continence, which is a medical sign rather than a laundry problem.
• Breathlessness lying flat, or needing more pillows than last month.
• Needing to pass urine repeatedly, or not passing urine at all.
• A fever, or being newly hot and clammy at night.
• Anything that follows a fall, since around a third of people aged 65 and over, and around a half of people aged 80 and over, fall at least once a year, and a fall is a symptom with a list of causes behind it (NICE guideline NG249, published 29 April 2025)⁶.
• A new or changed medicine. No page on this site gives a dose or tells you to give, stop or change anything, and that is the prescriber’s and the pharmacist’s job. What you can do is ask for the whole list to be reviewed: managing someone else’s medication covers how.

Spotting when someone is getting worse is the general method behind all of that. The rule of thumb is simple: a night that has changed shape in the last week or two is information about the person, and a night that has been like this for two years is a resourcing problem.

Which arrangements actually recover sleep?

The ones that transfer responsibility for a defined period to a named person. Everything else is a marginal gain on top of that, and most published advice consists entirely of the marginal gains.

Ranked roughly by how much sleep they return:

1. A paid waking night or a night sitter. The only intervention that reliably returns a whole night, because it removes both the tasks and the adjudication. It is also the most expensive thing in home care by a distance, which is why it tends to be used in crisis rather than in maintenance. Getting help in covers how each kind of help is triggered, and choosing a home care agency is the vetting method, since this site names no provider.
2. A second person on alternate nights. A partner, an adult child, a sibling. The gain is not half the nights; it is the two or three consecutive nights it becomes possible to string together, which is when the vigilance actually stands down. If there is a sibling at a distance, this is one of the few genuinely portable things they can own: see sharing care with siblings.
3. An agreed rule about what does and does not warrant getting up. Written, and agreed with a professional rather than invented by you at three in the morning. “Calling out once and settling does not need a response; calling out twice does” is a rule. “I will see how it goes” is not, and it costs a whole night either way.
4. A monitor set to a threshold rather than an open channel. An open audio feed makes the night worse: it delivers every rustle and converts sleep into continuous monitoring. A device that alerts on a defined event, such as somebody leaving the bed, is a different proposition. Adaptations and technology at home covers what this equipment realistically does and where it stops, including the consent and dignity questions that come with putting a sensor in another adult’s bedroom.
5. Moving where people sleep. Downstairs, into a room with a clear path to a bathroom, or into separate rooms if you are caring for a partner. This is a bigger intervention than it sounds and it often unlocks the others: see caring for a partner, not a parent for what changes when there is no shift change at all.
6. A continence plan built with the community nursing service, rather than a laundry routine built by you. Timing, products and skin care are all decisions somebody else is qualified to make.

Two things that are usually offered and rarely work on their own: general sleep hygiene advice aimed at people who are not listening for somebody else, and a suggestion that you nap in the day, which assumes a day with a gap in it.

What about the jobs that are supposed to happen at night?

This is where the honest answer is uncomfortable. Some care plans require things to happen overnight, and the plan usually assumes a carer who is awake.

Repositioning is the clearest example, and the guidance on it genuinely disagrees with itself. NICE CG179 sets minimum frequencies of at least every 6 hours for adults at risk and at least every 4 hours for adults at high risk (recommendations 1.1.8 and 1.1.9, published 23 April 2014 and never substantively revised)⁷. The fourth edition of the international pressure injury guideline, released chapter by chapter from 2025, suggests two hourly or three hourly intervals for most people at risk who are on an appropriate pressure redistribution full body support surface, and separately suggests not routinely extending intervals to four, five or six hourly, both as conditional recommendations resting on very low certainty evidence⁸. A carer being asked to turn somebody every two hours through the night is being asked to do something no single person can sustain for months.

Say so. The useful sentence to a district nurse is not “we are managing”, it is “the plan says every three hours overnight and I can do it for about a week”. That sentence gets the plan rewritten, or gets equipment, or gets a night service considered. Reporting that it is going well gets nothing, and it is the most common thing carers do. Preventing pressure sores covers the full picture, including why non-blanching redness is a call rather than a watch.

What do you say to your own doctor about this?

Tell them you are an unpaid carer, and ask for it to be recorded on your notes. It takes about ninety seconds and in several systems it changes what is offered to you.

Then describe the sleep the way you would describe somebody else’s symptom. Not “I am tired”, which every doctor hears twenty times a day, but: how many nights in the last fourteen were broken, how many times a night, at what times, what wakes you, whether you fall asleep and cannot stay asleep or cannot get to sleep at all, and what happens on the nights when somebody else covers. That last one is diagnostically useful and almost nobody offers it.

Be direct about the consequences too, including whether you are driving tired, whether you are making errors with somebody else’s medicines, and whether your mood has changed. Carer mental distress is measurable at population level: 20.5 percent of US carers reported frequent mental distress, meaning 14 or more bad mental health days in the past month, in 2021 to 2022, against 13.6 percent of non-carers, and 19.6 percent rated their own health as fair or poor against 16.9 percent (CDC, MMWR 2024, BRFSS caregiver module, 35 states and Puerto Rico, 92,461 carers against 353,242 non-carers)⁹. One caution, because the figure is often misused: sleep is not measured in that report, so no sleep figure attaches to it. The 36.7 percent insufficient sleep number comes from a different, older and narrower dataset, restricted to adults aged 45 and over in 2015 to 2017¹.

On the wider question of what sustained caring does to a carer’s health, the defensible summary is consistent evidence of worse mental health and worse morbidity among strained carers, and no good evidence that caring shortens your life. Carer burnout and your own health sets out both sides of that in full, including where the popular claims overstate it. And if the appointment itself is the obstacle, why carers skip their own appointments is about the cover problem rather than the motivation problem.

What the evidence says about the fix, which is less than you will be told

Be careful with what you are promised here. NICE guideline NG150 recommends that practitioners regularly discuss the value of a break with carers and explain the options, and that breaks should suit the carer in duration, timing, frequency and type and be arranged reliably. Its own evidence review states: “There was a lack of evidence on effectiveness or cost effectiveness to support a recommendation about the circumstances in which carers’ breaks should or shouldn’t be offered”, adding that “the evidence showed that carers’ breaks were often limited in nature, availability, quality and flexibility”¹⁰. The Cochrane review of respite care for people with dementia and their carers, four trials and 753 participants at very low quality, concluded that “current evidence does not demonstrate any benefits or adverse effects from the use of respite care for people with dementia or their caregivers”, while noting this may reflect a lack of good research rather than a lack of benefit¹¹.

Arrange the break regardless, and arrange it as consecutive nights rather than one. Among carers responding to Carers UK’s survey run June to August 2024, 65 percent felt overwhelmed because they had not been able to take a break and 49 percent said they needed more breaks or time off (State of Caring: mental health and social care, published February 2025)¹². What the thin evidence should change is your expectations, not your plan. If the first covered night produces no sleep, that is the state standing down slowly, not the arrangement failing.

This is a call today, not something to see how it goes

For the person you look after. Ring the same day, and say what has changed and when, if they:

• Are more confused than they were yesterday, or the confusion comes and goes through the day. Use the word confusion, and say what they were like last week⁵.
• Have become withdrawn, slow, drowsy or have stopped finishing meals, which is the quiet form of delirium and the one most often missed⁵.
• Have new or worsening pain, or new agitation at night in somebody who cannot tell you what hurts.
• Have had a fall, even one they got up from⁶.
• Have a sudden change in continence, a new fever, or have stopped drinking.
• Are newly breathless lying flat, or are needing more pillows than they did a month ago.

Who to ring: in the United Kingdom, their GP surgery, or NHS 111 out of hours. In Ireland, their GP or the out of hours GP service. In the United States, their own doctor’s office, or the nurse line their plan runs, or an urgent care center. In Australia, their GP, or healthdirect on 1800 022 222. In Canada, their family doctor or the provincial health line, 811 in most provinces.

Use the emergency number immediately for changed breathing, chest pain, one-sided weakness, changed speech, a first seizure, an unrousable person, or bleeding that will not stop: 999 or 112 in the United Kingdom and Ireland, 911 in the United States and Canada, 000 in Australia, 112 across the European Union.

For you. If you are falling asleep at the wheel, making errors with somebody else’s medicines, or if your own mood, your drinking, or thoughts of harming yourself or the person you care for are why you are awake, that is your own doctor or an urgent mental health line where you live, today. Not after the next respite booking. Say the words “I am an unpaid carer and I am not sleeping”, because in several systems that sentence changes what is offered.

The most dangerous thing about broken sleep in this job is how normal it becomes. Four years in, the question I wish somebody had asked me in 2022 is not “are you sleeping”, which I would have answered yes to out of habit. It is “how many of the last fourteen nights were unbroken”, which I could not have answered at all.