How to keep a simple sleep diary

What it is and where the consensus diary comes from

Sleep diaries have been used for decades in insomnia research, but there was no common template: many groups used their own.¹ In 2012, a group of specialists published one, the Consensus Sleep Diary, after bringing together 16 different diaries and testing the questions with 47 people who said they slept well, had insomnia or had sleep apnea.¹

The core version has nine questions, the last of them a space for comments; the morning rows of our template are based on them. A week fits on one sheet, and it's filled in within an hour of getting up. An expanded version adds, among others, questions about naps, caffeine, alcohol and medication.¹

Several participants found the form cluttered, and the request that came up most was for a place to write down, in their own words, what had affected the night: the numbers alone didn't describe their experience well. Its authors presented it as a living document, designed mainly for insomnia research, that still had to be validated.¹

In Spanish, the patient information of the Sistema Nacional de Salud guideline on insomnia includes a sleep–wake diary template with space for two weeks: it's filled in every morning with what you remember of the night and every evening with what happened during the day, such as naps.⁷

A template to copy

Our template follows the questions of the consensus diary¹ and adds those of the NHLBI diary—naps, caffeine, alcohol, exercise and daytime sleepiness—⁶ plus the type of day. Copy it into a notebook or onto your phone, with one column per day.

If that seems like too many rows, start with the morning ones: they're the core of the consensus diary.¹ The evening ones are useful when you want to relate your sleep to something specific in your day, such as an afternoon coffee.

Time asleep isn't asked about directly: as in the consensus diary, it's calculated from the other answers.¹⁴ You count from when you tried to fall asleep until the final awakening and subtract how long it took you to fall asleep and how long you were awake. In the table's example, from 11:40 p.m. to 6:45 a.m. is 7 hours and 5 minutes; minus 20 and 15 minutes, that leaves about 6 and a half hours.

For how many nights

Sleep varies quite a lot from one night to the next, so a single night says little.¹ How many are needed? There are few studies, and they're small:

• In Sweden, 54 workers at the same correctional facility, with an average age of 45, kept a diary and wore a research actigraph for seven nights in a row. For the average sleep quality written down in the diary to be stable, at least six nights were needed; counting only the nights from Monday to Thursday, four were enough.²
• In Brazil, 53 university students with an average age of 25 wrote down every morning, for four weeks, what time they fell asleep and woke up: average sleep time reached acceptable stability by the seventh night, and since they slept longer on weekends, the authors recommend including them.³

Before an appointment, the US NHLBI suggests keeping it for one to two weeks,⁶ and the template in the Spanish guideline has space for fourteen days.⁷

Our inference from these sources: if you just want to know your pattern, a full week with its weekend is a good starting point, and two give a more stable picture. If you're using it to test a change, the test sets the length; we explain how to set it up in how to test a change without fooling yourself. And once you have your answer, you can stop.

How to read it without obsessing

These tips, except the first, which comes from the consensus diary, are our own suggestions, not tested in trials.

• Write it down roughly, and in the morning. The instructions of the consensus diary ask for your best estimate, not exact minutes, and say not to look at the clock during the night; if you forget one day, leave that day blank rather than reconstructing it.¹
• Look at the week, not the night. One bad night is normal; what's informative is the average and what keeps repeating.
• Separate workdays from days off. If your schedule changes a lot from one to the other, you'll see the difference in the diary.
• Treat what you see as a question. If the bad nights coincide with the afternoon coffee, that doesn't prove it's the cause: it's something you can test.
• If it stresses you out, stop. The diary is a tool for answering a question, not a daily obligation.

In this example, what's useful isn't Tuesday night but seeing that on weekdays this person sleeps close to an hour and a half less than on days off. What to try in that case we cover in sleeping differently on weekends, and the same diary will help you see whether you notice a difference. If your schedule is about to change—a trip, the clock change or a one-off early start—see how to adjust your sleep when your schedule changes.

Diary and devices: two different perspectives

A diary records how you perceived the night; a device you wear estimates it from other signals, such as movement. They don't measure the same thing, and there's no reason they should match.

In an at-home study in the United States, 323 women aged 48 to 57 used three methods at once, for up to three nights: a diary, a research actigraph—a wristband that records movement, different from consumer watches—and polysomnography, the reference test, which records brain activity. The diary gave more time asleep than the other two measures (on average, about 14 minutes more than polysomnography and about 23 more than the actigraph) and less time awake during the night. And the worse the night, the less the three methods agreed with one another.⁴ In another study, of 223 middle-aged and older adults in Pittsburgh, the sleep written down in the diary was also longer than the measured sleep; according to its authors, 20 to 30 minutes longer than with polysomnography.⁵

That doesn't make it a worse measure, just a different one: the authors of the consensus diary present it as a valuable perspective in its own right—how each person experiences their sleep—which doesn't have to match objective measures.¹ In the Swedish study, the quality written down in the diary was barely related to what the actigraph recorded, and the authors conclude that the two methods capture different dimensions of sleep.²

If you wear a watch or a ring, our suggestion is not to try to decide every morning which one is right: the diary tells you how you experienced the night, and the device gives an estimate with its own errors, which we explain in what watches and rings that track sleep get right.

When to take it to a professional

A diary can't tell you whether you have a sleep disorder, and we give no figures here that would point to one: that's a health professional's job. What it does let you do is arrive at your appointment with data from one or two weeks instead of an impression; the NHLBI suggests taking it along to go over it with your doctor.⁶

The UK's health service advises seeing a professional if you've been sleeping badly for months, if it affects your daily life or if changing your habits hasn't helped.⁸ In Spain, the first step is your primary care center (centro de salud);⁷ if they give you their own diary template there, use that one. It's the same criterion we set out, along with the rest of our sleep guidance, in when habits aren't enough.