A seven-day sleep diary: wake times, caffeine and daytime sleepiness

Record estimated time asleep alongside daytime sleepiness, rather than trying to improve an app score. Seven days can reveal a pattern, but cannot diagnose a sleep disorder or prove an increase in REM sleep.

Eight hours in bed is not necessarily eight hours asleep. Subtract the time spent falling asleep and the time awake during the night. This guide helps a general adult use a paper diary or phone note to calculate sleep time and choose one habit to review next week.

What should you record to estimate actual sleep?

Separate bedtime, estimated sleep onset, total time awake overnight and final waking time. Label uncertain values as estimates. Do not repeatedly check the clock overnight just to collect data. Use the same approach each morning and record what you remember.

Suppose you get into bed at 23:30, fall asleep at 00:00, wake at 07:00 and spend 20 minutes awake during the night. Time in bed is 450 minutes. Estimated sleep is 420 minus 20, or 400 minutes: 6 hours 40 minutes. The ratio 400÷450×100 is about 88.9%, but it does not establish whether you have a disorder.

The 450-minute calculation above assumes that you leave bed immediately after your final awakening at 07:00. Waking and leaving bed can happen at different times. Following the Consensus Sleep Diary research, record when you enter bed, start trying to sleep, finally wake, and leave bed to start the day separately. For example, entering bed at 23:30, trying to sleep at 23:40 and falling asleep at 00:00 gives a 20-minute delay after trying. With final waking at 07:00, leaving bed at 07:30 and 20 minutes awake overnight, estimated sleep remains 400 minutes. Here we define the denominator as 23:30–07:30, or 480 minutes, giving 400÷480×100≈83.3%. This simple example assumes no time out of bed overnight. If another tool starts counting when you try to sleep, check its definition before comparing ratios. Do not shorten sleep or prescribe treatment for yourself to improve this number.

This table follows the second hypothetical example: estimated nighttime sleep is 400 minutes and time in bed is 480 minutes, giving about 83.3%. Do not mix it with the first example’s 450 minutes and 88.9%.

How do you calculate and compare seven days?

Convert each night's sleep to minutes before adding. The illustrative values 400, 430, 390, 420, 410, 480 and 450 total 2,980 minutes. Dividing by seven gives approximately 426 minutes, or 7 hours 6 minutes per night. That average can conceal short weekday nights behind longer weekend sleep. Also inspect the shortest night and differences in wake times.

CDC recommends at least seven hours for adults aged 18–60 and gives different recommendations for other ages. Repeated sleepiness or disrupted sleep despite enough time available deserves attention beyond the total. See CDC's sleep guidance and diary fields.

Record your usual pattern during the first seven days. During the next seven, choose one change, such as moving the last caffeinated drink earlier. Note workload, illness, medication and late shifts, too. Otherwise you might attribute every difference to the habit. This is personal observation, not an experiment proving cause and effect.

What is one useful change for the following week?

If waking time varies greatly, choose a more consistent time where feasible. If late coffee coincides with difficulty falling asleep, review its timing and amount. There is no reason to impose 14:00 on everybody: bedtime, sensitivity and shift work differ.

NHLBI notes that caffeine effects can last up to eight hours and recommends a quiet, dark, cool bedroom and a regular sleep schedule. This does not promise any specific increase in REM sleep. Use NHLBI's healthy sleep habits to select a manageable change.

Practical insight: keep unusual dates beside the average

Repeating Saturday's hypothetical 480 minutes on Sunday would raise the average, but would not establish that Monday sleepiness had improved. Attach the most tiring date and its work, nap and drink details to the average. If short sleep occurs only after Tuesday and Thursday overtime, reviewing driving and next-day commitments may matter more than adjusting a weekend alarm. The diary is not a personal scorecard. Its purpose is to provide a useful account when discussing habits or seeking care.

When should care come before further recording?

Discuss loud snoring with witnessed breathing pauses, persistent insomnia despite sufficient sleep opportunity, or ongoing daytime sleepiness with a healthcare professional. Do not drive or operate dangerous equipment when sleepy. Wearable sleep stages are estimates and do not replace a sleep study. Children, pregnant people, shift workers and people taking medication should not automatically follow a general adult schedule.

Frequently asked questions

Q. Do naps belong in the seven-day average?

Record them separately from nighttime sleep, then calculate a daily total if useful. Consider why naps increased and whether falling asleep at night became harder.

Q. What if I cannot remember how long I was awake?

Write an estimate in the morning. Do not record an unknown duration as definitely zero or watch the clock all night for the diary.

Q. Is setting an alarm in multiples of 90 minutes enough?

Sleep cycles are not a fixed timer. Do not shorten needed sleep to match a multiple; prioritize sufficient sleep opportunity and daytime functioning.

Q. Should a low wearable REM value lead to a supplement?

Do not choose medication or supplements from a device score. Discuss symptoms and existing medication with a clinician and obtain an appropriate assessment.

Q. Can one week's diary identify the cause?

It may suggest patterns but cannot prove a cause. Record changes such as illness and workload that could also explain the result.

Q. When should I bring the diary to a clinician?

Bring it when sleep problems recur or disrupt daily life. Do not wait to complete seven days if dangerous sleepiness or breathing pauses are present.