Most people who cannot fall asleep tonight do not have insomnia. They have a bad night, usually with a traceable cause. The NHLBI page on insomnia describes it as a common sleep disorder in which "you may have trouble falling asleep, staying asleep, or getting good quality sleep," and it draws a clear line between the short-term version and the chronic one.
That line matters, because the two have different answers. A rough fortnight after a stressful event is handled by fixing the obvious inputs. A pattern that has run for months has an established treatment path, and it is not the one most people reach for first. This piece sits within our complete guide to better sleep.
What counts as insomnia rather than a rough night?
The NHLBI splits it in two. Short-term insomnia stems from things like stress or a change in your schedule or surroundings, and lasts days to weeks. Chronic insomnia, by contrast, "occurs 3 or more nights a week, lasts more than 3 months, and cannot be fully explained by another health problem."
Three parts of that definition are worth separating out:
| Criterion | What it means in practice |
|---|---|
| 3 or more nights a week | Frequency, not severity. Three moderate nights count; one catastrophic night does not. |
| More than 3 months | Duration. This is what separates chronic insomnia from a stressful patch. |
| Not fully explained by another health problem | Insomnia here is the condition itself rather than a side effect of something else, which is part of why self-diagnosis has limits. |
If you are inside the first few weeks of a bad run, the useful move is to look at the everyday inputs below before concluding anything about a disorder.
The most common everyday causes
Stress and a busy head. The NHLBI names stress as a trigger for short-term insomnia directly, alongside changes to your schedule or surroundings. This is the version most people recognise: the body is tired, the mind is still filing paperwork.
An irregular schedule. CDC/NIOSH leads its sleep advice with going to bed and getting up "at about the same times every day, including days off." A wake time that moves by two hours at weekends leaves your body clock without a fixed reference point, and a clock with no reference point produces exactly this complaint.
Caffeine later than you think. This is the most underrated entry on the list. In a controlled trial published in the Journal of Clinical Sleep Medicine, Drake and colleagues gave 400 mg of caffeine at 0, 3 or 6 hours before bedtime and found that objectively measured total sleep time fell by 1.1 to 1.2 hours at every one of those time points, including six hours out. CDC/NIOSH advises avoiding caffeine, chocolate and nicotine for 5 or more hours before sleep. Our full look at the timing is in how long caffeine affects sleep.
A nightcap. Alcohol shortens the time it takes to fall asleep, which is why it gets mistaken for a solution. What it does afterwards is covered in does alcohol help or hurt sleep quality. CDC/NIOSH simply advises avoiding alcohol near bedtime.
Evening light. Light and dark are the strongest inputs to the body clock that governs when you feel sleepy, so a bright evening is not a neutral one. CDC/NIOSH recommends making the bedroom very dark and blocking out lights in the room, especially blue and white ones. See does blue light really ruin your sleep for what the evidence does and does not establish.
A room working against you. CDC/NIOSH puts a comfortable bedroom at roughly 65–68°F for most adults, and advises using the bed for sleep and intimacy only, with no TV, reading or work in the bedroom. A warm, bright room that doubles as an office is asking a lot of your nervous system.
When it is not just habits
Some causes are not habits at all, and no amount of checklist discipline will resolve them. Two signals are worth acting on rather than absorbing.
The first is time to fall asleep. CDC/NIOSH advises contacting a doctor if you consistently take 30 minutes or more to fall asleep. Consistently is the operative word.
The second concerns breathing. The NHLBI list of sleep apnea symptoms includes "breathing that starts and stops," "frequent loud snoring" and "gasping for air" during sleep, alongside daytime sleepiness and, notably, insomnia itself, which it lists as more common in women. If a partner has described any of those, that is information a clinician needs, not a habit to correct.
The NHLBI also notes that some prescription medicines used to treat other health conditions can increase the risk of insomnia. If your sleep changed after a medication did, that is a conversation to have with the prescriber rather than a coincidence to work around.
What actually treats chronic insomnia
Here the guidance is unusually direct. The NHLBI page on insomnia treatment says of cognitive behavioural therapy for insomnia: "This is usually recommended as the first treatment option for long-term insomnia and can be very effective."
The page describes it as a structured set of components rather than talking therapy in general:
- Cognitive therapy, aimed at reducing anxiety about sleep
- Relaxation or meditation therapy
- Sleep education covering good sleep habits
- Sleep restriction therapy, which sets a specific time in bed
- Stimulus control therapy, which works on a regular sleep and wake cycle
On medicines, the same page notes that one class of prescription sleep medicine "can be habit-forming and should be taken for only a few weeks." We are not going to name, compare or suggest any of them, including anything sold over the counter or as a supplement; that is a decision for you and a clinician who knows your history.
The practical order, then, is this. Fix the inputs first, because caffeine timing, light and a consistent wake time are free and fast, and they are collected in our sleep hygiene checklist. If the pattern still meets the NHLBI's chronic definition after that, ask a clinician about cognitive behavioural therapy for insomnia by name. It is the treatment the guidance points to first, and knowing what to ask for is often the difference between getting it and being handed something else.
