Cognitive Behavioral Therapy for Insomnia (CBT-I): The Treatment That Isn’t Sleep Hygiene Advice

There’s a particular way people talk about their sleep once they’ve been struggling with it for a long time, and I can usually recognize it within the first few minutes of a consultation. It’s apologetic. Someone will list everything they’ve already tried, the blackout curtains and the magnesium and the app with the rain sounds and the rule about no screens after nine that they kept faithfully for eleven days, and then they’ll land on some version of “so I know I’m probably just not being consistent enough.” They say it the way you’d confess to skipping the gym.

I want to take that off the table before we go any further, because I think it’s been quietly making things worse. If you’ve been lying awake most nights for months, the problem isn’t your willpower and it isn’t your commitment to the routine. What’s happened is that you were handed a set of general wellness suggestions and told they were a treatment, and then when general wellness suggestions failed to resolve a clinical condition, you drew the reasonable but incorrect conclusion that the failure was yours.

Sleep hygiene is fine as far as it goes, which is not very far. It’s a decent description of the conditions that support sleep in people who are already sleeping. What it is not is a treatment for chronic insomnia, any more than telling someone with a fractured ankle to wear supportive shoes is a treatment for the fracture. The actual treatment does exist, it’s been studied for decades, it usually takes a couple of months, and most of the people who need it have never heard its name. It’s called cognitive behavioral therapy for insomnia, shortened to CBT-I, and I’d like to walk you through what it actually involves, because I think the vagueness around it is a large part of why so few people ever get to it.

What is CBT-I?

Cognitive behavioral therapy for insomnia is a structured, time-limited treatment that targets the specific habits and thought patterns keeping insomnia going long after whatever started it has resolved. It typically takes somewhere around eight sessions. In that time we track your sleep carefully, change when and how you use your bed, deliberately narrow and then rebuild your sleep window, and work directly on the thinking that turns one difficult night into a difficult season.

The clearest way I know to put the distinction is this: sleep hygiene describes what a good sleep environment looks like, while CBT-I addresses the reason your body stopped reading that environment as a signal to sleep.

Why the insomnia outlasts whatever caused it

Nearly every case of chronic insomnia I see began with something entirely reasonable. A trial that ate four months. A closing that ran through the holidays. A new baby, a sick parent, a marriage coming apart, or just a punishing stretch of work where sleep was the only variable left to sacrifice. The insomnia made sense at the time, and everyone in the person’s life would have agreed. Of course you’re not sleeping. Look at what’s happening.

And then the trial ended, and the sleep didn’t come back.

That gap is what people find most bewildering, and it’s precisely what CBT-I was built around. Sleep researchers describe insomnia as having predisposing factors, a precipitating event, and perpetuating factors, and the genuinely useful insight buried in that framework is that the thing which starts insomnia is almost never the thing that keeps it running. What keeps it running is the very sensible set of adjustments you made in order to cope with it.

You started getting into bed at nine-thirty instead of eleven, on the theory that if sleep was scarce you should give it more runway. You began lying there awake for an hour or two, waiting, because getting up felt like conceding. You took calls from bed on the bad days, answered email from bed, did the worrying in bed, because that’s where you were. On weekends you slept until ten to catch up, which nudged your body clock a little further out of alignment each time. You started declining the Thursday dinner because you needed the rest, and then on top of everything else you were lonelier.

Every one of those decisions is defensible on its own. Taken together, over a few months, they teach your nervous system something very specific, which is that the bed is a place where you lie awake, think hard, and feel bad about it. Your body learns that association the same way it learns any other, through repetition, and it learns it thoroughly. That’s why the sleep doesn’t come back when the deadline does. It’s also, and I mean this encouragingly, why the whole thing can be unlearned.

What the treatment actually consists of

People are often surprised by how concrete this work is. It isn’t open-ended conversation about your relationship with rest. It’s closer to physical therapy, in the sense that there are specific exercises, they’re somewhat uncomfortable at first, and they work on a reasonably predictable timeline.

We start by finding out what’s actually happening

The first week or two is assessment, and it starts with a sleep diary. It’s simple, and it should take about ninety seconds each morning: roughly when you got into bed, roughly when you think you fell asleep, how many times you surfaced during the night, when you finally got up for the day. Rough estimates aren’t just acceptable here, they’re preferable to the data from your watch. Consumer wearables are genuinely unreliable at distinguishing light sleep from lying quietly awake, and in my experience their main clinical effect on an already anxious person is to hand them a number to be anxious about at six in the morning. I usually suggest people take the ring or the watch off for the duration and give it back to themselves at the end.

From the diary we calculate sleep efficiency, which is just the percentage of your time in bed that you spend actually asleep. If you’re in bed for nine hours and asleep for five and a half, that’s around sixty percent, and seeing it written down tends to explain a great deal about why you feel so much worse than the colleague who sleeps six hours cleanly and gets up. You’re not necessarily getting less sleep than you thought. You’re spending three and a half hours a night practicing wakefulness in the one place you need to associate with sleep.

Rebuilding what the bed means

The next piece is called stimulus control, which is a clinical name for a short set of rules that undo the conditioning I described above. Reserve the bed for sleep and sex. Go to bed when you’re sleepy rather than when you’re tired, which are genuinely different states and worth learning to tell apart. And if you find yourself awake for what feels like twenty minutes or so, get up, go somewhere else, do something undemanding in low light, and come back when actual sleepiness turns up.

Nobody enjoys the part where they leave a warm bed at three in the morning to go sit in a chair, and I’ve never presented it as pleasant. What I will tell you is that it’s an important component in the whole protocol, because as long as you keep lying in bed awake, you’re continuing to run the exact training program you’re trying to reverse. We work out the practicalities together, including where you’ll go, what you’ll do when you get there, and how to handle it if there’s a partner asleep next to you or a five-year-old down the hall, because a rule that doesn’t survive contact with your actual house isn’t a rule you’ll keep past Wednesday.

Sleep restriction, which is really sleep consolidation

This is the engine of the treatment, and it’s the main reason CBT-I tends to work better with a clinician than with an app. We temporarily narrow your time in bed until it’s close to the amount you’re genuinely sleeping, which builds up real physiological sleep pressure, and then, as your sleep efficiency climbs week over week, we widen the window back out.

I’m honest with people about this part in session, so I’ll be honest about it here. For roughly the first seven to ten days, you’ll be more tired rather than less. That’s the mechanism doing precisely what it’s meant to do, and it’s also, predictably, the point at which people abandon the treatment. So we plan for it. We look at your calendar before we start, and if there’s a hearing or a closing or a deposition on the horizon, we wait, because there’s no good reason to run the hardest week of this against the hardest week of your year. 

What tends to happen on the other side of it is that people are startled by the speed of the shift. Someone who spent two years falling asleep at two in the morning finds themselves genuinely sleepy at eleven and slightly suspicious of it.

The part that involves the thinking

Then there’s what happens in your head at two in the morning, which for most people is a very particular kind of arithmetic. If I fall asleep right now I can still get five hours. Now four and a half. If I don’t sleep, tomorrow is a write-off and someone will notice. I’ve been like this for a year and I’ll probably be like this forever.

Underneath that arithmetic there are usually a few beliefs doing the actual damage: that you need eight hours specifically or the day is lost, that a single poor night will visibly degrade your work, that your sleep is broken in some permanent structural way that isn’t going to change. We take those out and examine them properly, and we check them against what really happened on your worst nights, which is often that you got through the deposition perfectly competently on four hours and nobody said a thing.

This isn’t positive thinking, and I have very little patience for positive thinking. It’s closer to a careful cross-examination of your own assumptions, and people who reason for a living tend to find it satisfying once they realize that’s what we’re doing. The goal isn’t to feel cheerful about a bad night. It’s to stop treating a bad night as a catastrophe, because the catastrophizing is itself one of the things keeping you awake.

Where sleep hygiene finally earns its place

I don’t want to be unfair to sleep hygiene, so here’s where it genuinely belongs. Once the conditioning is repaired and the schedule is stable, the hour before bed starts to matter, and we build a wind-down that fits your real life rather than an aspirational version of it. For most of the people I work with that means something modest and achievable, like a genuine stop time for email and a buffer that doesn’t involve a screen, rather than a forty-minute ritual that nobody sustains past the second week.

Why I see so much of this in lawyers

I practiced employment defense litigation before I retrained as a therapist, so I have both a professional and a somewhat personal interest in this problem.

Legal training rewards a specific cognitive style. You learn to anticipate every adverse scenario, hold a dozen open threads at once, and never fully close a matter in your head, because the one you closed is the one that comes back. That skill is enormously valuable at four in the afternoon and quietly destructive at midnight, mostly because it doesn’t have an off switch. It just runs louder or softer.

The structural pieces stack on top of that. The billable hour means the day ends when you stop rather than when a shift ends, so there’s no external boundary to lean on, only the one you’re supposed to impose on yourself while everyone around you is failing to impose one. Email arrives from clients and partners across time zones, and the expectation of responsiveness is real rather than imagined. When a matter is live, the rational move is to work until you physically can’t, which means the final cognitive act of your day is high-stakes analytical work performed in a state of full alertness, roughly twenty minutes before you’d like to be unconscious.

That combination doesn’t just produce a few rough nights. It builds, with real efficiency, the exact perpetuating conditions CBT-I exists to unwind. If any of this is sounding familiar alongside what I’ve written about anxiety in lawyers or the kind of burnout that doesn’t look like falling apart, that isn’t a coincidence. These things arrive together far more often than they arrive alone.

When the insomnia is telling you about something else

I’d be doing you a disservice if I presented sleep as a self-contained mechanical problem, because a good deal of the time it isn’t.

Insomnia is often the first visible sign of depression or an anxiety disorder, and it’s very frequently the only symptom a high-functioning person is willing to bring to an appointment. That makes complete sense to me. Trouble sleeping feels like a plumbing issue, something happening to you rather than something about you, and it can be raised without admitting to anything you consider more serious. I have real respect for the fact that people come through the door this way. It isn’t avoidance so much as a sensible place to start.

One pattern is worth naming specifically. Waking at four in the morning and being unable to get back down, night after night, is a well-recognized feature of depression, and it can show up months before anyone would think to describe themselves as sad. If that’s your particular version of this, it’s worth reading what I’ve written about depression that doesn’t look like sadness, because in high achievers it very rarely does.

This is why I assess before I treat. CBT-I sits perfectly comfortably alongside treatment for depression or anxiety, and there are plenty of cases where getting the sleep back improves everything downstream of it. But if the insomnia is one thread in a larger weave, pulling on that thread alone is a slow route back to where you started.

Common questions about CBT-I

How long does CBT-I take? Most people finish in about eight sessions, assuming consistent adherence. Meaningful change usually shows up within the first three to four weeks, generally just after the most demanding stretch of the consolidation phase.

Does it work over telehealth? It does. The treatment is conversation, tracking, and scheduling, all of which translate to video without losing anything meaningful. I work this way with clients across California, Washington D.C., and Virginia.

Will it make me more tired at first? During the consolidation phase, for about a week to ten days, usually yes. That’s the treatment working rather than a sign that it isn’t, and we schedule it deliberately so that it doesn’t land on top of a trial or a closing.

I’ve had this for years. Is it too late? It isn’t, and duration matters far less than most people expect. Whether this started eight months ago or eight years ago, the factors keeping it going are the same handful of things, and those are exactly what the treatment addresses.

If you’ve been calling this a discipline problem

If you’ve spent a year or more believing that better habits would eventually fix your sleep, I’d gently suggest that the habits were never really the issue. Chronic insomnia is a specific, well-understood condition with a specific, well-tested treatment, and that treatment is not a longer list of rules about screens and caffeine.

I offer CBT-I as a genuine protocol within my work with lawyers and other people in demanding professions, which means we do it properly rather than spending the last four minutes of a session on sleep tips. If you’d like to know more about how I work, you can read about my background and approach, or you can schedule a consultation and we can talk about whether this is the right fit for what’s been happening. It’s a short conversation, there’s nothing to prepare, and you don’t have to turn up with a completed sleep diary.