If you walk into a GP clinic in Australia with chronic insomnia, there’s about a 70% chance you’ll walk out with a prescription for a benzodiazepine or a Z-drug (zopiclone, zolpidem). This happens despite every major clinical guideline — from the Australasian Sleep Association, the American Academy of Sleep Medicine, the European Sleep Research Society, and the UK’s NICE — recommending cognitive behavioural therapy for insomnia (CBT-I) as the first-line treatment.

The gap between what evidence supports and what patients receive is one of the most striking in sleep medicine. And it’s not because the evidence is ambiguous. It’s overwhelming.

What the Evidence Actually Shows

CBT-I has been studied in over 200 randomised controlled trials. A 2015 meta-analysis in the Annals of Internal Medicine, commissioned by the American College of Physicians, evaluated the entire body of evidence and concluded that CBT-I should be the initial treatment for chronic insomnia in all adults. The effect sizes were large and consistent.

Here’s what the data shows:

CBT-I reduces time to fall asleep by an average of 19 minutes. That might not sound dramatic, but for someone who’s been lying awake for 60-90 minutes nightly, reducing that to 40-70 minutes is meaningful — and the improvements continue to build over weeks.

Sleep efficiency improves by approximately 10%. Sleep efficiency is the ratio of time asleep to time in bed. Moving from 70% to 80% means an extra 45 minutes of actual sleep in an 8-hour bed period.

Effects are durable. This is where CBT-I decisively outperforms medication. At 6-month and 12-month follow-up, patients who completed CBT-I maintain their improvements. Patients who used medication and then stopped typically return to baseline insomnia severity within weeks of discontinuation.

Head-to-head comparisons favour CBT-I. In studies directly comparing CBT-I with pharmacotherapy, both produce roughly equivalent short-term improvements. But at 6 and 12 months, CBT-I patients continue improving while medication patients either maintain (if still taking medication) or relapse (if medication is discontinued).

What CBT-I Actually Involves

The name sounds intimidating — “cognitive behavioural therapy” suggests complex psychological intervention. In practice, CBT-I is a structured programme of 4-8 sessions that addresses the behavioural and thought patterns perpetuating insomnia. It has several core components:

Sleep restriction. This is the most counterintuitive and most powerful component. Patients are instructed to limit their time in bed to match their actual sleep time. If you’re sleeping 5.5 hours but spending 8 hours in bed, your initial prescribed time in bed might be 6 hours. This creates mild sleep deprivation that builds sleep pressure and consolidates fragmented sleep. As sleep efficiency improves, time in bed is gradually increased.

Sleep restriction is also the reason many patients initially struggle with CBT-I. Deliberately spending less time in bed when you’re already exhausted feels wrong. And for the first week or two, patients often feel worse — more fatigued, more irritable — before the consolidation effect kicks in. This temporary worsening requires careful patient education and ongoing support to prevent dropout.

Stimulus control. The bed becomes associated with wakefulness rather than sleep for chronic insomnia sufferers. Stimulus control breaks this association by establishing rules: use the bed only for sleep and sex. If you’re awake for more than 15-20 minutes, get up and do something quiet in another room. Return to bed only when sleepy. No reading, phone use, or TV watching in bed.

Cognitive restructuring. Many insomnia patients develop catastrophic thinking about sleep: “If I don’t sleep tonight, I won’t function tomorrow.” “My health is being destroyed.” “I’ll never sleep normally again.” These thoughts create anxiety that directly interferes with sleep onset, creating a self-fulfilling cycle. Cognitive restructuring helps patients identify and challenge these thought patterns.

Sleep hygiene. The supporting framework of CBT-I includes standard sleep hygiene recommendations — consistent wake times, limited caffeine after midday, comfortable sleep environment, reduced screen exposure before bed. Sleep hygiene alone isn’t effective for chronic insomnia, but it supports the other components.

Why GPs Prescribe Pills Instead

If CBT-I is this effective, why isn’t it standard practice?

Time constraints in general practice. A GP consultation in Australia is typically 10-15 minutes. Writing a prescription takes 30 seconds. Explaining CBT-I, assessing suitability, and arranging referral to a trained provider takes considerably longer. The economics of general practice work against time-intensive interventions.

Limited access to trained CBT-I providers. Australia has a shortage of psychologists and therapists specifically trained in CBT-I delivery. In metropolitan areas, wait times for CBT-I programmes can be 4-8 weeks. In regional and rural areas, in-person CBT-I may simply not be available.

Patient expectations. Many patients present expecting medication. They’ve heard about insomnia medication from friends, advertising, or previous medical encounters. When told that the recommended treatment is a behavioural programme that might make them feel worse initially and requires active effort over several weeks, some patients — understandably — prefer the faster-acting option.

GP training gaps. Sleep medicine receives minimal coverage in Australian medical training. Many GPs aren’t familiar with the CBT-I evidence base or don’t feel confident recommending it over medication. This isn’t a criticism of GPs — it’s a curriculum problem.

Digital CBT-I Is Changing Access

One of the most promising developments in insomnia treatment is the emergence of digital CBT-I programmes — app-based or web-based platforms that deliver the core components of CBT-I through guided self-help.

Sleepstation in the UK and similar platforms in Australia provide structured CBT-I programmes with sleep diary tracking, personalised sleep window recommendations, cognitive restructuring exercises, and progress monitoring. Some include human coaching or therapist check-ins; others are fully automated.

The evidence for digital CBT-I is strong, though effect sizes are slightly smaller than face-to-face delivery. A 2024 meta-analysis in Lancet Digital Health found that digital CBT-I produced clinically significant improvements in sleep onset latency, wake after sleep onset, and sleep efficiency compared with waitlist controls, with improvements maintained at follow-up.

Some organisations working on custom AI development for healthcare have explored how personalised algorithms could optimise CBT-I delivery — adjusting sleep restriction parameters in real time based on daily sleep diary data, rather than relying on weekly therapist reviews. The concept is promising, though clinical validation is still needed.

When Medication Has a Role

I don’t want to suggest that insomnia medication is always inappropriate. There are legitimate scenarios:

• Acute insomnia (triggered by bereavement, acute medical illness, or major life disruption) may benefit from short-term medication use (2-4 weeks) while the acute stressor resolves.
• CBT-I is contraindicated in some patients with untreated severe depression, active suicidal ideation, or certain medical conditions where sleep restriction could be dangerous (e.g., unstable epilepsy, bipolar disorder with mania risk).
• Combination therapy — medication for initial relief alongside CBT-I for long-term management — can be appropriate for severe chronic insomnia, with the medication tapered as CBT-I takes effect.

What’s not appropriate is long-term benzodiazepine or Z-drug prescriptions as the sole treatment for chronic insomnia, which is exactly what millions of Australians currently receive.

What I Tell My Patients

Chronic insomnia is a learned condition. Your brain has learned to be awake when it should be asleep, and it’s reinforced that learning through years of compensatory behaviours — going to bed earlier, lying in bed hoping sleep will come, napping to catch up, worrying about sleep.

CBT-I works because it directly retrains these learned patterns. It’s harder than taking a pill. It requires patience, commitment, and a willingness to feel temporarily worse. But it produces results that last — results that medication simply cannot match.

If you’ve been struggling with insomnia for more than three months, talk to your GP about a CBT-I referral. If in-person programmes aren’t available in your area, ask about digital CBT-I options. The treatment exists, the evidence supports it, and the barrier to accessing it is lower than it’s ever been.