The evidence base for chronic insomnia treatment continues to favour cognitive behavioural therapy for insomnia (CBT-I) over long-term medication, and the 2026 clinical picture continues to align with that evidence. The interesting story isn’t whether CBT-I is the recommended first-line treatment — it has been for years — but how clinical practice is finally catching up to the guidelines.

The basic evidence position hasn’t changed substantively in 2026. Multiple high-quality systematic reviews continue to show that CBT-I produces durable improvement in sleep onset latency, sleep efficiency, and waking after sleep onset. The effect sizes are clinically meaningful and persist beyond treatment, which medications generally don’t achieve. For chronic insomnia (lasting more than three months), CBT-I outperforms benzodiazepines, Z-drugs, and most newer pharmaceutical options on durability.

The medications remain genuinely useful for short-term insomnia, particularly situational insomnia where a clear external trigger is going to resolve. They’re also useful as a bridge while CBT-I is being initiated, and for patients with specific contraindications or comorbidities that complicate CBT-I delivery. The clinical position isn’t anti-medication. It’s pro-CBT-I as the durable solution for the chronic case.

What has changed in 2026 is access. The historical barrier to CBT-I implementation has been the limited number of trained providers and the time-intensive nature of the standard protocol. The combination of digital CBT-I products with reasonable evidence bases, telehealth delivery of CBT-I by trained providers, and primary care training programs that have built basic CBT-I capability outside specialist sleep clinics has materially improved access over the past three years. Patients in 2026 can access CBT-I more readily than they could in 2022.

The digital CBT-I products have a more robust evidence base than they did in 2022. Several products have completed proper randomised trials in patient populations relevant to general practice, and the outcome data is reasonable. The digital products are not equivalent to in-person CBT-I delivered by a trained psychologist for severe or complicated cases, but they’re a credible first step for patients with straightforward chronic insomnia. The MBS pathway in Australia continues to evolve in ways that broaden access.

The medication-prescribing pattern is shifting more slowly than guidelines suggest. Australian benzodiazepine and Z-drug prescribing for insomnia remains higher than the evidence base supports, particularly in older patient populations where the harm-benefit calculation is least favourable. The reasons are well-known and frustrating: short consultation times, patient expectations, the absence of immediately-available CBT-I alternatives in many primary care settings, and the operational difficulty of unwinding established prescriptions. The trends are slowly improving but the gap between guideline and practice is real.

The newer hypnotic options — orexin receptor antagonists in particular — have a place but haven’t displaced the older agents to the extent some enthusiasts predicted. The efficacy data is modestly favourable, the side-effect profile is generally better than older hypnotics, but the cost differential remains significant and the long-term safety data is still maturing. Clinicians in 2026 are using these agents more than they were in 2022, but they’re not yet first-line for most patients.

The melatonin question has settled into a more honest position. Standard melatonin has a small but real effect on sleep onset latency, particularly in older patients with documented melatonin deficiency. It’s not a sedative-hypnotic equivalent for general insomnia, and the early-2020s consumer enthusiasm has moderated as the realistic effect-size data has been more widely communicated. Melatonin still has a role; it’s not the universal solution it was sometimes pitched as.

The OSA-overlap question deserves attention. Patients presenting with insomnia complaints and occult obstructive sleep apnoea are common, and treating the apparent insomnia without first investigating the breathing-related sleep disorder produces poor outcomes. Sleep medicine is increasingly emphasising the screening question. The 2026 clinical practice is more disciplined about this than the 2020 practice was.

For patients dealing with chronic insomnia in 2026, the practical recommendation remains: pursue CBT-I, ideally with a trained provider, accept that it requires real engagement over weeks, and use medications for short-term bridging if needed. For clinicians, the remaining work is on access, training, and the patient-conversation skills required to redirect patients away from medication-first expectations toward the treatments with better long-term outcomes.

The good news in 2026 is that the access situation has genuinely improved. The implementation gap remains but it’s narrower than it was, and the trajectory is positive. Chronic insomnia continues to be enormously prevalent and enormously under-treated, and the work to close those gaps is real clinical and public-health work that produces measurable benefit.