Reimbursement gaps and workforce shortages limit CBT-I access

Q: CBT-I is the first-line treatment for insomnia as per the 2023 European guidelines, yet pharmacological treatment tends to be a frequent first choice by primary care practitioners (PCPs) and general practitioners (GPs). Why does this discrepancy exist, and how can it be addressed? What are the main challenges to wider CBT-I adoption?

Dr. Sweetman: It's a complex issue with barriers at multiple levels. Improving access to cognitive behavioral therapy for insomnia (CBT-I) requires a system-wide approach. There’s no single change that will fix this quickly. It will take many small improvements across different parts of the system sustained over time.

That said, there is momentum. Governments, clinicians, researchers, professional associations and industry partners are increasingly aligned with the goal of improving access. I’m hopeful we’ll get there.

In Australia, about nine out of ten patients presenting to primary care with insomnia are prescribed medication. That comes from Bettering the Evaluation of Care and Health (BEACH) data¹, which is a bit dated now, but more recent work by Dr. Jenny Haycock² shows that access to CBT-I still sits around just 1% of people with insomnia. So even a decade later, that number hasn’t really changed.

One of the biggest barriers is simply the workforce. In 2022, we tried to map psychologists across Australia who specialize in insomnia. We found just 65 across the entire country that specialize in CBT-I delivery. For an estimated 3 million people with chronic insomnia, that’s nowhere near enough.

This shortage drives high costs and long waitlists, sometimes six to twelve months. That, in turn, reinforces the prescribing behavior we see. People need help, and when CBT-I isn’t accessible, prescribing medication can become the default.

On top of that, digital CBT-I isn’t yet reimbursed under Medicare in Australia. We’d love to see that change. Just like GPs can refer to psychologists for in-person care under Medicare, we need similar item numbers to support access to digital CBT-I. That’s already happening in countries like the UK and Germany.

Q: The insomnia guidelines encourage short-term use of pharmacological treatments to provide immediate relief while longer-term strategies are developed. Yet many patients stay on medication long-term or indefinitely. Why do you think this happens?

Dr. Sweetman: Sedative-hypnotic medications absolutely have a place in insomnia treatment. Treatment shouldn’t be framed as a choice between either CBT-I or medication as they both have roles to play.

For example, in cases of acute insomnia lasting a week or two due to things like bereavement, grief or surgery recovery, sedatives can provide quick relief. In those situations, CBT-I might not be the right approach. You wouldn’t want to restrict someone’s sleep further when insomnia is driven by something temporary or situational.

But while medications are recommended for short-term use, we often see them used long-term in practice. One reason is the lack of access to CBT-I. When the most effective long-term treatment isn’t available, short-term solutions become the default. There’s also the issue of dependence. Several sedatives are associated with dependence and people can develop this through psychological or physiological mechanisms. For instance, withdrawal or rebound insomnia can occur when trying to reduce or stop long-term use of medications like benzodiazepines or Z-drugs.

So, the pattern becomes: When I take the sedative, I sleep and when I stop, my insomnia returns. That reinforces continued use. Initially, the medication provides real therapeutic benefits. But over time, the balance shifts, benefits give way to dependence, withdrawal symptoms and an increased risk of side effects.

There’s a meta-analysis by Winkler³ from about ten years ago that looked at randomized, placebo-controlled trials of benzodiazepines for insomnia. It found that about 64% of the observed effect was also seen in the placebo group. That suggests two-thirds of the benefit may come from expectation effects, not the drug itself.

Q: What are some of the risks or downsides that might make pharmacological interventions not always in a patient’s best interest?

Dr. Sweetman: Beyond dependence, there are risks of side effects and adverse events.

There’s a fair bit of epidemiological data showing that sedative-hypnotics are linked to cognitive and psychomotor impairment, increased risk of falls and even car accidents. These risks are particularly pronounced in older adults.

One standout study by Glass⁴ from about 20 years ago found that the number needed to harm was half the number needed to treat for benzodiazepines. That led to recommendations against using these medications in older adults.

So the concerns is not just about dependence. It’s about real-world adverse outcomes, particularly in certain groups.

Q: Are there healthcare systems where reimbursement structures support insomnia care in primary care?

Dr. Sweetman: Yes. In Australia, we have Medicare, a universal healthcare system that subsidizes GP consultations. Patients often pay a small gap fee, but most of the cost is covered. GPs can now refer directly for sleep studies, which previously required a specialist referral. That change has reduced wait times and improved access.

Sleep studies themselves are also subsidized under Medicare, through different item numbers that help reduce out-of-pocket costs.

Medicare also supports access to psychological care. Patients can receive up to 10 subsidized sessions with a psychologist each year. There’s sometimes a gap fee, but it significantly lowers the barrier to care.

Outside Australia, I’m less familiar with all the details, but I know countries like Germany have made strong progress. They’ve implemented programs like Somnio, which gives patients access to digital cognitive behavioral therapy for insomnia (dCBT-I) through Germany’s “app on prescription” system, with costs reimbursed by statutory health insurers. Similarly, the UK’s National Health Service offers access to programs like Sleepio, a dCBT-I solution recommended by the National Institute for Health and Care Excellence (NICE) that is free in Scotland and available in selected regions of England via self-referral.

In the U.S., there are similar offerings, though the U.S. healthcare system has its own access and subsidy challenges.

Germany has been especially impressive in scaling access. By subsidizing digital CBT-I, they've made it available to tens of thousands of people. That’s a major step toward increasing access and moving beyond the 1% who typically receive CBT-I.