Effectiveness of group-delivered cognitive behavioural therapy for insomnia in primary care: a pragmatic, multicentre..., 2025, Hrozanova et al

[rainy]

Effectiveness of group-delivered cognitive behavioural therapy for insomnia in primary care: a pragmatic, multicentre randomised controlled trial

https://www.sciencedirect.com/science/article/pii/S138994572500156X

Abstract​

Insomnia is a prevalent disorder in the general population. Group delivery increases access to the treatment of choice for insomnia, cognitive-behavioural therapy for insomnia (CBT-I). The aim of this pragmatic randomized controlled trial was to investigate the effectiveness of low-threshold group-delivered CBT-I implemented in Norwegian primary care. The trial was registered in the ISRCTN registry (ISRCTN16185698), and conducted across 26 Healthy Life Centres. Inclusion criteria were age ≥ 18 years and insomnia symptoms with a significant impact (Insomnia Severity Index (ISI) ≥ 12). Eligible participants were randomized to group-delivered CBT-I or to a waiting list according to a 2:1 ratio. The intervention consisted of four 2-h sessions over 4 weeks, with 5–15 participants per group. Primary outcome was self-reported insomnia severity at 3-months follow-up, measured by the ISI. Participants were not blinded to group assignment, but those responsible for analyses and interpretation of the results were. In total, 308 participants were randomly allocated to group-delivered CBT-I (n = 181) or waiting list (n = 127). The mean difference in ISI score between groups at 3-months follow-up was −3.4 (95 % CI: 4.3 to −2.5), favoring the group-delivered CBT-I. Of those receiving group-delivered CBT-I, 51 (33.6 %) reported a clinically relevant improvement in insomnia severity (i.e., ≥ 8 points on ISI), compared to 15 (13.2 %) on the waiting list (OR 3.4, 95 % CI: 1.7 to 6.8). Thus, group-delivered CBT-I reduced insomnia severity at 3-months follow-up for adults with insomnia. Group-delivered CBT-I offered at Healthy Life Centres is a suitable low-threshold treatment for insomnia in primary care.

Results​

At 3-months follow-up, the estimated mean difference in the ISI score between group-delivered CBT-I and waiting list was −3.4 (95 % CI: 4.3 to −2.5, p < 0.001). The estimated marginal mean ISI score at 3-months follow-up was 10.7 (95 % CI: 10.1 to 11.4) for group-delivered CBT-I, and 14.1 (95 % CI: 13.4 to 14.8) for waiting list, indicating that participants receiving group-delivered CBT-I reported lower insomnia severity.

(....)

For the sleep diary data at 3-months follow-up, group-delivered CBT-I led to significantly greater increase in daytime functioning, improved sleep quality and increased sleep efficiency, as well as a significantly greater reduction in wake after sleep onset, when compared to the waiting list. The estimated between-group differences and their statistical significance were sustained at 6-months follow-up.
At 3-months follow-up, group-delivered CBT-I led to a significantly greater reduction in use of prescription and non-prescription sleep medication than the waiting list, but this effect was attenuated and no longer statistically significant at 6-months follow-up. There were only small and not statistically significant or no differences between the groups regarding sleep duration, sleep onset latency, number of awakenings, and napping frequency at any of the follow-ups.

At 3-months follow-up, the estimated mean difference in the ISI score between group-delivered CBT-I and waiting list was −3.4 (95 % CI: 4.3 to −2.5, p < 0.001). The estimated marginal mean ISI score at 3-months follow-up was 10.7 (95 % CI: 10.1 to 11.4) for group-delivered CBT-I, and 14.1 (95 % CI: 13.4 to 14.8) for waiting list, indicating that participants receiving group-delivered CBT-I reported lower insomnia severity.

(....)

Conclusion​

Group-delivered CBT-I reduced insomnia severity with sustained between-group differences at 6-months follow-up. Group-delivered CBT-I offered at Healthy Life Centres is a suitable low-threshold treatment for insomnia in primary care. Future research should investigate group-delivered CBT-I contrasting its effectiveness with other low-threshold CBT-I modalities, such as digital CBT-I and sleep restriction therapy.

 

[rainy]

They didn’t find an improvement in sleep duration, sleep onset latency, number of awakenings, and napping frequency.

From their secondary outcomes they mention increased sleep efficiency. What they mean by sleep efficiency is what proportion of the time spent in bed is spent asleep.
But a main part of the treatment was what they called sleep restriction, or bed restriction, where the point was to spend as little time in bed as possible. This was demonstrated in a video documenting the trial. At first the participants kept a sleep diary of how much they slept on average, and then they were instructed to only spend as much time in bed as how much they slept on average. So if they slept 6 hours, that’s how long they were allowed to be in bed, and if they managed to sleep those hours they would have 100% sleep efficiency.

So when “sleep efficiency” improved, but sleep duration didn’t, wouldn’t that just mean that the participants spent less time in bed? The reason people spend those hours awake in bed is because they wish to be sleeping those hours.

 

[Utsikt]

So when “sleep efficiency” improved, but sleep duration didn’t, wouldn’t that just mean that the participants spent less time in bed? The reason people spend those hours awake in bed is because they wish to be sleeping those hours.

Yes, this is a common problem where the outcomes are targeted directly by the intervention.

The same logic has been used to argue that asking GPs to be more restrictive with sick leave causes less illness because the number of sick leaves decreased.

 

[rvallee]

the treatment of choice for insomnia, cognitive-behavioural therapy for insomnia (CBT-I)

Hmm, no, I don't think it is.

In total, 308 participants were randomly allocated to group-delivered CBT-I (n = 181) or waiting list (n = 127).

Big trial.

There were only small and not statistically significant or no differences between the groups regarding sleep duration, sleep onset latency, number of awakenings, and napping frequency at any of the follow-ups.

What even is insomnia if not sleep duration and remaining asleep after falling asleep and not taking too long to achieve it? This is all basically like a bag of food that says "33% bigger" except only the bag is bigger, there isn't any more food in it. Who does this even serve?

From their secondary outcomes they mention increased sleep efficiency. What they mean by sleep efficiency is what proportion of the time spent in bed is spent asleep.

But this is nonsense, no change in sleep latency or duration means this can't also be true.

So when “sleep efficiency” improved, but sleep duration didn’t, wouldn’t that just mean that the participants spent less time in bed?

The only plausible explanation for that is less daytime napping because nothing actually changed. Although there is odd data in there because daytime napping, a simple yes/no, went from 140/64 to 137/75 so the numbers don't match unless they added participants in the interval. For improvements things moved up and down between months 1, 3 and 6, so there's fluctuating noise in there. I guess they had data gathering problems.

Prescription medication also barely changed, also a simple yes/no.

Once more all they're proving is that they can find people who will do everything they ask and despite no actual benefits will seem to report them anyway, at least according to the questionnaires they use. What even is this nonsense? Why the total lack of ambition here?

Future research should investigate group-delivered CBT-I contrasting its effectiveness with other low-threshold CBT-I modalities, such as digital CBT-I and sleep restriction therapy.

Ah, the traditional "pay us to do it again, identically, again and again". This is clearly who is served here.

 

[rainy]

The thing I found the most confusing about this trial is a short film they made about it.

The film was supposed to “present the results in an understandable and applicable way to reach a much larger target group than research articles usually do.”

The video follows two participants that are being interviewed.

At the end of the video they present some data from the two participants sleep diaries as text on the screen:

____

Participant A:

Sleep duration
week 1 - 5,10 h
week 14 - 6,10 h

Sleep quality
week 1 - 65%
week 14 - 81,9%

____

Participant B

Sleep duration
week 1 - 4 h
week 14 - 6,43 h

Sleep quality
week 1 - 59%
week 14 - 84%

____

This makes it seem like an effect of the trial is that the participants improved their sleep duration, but the trial paper itself states that they did not improve sleep duration.

From the creators report on the film project:
«The film was as we hoped, with documented good results on the two course participants that we follow in the film. This was an x-factor we could not control. But here it is, as we hoped, a confirmation of the results from the research project.»



Also, I feel like the “81,9% sleep quality” in the video makes it seem like they’ve measured something specific. If I just saw this video, without reading the paper, I would probably have thought it was some sort of measurement from a device.

But I think this is how they actually measured sleep quality in the trial:

“Sleep quality: Use the scale in your sleep diary to indicate how you experienced the quality of your night's sleep.

How was your sleep overall last night? 1 = very light, 2 = light, 3 = medium, 4 = deep, 5 = very deep”