Insomnia – do the drugs work?

Phil is 75 and can’t sleep. Or rather he does sleep, just not enough. He doesn’t have chronic pain, depression or acute stress. He doesn’t have arthritis, obstructive sleep apnoea or heart failure. He isn’t on ADHD stimulants, steroids or SSRIs. He just can’t sleep. 

In fact he gets to sleep just fine, but then manages three or four hours of broken sleep, and that’s it. Awake, tired, frustrated. Any of you that have ever had insomnia will know the feeling – hours of your life lying in the dark, watching the clock tick by until you eventually accept that you might as well get up. Anyone who has ever been troubled by insomnia will empathise with why patients seek help because of it.

This isn’t a blog about sleep hygiene or cognitive behavioural therapy for chronic insomnia, which is defined by ICSD-3 (the International Classification of Sleep Disorders, 3^rd edition) as difficulty with sleep having negative consequences, on at least 3 night per week for >3 months. CBTi has high quality evidence demonstrating its efficacy, with one study reporting a positive treatment response in 60% and remission in 40%. 

Impressive, yes, but the flip side of these figures is that clearly CBTi doesn’t work for everyone. So, what can we offer people who are still struggling to sleep? 

In the past doctors may have turned to benzodiazepines, but we now know the limitations and dangers of the drugs and avoid anything but very short-term prescribing. A 2021 BJGP paper examined the trend in prescription drugs for insomnia between 2011 and 2018 in Australian general practice, showing a fall in benzo and z-drug prescribing, but the consequences of this is a rise in prescribing of alternative treatments for insomnia, including amitriptyline, mirtazapine, quetiapine and melatonin.

Are these medications a suitable alternative? The problem we’ve had until now is a lack of evidence to guide clinicians as to whether these are actually helpful options. 

This knowledge gap is something a group of researchers from the Netherlands are trying to address and have just published in this month’s BJGP. 

This was a randomised controlled trial undertaken in general practice in and around Amsterdam, recruiting 80 adults aged 18-85yo with insomnia and sleep maintenance issues where non-pharmacological treatment (such as sleep hygiene and CBTi) failed to improve symptoms sufficiently. Participants were randomised to either off-label low dose mirtazapine (7.5 – 15mg nocte), amitriptyline (10 – 20mg nocte) or placebo for 16 weeks. The effects of insomnia were measured using the Insomnia Severity Index, checked at baseline, 6 weeks (the primary outcome), 12, 20 and 52 weeks.

For clinicians the results are good and bad news.

The good news is that at 6 weeks, patients taking mirtazapine had a statistically significant AND clinically relevant improvement in their insomnia scores compared to placebo (52% vs 14%) and recovery rates (56% vs 14%). For context, on average patients reported sleep duration was 1 hour longer on mirtazapine than placebo. Amitriptyline wasn’t too far behind at 40% improvement and 36% recovery but due to the small sample size this failed to reach statistical significance. Reassuringly no serious adverse events were identified as a result of the treatments.

The bad news is that by 12 weeks and any point thereafter, there was no difference between medications or placebo – all methods showed significant improvement in insomnia scores. Of course, we could view this as good news! 53% improved with placebo and 45% recovered by week 12.

This data is reassuring then that mirtazapine could provide an effective short-term option where this was felt necessary, and that around half of patients will have improved regardless of what we do. It does raise an important issue. 

The authors counsel against longer-term use of mirtazapine given the lack of benefit over placebo, but would we expect the same improvements with ‘usual care’? This data cannot answer that question but one suspects not. Does this inevitably lead us to prescribe such medications longer-term? Like much of general practice, this is a difficult area with no easy answers.

On the subject of no easy answers, what about the half of patients who continue to have issues with their sleep? Clearly, there is more work to be done for science in the arena of insomnia. Try not to lose sleep over it.