From the Editor
After completing a course of psychotherapy, he felt better and began a part-time job. The employment gave him a sense of purpose. As a physician, I could both sympathize and empathize; for many, work is a meaningful part of life, after all.
But does psychotherapy necessarily result in employment? Does therapy pay for itself with economic benefits? Otto R. F. Smith (of NLA University College) and his co-authors attempt to answer these questions in an impressive new paper in Psychological Medicine. They report on an RCT involving more than 700 Norwegian participants who were randomized to a psychotherapy program (modeled after the UK’s IAPT service) or to treatment as usual. The authors used administrative databases to analyze employment, income, and the economic benefit. “The results support the societal economic benefit of investing in IAPT-like services.” We consider the paper and its implications.
In the second selection, Dr. Calina Ouliaris (of Macquarie University) and her co-authors look at the zero suicide approach. In a Commentary for The British Journal of Psychiatry, they argue that – despite being studied and implemented in several places – it lacks evidence. “The Zero Suicide Framework is an arguably vague framework with a scant evidence base, particularly for application in healthcare settings. Despite this, the concrete goal championed… that of ‘zero suicides’, is appealing and has been widely promulgated in mental health services, ahead of evidence for the same.”
Finally, in the third selection, Dr. Chris Y. Kim (of the University of Toronto) and his co-authors weigh the use of melatonin for children and adolescents. In The Canadian Journal of Psychiatry, they are cautious, in part because of the lack of consistency of over-the-counter melatonin. “Melatonin used as a hypnotic agent for the treatment of insomnia is controversial.”
DG
Selection 1: “Cost–benefit of IAPT Norway and effects on work-related outcomes and health care utilization: results from a randomized controlled trial using registry-based data”
Otto R. F. Smith, David M. Clark, Gunnel Hensing, et al.
Psychological Medicine, 13 March 2025
The English ‘NHS Talking Therapies for Anxiety and Depression’ programme (formerly known as IAPT) represents one of the largest initiatives to increase access to evidence-based care for depression and anxiety, through a substantial investment in training new therapists and implementing a stepped-care model of empirically supported treatments nationwide. Inspired by its impressive results and scalability, similar models are now being rolled out in several other, including the ‘Prompt Mental Health Care’ (PMHC) program in Norway. Research have shown that both the IAPT and PMHC treatments are associated with substantial improvement in mental health symptoms and functioning…
A key argument for the viability and scalability of IAPT is that it will pay for itself through gains achieved by increased work participation, tax receipts, and reduced healthcare utilization. Return on investment analyses provide a strong case for investing in treatment for anxiety and depression. However, these are partly based on modeling and assumptions that have not been fully tested. Efficacy trials of psychological interventions for depression outside of IAPT-like services show that reduction in symptoms does not necessarily translate to increased work participation. According to a recent Cochrane review, there is only low-certainty evidence that such interventions may reduce sick leave days compared to usual care among employees with depression…
So begins a paper by Smith et al.
Here’s what they did:
- They conducted an RCT with parallel assignments for access to psychotherapy and treatment as usual to analyze work-related outcomes (including income) and healthcare use, and to estimate the cost-benefit of that program.
- To do this, they used several administrative databases, drawing data for up to five years following the intervention. (!)
- The intervention consisted of group-based psychoeducation (the primary treatment), individual CBT, guided self-help, or some combination of these.
- Primary outcome: employment.
- Statistical analyses included Bayesian estimation with 90% credibility intervals and posterior probabilities. They also did cost-benefit analysis.
Here’s what they found:
- 527 patients were randomized to the PMHC condition and 247 to the TAU.
- “The PMHC group was more likely than the TAU group to be in regular work without receiving welfare benefits in 2019–2022 (1.27 ≤ OR ≤ 1.43).” See figure below for income data.
- There was some evidence that the PMHC group spent less on healthcare.
- The estimated average cost per PMHC patient was 17,054 NOK; results from the regression analysis indicated that the average economic gain in favor of PMHC was 89,680 NOK. Thus, the benefit–cost ratio (that is, economic gain relative to intervention cost) was estimated at 5.26. (!!)
A few thoughts:
1. This is an impressive study, with good data, published in a solid journal.
2. The main finding in a sentence: participants who received the intervention were more likely to work and to earn more, and less likely to receive welfare benefits; overall, PMHC was cost-effective.
3. The results aren’t surprising – it makes sense that care, including psychotherapy, would result in more wages and less disability – but it is reassuring.
4. Smith and his colleagues are walking on a familiar path, of course. Many have done estimations of the cost-benefit of mental health services. That said, this study is an RCT. (!)
5. There are many reasons to fund mental health services. This paper strengthens the economic argument, something particularly important for policymakers.
6. Like all studies, there are limitations. The authors note several, including some estimations with wage data. And we shouldn’t overgeneralize. They showed the economic advantages of a specific program in one country. Still…
The full Psychological Medicine paper can be found here:
Selection 2: “The zero suicide approach: style over substance”
Calina Ouliaris, Pramudie Gunaratne, Christopher Ryan, et al.
The British Journal of Psychiatry, May 2025
The aspirational target of ‘zero suicides’ has been used to drive mental health system reform in many countries with the belief that all deaths from suicide in healthcare settings are preventable. Such approaches and models narrowly focus on suicide prevention as a key indicator for mental health system performance. One of the most well-known approaches, the Zero Suicide Framework (ZSF), is both an aspirational target and a system-wide approach for suicide prevention within health services that was first developed in the USA. The ZSF has been lauded as an evidence-based approach to suicide prevention within health services.
So begins a paper by Ouliaris et al.
They analyze two zero suicide programs:
- US Air Force. “An early pre–post-secondary analysis of the US Air Force administrative data-sets indicated a 33% reduction in suicide deaths during the intervention (1997–2002) relative to the period before (1990–1996). However, these gains were not sustained: there was a spike in suicide rates in 2004 and a general upward trend of suicide rates in the 2010s. The US Air Force Program was gradually phased out in a few years following 2002.”
- Henry Ford. “The Perfect Depression Care initiative was developed by the Henry Ford Health System with the key goal of suicide elimination… The initiative was described as ‘landmark’, leading to ‘zero suicides for 18 months in 2009–2010, and a statistically significant reduction in suicide rates within Henry Ford from its inception’. The accompanying study, a pre–post-secondary analysis of administrative data from a single organisation in Detroit, reported that the rate of suicide in the patient population decreased by 75%, from 89 per 100 000 at baseline (in 2000) to 22 per 100 000 for the 4-year follow-up interval (the average rate for 2002–2005)…” They note problems with the data, adding: “earlier correspondence by the authors to the editor of the JAMA revealed a total of 35 suicides across the decade. This indicates that the dramatic decline illustrated in the [literature] approximately represents a reduction from eight to zero suicides… “
Thus, with small numbers, they are cool to the argument of evidence for zero suicide. “There is scant evidence demonstrating the success of any individual programme in achieving the stated aim of suicide reduction.”
Despite that, they write about the enthusiasm for the concept in Australia. “The Queensland State Government has explicitly adopted the ZSF while other states, such as New South Wales, Victoria and South Australia, have implemented strategies with similar aspirations to work towards zero suicides.” They note the Australian Federal Government allocated $46 million dollars in the 2022–2023 budget to “associated initiatives.”
Again, they aren’t persuaded by the data. They focus on an experiment in Queensland built on a pathway that “lasted, on average, for 16 days and was found to be most efficacious in the first 14 days, with higher rates of repeated suicide attempts at 90 days in both groups. Notably, this study did not demonstrate evidence of decrease in actual suicides, only decreased re-presentations with suicide attempts to the hospital emergency department, as one would arguably expect when receiving face-to-face follow-up care following an attempt.”
Importantly, they question the approach itself. “When suicide is selected as an end-point driver of mental health care reform, there are many potential adverse consequences that we must consider. First, given the broad objective of suicide prevention, there is a risk of surrogation effect in which concrete metrics of suicide rates, suicide attempts and suicidal thoughts replace the ultimate goal of suicide prevention, particularly where healthcare systems readily accept such substitution. This would then misguide the direction of mental healthcare provision and reforms. For instance, where suicidal ideation or attempts are designated as key items that drive access to acute care and complex services, patients who are not suicidal but equally require specialist mental health support (e.g. those with severe anxiety or depression) are likely to experience barriers in accessing care. This may have the paradoxical effect of exacerbating mental ill health, a risk factor for suicide.”
“Suicide prevention requires more than healthcare solutions… Suicide prevention efforts must be multifaceted and involve a whole-of-government and whole-of-community approach given the important social determinants of mental ill health and suicide. These are the factors that affect social, economic and physical environments; for example, poverty, unemployment, homelessness, substance use and domestic violence. Thus, strategies must engage all levels of the health ecosystem and non-health sectors, including media regulations, means restriction, drug and alcohol laws and gatekeeper training.”
A few thoughts:
1. This is a well-argued Commentary.
2. The above summary doesn’t quite capture their nuanced review of the Air Force and Henry Ford Health System experiences.
3. Is the very concept itself unhelpful? The authors argue yes. Should an oncology ward aim for zero cancer deaths? By creating an unrealistic goal, are we blinding us to important aspects of care?
4. To play the Devil’s advocate: can a goal (even if not fully possible) help clarify purpose and thus focus efforts? After all, isn’t every suicide a tragedy?
5. Suicide prevention has been considered in past Readings. In late 2024, we analyzed the CJP paper on a suicide barrier and completed suicides by Sinyor et al. That Reading can be found here:
The full British Journal of Psychiatry commentary can be found here:
Selection 3: “The Use and Misuse of Over-the-Counter Melatonin in Children and Adolescents: A Commentary”
Chris Y. Kim, Persis Yousef, Royi Gilad, et al.
The Canadian Journal of Psychiatry, 21 May 2025 Online First
Melatonin is a naturally occurring hormone in humans produced primarily by the pineal gland. The natural rise of melatonin occurs in the early evening and ‘sends a message’ for the person to go to sleep in approximately three hours. Melatonin is a versatile neurohormone that has been recognized to have multiple functions in children and adolescents. The purified forms of melatonin are widely used as prescription medications in sleep clinics around the globe (but not widely available in Canada) to treat Delayed Sleep Phase Syndrome (DSPS), which affects up to 16% of teenagers. In children, it is primarily used to treat insomnia and circadian rhythm disorders. Over-the-counter (OTC) and off-label melatonin have been increasingly used to treat pediatric insomnia in North America because of its supposed benign safety profile.
So begins a paper by Kim et al.
They note the importance of a good history. “Pediatric insomnia may be exacerbated by another sleep disorder such as obstructive sleep apnea, restless legs syndrome, periodic leg movement disorder and comorbid medical and psychiatric disorders. Before using melatonin to treat pediatric insomnia, it is reasonable to rule out other disorders, which may precipitate and perpetuate sleep disruptions.
They draw on recommendations. “If an adolescent shows signs of DSPS, such as frequently getting up late in the morning and catching up on sleep on weekends, it is important to evaluate them using the Dim Light Melatonin Onset (DLMO) test to gauge the extent of the phase shift. Treating DSPS with melatonin, with close monitoring by a physician, improves sleep. For optimal chronobiotic use, melatonin should be taken about three hours before bedtime and it can be effective in DSPS when given for 4 to 6 weeks at 7 PM.”
But they offer a note of caution. “There are limited data on long-term effects of melatonin in children and adolescents. Notably, the effects of melatonin on bones and sexual maturation are not adequately understood.”
They comment on the uneven quality of melatonin.
- Supplements. “Melatonin is available as an unregulated OTC supplement often used in children and adolescents in Canada and the United States (US). Erland and Saxena tested 31 OTC melatonin supplements purchased from grocery stores and pharmacies in Guelph, Ontario, Canada, and analyzed their content. The study showed that there is a high variability in melatonin concentrations, ranging from −83% to +478% of the labelled concentration.”
- Gummy products. “More recently, Cohen et al. analyzed 25 melatonin gummy products available in the US market. Of these, one product did not contain detectable levels of melatonin, but had 31.3 mg of Cannabidiol (CBD). In the remaining products, the quantity of melatonin ranged from 1.3 mg to 13.1 mg per serving size. In products that contained melatonin, the quantity of melatonin ranged from 74% to 347% of the labeled quantity…”
A few thoughts:
1. This is an excellent and practical summary.
2. Working with Ilana Rosen, they produced a helpful visual abstract:
3. They are clear in their advice: “From a clinician’s perspective, such variability in quantity and the presence of contaminants in OTC melatonin limit its use as a reliable therapeutic option for consistent outcomes in both children and adults.”
4. This paper is about melatonin for children and adolescents – but the larger point about the unregulated market for supplements and edibles applies to all age groups (and, of course, to more than just melatonin).
The full Canadian Journal of Psychiatry paper can be found here:
https://journals.sagepub.com/doi/full/10.1177/07067437251340683
Reading of the Week. Every week I pick articles and papers from the world of Psychiatry.
Recent Comments