A core challenge in research is translating the results of the studies into actual policies and services. Many excellent interventions are developed and tested, but then never integrated into healthcare services (Wainberg et al., 2024) – why?
Let’s take an imaginary intervention we might test, for depression. Like most studies, we might examine symptom remission as our primary outcome. Unlike many studies, we may find exceptional evidence, that depressive symptoms decrease after our studied intervention. However, what is the next step, how do you motivate public (or indeed, private) investment from this?
Health services and governments around the world are prioritising mental health now more than ever (WHO, 2004; Department of Health (Ireland), 2024). One reason is because of the significant economic-costs associated with mental disorders – estimated to be about a 6.5% loss of the gross domestic product in Western Europe (Arias, Saxena and Verguet, 2022). Many excellent studies (including our imaginary one), may have found solutions to reducing the rates of mental disorders – but governments and services must consider costs, because investment in our study is a trade-off to not funding other interventions. Given that experimental and exploratory research doesn’t often prioritize economic considerations – our study, like many others, may never be brought forward to healthcare services.
This phenomenon, is a key target of this recent paper by Ullah et al (2025). They ran a cost-effectiveness study examining a cognitive behavioural therapy CBT) intervention, targeting postnatal depression in British South Asian women (the ROSHNI-2 trial). We blogged about the clinical effectiveness of this intervention yesterday, so you can read about that as well if you want: Culturally adapted CBT may lead to recovery from postnatal depression in British South Asian women.
Many excellent interventions are developed and tested, but then never integrated into healthcare services. Why?
Methods
The ROSHNI-2 trial is a multicentre randomised controlled trial (RCT) in the UK. The study participants were British South Asian women who had a child aged less than a year-old, and who met DSM-5 criteria for depression.
The intervention was the Positive Health Programme (PHP), a CBT intervention that involved 12 group sessions over 4 months. CBT focuses on changing negative thoughts, beliefs and behaviours, and is considered the “gold standard” of psychological therapies. For this study, the authors developed a culture specific CBT, so the sessions were delivered by two trained bilingual facilitators, and focused on the pressures of being a South Asian mother. The study was also developed by working with Patient and Public Involvement (PPI), who were involved at all stages of developing the intervention.
A unique challenge was faced by the study, as it ran in 2020, and had to move online due to the Covid-19 pandemic. Incidentally, this allowed the team to further calculate the costs of in-person and online session delivery.
A robust range of cost outcomes were collected, all through self-report at 4-months and 12-months. They measured quality of life (cost-utility), remission from depression (cost-effectiveness), healthcare service and social support use (resource use), and the costs of running the PHP, e.g. venue hire, staffing.
Three key analyses were conducted:
- Incremental costs (monetary expenses of the treatment): using γ-distributed generalised linear models.
- Incremental effects (health benefits/outcomes: using β-distributed generalised linear models.
- The incremental cost-effectiveness ratio (ICER): ICER considers the value for money, of an intervention per quality-adjusted life year. In the UK, the National Institute of Health and Care Excellence (NICE) considers between £20,000-£30,000 cost-effective.
This study design benefited from Patient Public Involvement and cost-effectiveness considerations.
Results
A total of 732 mothers participated, who were divided into groups of 368 receiving PHP, 364 receiving treatment as usual (TAU). Attendance of sessions was mixed, while ~52% attended most/all sessions, 13% attended none, and another 14% attended only a few sessions.
Read the clinical effectiveness findings in yesterday’s blog: Culturally adapted CBT may lead to recovery from postnatal depression in British South Asian women.
For this paper, data was gathered on the following outcomes:
- Resource use: Interestingly, resource use was higher in the PHP group, compared to TAU. The PHP group by chance already had slightly higher costs at baseline (PHP =£608, TAU=£596). Both had reduced significantly at 12-month follow-up (PHP=£274, TAU=£264), but the PHP group were still using more healthcare and social support services. The primary difference during the trial and follow-up was that the PHP group had a higher number of outpatient appointments.
- Running costs: the PHP cost £4063 per group, and £463 per person. The key costs were facilitator pay, crèche services (an important cost to facilitate the mothers attending the services), and venue hire.
Cost-utility and cost-effectiveness were calculated, as follows:
- Quality of life improvements: PHP showed a higher Quality-adjusted life-years (QALY) gain (0.036 95%CI [0.006 to 0.067]), at a mean cost difference of £712, relative to TAU. The ICER for the QALY was £19,601
- Symptom remission: Remission from depression was also higher in the PHP group at 4-months (0.126, 95%CI [0.049 to 0.203]), with a mean cost difference of £692 relative to TAU. The ICER for depression remission was £5,509.
- Platform differences: The authors examined online versus in-person PHP, compared to TAU, and found that online was much cheaper to run (£202 vs £22,295). They examined the ICER for non-attendance at any sessions, and relative to TAU this is a significantly higher cost.
Given that the ICER was £19,601 for one additional quality of life and £5,509 for remission from depression, on a willingness-to-pay basis, the PHP meets the NICE criteria for a cost-effective intervention on both counts.
The Positive Health Programme delivered to South Asian UK women with postnatal depression, showed promise as both a clinically effective and cost-effective intervention.
Conclusions
The Positive Health Programme is a promising intervention, which demonstrated effective improvements in quality of life and rates of postnatal depression, in a population which has traditionally been difficult to get engaged in treatment (Husain et al, 2024). This study is a vital part of moving such treatments from promising research, into practically delivered healthcare services, where they can do the most good.
Overall, PHP is more expensive to run than TAU. This is expected, as it’s an additional service with more costs associated with it. However, the PHP falls well within the NICE assessment (£20,000-30,000) for cost effectiveness, £19,601. They considered a number of challenges e.g. non-attendance, under-estimates of costs and over-estimates of utilities for missing data, and all still showed that PHP is a cost effective intervention.
Strengths and limitations
This study is a positive and pragmatic step forward, into real-world evaluation of psychological interventions. Many studies only focus on whether an intervention affects symptom remission, without taking a practical approach to the delivery of an intervention that may show promise. By contrast, cost-effectiveness measures were incorporated by the authors here from the outset of design and data collection, and future studies should consider this blueprint.
The second important factor is that this is a timely and cost-effective study. The researchers are now in a strong position to move forward with the PHP intervention as a larger scale trial e.g. nationwide. By examining all costs, including risks e.g. non-attendance, they can present a realistic estimate to governmental organisations, and provide information such groups need to consider implementation.
Finally, there were some interesting, albeit preliminary evidence, for the potential of the PHP as an online intervention. Because this was a transition of necessity (COVID-19 pandemic), the measurements were based on those who had no-online, some-online and all online. This isn’t a perfect RCT, and group allocation was driven by time of enrolment more than other key factors. However, it was more affordable, with relatively similar outcomes. It offers a promising insight into the value of adaptation of interventions e.g. one of the highest costs for running the PHP was childcare while the mothers were in the session, an issue which may be lessened in an online session. However, the study didn’t examine factors such as whether having children at home while doing online sessions was an issue of the treatment. More research on PHP as an online intervention needs to be done, but this is still a promising result.
There are some limitations. First, this is a relatively small sample (n=732), and there is still a need for a larger study e.g. a regional nationwide trial, before this intervention may move forwards as an available treatment in the healthcare service. Secondly, cost-effectiveness analysis doesn’t provide all information of associated costs with scaling the intervention.
For example, training costs of the facilitators was a significant but acceptable cost in this intervention, however in a larger scale study, you would need to train more facilitators, who either would be employed as full-time staff, or as special trained currently employed staff. While full-time staff should be more freely available, each changeover of staff would need more training, so it would be desirable to have a bank of permanent full-time in person and online trained facilitators – which can be difficult to achieve among staff already employed within the healthcare system with other responsibilities.
Finally, it is worth considering the significant benefit of the patient and public involvement group to the development of this study – it is likely that for regional service provision or service provision to particular social communities, truly competent rollout will require further patient and public input, which raises its own costs and time challenges.
You can read more about the strengths and weaknesses of the clinical effectiveness ROSHNI-2 paper in yesterday’s blog: Culturally adapted CBT may lead to recovery from postnatal depression in British South Asian women.
While the intervention has been tested for online delivery, facilitator training and costs remain to be assessed.
Implications for practice
This cost-effectiveness study, and its predecessor examining the clinical effectiveness of the ROSHNI-2 trial PHP intervention (Husain et al, 2024), both demonstrate that it’s possible to create and deliver effective therapies that are developed by and with ethnic minority groups. This is an important step forward to making mental health services valuable and appropriate for everyone, not just the majority. It’s important for researchers and clinicians to be aware of the harms when we label groups as “difficult to reach”, and then don’t attempt to reach them. Ullah and colleagues showed how it is possible, and why, when you do it, you need to consider all aspects of the intervention, to improve the possibility of it moving from research to clinical services.
While there are many steps we need to take between this study and it being available generally in clinical services, it does offer promising insights into the possibility of implementing such programmes. £19,601 above currently available treatment is not a small-cost, but at the wide-scale that NICE measures cost-effectiveness, it’s an affordable one. Postnatal depression can have serious health implications for mother and infant (Slomian et al, 2019; Suryawanshi & Pajai, 2022), including developmental and physical health complications for the child, and the long-term physical and mental health of the mother. Interventions that address this, and focus on those our healthcare system can miss, are some of the most important areas of mental health research. The PHP intervention and this study, are an important step in bridging this gap.
Effective and affordable interventions for postnatal depression can help both mothers and their families.
Statement of interests
No conflicts of interest
Links
Primary paper
Ullah A, Lunat F, Brugha T, et al. Cost-effectiveness of a group psychological intervention for postnatal depression in British South Asian women: an economic evaluation from the ROSHNI-2 trial. Lancet Psychiatry. Published online 2025.
Other references
Wainberg ML, Gouveia L, McKinnon K. Generating better implementation evidence to improve mental health care everywhere. Lancet Psychiatry. 2024;11(5):317-319. doi:10.1016/S2215-0366(24)00090-7
World Health Organization. Prevention of mental disorders: Effective interventions and policy options: Summary report. Published online 2004.
Department of Health. Sharing the Vision; A Mental Health Policy for Everyone. Department of Health, Ireland; 2020.
Arias D, Saxena S, Verguet S. Quantifying the global burden of mental disorders and their economic value. eClinicalMedicine. 2022;54. doi:10.1016/j.eclinm.2022.101675
David D, Cristea I, Hofmann SG. Why Cognitive Behavioral Therapy Is the Current Gold Standard of Psychotherapy. Front Psychiatry. 2018;9:4. doi:10.3389/fpsyt.2018.00004
Husain N, Lunat F, Lovell K, et al. Efficacy of a culturally adapted, cognitive behavioural therapy-based intervention for postnatal depression in British south Asian women (ROSHNI-2): a multicentre, randomised controlled trial. Lancet Lond Engl. 2024;404(10461):1430-1443. doi:10.1016/S0140-6736(24)01612-X
Slomian J, Honvo G, Emonts P, Reginster JY, Bruyère O. Consequences of maternal postpartum depression: A systematic review of maternal and infant outcomes. Womens Health. 2019;15:1745506519844044. doi:10.1177/1745506519844044
Suryawanshi O, Pajai S. A Comprehensive Review on Postpartum Depression. Cureus. 2022 Dec 20;14(12):e32745. doi:10.7759/cureus.32745
