In clinical practice, service-users meeting criteria for just one psychiatric diagnosis are rare – sometimes we wonder if unicorns might be more common! These high rates of comorbidity are just one reason for a groundswell of opinion that diagnostic categories may not be the optimal way to characterise mental health problems, despite their heuristic utility in the healthcare system. This has led to a burgeoning of so-called ‘transdiagnostic research’ in the past 15 or so years, investigating novel classification systems, fundamental biopsychosocial processes, and new clinical interventions, that all cut across traditional diagnostic boundaries.
Recent Mental Elf Blogs have highlighted some of this research, including reviews demonstrating equivalent treatment effects between transdiagnostic and disorder-specific treatments for anxiety, summaries and theoretical models for transdiagnostic processes across common mental health problems, and similarities between processes in anxiety and depression. The transdiagnostic field is at a critical juncture in terms of advancing our understanding of mental health problems and using this knowledge to improve interventions. Systematic reviews and critiques play an important role in such emerging areas of research that may represent a paradigm shift, and can provide constructive guidance for improving the field. This is why the recent review of transdiagnostic approaches in World Psychiatry led by Fusar-Poli et al. (2019a) is potentially so important. Unfortunately, in our view the review falls short of meeting these ambitions.
Fusar-Poli et al. set out to present a “broad systematic review of transdiagnostic psychiatry”, in order to examine “whether transdiagnostic research meets such high expectations for delivering a radical paradigm shift that impacts classification and clinical care.” Since the publication of this initial review, Patrick McGorry and Barnaby Nelson (McGorry and Nelson, 2019), as well as Warren Mansell (Mansell, 2019), have replied in the same journal with Letters to the Editor offering critiques of the original paper. The lead author of the review, Paolo Fusar-Poli has provided a rejoinder to the editor expanding on the original rationales for the review’s conclusions (Fusar-Poli et al., 2019b). In this blog we evaluate the review, the responses and the rejoinder.
We begin by summarising the Fusar-Poli et al. target article (Fusar-Poli et al., 2019a).
A systematic review was conducted, restricted to studies listed on Web of Science that included “transdiagnostic” in the study title, and that were filtered for ‘Psychiatry’. The authors also searched the reference lists of articles for additional studies.
Review inclusion criteria:
- Original individual articles since 2004, with no restriction on study design or topic
- Clear and primary focus on transdiagnostic approach, demonstrated by using the word “transdiagnostic” in the title
The authors compared studies to what they termed the “Mansell criteria” to define transdiagnostic approaches in psychiatry. They are:
- Presence of a clinical population
- Presence of at least four different mental disorders
- Presence of a nonclinical sample, and
- Demonstration of the transdiagnostic construct in all mental disorders investigated.
Overall, the analyses in the review were largely descriptive and quantitative comparisons were deemed not possible given the variety of methods in the included studies.
The search identified 627 studies for title/abstract screening, 239 of these for full text screening, and a final sample of 111 studies (original research published 2004-2018). The review authors identified a heterogenous mix of interventional studies, investigations of psychosocial processes, neuroscientific studies, prediction, and classification. The studies focused primarily on mood and anxiety disorders.
The identified studies were then evaluated according to the “Mansell criteria”. By that metric, only 3 out of the 111 studies were classified as being “truly transdiagnostic”. The authors reported that the quality of the studies was generally low, citing lack of pre-registration and biased/over-fitted models (Fusar-Poli et al., 2019a). In addition, only a few findings were externally replicated.
Conclusions of the review
The authors (Fusar-Poli et al., 2019a) stated that “no clear transdiagnostic construct that could be valid across all mental disorders and diagnostic spectra was identified”, and that the “transdiagnostic literature is heterogeneous and intrinsically incoherent”.
The so-called “TRANSD” recommendations made by the authors on the back of this review aim “to improve the consistency and quality of the next generation of transdiagnostic research” (Fusar-Poli et al., 2019a).
Reflections on the review and published responses in Letters to the Editor
What are systematic reviews for? In our view they should aspire to four things:
- To be firmly grounded in, and motivated by, the existing literature
- To meet the highest standards for capturing the relevant studies within the literature
- To evaluate the identified studies against fair, reasonable and appropriate criteria
- To draw balanced and constructive conclusions.
Here, we consider the review and responses in Letters to the Editor in these terms.
Reviews should reflect cutting edge developments in emerging research areas
The first requirement for a systematic review of an emerging field is that the review should be firmly grounded in the extant literature and scholarship within that domain. The good review will summarise the state-of-the-art and provide a platform for further development.
As anyone who has been even somewhat immersed in the transdiagnostic research literature appreciates, the field encompasses a number of related but also separable endeavours under the same transdiagnostic umbrella. At least three sub-domains stand out: efforts to develop and validate alternative nosological systems to the traditional diagnostic approach; efforts to identify and elucidate transdiagnostic biopsychosocial processes at various levels of analysis (genetic, molecular, neural, behavioural); and finally efforts to develop transdiagnostic clinical interventions.
Any systematic review of the field that uses broad search criteria should, at minimum, organise itself around these discrete themes. If not, the review will simply obscure the scientific and intellectual delineations that define the area of study (and indeed that define psychiatric scholarship more broadly).
The current systematic review by Fusar-Poli et al. merely pays lip service to these different areas of the transdiagnostic literature in its Introduction. These fleeting insights appear to play no role in guiding the literature search process, the evaluative criteria applied to the search results, nor the conclusions that are drawn from the review. This would perhaps be more understandable if these different transdiagnostic domains were hard to identify in the literature but this is far from the case. Nosological efforts include much cited work on the so called general psychopathology or ‘P’ factor (Caspi et al., 2014) as well as the efforts of the HiToP group – a consortium of over seventy distinguished international scholars (Kotov et al., 2017). Work on biopsychosocial processes includes the high-profile Research Domains Criteria (RDoC) initiative of the US National Institute of Mental Health. Finally, there already exist numerous systematic reviews and meta-analyses of transdiagnostic clinical interventions (e.g., Newby et al., 2015; Pearl & Norton, 2017).
Instead of presenting a review informed by these insights, Fusar-Poli et al., instead eschew any such organising principles to conduct what they term ‘an epistemological test’ of the transdiagnostic literature, which has unfortunately conflated nosology, mechanism and intervention.
Reviews should systematically review the literature
Any systematic review should, well, systematically review the pertinent literature.
However, Fusar-Poli et al. limited their search to studies that contained the word ‘transdiagnostic’ in the article title, ignoring guidelines for the proper conduct of systematic reviews (e.g. Atkinson & Cipriani, 2018). As both Mansell and McGorry & Nelson in their Letters argue, this is far too narrow. Imagine if one wanted to do a systematic review in the area of ‘mood disorders’ and decided to only review articles with that term in their title, setting aside literature only referring to, say, depression or bipolar disorder in the titles?
As a consequence, the review misses multiple studies and themes within the literature that are recognised by researchers as prototypical within the transdiagnostic domain.
Reviews should evaluate the findings appropriately
The third aspiration for any systematic review worth its salt is to evaluate the articles that are returned against fair and appropriate criteria. Again, serious questions must be asked about the approach to evaluation adopted by Fusar-Poli et al.
To illustrate, a key component of this evaluation concerned the perceived ‘quality’ of the identified articles. This involved establishing whether a pre-study protocol was available, if industry funding was involved, and if the findings had been externally and independently replicated. Within the domain of clinical science, these are perplexing indices of study quality. Rarely would studies, other than clinical trials, have a pre-published protocol (although this is changing with an increase in preregistration). Industry involvement outside of pharmacological interventions would be very unusual and if anything a potential source of concern due to putative conflicts of interest, rather than a marker of quality. Finally, independent replication, while a laudable aspiration, is a high hurdle for a relatively novel field and of course much harder to truly evaluate when the review search criteria miss much of the literature anyway.
Perhaps most perplexing though was the ‘literature analysis’ aspect in the review where the identified articles are benchmarked against what the authors term the “Mansell criteria”. These criteria were proposed in a 2004 book by Allison Harvey et al., on transdiagnostic cognitive-behavioural processes, identifying 12 processes that satisfied the criteria (e.g. that the process is implicated in at least 4 separate clinical disorders). According to Fusar-Poli et al.’s analysis only three of their identified 111 papers satisfied the “Mansell criteria”, leading them to conclude that the “transdiagnostic designation is applied in a loose and unstandardized way”.
It is unfortunate that criteria initially introduced in a publication first-authored by a senior female scientist (Allison Harvey) are named after her male co-author (albeit based on a later paper). More pertinently, it seems very strange to apply criteria explicitly developed to evaluate cognitive-behavioural processes to outputs from a systematic review which is not limited to reviewing such processes but also includes interventions, non-cognitive processes, and classification initiatives. As the review authors state, 45% of the 111 identified studies were intervention studies for which, what we might call the “Harvey Criteria”, are wholly inappropriate. Even for the 28% of studies identified in the review that do focus on processes, the application of the Harvey Criteria to individual studies seems an odd approach as the criteria were clearly designed to be applied across studies in order to identify transdiagnostic processes in the literature taken as a whole.
Reviews should provide constructive recommendations that reflect state-of-the-art science and practice
Fusar-Poli et al. sign off their review by concluding that “transdiagnostic research in psychiatry has, to date, been overenthusiastic and undercritical, heterogeneous, intrinsically incoherent and predominantly focused on a limited subset of mental disorders. It is grounded more in rediscoveries than true innovations, and it is demonstrably affected by conceptual biases”. Their resultant view is that “transdiagnostic research does not represent a credible paradigm shift that can impact the classification of or clinical care for mental disorders”. These conclusions are based on a lengthy list of very negative evaluations of the field in the Discussion of their review which form the basis of a set of ‘TRANSD recommendations’ for future work.
By any metric, this is a damning indictment indeed and we agree with Mansell and McGorry & Nelson that it seriously misrepresents the current state-of-the art of the transdiagnostic field. It is tempting to go through each concluding point in the review and outline why we feel it is misguided. However, in fact, the review’s conclusions were inevitable from the start. Conducting a systematic review where no attempt is made to organise the literature along established grounds within the discipline, and then carrying out a literature search based on a single key term being requisite in the article titles, will inevitably generate a highly selective and heterogenous mishmash of studies. Evaluating those studies against very specific criteria relevant only to a subdomain of the field (and thus to only a minority of identified studies in the review) then condemns most studies to fail to make the grade from the outset, before we even consider the list of other inappropriate quality metrics that are applied.
Last year, we had the pleasure to host a vibrant, intellectually sharp, and diverse meeting, The 2018 Conference on Transdiagnostic Approaches to Mental Health Challenges, which brought together researchers from all over the world and critically examined the relevant issues in the field. The #Transdx2018 summary and opinion podcasts captured by Mark Brown from the #BeyondTheRoom team can be found here, or you can listen to them below. Rather than imbibe the conclusions of this disappointing systematic review, we recommend anyone interested in a more balanced evaluation of the difficulties and challenges facing the transdiagnostic field going forward to perhaps start there.
Implications for practice
- Transdiagnostic research has the potential to better represent the clinical and scientific reality of mental health problems, reflecting the complexity and comorbidity that is the norm in clinical practice.
- The reality is that transdiagnostic approaches are already integrated into practice and reflect the clinical and personal realities of service-user preferences and treatment delivery.
- It is the responsibility of researchers to determine how the evidence can integrate this reality to facilitate improvement in understanding the structure of mental health problems and interventions.
Conflicts of interest
Melissa Black and Tim Dalgleish work in the Cognition, Emotion and Mental Health Programme at the MRC Cognition and Brain Sciences Unit, University of Cambridge, and are affiliated with Cambridgeshire and Peterborough NHS Foundation Trust. We have an overlap in research interests with the review authors, hosted the 2018 Conference on Transdiagnostic Approaches to Mental Health Challenges, and are currently working on a trial of a novel transdiagnostic psychological therapy (protocol can be found here).
Fusar-Poli P, Solmi M, Brondino N, Davies C, Chae C, Politi P, Borgwardt S, Lawrie SM, Parnas J, McGuire P (2019a) Transdiagnostic psychiatry: a systematic review. World Psychiatry. 18(2):192-207
McGorry PD, Nelson B. (2019). Transdiagnostic psychiatry: premature closure on a crucial pathway to clinical utility for psychiatric diagnosis. World Psychiatry. 18(3): 359-360
Mansell W. (2019) Transdiagnostic psychiatry goes above and beyond classification. World Psychiatry. 18(3): 360-361
Fusar-Poli P. (2019b). TRANSD recommendations: improving transdiagnostic research in psychiatry. World Psychiatry. 18(3): 361-362
Atkinson, L.Z., & Cipriani, A. (2018). How to carry out a literature search for a systematic review: a practical guide. BJPsych Advances, 24: 74-82.
Brown TA , Campbell LA , Lehman CL , et al. (2001). Current and lifetime comorbidity of the DSM-IV anxiety and mood disorders in a large clinical sample. Journal of Abnormal Psychology, 110:585–99. [APA abstract]
Caspi, A., Houts, R.M., Belsky, D.W., Goldman-Mellor, S.J., Harrington, H., Israel, S., Meier, M.H., Ramrakha, S., Shalev, I., Poulton, R., & Moffit, T.E. (2013). The p Factor: One General Psychopathology Factor in the Structure of Psychiatric Disorders? Clinical Psychological Science, 2: 119-137.
Kessler RC , Berglund P , Demler O , et al. (2005). Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Archives of General Psychiatry. 62:593–602. [PubMed abstract]
Harvey, AG, Watkins, E, Mansell, W, & Shafran, R. eds. (2004). Cognitive behavioural processes across psychological disorders: A transdiagnostic approach to research and treatment. Oxford University Press: Oxford.
Insel, T, Cuthbert, B, Garvey, M, Heinssen, R, Pine, DS. (2010). Research Domain Criteria (RDoC): Toward a New Classification Framework for Research on Mental Disorders. The American Journal of Psychiatry, 167, 7.
Kotov, R., Krueger, R. F., Watson, D., Achenbach, T. M., Althoff, R. R., Bagby, R. M., … & Eaton, N. R. (2017). The Hierarchical Taxonomy of Psychopathology (HiTOP): A dimensional alternative to traditional nosologies. Journal of Abnormal Psychology, 126, 454-477.
Mansell, W; Harvey, A; Watkins, E; Shafran, R; (2009) Conceptual foundations of the transdiagnostic approach to CBT. Journal of Cognitive Psychotherapy, 23 (1) pp. 6-19. [Springer abstract]
Newby, J.M., McKinnon, A., Kuyken, W., Gilbody, S., & Dalgleish, T. (2015). Systematic review and meta-analysis of transdiagnostic psychological treatments for anxiety and depressive disorders in adulthood. Clinical Psychology Review. 40: 91-110.
Pearl, S. & Norton, P. J. (2017). Transdiagnostic versus diagnosis specific cognitive behavioural therapies for anxiety: A meta-analysis. Journal of Anxiety Disorders, 46, 11-24.
Sauer-Zavala, S, Gutner, CA, Farchione, TJ, Boettcher, HT, Bullis, JR, Barlow, DH. (2017). Current Definitions of “Transdiagnostic” in Treatment Development: A Search for Consensus. Behavior Therapy, 48, 128-138.