Hostname: page-component-77c89778f8-cnmwb Total loading time: 0 Render date: 2024-07-22T10:42:01.688Z Has data issue: false hasContentIssue false

Psychotherapies should be assessed for both benefit and harm

Published online by Cambridge University Press:  02 January 2018

Jan Scott*
Department of Academic Psychiatry, Institute of Neuroscience, Newcastle University, UK
Allan H. Young
Centre for Affective Disorders, Department of Psychological Medicine, Institute of Psychiatry, Psychology and Neurosciences, King's College London, UK
Professor Jan Scott, Academic Psychiatry, Wolfson Unit, Campus for Vitality & Ageing, Westgate Road, Newcastle upon Tyne NE4 6BE, UK. Email:
Rights & Permissions [Opens in a new window]


The past four decades have witnessed a transformation in research on the benefits of psychological therapies. However, even though therapists highlight that negative and adverse effects are seen in day-to-day practice, research on the negative effects of psychotherapy is insufficient. Given the unrelenting popularity of therapies, the argument for examining the adverse effects of psychotherapy would seem to be compelling. Such a strategy would extend beyond supervision of individual therapists to the introduction of monitoring systems that allow for a more systematic examination of failed psychotherapy interventions (such as exist for medication prescribing). The starting point could be the development of a consensus on how to define, classify and assess psychotherapy side-effects, unwanted events, adverse reactions, etc. This would provide a conceptual framework for communication, monitoring and research. This approach should not be viewed as an attack on therapies: every branch of medicine learns from mistakes, the same must surely be true for psychological treatments.

Copyright © Royal College of Psychiatrists, 2016 

Over the past 40 years, enormous progress has been made in psychotherapy research and there have been repeated demonstrations of positive effects for a range of psychological interventions, especially in the treatment of depression and anxiety disorders. Reference Cuijpers, Sijbrandij, Koole, Huibers and Berking1,Reference Clarke, Mayo-Wilson, Kenny and Pilling2 However, the study of adverse effects has not received the same level of attention. Reference Barlow3 This is despite the fact that even the earliest meta-analyses identified that, of nearly 500 outcomes studied, about 10% of the effect sizes were negative. Reference Smith, Glass and Miller4,Reference Shapiro and Shapiro5 Therapists highlight that negative or adverse effects are seen in day-to-day practice; Reference Castonguay, Boswell, Constantino, Goldfried and Hill6,Reference Nutt and Sharpe7 thus the case for systematically examining the adverse effects of psychotherapy would seem to be compelling.

It is unsurprising that psychotherapy can have negative outcomes; indeed, it simply indicates that psychological treatments are similar to almost every other therapeutic intervention used in medicine or mental health, from aspirin to brain surgery. Reference Mohr8 However, what sets psychotherapies apart is that they are frequently recommended as the treatment of choice for a wide variety of patients, and often in preference to medications, because of the presumption that therapy will be helpful to all patients, or at the very least will not have any adverse effects. Reference Nutt and Sharpe7,Reference Mohr8 This is in marked contrast to prevailing social and media representations of psychotropic medications. Reference Nutt and Sharpe7

One reason for the relative lack of scrutiny of therapies and the rigorous scrutiny of medications is the values that are brought to bear by three ‘interested parties’: society, the individual, and health professionals Reference Strupp and Hadley9 . Increasingly, publications used to inform patient choice on available interventions for common mental disorders suggest that therapies and medications are equally efficacious, but that medications have side-effects, adverse effects or are addictive, whereas psychotherapy is nearly always portrayed with the absence of such issues. Reference Nutt and Sharpe7

Negative perceptions of medications are probably increased by regulatory requirements to identify drug harms as well as benefits in any scientific communications, and this feeds into the view that they are dangerous. The US Food and Drug Administration (FDA) requires side-effects or adverse effects to be listed alongside the benefits of a drug, and the UK regulators who undertake post- marketing surveillance of medications (the Committee on Safety of Medicines) amended the ‘yellow card’ scheme for reporting adverse effects of drugs, so that patients have been able to use them since about 2004. Reference Nutt and Sharpe7 Although these strategies are entirely reasonable, no such systems or requirements are in place for therapies. Reference Barlow3,Reference Nutt and Sharpe7

One problem in trying to examine the issues of negative effects of psychological therapy is that the research approaches that are suitable for identifying positive effects often obscure negative effects, Reference Barlow3,Reference Dimidjian and Hollon10,Reference Linden11 an issue that is compounded by the lack of a theoretical concept on how to define, classify and assess psychotherapy side-effects. Reference Dimidjian and Hollon10,Reference Linden11 This does not mean that therapy researchers are complacent or negligent and indeed when problems have been identified changes in clinical practice have been instigated. One of the best examples is the transformation in perception of ‘critical incident stress debriefing’ (CISD), which was used routinely to help people immediately after a traumatic event. Over time it became apparent that significant subgroups of people who were treated with CISD actually experienced deterioration rather than an improvement in their psychological well-being, leading to changes in clinical guideline recommendations. Reference Lilienfeld12 However, in the absence of empirical predictors of likely adverse effects of therapy, placing a greater emphasis on clinical monitoring may provide a means to better understand the causes of negative or iatrogenic effects and to differentiate them from lack of therapeutic efficacy or from short-lived (but distressing) therapy side-effects. Reference Barlow3

In the paper by Crawford et al, Reference Crawford, Thana, Farquharson, Palmer, Hancock and Bassett13 about 1 in 20 adults offered therapies via the National Health Service or Improving Access to Psychological Therapies reported negative effects. Younger adults, individuals from minority ethnic groups (especially from Black, Asian and Chinese populations) and ‘non-heterosexuals’ (from lesbian, gay, bisexual, trans-sexual and transgender populations) were up to three times more likely to report negative effects than other individuals; unfortunately, other characteristics that can predict negative experiences, such as education and income, Reference Lambert and Barley14 were not examined. In terms of types of therapy, self-reported rates of negative effects ranged from 4% for cognitive–behavioural therapy to 9% for psychodynamic therapy. Number of therapy sessions did not predict negative effects, but insufficient information about the nature of the therapy or rationale for the interventions was associated with negative experiences of therapy.

Crawford and colleagues go to some lengths to differentiate negative effects from side-effects or lack of efficacy, but the survey did not examine whether any therapist characteristics contributed to the patients' negative experiences of therapy. For example, it is known that therapeutic alliance consistently predicts satisfaction with the relationship and with clinical outcome, and also that patients who report negative effects are often seen by therapists who are slow to recognise the severity of the patients' disturbance and their magnitude of need. Reference Scott15 This is important, as it is clear that expertise (i.e. clinical skilfulness in delivering the therapy) rather than experience (years of practicing a therapy) can make a critical difference to a patient's experience. Reference Imel, Baldwin, Atkins, Owen, Baardseth and Wampold16 Also, several studies high-light that the general competence of a therapist is not synonymous with either their cultural competence Reference Owen, Drinane, Tao, Adelson, Hook and Davis17,Reference Campos and Goldfried18 or level of comfort in discussing sexuality. Reference Harris and Hays19,Reference McHugh, Whitton, Peckham, Welge and Otto20


Empirical research on the negative effects of psychotherapy is insufficient, partly because there is a lack of a coherent framework for defining, discussing and monitoring issues such as unwanted events, adverse reactions, malpractice reactions or deterioration of illness. Reference Linden11 However, given the unrelenting popularity of therapies as a treatment for common mental disorders in primary and secondary care, it is important to develop a strategy for examining failed psychotherapy interventions. Reference Barlow3,Reference Mohr8 This is not an attempt to undermine therapies, but is likely to be beneficial. First, because it is likely to lead to improvements in techniques and practice, Reference Barlow3,Reference Mohr8 and second, because every branch of medicine learns from its mistakes; Reference Strupp and Hadley9,Reference Dimidjian and Hollon10 it is inconceivable that the same is not true for psychotherapies.


See pp. 210–212 and 260–265, this issue.

Declaration of interest

J.S. has undertaken research on the use of psychological interventions, especially cognitive–behavioural therapy (CBT) either alone or in combination with medications, for common and for severe mental disorders. She is a Distinguished Founding Fellow of the Academy of Cognitive Therapy and has contributed to several committees examining psychotherapy practice, such as the recent proposals on supervision competencies for psychotherapists (see A.H.Y has received payment for lectures and advisory boards for all major pharmaceutical companies with drugs used in affective and related disorders.


1 Cuijpers, P, Sijbrandij, M, Koole, S, Huibers, M, Berking, M, Anderss Psychological treatment of generalized anxiety disorder: a meta-analysis. Clin Psychol Rev 2014; 34: 130–40.CrossRefGoogle Scholar
2 Clarke, K, Mayo-Wilson, E, Kenny, J, Pilling, S. Can non-pharmacological interventions prevent relapse in adults who have recovered from depression? A systematic review and meta-analysis of randomised controlled trials. Clin Psychol Rev 2015; 39: 5870.CrossRefGoogle ScholarPubMed
3 Barlow, DH. Negative effects from psychological treatments. Am Psychol 2010; 65: 13–9.CrossRefGoogle ScholarPubMed
4 Smith, ML, Glass, GV, Miller, TI. The Benefits of Psychotherapy. Johns Hopkins University Press, 1980.Google Scholar
5 Shapiro, DA, Shapiro, D. Meta-analysis of comparative therapy outcome studies: a replication and refinement. Psychol Bull 1982; 92: 581604.CrossRefGoogle ScholarPubMed
6 Castonguay, L, Boswell, J, Constantino, M, Goldfried, M, Hill, C. Training implications of harmful effects of psychological treatments. Am Psychol 2010; 65: 3449.CrossRefGoogle ScholarPubMed
7 Nutt, DJ, Sharpe, M. Uncritical positive regard? Issues in the efficacy and safety of psychotherapy. J Psychopharmacol 2008; 22: 36.CrossRefGoogle ScholarPubMed
8 Mohr, DC. Negative outcome in psychotherapy: a critical review. Clin Psychol Sci Pract 1995; 2: 127.CrossRefGoogle Scholar
9 Strupp, H, Hadley, S. A tripartite model of mental health and therapeutic outcomes: with special reference to negative effects in psychotherapy. Am Psychol 1977; 32: 187–96.CrossRefGoogle ScholarPubMed
10 Dimidjian, S, Hollon, S. How would we know if psychotherapy were harmful? Am Psychol 2010; 65: 2133.CrossRefGoogle ScholarPubMed
11 Linden, M. How to define, find and classify side effects in psychotherapy: from unwanted events to adverse treatment reactions. In Clin Psychol Psychother 2012; 20: 286–96.Google ScholarPubMed
12 Lilienfeld, S. Psychological treatments that cause harm. Perspect Psychol Sci 2007; 2: 5370.CrossRefGoogle ScholarPubMed
13 Crawford, MJ, Thana, L, Farquharson, L, Palmer, L, Hancock, E, Bassett, P, et al. Patient experience of negative effects of psychological treatment: results of a national survey. Br J Psychiatry 2016; 208: 260–5.CrossRefGoogle ScholarPubMed
14 Lambert, MJ, Barley, DE. Research summary on the therapeutic relationship and psychotherapy outcome. Psychotherapy 2001; 38: 357–61.CrossRefGoogle Scholar
15 Scott, J. Cognitive–behavioural therapy for severe mental disorders: back to the future? Br J Psychiatry 2008; 192: 401–3.CrossRefGoogle ScholarPubMed
16 Imel, ZE, Baldwin, S, Atkins, DC, Owen, J, Baardseth, T, Wampold, BE. Racial/ethnic disparities in therapist effectiveness: a conceptualization and initial study of cultural competence. J Couns Psychol 2011; 58: 290–8.CrossRefGoogle ScholarPubMed
17 Owen, J, Drinane, J, Tao, KW, Adelson, JL, Hook, JN, Davis, D, et al. Racial/ethnic disparities in client unilateral termination: the role of therapists' cultural comfort. Psychother Res 2015; 21: 110.Google Scholar
18 Campos, PE, Goldfried, MR. Introduction: perspectives on therapy with gay, lesbian, and bisexual clients. J Clin Psychol 2001; 57: 609–13.CrossRefGoogle Scholar
19 Harris, SM, Hays, KW. Family therapist comfort with and willingness to discuss client sexuality. J Marital Fam Ther 2008; 34: 239–50.CrossRefGoogle ScholarPubMed
20 McHugh, RK, Whitton, SW, Peckham, AD, Welge, JA, Otto, MW. Patient preference for psychological vs pharmacologic treatment of psychiatric disorders: a meta-analytic review. J Clin Psychiatry 2013; 74: 595602.CrossRefGoogle ScholarPubMed
Submit a response


No eLetters have been published for this article.