Key Takeaways
What Is Bad Therapy?
"Bad therapy" has two common meanings: psychological treatment that leaves people worse off, and the title of journalist Abigail Shrier's book Bad Therapy: Why the Kids Aren't Growing Up, released in early 2024[7]. Both raise the same research question: how often does talk therapy cause harm, and which approaches carry the most risk?
A national survey in England and Wales offers the clearest estimate: 763 of 14,587 people treated for common mental disorders, or 5.2%, reported lasting bad effects from their therapy[1]. In the U.S., 21.6% of adults, or 57.3 million people, received mental health treatment in 2025[8]; how many adults have any mental illness is covered on our pillar page.
The evidence below separates three things that often get blurred: patients who worsen during therapy, specific treatments shown in trials to backfire, and therapists whose conduct breaks professional rules.
Iatrogenic harm
Harm caused by a treatment itself rather than by the underlying condition. In psychotherapy research it is distinguished from deterioration, which can also reflect the natural course of a disorder in people for whom treatment was ineffective[2].
Related Terms
Source: Foulkes and Stringaris, BJPsych Bulletin (2023)
Can Therapy Make You Worse?
Yes, for a minority of people. In clinical settings, roughly 3% to 10% of patients experience symptom deterioration during therapy[2].
That range overstates harm caused by therapy itself. Foulkes and Stringaris note it mixes iatrogenic harm with the natural course of a disorder in people the treatment did not help[2].
Lasting harm is a narrower measure. In the England and Wales survey, respondents who were unsure what type of therapy they had received were more likely to report lasting bad effects (odds ratio 1.51), while those who said they got enough information before therapy began were less likely to (odds ratio 0.65)[1].
People over 65 were less likely to report such effects, and sexual and ethnic minority respondents were more likely to[1]. The authors concluded that clinicians should discuss potential negative as well as positive effects before therapy starts[1].
The field was slow to study this. In a 2007 review, psychologist Scott Lilienfeld wrote that psychologists had until recently paid little attention to hazardous treatments, and argued for more emphasis on potentially dangerous therapies than on empirically supported ones[9].
People treated in England and Wales who reported lasting bad effects from psychological therapy
That is 763 of 14,587 survey respondents, or one in 20. The figure is self-reported, so it measures perceived lasting harm rather than harm confirmed by a clinician or a control group.
"One in 20 people responding to this survey reported lasting bad effects from psychological treatment."
Which Therapies Have Been Shown to Cause Harm?
Lilienfeld's provisional list of potentially harmful therapies (PHTs) sorted treatments into Level I, "probably harmful for some individuals," and Level II, "possibly harmful for some individuals"[9]. Both labels refer to harm in some individuals, not in every person who receives the treatment.
The strongest evidence comes from randomized trials. A 2002 Cochrane review of 15 trials found that single-session individual psychological debriefing after a traumatic event did not prevent post-traumatic stress disorder or reduce distress[3].
In one of those trials, people who were debriefed had a significantly higher risk of PTSD at one year (odds ratio 2.51)[3]. The review found no evidence that single-session debriefing is a useful way to prevent PTSD[3].
A 2013 Cochrane review of "Scared Straight" and similar juvenile awareness programs pooled seven studies and found more officially recorded offending among participants than among youths who received nothing (odds ratio 1.68)[4]. The reviewers concluded these programs increase delinquency relative to doing nothing at all[4].
Other entries rest on rare events documented in replicated case reports, including deaths and serious injuries in attachment therapies and false memories of trauma produced by recovered-memory techniques[9]. Practices that try to change sexual orientation or gender identity are covered in our explainer on conversion therapy.
Potentially Harmful Therapies and the Evidence Behind Them
| Intervention | Potential harm | Main evidence |
|---|---|---|
| Critical incident stress debriefing | Higher risk of posttraumatic stress symptoms | RCTs |
| Scared Straight programs | Worse conduct problems | RCTs |
| Facilitated communication | False abuse accusations against family | Case reports |
| Attachment therapies (e.g., rebirthing) | Death and serious injury to children | Case reports |
| Recovered-memory techniques | False memories of trauma | Case reports |
| DID-oriented therapy | Induction of "alter" personalities | Case reports |
| Grief counseling for normal bereavement | Increased depressive symptoms | Meta-analysis |
| Expressive-experiential therapies | Heightened painful emotions | RCTs |
| Boot camps for conduct disorder | Worse conduct problems | Meta-analysis |
| DARE programs | More alcohol and other substance use | RCTs |
Level I ("probably harmful for some individuals") treatments on Lilienfeld's provisional list of potentially harmful therapies, Perspectives on Psychological Science, 2007 [9].
1Level II ("possibly harmful") entries were peer-group interventions for conduct disorder and relaxation treatments for panic-prone patients [9].
2"Case reports" means rare events documented in replicated case reports. A listing indicates harm in some recipients, not in every recipient.
Sources
- •Lilienfeld, Perspectives on Psychological Science, 2007
What Does the Bad Therapy Book Argue, and What Do School Trials Show?
Among the book's targets is school-based social-emotional learning (SEL), which Shrier characterizes as a harmful, overly therapeutic practice that contributes to the youth mental health crisis[7]. A 2026 review in the journal Social and Emotional Learning: Research, Practice, and Policy notes that this characterization rests on unsystematically selected interviews and an examination of school policies and reports[7].
Researchers have raised a related but narrower concern in peer-reviewed form. In a 2023 BJPsych Bulletin commentary, Lucy Foulkes and Argyris Stringaris wrote that a growing body of quantitative research indicates some school-based mental health interventions increase distress or clinical symptoms relative to control activities[2].
They cited a meta-analysis of anti-bullying programs in which, in some studies, students taught CBT skills had more internalizing symptoms than control groups, and a randomized trial of a CBT-based school program with the same result[2].
The MYRIAD trial tested school-based mindfulness training in 84 secondary schools with 8,376 students aged 11 to 13[5]. Compared with usual social-emotional teaching, students already at risk of mental health problems had worse depression-risk and well-being scores after the program and at one-year follow-up, though the differences were small and not clinically relevant[5].
The trial authors concluded that the program, as delivered, is not indicated as a universal intervention and may be contraindicated for students with existing or emerging symptoms[5].
The same commentary that flags these harms also states that, for some adolescents, school efforts are essential and reduce clinical symptoms, and that trial evidence shows school-based interventions can reduce mental health problems, with generally small effects[2]. Its proposed mechanisms (rumination on negative thoughts, labeling negative thoughts and emotions in psychiatric terms, and peer influence in group sessions) are described by the authors as hypotheses with very little investigation so far[2].
Trends in adolescent mental health are tracked on our youth mental health statistics page.
Bad Therapy or Bad Therapist? What Research and Ethics Codes Say
Part of the variation in therapy outcomes traces to the individual clinician. A 2026 scoping review of 29 systematic reviews and meta-analyses reports that the therapist effect, the share of outcome variation attributable to individual therapists, is consistently estimated at 5% to 8%[6].
Interpersonal skills showed the most consistent link with outcomes, though the authors flag common methodological weaknesses in the underlying reviews[6]. Strains in the therapy relationship, including subtle hostility from a clinician, are covered in our post on the passive-aggressive therapist.
Professional ethics codes draw firm lines. The American Psychological Association's code, last amended effective January 1, 2017, states: "Psychologists do not engage in sexual intimacies with current therapy clients/patients"[10].
The same code requires psychologists to end therapy when it becomes reasonably clear the client no longer needs the service, is not likely to benefit, or is being harmed by continued service[10]. Under that standard, a client who is not benefiting is a reason to end treatment, not to extend it.
Licensing, training and prescribing rules differ across counselors, social workers, psychologists and psychiatrists; our guide to kinds of therapists lays out who does what.
Dropping Out, Ending Therapy and Filing Complaints
Many people stop before therapy is finished. A meta-analysis of premature discontinuation, or dropout, pooled 669 studies covering 83,834 clients and found a weighted dropout rate of 19.7%, about 1 in 5[11].
Dropout varied with diagnosis, age, provider experience, treatment setting and how dropout was defined, but not with the therapy's theoretical orientation or whether it was delivered individually or in a group[11].
Formal complaints about a psychologist go to the state, provincial or territorial psychology licensing board, which controls whether that person can practice[12]. The APA accepts complaints against its members only when no other forum can hear them[12].
As a membership organization, the APA cannot revoke a license; its most serious sanction is expelling a member[12].
Frequently Asked Questions
Sources & References
- Crawford MJ, Thana L, Farquharson L, et al. Patient experience of negative effects of psychological treatment: results of a national survey. Br J Psychiatry. Published 2016. Accessed October 2026. https://pubmed.ncbi.nlm.nih.gov/26932486/ [Link]
- Foulkes L, Stringaris A. Do no harm: can school mental health interventions cause iatrogenic harm? BJPsych Bull. Published 2023. Accessed October 2026. https://pmc.ncbi.nlm.nih.gov/articles/PMC10764817/ [Link]
- Rose SC, Bisson J, Churchill R, Wessely S. Psychological debriefing for preventing post traumatic stress disorder (PTSD). Cochrane Database Syst Rev. Published 2002. Accessed October 2026. https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD000560/full [Link]
- Petrosino A, Turpin-Petrosino C, Hollis-Peel ME, Lavenberg JG. 'Scared Straight' and other juvenile awareness programs for preventing juvenile delinquency. Cochrane Database Syst Rev. Published 2013. Accessed October 2026. https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD002796.pub2/full [Link]
- Montero-Marin J, Allwood M, Ball S, et al. School-based mindfulness training in early adolescence: what works, for whom and how in the MYRIAD trial? Evid Based Ment Health. Published 2022. Accessed October 2026. https://pubmed.ncbi.nlm.nih.gov/35820993/ [Link]
- Alfonsson S, Hedman-Lagerlöf M, Högberg Ragnarsson E, et al. The therapist effect in adult psychotherapy: a systematic scoping review of reviews. Psychother Res. Published 2026. Accessed October 2026. https://pubmed.ncbi.nlm.nih.gov/42240233/ [Link]
- Strambler MJ. The crossroads of social and emotional learning and critical social justice. Social and Emotional Learning: Research, Practice, and Policy. Elsevier. Published 2026. Accessed October 2026. https://www.sciencedirect.com/science/article/pii/S2773233925001007 [Link]
- Substance Abuse and Mental Health Services Administration. SAMHSA releases annual National Survey on Drug Use and Health. SAMHSA. Published 2026. Accessed October 2026. https://www.samhsa.gov/newsroom/press-announcements/20260727/samhsa-releases-annual-national-survey-on-drug-use-and-health [Link]
- Lilienfeld SO. Psychological treatments that cause harm. Perspect Psychol Sci. Published 2007. Accessed October 2026. https://pubmed.ncbi.nlm.nih.gov/26151919/ [Link]
- American Psychological Association. Ethical principles of psychologists and code of conduct. American Psychological Association. Published 2017. Accessed October 2026. https://www.apa.org/ethics/code [Link]
- Swift JK, Greenberg RP. Premature discontinuation in adult psychotherapy: a meta-analysis. J Consult Clin Psychol. Published 2012. Accessed October 2026. https://pubmed.ncbi.nlm.nih.gov/22506792/ [Link]
- American Psychological Association. Complaints regarding APA members. American Psychological Association. Accessed October 2026. https://www.apa.org/ethics/complaint [Link]