The Hidden Risks of the "Talking Cure": New Research Unveils the Prevalence and Complexity of Negative Effects in Psychotherapy

For decades, psychotherapy has been framed in the public consciousness as a near-universal good—a "talking cure" that, even if it fails to help, is generally assumed to be harmless. However, a growing body of psychological research is beginning to dismantle this assumption. Recent studies published in 2026 suggest that psychotherapy, like any potent medical intervention, carries significant risks of side effects, adverse events, and "nocebo" responses that are often overlooked by practitioners and underestimated by patients.

A series of new investigations, highlighted by the advocacy and research organization Mad in America, provides a sobering look at the darker side of clinical intervention. From a narrative review of global literature to specific Dutch studies on clinician attitudes, the data reveals a systemic failure to define, monitor, and discuss the potential harms of psychological treatment.

Main Facts: The Triple Threat of Therapeutic Harm

The emerging research categorizes the negative outcomes of therapy into three distinct but overlapping domains. Understanding these is critical for both the practitioner’s ethical responsibility and the patient’s informed consent.

1. The Prevalence of Dissatisfaction: When defined broadly—including unmet expectations and general dissatisfaction—an astonishing 95.6% of patients report experiencing some form of negative event during their therapeutic journey.
2. The "Nocebo" Effect: Just as a placebo can heal through expectation, the "nocebo" effect can harm. Negative expectations, poor clinician communication, and past traumatic experiences with the mental health system can lead to a measurable worsening of symptoms.
3. The Clinician Gap: There is a profound disconnect between patient experience and provider perception. Research indicates that many clinicians dismiss patient-reported harms as being unrelated to the therapy itself, or they view worsening symptoms as an "unavoidable" part of the process, often failing to warn patients of these risks beforehand.

Chronology of Findings: From Literature Reviews to Empirical Surveys

The current shift in perspective is driven by three pivotal studies that examine the issue from the macro level of global literature down to the micro level of individual clinician-patient interactions.

The Global Landscape: Ram et al. (2026)

Led by Dushad Ram of Shaqra University, a narrative review published in the Industrial Psychiatry Journal sought to synthesize the existing literature on adverse events. This review established a foundational hierarchy of harm:

• Predictable Side Effects (33%–41%): These are negative experiences resulting from correctly applied treatment, such as the temporary spike in anxiety during exposure therapy.
• Adverse Effects (5%–10%): These are unexpected and harmful results, such as the deterioration of original symptoms or the strain of family relationships.
• Malpractice (2.6%): This involves severe ethical breaches, including misdiagnosis and boundary violations.

The Psychological Mechanism: Evers (2026)

Following the "what" and "how many" established by Ram, Andrea W. M. Evers of Leiden University published an exploration of the "why" in Current Opinion in Psychology. Her work focuses on the nocebo effect, arguing that the therapeutic context itself can be pathogenic. She identified that "evolutionary bias"—the human tendency to prioritize negative experiences over positive ones—makes past therapeutic failures a primary driver of future ones.

The Practitioner’s Perspective: Houben et al. (2026)

The most recent empirical data comes from Maastricht University, where Sanne T. L. Houben and colleagues conducted two studies on Dutch clinicians. Published in Clinical Psychology & Psychotherapy, these studies moved the focus to the "room where it happens," revealing that even when therapists are aware of negative effects, they often lack the tools or the willingness to address them with their clients.

Supporting Data: Factors Contributing to Therapeutic Failure

The data suggests that therapeutic harm is rarely the result of a single factor but is instead an intersection of therapist, patient, and systemic variables.

Therapist-Driven Factors

The research identifies "therapist factors" as a primary source of preventable harm. These include:

• Inflexibility and Lack of Empathy: Rigid adherence to a specific manual regardless of the patient’s response.
• Weak Therapeutic Alliance: The failure to build a "bond" or agree on "tasks" and "goals," which is often the strongest predictor of a poor outcome.
• Inadequate Informed Consent: Only about 32% of clinicians with low familiarity with negative effects bother to discuss them with patients. Even among those "familiar" with the risks, only 70% actively disclose them.

Vulnerable Populations

The literature review by Ram et al. highlights that certain groups are disproportionately at risk:

• Youth: More likely to experience symptom exacerbation and a weak alliance.
• Trauma Survivors: Highly susceptible to emotional "re-traumatization" if therapy is rushed or insensitive.
• Cultural Minorities: Frequently harmed by "Westernized" models that do not align with their cultural values, combined with a lack of cultural competence from the therapist.

The Nocebo Triggers

Evers’ research details how the "nocebo" effect manifests through five specific triggers:

1. Negative Expectations: Believing therapy will be painful.
2. Social Learning: Hearing horror stories about therapy from peers or social media.
3. Contextual Factors: The stress of long waiting lists can cause symptoms to deteriorate before the first session.
4. Communication: Therapists who overemphasize the difficulty of the work or provide a poor rationale for the treatment.
5. Previous Experience: Chronic cases where past failures haunt the current attempt.

Official Responses and Professional Disagreement

The research by Houben et al. reveals a startling lack of consensus within the profession. When Dutch clinicians were asked to classify patient-endorsed negative effects, the results showed a field in conflict:

• The "Unavoidable" Label: A significant portion of clinicians believe that effects like "feeling gloomy," "worsening relationships," and "excessive self-reflection" are simply unavoidable. If a clinician views a harm as unavoidable, they are statistically less likely to take corrective action to mitigate it.
• Dismissal of Patient Experience: In the Houben study, 48.8% of participants dismissed a patient "feeling worse" as being unrelated to the therapy. Similarly, nearly half of the clinicians dismissed "reduced labor market opportunities" or "stigma" as non-existent effects of the therapeutic process.
• The Informed Consent Crisis: The data suggests a widespread ethical lapse. Informed consent is a cornerstone of medical ethics, yet in psychotherapy, the potential for "deterioration" is rarely presented with the same gravity as side effects for a pharmaceutical drug.

While professional bodies like the American Psychological Association (APA) and the British Psychological Society (BPS) have long maintained ethical codes regarding "doing no harm," these studies suggest that the definition of harm is so inconsistently applied that the codes may be ineffective in practice.

Implications: A Call for Standardized Monitoring

The implications of this research are far-reaching, suggesting a need for a fundamental shift in how psychotherapy is practiced, taught, and regulated.

1. Mandatory Standardized Monitoring

The authors of all three studies converge on a single recommendation: the need for standardized monitoring of adverse events. Just as a surgeon monitors for post-operative infection, a psychotherapist must use validated tools—such as the "Positive and Negative Experiences in Psychotherapy" survey—to check for harm at every session. This moves the burden of reporting away from the patient and integrates it into the clinical routine.

2. Radical Transparency in Informed Consent

Patients must be treated as partners in their care. This requires therapists to move beyond a "sales pitch" for their specific modality and provide a balanced view of potential outcomes. Patients should be told, before beginning, that symptoms may temporarily worsen and that certain modalities may not be a fit for their specific cultural or personal history.

3. Reconditioning the Nocebo Effect

For patients with a history of "treatment resistance" or past therapeutic trauma, Evers suggests that therapy should begin with "reconditioning." Before addressing the primary psychological issue, the clinician must address the patient’s fear of the clinical setting itself.

4. Training and Cultural Competence

The findings regarding cultural minorities suggest that "one-size-fits-all" Western therapy can be a form of systemic harm. Training programs must prioritize cultural humility and the recognition of "therapist-patient mismatch" as a valid reason to terminate or refer a client, rather than viewing it as a failure of the patient to "engage."

5. Challenging the "Unavoidable" Narrative

Perhaps the most critical implication is the need to challenge the clinician’s belief that harm is unavoidable. By labeling worsening symptoms as a "necessary part of the process," clinicians may be ignoring warning signs of a failing intervention. Differentiating between "productive struggle" and "harmful deterioration" is a skill that must be sharpened in clinical training.

In conclusion, the "talking cure" is a powerful tool, but as these studies demonstrate, power always comes with the potential for misuse and unintended consequences. By acknowledging the prevalence of negative effects and the reality of the nocebo response, the field of psychotherapy can move toward a more honest, ethical, and ultimately more effective model of care. The goal is not to discourage therapy, but to ensure that when a patient seeks help, they are not met with an unexpected and unacknowledged harm.