France Launches Major Trial to Test Whether Talking Therapy for Voices Is Worth the Cost

Hearing voices that others cannot hear is one of the most distressing and destabilizing experiences in psychiatry. For between 60 and 80 percent of people living with schizophrenia, auditory verbal hallucinations are a daily reality, bringing distress, disrupting concentration and relationships, and often triggering relapse and hospitalization. For decades, the standard response has been pharmacological: second-generation antipsychotic medications that dampen, but rarely eliminate, the voices, while carrying a substantial burden of side effects ranging from metabolic problems to sedation. Now a large French research consortium has unveiled the design of what it describes as the first large-scale medico-economic trial in France to test whether a structured cognitive-behavioural therapy program for distressing voices can deliver real clinical value, and whether that value justifies its cost to the health system.

The study protocol, published in BMC Psychiatry by a team led by Stephane Raffard, Erica Fongaro and Delphine Capdevielle of the University Department of Adult Psychiatry at CHU Montpellier, together with collaborators including Mark Hayward of the Sussex Partnership NHS Foundation Trust in the United Kingdom, lays out an ambitious randomized controlled trial. The trial is sponsored by CHU Montpellier and funded through the French government’s national hospital clinical research grant program, known as PHRC-N, administered by the Directorate General of Health Care Offer. It has been registered on ClinicalTrials.gov under the identifier NCT07134816, with registration dated February 4, 2025, and recruitment is expected to begin in July 2026.

At the heart of the trial is a deceptively simple question with profound policy implications: if cognitive-behavioural therapy genuinely reduces the severity and distress of auditory hallucinations, does the improvement translate into enough additional quality-adjusted life to make the therapy economically worthwhile? Health economists answer such questions with a metric called the incremental cost-utility ratio, expressed as the cost per quality-adjusted life year, or QALY. A QALY combines survival time with a weighting for health-related quality of life, where one year in perfect health equals one QALY and a year lived with significant impairment counts for less. By measuring both the extra costs of providing therapy and the extra QALYs it generates, researchers can compare the intervention against other uses of scarce healthcare resources.

The technical machinery of the trial reflects this economic focus. A total of 240 patients aged 15 to 60 years, all experiencing distressing auditory verbal hallucinations for at least three months, will be recruited from outpatient psychiatric services across France. Participants will be randomly assigned in a one-to-one ratio to one of two arms. The intervention group will receive 16 weekly one-hour individual sessions of cognitive-behavioural therapy layered on top of treatment as usual, meaning standard psychiatric follow-up and pharmacological care continue unchanged. The control group will receive treatment as usual alone. The primary outcome is the incremental cost-utility ratio at 12 months, with quality of life measured using the EuroQol five-dimension five-level questionnaire, a widely used instrument that captures mobility, self-care, usual activities, pain and discomfort, and anxiety and depression.

The therapy itself is not a single technique but a structured program that integrates several evidence-based strands of cognitive-behavioural practice. It combines Coping Strategy Enhancement, an approach that helps patients identify and systematically strengthen the strategies they already use, however informally, to manage their voices; relational therapy, which works on the patient’s relationship with the voices, often conceptualized as a social dynamic in miniature; third-wave CBT approaches, which draw on mindfulness, acceptance and compassion-focused methods to change how patients relate to distressing experiences rather than trying to suppress them; and imagery-focused techniques that target the vivid mental images frequently intertwined with hallucinations. The rationale is that voices are not simply sensory noise but experiences shaped by beliefs, emotions and relationships, and that intervening on multiple fronts simultaneously may achieve reductions in intensity and distress that medication alone cannot.

The trial’s secondary outcomes read like a comprehensive map of what matters in recovery from psychosis. At four, eight and twelve months, assessors will track changes in the severity of auditory hallucinations, depressive symptoms, suicidal ideation, quality of life, functioning, sleep quality, substance use, psychiatric comorbidity, exposure to pharmacological treatment and medication adherence. The breadth is deliberate: voices rarely exist in isolation. They are entangled with depression, insomnia, substance use and suicide risk, and a therapy that improves only one symptom while leaving the wider web untouched would offer limited real-world value. Conversely, if CBT improves sleep, reduces suicidal thinking or helps patients adhere to their medication, the economic case strengthens well beyond the direct effect on hallucinations.

Methodological rigor has been engineered into the design at several levels. All endpoint assessments will be conducted by independent evaluators blinded to treatment allocation, preventing the expectations of assessors from contaminating the outcome data. The randomization will be stratified by illness severity, ensuring that patients with more severe illness are balanced across the two arms, so that any observed effect cannot be explained by one group starting out sicker than the other. The trial has received ethical approval from the CPP Ouest IV ethics committee, and informed consent will be collected from participants, or from parents or legal representatives for younger participants, before any trial-specific procedure begins, using an online consent signature. Participation is voluntary and can be withdrawn at any time.

Why does France need such a trial at all? The protocol’s authors are candid about the barriers. Although cognitive-behavioural therapy for psychosis has accumulated decades of supporting evidence, showing reductions in voice-related distress and relapse risk, its use remains limited in many countries, France among them. The reasons are structural rather than scientific: insufficient clinician training in CBT for psychosis and severe time constraints in routine psychiatric services mean that few patients ever receive it. Treatment therefore continues to rely mainly on second-generation antipsychotics, which the authors characterize as having moderate efficacy and notable adverse effects. In effect, an entire national mental health system is making an implicit economic decision every day, choosing medication over therapy, without the cost-utility data needed to know whether that choice serves patients or budgets best.

The economic analysis will draw on France’s unusually rich health data infrastructure. The study’s abbreviation list points to the SNDS, the Système National des Données de Santé, the French national health data system, along with the CNAM, the national health insurance fund, suggesting that healthcare consumption and costs can be traced through administrative records rather than relying solely on patient recall. This matters because cost-utility studies live or die on the quality of their cost data. Hospitalizations, outpatient visits, medication dispensations and emergency contacts all carry price tags, and capturing them systematically allows the incremental cost of adding 16 hours of therapy to be weighed against any savings from avoided relapses, reduced hospital stays or lower medication use.

The stakes extend well beyond France. Health systems everywhere are searching for ways to improve outcomes in schizophrenia without spiraling costs, and psychological therapies are frequently proposed as part of the answer but rarely subjected to the same economic scrutiny as new drugs. If this trial demonstrates that CBT for distressing voices delivers QALY gains at a cost per QALY that compares favorably with accepted thresholds, it would give policymakers a concrete, quantified argument for investing in clinician training and service redesign. If it shows the opposite, that too would be valuable, redirecting resources toward alternatives. Either way, the study represents a maturing of psychiatric research, in which the question is no longer only whether a treatment works, but whether it works well enough, for enough people, at a price a health system can afford. Results, once recruitment begins in mid-2026 and follow-up extends over twelve months, will be awaited closely by clinicians, economists and, above all, the patients who live with the voices every day.

Subject of Research: Cost-utility and clinical effectiveness of cognitive-behavioural therapy for auditory verbal hallucinations in schizophrenia

Article Title: Cost-utility study of cognitive-behavioural management of auditory hallucinations in schizophrenia: a study protocol

Article References: Cost-utility study of cognitive-behavioural management of auditory hallucinations in schizophrenia: a study protocol. (n.d.). https://doi.org/10.1186/s12888-026-08557-1

Image Credits: AI Generated

DOI: 10.1186/s12888-026-08557-1

Keywords: schizophrenia, auditory hallucinations, cognitive-behavioural therapy, cost-utility, QALY, randomized controlled trial, antipsychotics, mental health policy, France, psychiatry, health economics, CBT for psychosis