Read our COVID-19 research and news.

New hope for schizophrenia sufferers? The mental illness has been described as a waking nightmare, as conveyed in this artist's conception of the experience

New hope for schizophrenia sufferers? The mental illness has been described as a waking nightmare, as conveyed in this artist's conception of the experience.

Louise Williams/Science Photo Library

Schizophrenia: Time to Flush the Meds?

Schizophrenia and related mental illnesses can have a devastating effect on people who suffer from them, often making it impossible for them to work or maintain normal social relationships. Antipsychotic drugs are usually the first line of defense, but they can have serious side effects. A new study concludes that psychological approaches could be an alternative for patients who either can’t or won’t take medication, although some critics continue to question the effectiveness of these interventions.

Schizophrenia, which can involve hallucinations, delusions, paranoia, emotional problems, and severe difficulty focusing on daily tasks, affects about 1% of populations worldwide. More than 20 antipsychotic medications, such as risperidone, haloperidol, and clozapine, are now on the market, and they are often effective in temporarily relieving the worse symptoms. But when taken for extended periods, such drugs can cause uncontrollable muscle movements, serious weight gain, and higher risk of heart attacks. In recent years, a number of psychiatrists and psychologists have begun to advocate psychological approaches, including an approach called cognitive behavioral therapy (CBT), as an adjunct to antipsychotic drugs.

With CBT, which has long been shown to be effective for depression and anxiety disorders, a therapist takes the subject through a series of guided steps designed to explore alternative interpretations and explanations of what he or she is experiencing, with the goal of changing both outlook and behavior. A schizophrenic patient who is having hallucinations might be encouraged to stop trying to fight them off or suppress them, for example, or to stop engaging with voices in his or her head, to test how strong such symptoms really are and how much control they exert over the subject’s life. The technique also involves what practitioners call “normalization”: The patient might be reassured that hearing voices and seeing things that are not there is an experience that many normal people have from time to time, thus reducing some of the anxiety that makes sufferers feel distressed and isolated.

In recent years, researchers have conducted several dozen clinical trials for treating schizophrenia with CBT. Most of them, including a number of so-called meta-analyses that look at many trials at a time, have claimed modest success in reducing psychotic symptoms. As a result, health authorities in the United Kingdom, where the use of CBT for psychosis has been most studied, have been recommending for a number of years that it be offered to all people suffering from schizophrenia. Nevertheless, some skeptics question these recommendations: In the January 2014 issue of The British Journal of Psychiatry, for example, a research team published a new meta-analysis, concluding that past trials showing CBT’s effectiveness for schizophrenia were seriously flawed.

In the latest study, published online today in The Lancet, a team led by Anthony Morrison, a clinical psychologist at the University of Manchester in the United Kingdom, conducted a randomized controlled trial of patients diagnosed with schizophrenia and related disorders who had chosen not to take antipsychotic drugs. The team started with 74 such individuals, who were randomly assigned by a computer to receive either “treatment as usual” (TAU)—meaning regular monitoring by mental health services, sympathetic face-to-face contact with doctors and nurses, crisis management as needed, and other such interventions that are standard in the United Kingdom—or TAU plus 26 weekly sessions of CBT over 9 months, plus as many as four additional CBT sessions over the next 9 months, for a total of 18 months.

The study began with 37 patients in each group, although over the course of the trial a number of subjects dropped out of both groups for various reasons—including two deaths in the TAU-only cohort that appeared to be unrelated to their mental illness. The researchers ended up with final tallies of 25 subjects in the TAU-only group and 26 in the TAU-plus-CBT group. Every 3 months during the trial, the subjects were given a standard battery of tests to measure their psychotic symptoms, along with tests designed to measure their levels of emotional distress and social functioning.

The team found that at the end of the 18-month trial, the group given CBT had lower overall levels of psychotic symptoms than the TAU-only group, although the differences were modest: The overall “effect size,” a statistical measure of the differences between two groups in a clinical trial, was 0.46 (on a scale where 0.2 is considered low, 0.5 is considered moderate, and anything over 0.8 is considered to be high). Nevertheless, the researchers argued, the effect size they found was equivalent to that typical of most antipsychotic drugs when compared with placebos (dummy pills that contain no medication).

The authors caution that despite what they see as encouraging results, the findings should not be interpreted to mean that all patients suffering from schizophrenia can or should go off their meds; the two groups they studied were comparatively high-functioning patients who were not hospitalized and posed no danger to themselves or their communities. Nevertheless, Morrison says, other studies have shown that up to 50% of schizophrenia sufferers fail to take drugs over the long term, and that this figure can rise to more than 70% over any particular 18-month period. “It seems that offering people choices” about whether to take drugs or not “is a sensible thing to do,” he says.

The results are “utterly convincing,” especially in light of recent studies suggesting that some antipsychotic drugs are actually less effective than placebos in children and young adults, says Max Birchwood, a psychologist at the University of Warwick in Coventry, U.K. But he adds that the sample studied by the Morrison team is small and needs to be expanded to a larger population, something the authors themselves advocate.

Kate Hardy, a psychologist at the University of California, San Francisco, who treats schizophrenia patients with CBT—one of only a small number of practitioners in the United States using the technique in treating psychosis—says that the results may give more choices to patients about their own treatment. Up to now, Hardy says, the only choice psychosis patients have had is whether “to take medications or not.”

But Keith Laws, a psychologist at the University of Hertfordshire in Hatfield, U.K. and a co-author of The British Journal of Psychiatry meta-analysis critical of CBT for schizophrenia, questions the team’s analysis of its own data. He argues that there was no real difference between the CBT and control groups at the end of the 9 months of intense CBT treatment and that the differences at the end of 18 months were only marginal. “They are not in a position to make grand claims,” Laws says. He also says that the number of “adverse events” during the trial—several patients in both groups got worse and a few had to be hospitalized—were much greater than in other CBT trials in which patients were also given medications, making such drug-free approaches “very risky.”