Most schizophrenia patients able to consent

Long-term study found that 96 percent still scored well on capacity assessments

An overwhelming majority of research subjects with schizophrenia monitored over the course of a long-term medication study retained the capacity to give consent for research.

The results, published recently in the journal Schizophrenia Research, are the first large-scale investigation of longitudinal consent abilities for subjects with schizophrenia. But they confirmed what researchers had already believed about doing research with this population, says lead author T. Scott Stroup, MD, MPH, director of the Program for Intervention Effectiveness Research at Columbia University's New York State Psychiatric Institute in New York City.

"We monitored it a little bit more formally than most people do with this structured assessment," Stroup says. "But I don't think this was a surprise to investigators. That's the great thing about our results — that they make sense."

The assessments were part of the larger Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) study, of which Stroup was the co-principal investigator. Stroup says the group conducted periodic assessments of subjects in the CATIE study to see how the ability to consent changed over time among people with schizophrenia.

More than 1,100 subjects were assessed at their initial consent with several different instruments which helped gauge their understanding and reasoning abilities, neurocognitive function and symptoms including delusions, hallucinations and disorganized thinking. They were assessed again after 6 and 18 months of study and at any time that a study medication was discontinued.

In order to be randomized to a medication in the CATIE study, subjects had to score at least 16 points or higher on the understanding portion of the MacArthur Competence Assessment Tool-Clinical Research (MacCAT-CR) a commonly used screening tool for consent capacity. The team measured how that score changed over time, as well as how symptoms and other measures of functioning changed.

Over the course of the study, 96 percent of subjects maintained a score of 16 or greater on the MacCAT-CR understanding portion. Of the entire group, 20 percent actually saw a notable increase in their score, while 24 percent declined and 56 percent showed no substantial change.

The 4 percent who did fall below the initial threshold shared many characteristics, including borderline initial scores, worse-than-average neurocognitive functions and more severe symptoms at the outset. In addition, people whose symptoms worsened over time also were found to have greater declines in their later MacCAT scores.

These factors provide investigators with a better picture of whom to watch more carefully in a study of this kind, Stroup says.

"I think investigators should know who to monitor — those with marginal capacity whose symptoms are worse at the outset and those who worsen," he says.

Stroup doesn't believe this type of intense structured assessment is necessary for all people with schizophrenia who participate in research, based on the small number of overall participants whose scores fell below the initial threshold.

He notes that the CATIE study was minimal risk, participants receiving approved medication at approved doses. For a more experimental treatment, or another type of higher-risk protocol, more structured assessments may be necessary. In addition, Stroup says researchers can use other safeguards, such as a subject advocate — a friend or family member who can assist in determining whether a subject continues to have capacity to consent.

"The lost capacity (in this study) was so rare and the effort for longitudinal structured interviews and subject advocates may exceed what's necessary," he says. "For things with minimal risk or no more than usual risk, I don't think that's necessarily needed. That said, I do think it's important for investigators to remain vigilant."


Stroup TS, Appelbaum PS, Gu H. Longitudinal consent-related abilities among research participants with schizophrenia: Results from the CATIE study. Schizophr Res 2011 May 9 Epub.