Informed Consent: Assessment of Comprehension

Our data were collected retrospectively from patients included in 10 double-blind clinical research protocols conducted at the West Los Angeles Veterans Affairs (VA) Medical Center (table 1). Each protocol and its informed consent form was approved by the Institutional Review Board of the center. All subjects included in this report gave written informed consent to participate in one of these studies. Moreover, the informed consent procedure described in this report was approved as mandated by the Institutional Review Board and by the Data and Statistical Monitoring Board at the center (which monitors the progress of schizophrenic, Alzheimer’s disease, and other high-risk patients in research studies) for all schizophrenic subjects participating in our research program.

Schizophrenic patients at the West Los Angeles VA Medical Center were approached by their treating physicians regarding their interest in participating in a randomized clinical trial. In cases where the patient was interested, the attending physician described the protocol in some detail. If the patient expressed further interest in being a participant, the study coordinator then thoroughly read the informed consent form to the participant, frequently pausing to assess the level of understanding and answer necessary questions. Upon completion of this procedure, a questionnaire assessing the patient"s comprehension of the informed consent material was administered, and responses were recorded.

This questionnaire, the Informed Consent Survey (appendix 1), was designed by us and incorporates the previously discussed four “legal standards” for assessing patients’ capacities to understand informed consent. For each clinical study, we designed a unique Informed Consent Survey. Items assess patients’ knowledge of the following: specific details and overall goals of the study, patients’ choices as participants in the study, doctors’ responsibilities to the study, and the potential risks of antipsychotic medications. Also, the Informed Consent Survey inquires about the subject’s understanding of his or her physician’s dual role as both clinician and researcher. In cases where the patient did not respond correctly to an item of the questionnaire, that portion of the consent form was re-explained to him or her, and the Informed Consent Survey was then readministered. This procedure was repeated until the patient answered all items of the survey correctly, at which point the informed consent form was signed. The Informed Consent Survey was readministered 7 days later. If a patient failed to understand a critical item at this 1-week postconsent quiz, the item was explained until the patient reported that he or she grasped the concept and answered the survey item correctly. If a patient could not answer the critical questions correctly, he or she was excluded from participation in the research protocol. All subjects were administered the Brief Psychiatric Rating Scale (BPRS) at the completion of the first administration of the Informed Consent Survey. Statistical analyses of the data were performed with SAS software (11).

The 49 subjects were inpatients and outpatients, often in transition from one phase to another. Their mean age was 46.5 years (SD=8.5), and their mean age at onset of illness was 26.5 years (SD=6.4). The majority of the subjects were single (72%, versus 9% married and 19% divorced), were male (94%, versus 6% female), and were veterans (98%, versus 2% nonveterans). The subjects were predominately Caucasian (47%) and African American (39%), with smaller portions being Latino (8%) or Asian (6%). The mean level of education was 13.0 years (SD=1.8). All subjects were free of conservatorship or guardianship and were legally competent to give consent to participate in treatment research; i.e., none of the subjects had had a judicial process to determine competency.

The patients’ median score on the first trial was 80% correct. Fifty-three percent (N=26) of the patients required a second trial to obtain 100% correct, and 37% (N=18) required three or more trials. For the entire study group, scores improved between the first trial and the day 7 trial (McNemar χ²=9.8, df=1, p=0.02). There was a relationship between conceptual disorganization as rated on the BPRS and the percentage of correct responses on the day 7 postconsent trial of the Informed Consent Survey (table 2). Open-ended questions regarding participation in the research were answered as follows: 65% (N=32) of the subjects claimed that they were participating for personal benefits or altruistic reasons, 35% (N=17) claimed that they were participating at the suggestion of others, and 96% (N=47) felt that they were adequately informed.

This retrospective study demonstrated that through the implementation of systematic and thorough informed consent procedures, many psychotic patients—but not necessarily all—are able to comprehend and retain critical components of informed consent. Our results may reflect some preselection bias, since we did not approach patients who were so ill that they could not possibly give consent. A majority of the patients required two or more reiterations of the questionnaire in order to demonstrate complete mastery of the material.

We also found that conceptual disorganization was correlated with impaired performance on the Informed Consent Survey when it was readministered 7 days after consent had been given, supporting the findings of a previous study of informed consent in schizophrenia (7). It is notable that even high scores on other psychosis items of the BPRS (i.e., unusual thought content, hallucinations, suspiciousness) did not appear to impair subjects’ learning and retaining key information on the procedures, risks, and benefits of the research protocol (table 2). These findings support previous studies which reported that psychiatric patients, when exposed to the informed consent process on just one occasion, are able to retain only a portion of the informed consent material (5, 7), as well as studies which demonstrated that informed consent procedures that incorporate patient education result in increased comprehension and retention of information (3, 8, 12). Further, these findings suggest that psychiatric symptoms such as hallucinations and delusions do not interfere with competency. Our findings also suggest that there are large differences in the ability of patients with schizophrenia to understand the important elements of consent.

On the basis of these findings, we plan to further evaluate patients’ ability to retain informed consent information over longer periods of time and to correlate their ability to comprehend this information with concurrent assessments of their neurocognitive functioning, particularly, objective measures of verbal memory and abstract reasoning. In addition, we will design and validate training methods that result in enhanced acquisition and durability of the information that is required for genuine informed consent by patients participating in trials of treatments for schizophrenia.

Presented as a poster session at the 149th annual meeting of the American Psychiatric Association, New York, May 4–9, 1996Received May 7, 1997; ; revision received March 23, 1998; accepted April 9, 1998. From the Department of Psychiatry, West Los Angeles VA Medical Center; and the Department of Psychiatry and Biobehavioral Science, School of Medicine, University of California at Los Angeles. Address reprint requests to Dr. Donna A. Wirshing, West Los Angeles VA Medical Center, 11301 Wilshire Blvd., Bldg. 210, Room 15, Los Angeles, CA 90073.