April 1983

Reliability of Life-Event Interviews With Outpatient Schizophrenics

Author Affiliations

From the Gertrude H. Sergievsky Center, Columbia University Faculty of Medicine, and the Epidemiology of Brain Disorders Department, New York State Psychiatric Institute, New York.

Arch Gen Psychiatry. 1983;40(4):378-383. doi:10.1001/archpsyc.1983.01790040032005

• Reliability of life-event reports was examined in 18 outpatient schizophrenics, using a 102-item structured checklist. Patients and one close relative were interviewed regarding events during the 12 preceding months. Interviewers were randomly assigned to patients, pair members were seen by different individuals. Mean intrapair agreement for all events combined was .22. Objective items exhibited greater reliability than subjective ones, but item ambiguity, event recency, and stressfulness did not appear to influence agreement. Pairs evidenced low concordance on event dates. A greater number of events were recalled for the more recent six months than for the earlier period, and a pronounced interviewer effect on the number of reported events emerged. These findings, together with those from other reliability investigations, cast doubt on the validity of some retrospective studies of the relationship of life events to illness onset.