Factors determining interrater agreement with rating global change in dementia: The cibic‐plus

Abstract

SUMMARYGlobal ratings of change were obtained from 17 clinicians on eight patients, as though participating in an antidementia drug trial. Simultaneous rating was made possible by the use of videotaped semi-standardized interviews, and clinical change was simulated by providing raters with adjusted initial assessment documentation representing the situation 6 months earlier. In a second study, seven more ratings were obtained on the two patients who had commanded best and worst agreement in order to clarify important variables for levels of agreement. The main finding was that levels of interrater agreement on a global rating of change (which included carer information) varied widely from patient to patient. An important contributor to this variation was carer information. The order in which information for rating had been presented was found to be of overriding importance; if carer information was presented before patient information, considerably higher levels of agreement were achieved on the global rating than if the reverse order was used. KEY w o m d e m e n t i a ; global change; CIBIC; interrater agreement

Galantamine Prolonged-Release Formulation in the Treatment of Mild to Moderate Alzheimer’s Disease

“…First, the CIBICplus rating scale is subjective by design and is susceptible to cultural and geographic variations related to caregiving practices. These factors cause inter-rater variability and may have played a role in the inconsistent effects on the CIBIC-plus in this trial [22] . There are many potential explanations for why the centers in the United States differed from those of the rest of the world in results, including cultural factors related to caregiving or interviewing styles [23] .…”

Reliability Study on the Japanese Version of the Clinician’s Interview-Based Impression of Change

“…The videos of 8 patients were excluded while the remaining 13 were edited to remove unnecessary information for the reliability study and to shorten the viewing time. However, the CGIC of the remaining 13 patients was somewhat limited and ranged from 'minimally improved' (3) to 'moderately worsened' (6). In other words, scores of 'markedly improved' (1), 'moderately improved' (2) and 'markedly worsened' (7) were lacking or scarce.…”

“…Because of this subjective nature of global measures, reliability has been an issue of concern, and some studies have examined this issue. Dahlke et al [3] , Boothby et al [6] and Quinn et al [12] studied interrater reliability. However, satisfactory results were not always obtained with respect to the reliability of global change measures, and the interrater reliability of the CIBIC-plus J has not yet been studied.…”

“…The Alzheimer's Disease Cooperative Study has developed a semistructured version of the CIBIC-plus called the ADCS-CGIC [6] . The domains of the ADCS-CGIC are cognition, behavior and social and daily functioning.…”

Reliability Study on the Japanese Version of the Clinician’s Interview-Based Impression of Change

“…Methods: Eleven raters who were clin ical physicians watched videotapes of 20 patients with cal global assessment in 1990 as the primary outcome measure for use in demented patients, global change ratings using clinician interviews have been called Clinician's Interview-Based Impression of Change (CIBIC) and regarded as the measure for detecting 'clinically meaningful change ' [4-6] . The rating instrument, which includes a patient interview 'plus' a caregiver interview, is called 'CIBIC-plus' [7] .CIBIC-plus J, the subject of this study, is the Japanese version of the CIBIC-plus. It has semistructured interview formats for both caregivers and patients and its subscales include questions about the activities of daily living, behavioral and psychiatric symptoms, impaired cognitive functions, motivation, and emotional disturbances of dementia patients, and the clinician assesses changes in the condition of the patient to make a global judgment concerning CGIC using a 7-point Likert-type scale [8] .…”