Clustering and the design of preference-assessment surveys in healthcare

Objective. To show cluster analysis as a potentially useful tool in definin g common outcomes empirically and in facilitating the assessment of prefere nces for health states. Data Sources. A survey of 224 patients with ventricular arrhythmias treated at Kaiser Permanente of Northern California. Study Design/Methods. Physical functioning was measured using the Duke Acti vity Status Index (DASI), and mental status and vitality using the Medical Outcomes Study Short Form-36 items (SF-36). A "k-means" clustering algorith m was used to identify prototypical health states, in which patients in the same cluster shared similar responses to items in the survey. Principal Findings. The clustering algorithm yielded four prototypical heal th states. Cluster 1 (21 percent of patients) was characterized by high sco res on physical functioning, vitality, and mental health. Cluster 2 (33 per cent of patients) had low physical function but high scores on vitality and mental health. Cluster 3 (29 percent of patients) had low physical functio n and low vitality but preserved mental health. Cluster 4 (17 percent of pa tients) had low scores on all scales. These clusters served as the basis of written descriptions of the health states. Conclusions. Employing a clustering algorithm to analyze health status surv ey data enables researchers to gain a data-driven, concise summary of the e xperiences of patients.