No abstract

Background: Personality traits are stable factors which can aid in understanding patient health behaviour. The five-factor model of personality can serve as a theoretical background for exploration and understanding of health behaviours. Extant studies report the implication of domains of Neuroticism and Conscientiousness in diabetes self-care behaviours. However, no study exists that take in to account a specific set of self-care behaviours in a chronic illness population and connect these to possible personality profiles. This paper presents an analysis of clustering of personality domains with relevant specific diabetes self-care behaviours, relevant demographic factors and blood glucose related outcomes in adult patients. Methods: Self-report questionnaire data were collected from a sample of 295 patients of diabetes which included two types of diabetes (Type1= 52; Type 2 = 243). Tools included a measure of patient information schedule, 60-items of NEO-FFI personality inventory and Summary of Diabetes Self-care Activities (SDSCA). Following simple linear correlation, regression analysis, clustering process was initiated via Two-Step Cluster Analysis which identified two clusters as an optimal solution. Unsupervised k-means segmentation helped deduced two significantly different clusters of patients. Results: Two clusters were found to be significantly different with respect to four domains of personality: Extraversion, Openness to Experience, Agreeableness and Conscientiousness, three critical self-care behaviours -following a general diet, following recommended diet of fruits and vegetables and performing exercise behaviour and two of the physiologic measures of blood glucose control -Fasting Blood Sugar (FBS) and Post Prandial post meal measure (PP). Conclusion: The results show personality traits tend to cluster. This indicates that specific traits can be grouped together. In addition, specific self-care behaviours were associated with these trait combinations. Irrespective of individual factors such as age, duration of illness and gender the obtained patient profiles were distinct to one another. These findings have significant meaning for future diabetes related health programs. Findings can help in development of tailor-made intervention programs with use of such knowledge of trait clustering.