Characterizing explanatory models of illness in healthcare: Development and validation of the CONNECT instrument

Abstract

Objective: A growing body of qualitative and quantitative research suggests that individual patients and physicians often have differing perspectives, or 'explanatory models,' regarding the patient's health condition or illness. Discordance between explanatory models may lead to difficulties in communication and poor disease outcomes. However, due to a lack of tools to systematically measure concordance in patient and physician explanatory models, a large-scale study of explanatory models of illness has not been previously possible. The objective of this project was to develop and pilot-test a survey-based tool (the CONNECT Instrument) that measures salient aspects of explanatory models of illness. Methods: We conducted a multi-method survey development project that included qualitative and quantitative item development, refinement, pilot testing, and psychometric evaluation. We evaluated the instrument in two unique, consecutive cohorts of primary care patients in a variety of private and public settings in Houston, TX. We also used the instrument to examine concordance between patient and physician explanatory models in the second cohort. Results: The final version of the CONNECT Instrument contains nineteen items that focus on six dimensions of explanatory models. Cronbach alphas ranged from 0.65 to 0.89 for the six CONNECT dimensions. The instrument demonstrated evidence of criterion-related validity when individual CONNECT dimension scores were compared with scores from previously published instruments, and demonstrated expected differences between patients 'and physicians' explanatory models of illness. Conclusion: The CONNECT instrument is a tool with good psychometric properties that enables researchers to measure important aspects of patients 'and physicians' explanatory models of illness. Our continuing work will focus on gathering additional validity evidence and evaluating associations between explanatory model concordance and health outcomes. Practice implications: The CONNECT instrument can be used to improve quality in clinical practice and medical education by measuring an important intermediate outcome in the chain of factors leading to patient trust, satisfaction, and adherence.