Patient Preferences for Attributes that Characterise Alternative Models of Care in Gastroenterology: A Discrete Choice Experiment.

Objectives: Increased demand for gastroenterology services has resulted in growing waitlists, with patients at risk of exceeding clinically recommended wait-times. Given limited healthcare resources, expanded scope models of care are an option to help address this demand, but little is known about patient preferences for these models of care.

Methods: Low-risk gastroenterology patients (n = 1198) referred to an outpatient tertiary service in Australia over a 2-year period were invited to participate in an unlabelled discrete choice experiment with seven attributes: primary healthcare professional, wait-time, continuity of care, consultation length, manner and communication skills, reassurance, and cost. These were developed using qualitative research, literature review, and stakeholders' experiences. A d-efficient fractional design was used to construct four blocks of 12 choice sets, with two alternatives. A 13th choice set was included as a data and quality check. Latent class and mixed logit regression were used for analysis. The resulting preference parameters for individual attributes were then used to calculate willingness to pay and willingness to wait.

Results: Overall, the model based on the 347 respondents suggested no strong preference for professional background. All other attributes were statistically significant predictors of preference (p < 0.001), with respondents willing to make significant trade-offs (time and cost) before accepting deterioration in attributes. There was strong emphasis on manner and communication skills, with a clinician who listens and provides good explanations overwhelmingly the most important attribute. Latent class analysis identified two patient segments who differed in their preference for the primary treating healthcare professional (doctor or dietitian) based on exposure to either traditional medical or non-medical professional role substitution model.

Conclusions: Patients have strong but varied preferences for gastroenterology services based on whether they have been exposed to expanded scope models of care. Design and implementation of new models of care need to consider strategies to overcome any perceived loss in utility or deterioration in healthcare quality for those unfamiliar with professional role substitution.