Uncovering Preference and Scale Heterogeneity in Digital Neurorehabilitation: Results From a Discrete Choice Experiment

OBJECTIVES: Stroke, the leading cause of persistent disability globally, necessitates effective rehabilitation strategies. Digital technologies present opportunities for enhancing the efficiency and effectiveness of these strategies. However, the successful adoption of these innovations heavily depends on patient acceptance. Therefore, a health preference research study was conducted to explore patient preferences for digital interventions in neurorehabilitation, addressing both preference heterogeneity and scale heterogeneity.

METHODS: The study utilized a discrete choice experiment to examine preferences regarding (1) the explanation and presentation of therapy exercises, (2) information in therapy, (3) contact with healthcare professionals, (4) patients’ choices in the therapy process, (5) data processing, (6) therapy success within 6 months, and (7) copayment per month. The analysis focused on preference and scale heterogeneity, employing mixed logit regression models (MXL) for mean utilities and latent class analysis (LCA) to identify subgroups with distinct preference patterns. Scale heterogeneity analysis used a heteroscedastic conditional logit model (HET) to observe the influence of unobservable variables on decision behavior variance.

RESULTS: The study included 1,055 participants, with a mean age of 52 years and an equal gender distribution. MXL revealed significant heterogeneity in the technical aspects of explanation and presentation, information, patients’ choice, and data processing. LCA identified three distinct preference classes, each with unique priorities. Class 1 (N = 225) focused on copayment. Class 2 (N = 340) prioritized therapy success. Class 3 (N = 490) placed the greatest importance on technical aspects. The HET investigated the relationship between error variance and user-defined variables, finding that health status, among other therapy and health-related dimensions, impacted decision-making consistency.

CONCLUSIONS: Integrating analyses of preference and scale heterogeneity offers nuanced insights into the complexities of healthcare decision-making. The insights gained can improve neurorehabilitative care through personalized digital solutions and informed decision-making, benefiting individuals in rehabilitative contexts.