Patients' perceptions and treatment effectiveness: A reassessment using generalized maximum entropy
Diabetes Mellitus, Type 2
Recent research using Prospect Theory (PT) suggests that treatment history contributes to a frame of reference that introduces heterogeneity in patients' perceptions of how further treatment will improve their health status and carries over into an assessment of how effective the recent treatment was. Analysis using the Monotone Rank Estimator (MRE), a semi-parametric classical statistical technique that allows for heterogeneity across individual responses, supported this idea. This article checks whether the MRE results are accurate using an alternative technique, Generalized Maximum Entropy (GME), which more effectively incorporates heterogeneity through the assignment of support points (based on prior information) for the error term. The results are compared to those obtained previously with the MRE. In the new results, prior treatment outcomes do not appear to significantly affect patients' perceived outcomes for subsequent treatments, which contradicts the earlier findings. © 2012 Copyright Taylor and Francis Group, LLC.
Digital Object Identifier (DOI)
Additional Document Info
has global citation frequency