Ambiguity: What's Your Tolerance?

Abstract

Background and AimsWithin the anatomical sciences and authentic clinical experiences, learners are regularly exposed to the realities of variation and medical ambiguities. Doctors with low ToA may have increased levels of burnout, anxiety, and depression and these doctors may avoid complex patients or overtest patients in an effort to minimize ambiguities. Despite the potential significance of ToA, the use of ToA scales has more recently been called into question. This research sought to (1) study the dimensionality of the healthcare‐specific Tolerance of Ambiguity in Medical Students and Doctors (TAMSAD) scale, (2) determine whether its validity transfers to the context of undergraduate Australian medical students, and (3) evaluate its stability over time.MethodsFirst and second‐year medical students at Monash University electively participated in validity testing of existing healthcare‐related ToA scale items. Phase I of this study tested the dimensionality of the TAMSAD scale using confirmatory factor analysis (CFA). Phase II of this study sought to establish an optimized healthcare‐related ToA inventory using a combination of items derived from other published works. The reliability and validity of this optimized scale were assessed using Cronbach's alpha and exploratory factor analysis (EFA), respectively. To assess the stability of the identified factors, multiple administrations of the scale were psychometrically evaluated over time.ResultsUpon confirmatory factor analysis of medical students' (n=102) scores, the inadequacy of model‐fit indices revealed that TAMSAD scores did not conform to a unidimensional model as proposed in the literature. Due to a lack of evidence supporting the construct validity of the TAMSAD scale (as deduced from CFA and EFA findings), the research team questioned whether other ToA inventories used in healthcare settings might have stronger construct validity when administered to a medical student population.Phase II of this study sought to establish an optimized healthcare‐related ToA inventory using items derived from other published works. Pilot testing with second‐year medical students (n=63) indicated this inventory also had weak construct validity and was highly unstable across multiple distributions. Upon comparing how items loaded on each factor, 36% of items (8 of 22) crossed factors following the second distribution of the instrument, meaning these items did not load consistently on the same factors across the two distributions of the inventory. This finding suggests the latent constructs measured by the optimized ToA inventory are highly unstable making the dependability of scores derived from ToA inventories suspect.ConclusionOverall, ToA appears to be a complex construct which is likely multi‐faceted. Our findings, coupled with a review of others' ToA scale outcomes, suggest that the field would benefit from studies rigorously exploring the conceptual underpinnings of ToA to better inform the measurement of ToA.This abstract is from the Experimental Biology 2019 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.