274 - Into the Unknown: Teaching Residents to Discuss Diagnostic Uncertainty

Publication Number: 274.5537

Doha K. Aboul-Fotouh, McGovern Medical School at the University of Texas Health Science Center at Houston, Houston, TX, United States; Shivani Briggs, McGovern Medical School at the University of Texas Health Science Center at Houston, Pearland, TX, United States; Autumn Atkinson, McGovern Medical School at the University of Texas Health Science Center at Houston, Houston, TX, United States

Background: Diagnostic uncertainty is the personal perception that the provider cannot give an accurate representation of the patient's health problems during a given encounter. It is a separate entity from technical and personal uncertainty, defined respectively as a lack of concrete knowledge or ability and uncertainty about the physician-patient relationship. While present amongst all specialties, diagnostic uncertainty is especially prevalent in pediatrics. Our limited, existing communication regarding diagnostic uncertainty contributes to reluctance to withdraw therapy, increased healthcare costs, and poor patient adherence. Furthermore, diagnostic uncertainty can contribute to trainee burnout and physician anxiety.

Objective: To determine if instruction on identifying and approaching diagnostic uncertainty changed pediatric resident response to uncertainty.

Design/Methods: Through literature review and a systematic instructional design methodology, we created instruction for pediatric residents on how to identify and discuss diagnostic uncertainty. During their sessions, residents were taught to set expectations, explain a differential and workup, and move forward with their patient by validating symptoms and discussing next steps. The effects of the instruction were measured using the Physician Response to Uncertainty scale pre instruction, post instruction, and then 4-6 weeks after instruction.

Results: Answers to the Physician Response to Uncertainty scale were similar pre, post, and delayed post instruction, except for responses to “I usually feel anxious when I am not sure of a diagnosis.” Using a Kruskal Wallis and Wilcoxon Rank Tests to examine unpaired and paired responses, anxiety in response to uncertainty about a diagnosis significantly decreased (p value 0.004, p value 0.046) in both analyses. There was, however, a slight rebound effect, as answers in the delayed responses showed an increase in anxiety compared to the immediate post responses, although still insignificantly less than pretest responses.

Conclusion(s): This medical education study showed that instruction on diagnostic uncertainty can help allay trainee anxiety about uncertainty, although without further reinforcement, that effect can fade. Diagnostic uncertainty is another type of difficult conversation we have with families, and like other difficult conversations, can be taught to the benefit of trainees and patients alike.

Table 1
Paired analysis of pre and post intervention survey responses.

Figure 1
Paired comparison of pre and post intervention survey responses to "I usually feel anxious when I am unsure of a diagnosis"

Figure 2
Unpaired comparison of pre,pPost, and delayed post intervention survey responses to "I usually feel anxious when I am unsure of a diagnosis"