PS-119 Diagnostic Decision Making In Paediatrics

Abstract

Background and aims The cognitive processes underlying diagnostic thinking are complex and the strategies used by paediatricians to arrive at a diagnosis are poorly described. The aim of this study is to identify the most frequently used diagnostic strategies in hospital paediatrics. Methods The online survey employed a 7-point Likert scale and was designed using a three-stage model of diagnostic reasoning. It was pre-tested and administered using SurveyMonkey, an online survey tool. Participants were invited to participate by email. Results The overall response rate was 38% (118/310). Respondents included paediatric consultants (31.6%), registrars (44.5%) and senior house officers (21.4%). Respondents were practicing paediatrics for a median of 7 years. Figure 1 summarises the frequency of use of different diagnostic strategies. Diagnostic strategies are frequently combined within a single consultation to both initiate and refine a diagnosis. Diagnostic strategies were generally equally utilised among different groups, regardless of level of experience. Trainees use a ‘test of treatment’ more frequently to define a diagnosis compared to consultants (0 = 0.35). Conclusion Restricted rule-outs, a strategy aimed at preventing errors in clinical practice, is commonly used to refine a diagnosis. Probabilistic reasoning, using a clinical sign or test to ‘rule in’ or ‘rule out’ a diagnosis, is also frequently utilised but is prone to diagnostic error. It requires awareness of the diagnostic accuracy of tests and the impact of false positive and false negative results on the probability of disease. Understanding the cognitive processes underlying diagnostic thinking can improve decision-making and decrease diagnostic error.