The diagnostic accuracy of physical signs: practice experience compared with research evidence

Background: Physical diagnosis is the first course of clinical medicine and the first contact with patients during clinical practice. Evidence-based physical diagnosis is a helpful tool in both education and practice. We aim to compare the accuracy of physical signs based on practice experience and research evidence. Methods: A questionnaire was designed and administered using convenience sampling to collect empirical estimates of physical signs. The literature search was conducted using Medline, EMBASE, and the Cochrane Library up to December 2023. Included studies independently compared diagnostic performance between physical signs and acceptable diagnostic standards. Hierarchical summary receiver operating characteristic models were used to summarize test accuracy data across studies. The pooled sensitivity, specificity, likelihood ratios, and diagnostic odds ratio were calculated. Results from the two parts were compared. Results: A total of 1971 respondents completed the questionnaire, and 37 studies reporting the performance of physical signs were included in the review. Compared with synthesized results of meta-analysis, estimations of palpable liver (sensitivity and specificity), palpable spleen (sensitivity and specificity), Traube space dullness (sensitivity and specificity), shifting dullness (sensitivity), right upper quadrant (RUQ) tenderness (sensitivity and specificity), Murphy sign (specificity), and distended abdomen (sensitivity) had relatively good concordance. The sensitivity of bulging flanks and the specificity of the distended abdomen were underestimated. The specificity of bulging flanks, shifting dullness, and sensitivity of the Murphy sign were overestimated. Conclusions: Variable results were observed in the comparison between clinical experience and evidence-based medicine. Evidence-based physical diagnosis should be introduced during teaching and emphasized throughout a career as a guide for accumulating expertise.