Reliability of clinical examinations for pediatric skin and soft-tissue infections.

Abstract

To determine the interrater reliability of clinical examination by pediatric emergency medicine physicians for the diagnosis of skin and soft-tissue infections (SSTIs). A cross-sectional study of patients presenting to a pediatric emergency department with SSTIs was performed. Each lesion was examined by a treating physician and a study physician (from a pool of 62 physicians) at the bedside during the emergency department visit. The primary outcome was reliability, as measured with the weighted κ statistic, for determining whether the lesion was an abscess and whether the lesion required a drainage procedure. A total of 371 lesions were analyzed for interrater reliability. The weighted κ value for diagnosis of the lesion as an abscess was 0.39 (95% confidence interval: 0.32-0.47), and that for assessment of the need for drainage was 0.43 (95% confidence interval: 0.36-0.51). Agreement was statistically more likely for lesions in children ≥ 4 years of age but was not more likely for lesions in nonblack patients, lesions in patients with a history of or exposure to a close contact with a SSTI, or lesions examined by 2 experienced pediatric emergency medicine physicians. Among the 62 participating physicians at our site, the reliability of the clinical examination was poor. This may indicate that improved education and/or more-objective means for diagnosing these infections in the acute care setting are warranted. Additional studies are needed to determine whether these results are generalizable to other settings.