Comparison of the 4-digit diagnostic code and the Hoyme diagnostic guidelines for fetal alcohol spectrum disorders.

OBJECTIVE The 4-Digit Diagnostic Code for fetal alcohol spectrum disorders and the Hoyme fetal alcohol spectrum disorders diagnostic guidelines differ markedly. The performances of the 2 diagnostic systems were compared. METHODS The fetal alcohol syndrome diagnostic criteria from the 4-Digit Code and Hoyme guidelines were applied to 952 patients who had received an interdisciplinary, fetal alcohol spectrum disorders, diagnostic evaluation at the University of Washington with the 4-Digit Diagnostic Code and 16 children with confirmed absence of prenatal alcohol exposure. RESULTS The prevalence of fetal alcohol syndrome was 3.7% with the 4-Digit Code and 4.1% with the Hoyme guidelines. Although the prevalences were similar, the patients identified were not. Only 17 individuals met the fetal alcohol syndrome criteria for both systems. An extraordinary number of patients (35%) met the Hoyme criteria for the fetal alcohol syndrome facial phenotype, but only 39 of those 330 patients met the Hoyme criteria for fetal alcohol syndrome. Even some children with no alcohol exposure (25%) had the Hoyme fetal alcohol syndrome face. The specificities of the Hoyme fetal alcohol syndrome face for the Hoyme fetal alcohol syndrome diagnosis and prenatal alcohol exposure were low in these populations. CONCLUSIONS Without a specific facial phenotype, a valid diagnosis of fetal alcohol syndrome cannot be rendered for patients with prenatal alcohol exposure, because a causal link between their outcomes and exposure cannot be established, and a valid diagnosis of fetal alcohol syndrome cannot be rendered for patients with unknown alcohol exposure, because the face cannot serve as a valid proxy measure for alcohol exposure. Diagnostic guidelines must confirm the specificity of their fetal alcohol syndrome facial criteria to validate their diagnostic criteria.