Eating disorders (EDs) represent a significant public health concern, with a high prevalence among adolescent girls. The COVID-19 pandemic has led to a marked increase in EDs, underlining the need for a brief screening tool in clinical settings. The present study compared ICD-10-based ED-diagnoses with SCOFF-identified at-risk female patients in a clinical setting and assessed the tool's reliability. Routine data (n = 955 participants) from a child and adolescent psychiatric hospital were evaluated regarding the incidence of ED-diagnoses before (n = 312) and during the pandemic (n = 643) and compared with SCOFF results. The SCOFF's sensitivity and specificity were examined. ED-diagnoses significantly increased, almost doubling from before to during the pandemic. The number of at-risk patients identified by the SCOFF likewise differed between the two time points. The SCOFF showed high sensitivity at cut-offs of 2 (pre-pandemic) and 3 (peri-pandemic) but low specificity. Internal consistency was poor. The findings align with previous research indicating an elevated rate of admission to clinical care for ED-diagnoses during the pandemic, a trend also reflected in the SCOFF results. However, the SCOFF's low specificity and poor internal consistency raise concerns about its applicability among adolescent psychiatric inpatients.