Distinct mental health and substance use patterns in youth who endorse eating disorder risk using the eating disorder screen for primary care

Background Comorbidity of eating disorders (EDs) and substance use disorders (SUDs) in youth is common, although ED symptomology is not often captured in substance use research. Accounting for ED presentations may explain variability in substance use outcomes, as well as mental health comorbidities that contribute to substance use. ED screening tools may be well-positioned to capture ED risk among youth recruited for substance use research due to their brief nature. The current study assessed the utility of an existing ED screening tool, the Eating Disorder Screen for Primary Care (ESP), for detecting ED risk in adolescents and young adults with varying levels of substance use. Additionally, patterns of mental health vulnerabilities and substance use were assessed between ESP risk groups to understand mechanisms that may contribute to concurrent ED and substance use presentations in youth. Methods Data were collected from 682 youth (ages 15–25) across 722 visits using a standardized intake process. The ESP and a battery of other measures assessing emotional dysregulation, impulsivity, depression, anxiety, substance use consequences, and past 30-day substance use were collected. A range of ESP cut-off scores (2–5) were assessed to measure the sensitivity, specificity, positive-and-negative predictive value, and the resulting receiver operator characteristic (ROC) for classifying EDs. After identifying an optimal ESP cut-off for our sample, Wilcoxon rank sum tests were performed to analyse the difference between ED risk groups on the mental health, substance use consequences, and past 30-day substance use outcomes. Results The ESP yielded high sensitivity but poor specificity for detecting ED diagnoses using the clinical cut-off score ≥ 2. ROC curves indicated a score of ≥ 3 improved the balance of sensitivity and specificity for detecting ED diagnoses. Individuals who were classified at ED risk (ESP score ≥ 3) reported significantly higher emotional dysregulation, negative urgency, depression and anxiety, and problems related to cannabis (MAPI) and nicotine (PSECDI) use, in addition to lower quantify and frequency of alcohol use, yet higher cannabis use frequency relative to those not at risk. Conclusions Our tailored ESP cut-off (≥ 3) demonstrated acceptable diagnostic accuracy for detecting ED diagnoses in youth recruited for substance use research. The revised cut-score indicated individuals who endorse ED risk may be more vulnerable to experiencing maladaptive thought patterns and behaviours, as well as greater frequency of cannabis use, and greater problems related tocannabis and nicotine use.