Development and validation across trimester of the Prenatal Eating Behaviors Screening tool

Abstract

Rapid screening tools are useful for identifying at-risk patients and referring them for further assessment and treatment, but none exist that consider the unique medical needs of pregnant women with eating disorders (EDs). There is a need for a rapid, sensitive, and specific screening tool that can be used to identify a potential ED in pregnancy. We started with a set of 25 questions, developed from our qualitative work along with other ED screening tools, and tested on a development (n = 190) and validation sample (n = 167). Statistical analysis included factor analysis and logistic regressions with ROC curves. Development and validation samples were combined for trimester analysis (n = 357). Refining the tool to 12 items demonstrated strong internal reliability (development alpha = 0.95, validation alpha = 0.91). With correlated errors, questions demonstrated acceptable CFA fit (development: GFI: 0.91, RMSEA: 0.10, NNFI: 0.95; validation: GFI: 0.85, RMSEA: 0.14, NNFI: 0.86). Similar fits were seen by trimester: first trimester n = 127, GFI: 0.89, RMSEA: 0.12, NNFI: 0.94; second trimester n = 150, GFI: 0.83, RMSEA: 0.14, NNFI: 0.88; third trimester n = 80, GFI: 0.99, NNFI: 0.99. Validation against current ED diagnosis demonstrated acceptable sensitivity and specificity using a cutoff of 39 (development sensitivity = 80.7%, specificity = 79.7%, OR = 16.42, 95% CI: 7.51, 35.88; validation sensitivity = 69.2%, specificity = 86.5%, OR: 17.43, 95% CI: 6.46, 47.01). Findings suggest the PEBS tool can reliably and sensitively detect EDs across pregnancy trimesters with 12 questions. A further implication of this work is to reduce health and mental health treatment disparities through this standard and rapid screening measure to ensure early identification and treatment.