Abstract
BACKGROUND: A long duration of untreated illness (DUI) is an unfavorable prognostic factor in anorexia nervosa (AN) and is associated with chronic illness progression. Although previous preventive measures aimed at reducing DUI and thus improving short- and long-term treatment outcomes have been partially successful, a better understanding of the factors involved in the sensitive phase prior to treatment initiation is needed. To date, there is no validated instrument available to assess these factors specifically for patients with AN. The FABIANA-project (Facilitators and barriers in anorexia nervosa treatment initiation) aims at identifying predictors of the DUI in order to target preventive measures better in the future. As part of this project, the FABIANA-checklist was developed, based on a multi-informant perspective and a multimodal bottom-up approach. The present study focusses on the process of item generation, item selection and psychometric validation of the checklist. METHODS: Based upon a previous qualitative study, an initial set of 73 items was generated for the most frequently mentioned facilitators and barriers of treatment initiation in AN. After a process of consensual rating and cognitive pre-testing, the resulting 25-item version of the FABIANA-checklist was provided to a sample of female patients (N = 75), aged ≥ 14 years with AN that underwent their first psychotherapeutic treatment in the last 12 months. After item analysis, dimensionality of the final version of the FABIANA-checklist was tested by Principal Component Analysis (PCA). We evaluated construct validity assuming correlations with related constructs, such as perceived social support (F-SozU), support in the health care system (PACIC-5A), illness perception and coping (BIPQ). RESULTS: We included 54 adult and 21 adolescent patients with AN, aged on average 21.4 years. Average BMI was 15.5 kg/m(2), age of onset was 19.2 years and average DUI was 2.25 years. After item analysis, 7 items were excluded. The PCA of the 18-item-FABIANA-checklist yielded six components explaining 62.64% of the total variance. Overall internal consistency was acceptable (Cronbach's α = .76) and construct validity was satisfactory for 14 out of 18 items. Two consistent components emerged: "primary care perceived as supportive and competent" (23.33%) and "emotional and practical support from relatives" (9.98%). With regard to the other components, the heterogeneity of the items led to unsatisfactory internal consistency, single item loading and in part ambiguous interpretability. CONCLUSIONS: The FABIANA-checklist is a valid instrument to assess factors involved in the process of treatment initiation of patients with AN. Psychometrics and dimensionality testing suggests that experienced emotional and practical support from the primary health care system and close relatives are main components. The results indicate that a differentiated assessment at item level is appropriate. In order to quantify the relative importance of the factors and to derive recommendations on early-intervention approaches, the predictive effect of the FABIANA-items on the DUI will be determined in a subsequent study which will further include the perspective of relatives and primary caregivers. Trial registration Clinical Trials.gov Identifier: NCT03713541: https://clinicaltrials.gov/ct2/show/NCT03713541 .