Understanding dropout and non-participation in follow-up evaluation for the benefit of patients and research: evidence from a longitudinal observational study on patients with eating disorders.

Treatment outcomes in eating disorders (EDs) are still an open field for clinicians and researchers. Besides difficulties in egosyntonic-linked treatment engagements, dropout is one of the most crucial elements that cause a reduction in the treatment efficacy. Thus, the aim of this study is to evaluate factors that could contribute to high dropout rates and non-participation in follow-up evaluation in patients with ED. This study used a large sample of patients from a specialized ED ward and day hospital (DH). A sample of 428 individuals was recruited for this study. Psychological and demographic data were collected at the time of hospitalization and discharge from the facilities. These data were used to explore a possible link between dropout and follow-up non-participation. Specially, the random forest was used to rank demographic and psychological features in importance and evaluate the top results with regression analyses for statistical significance. A dropout rate of 12.14% during inpatient and DH treatment was found. Anger-hostility and general psychopathology were found to be predictors of dropout during treatment, while the duration of the hospitalization predicted non-participation at the six-month follow-up. Specific psychological features should be considered before and during treatments for patients with EDs to reduce dropout rates. The duration of the hospitalization should also be evaluated as a relevant healthcare element that could affect engagement and, accordingly, outcome.