To predict the prognosis of adolescents with anorexia nervosa leaving inpatient treatment: the development and initial evaluation of a novel tool to be used by a multi-disciplinary team.

Abstract

Purpose: This study aimed to develop and evaluate the Newbridge Prognosis Score (NPS), a prognostic questionnaire designed for use by a multidisciplinary team (MDT) when an anorexia nervosa (AN) patient is discharged from an inpatient treatment facility for eating disorders. The tool integrates the collective judgment of MDT members to predict short-term outcomes in adolescents with AN, focusing on physical, psychological, and behavioural factors and systematically assessing psychopathology, weight, and continued treatment needs. This information is intended to guide individualized community support, enhance post-discharge recovery, and aid in the allocation of limited community-based resources.

Methods: A group of adolescent girls and boys undergoing inpatient treatment for AN participated in the study. In addition, a matched control sub-sample was created. Upon discharge, the multidisciplinary team scored the Newbridge Prognosis Score (NPS), and follow-up data were collected 6 months later from the young person and/or their parents or carers.

Results: The NPS significantly correlates with key recovery items, such as weight maintenance, lower eating disorder psychopathology, subsequent need for readmission and engagement in follow-up. Higher NPS scores predict poorer outcomes, such as more severe psychopathology, lower weight or weight loss and higher readmission risk at follow-up. Psychological factors are strong predictors of post-discharge prognosis, with the suggestion that those sufferers may require enhanced psychological support. The NPS is more effective at predicting long-term readmission risk than short-term outcomes. However, the NPS explained only a proportion of the variance in these outcomes and sensitivity in predicting readmission within the matched control sample was mixed.

Conclusions: The NPS shows promise as a potential tool for predicting short-term outcomes following AN treatment. While promising, further refinement of the tool is needed, prior to being validated for use in clinical practice.

Level of evidence: Level III. Evidence obtained from a well-designed cohort or case-controlled analytic studies.