Defining and understanding treatment non-completion and non-response in family based treatment, for anorexia nervosa or atypical anorexia nervosa in children and adolescents: a systematic review

Abstract Background Family-Based Treatment (FBT) is a first-line treatment for anorexia nervosa in childhood and adolescence, however, remission rates following treatment remain modest. This systematic review aimed to examine definitions of, and identify factors that may influence, non-completion or non-response to FBT in children and adolescents with anorexia nervosa or atypical anorexia nervosa (AN/AAN). Methods A systematic review was conducted according to PRISMA guidelines. Three databases (APA PsycINFO, Embase, and Web of Knowledge) were searched from 2005 to 4th August 2025, supplemented by citation searching. Inclusion criteria were studies in any language examining FBT for AN/AAN, in participants aged 18 years or under, in outpatient settings. Quality assessment was completed using the Joanna Briggs Institute critical appraisal tools. The review was prospectively registered (PROSPERO CRD420251114023). Results Searches identified 1794 articles, 988 following removal of duplicates. Fifty-five papers examined treatment non-completion and/or non-response (25 primary studies; 30 secondary analytic studies; N = 2,820). Definitions of treatment non-completion and non-response varied significantly. Non-completion rates ranged from 2% to 55% but in more than half of papers, this group was not characterised further. Non-response rates ranged from 14% to 71%. Factors associated with treatment non-response included lower baseline weight and absence of early weight gain, identified across seven cohorts each. Older age, longer illness duration, and higher parental expressed emotion were associated with poorer treatment response in several cohorts. Stronger therapeutic alliance and greater parental self-efficacy were associated with better treatment outcomes. Factors associated with non-completion were inconsistent, though weaker therapeutic alliance and higher parental criticism showed associations. Conclusions Heterogeneity in definitions, outcome measures and study designs limited the ability to draw definitive conclusions about treatment non-completion and non-response in FBT. Studies often lacked analysis of treatment non-completion, representing a gap in the literature. However, older age, lack of early weight gain, higher parental expressed emotion, and weaker therapeutic alliance were associated with treatment non-response. Standardisation of definitions and outcome reporting would facilitate future reviews and synthesis of findings across studies.