Comparing family-based treatment with inpatient treatment in youth with anorexia nervosa eligible for hospitalization: A 12-month feasibility study.

OBJECTIVE: Family-based treatment (FBT) for youth with anorexia nervosa (AN), has not been compared to inpatient, multimodal treatment (IMT). METHOD: Prospective, non-randomized pilot feasibility study of adolescents with AN receiving FBT (n = 31), and as a reference point for exploratory outcome comparisons IMT (n = 31), matched for baseline age and percent median BMI (%mBMI). Feasibility of FBT in youth fulfilling criteria for IMT was assessed via study recruitment and retention rates; acceptability via drop-out and caregiver strain; safety via adverse events; preliminary treatment effectiveness between groups was assessed via a change in %mBMI, AN psychopathology (Eating Disorder Examination-Questionnaire, EDE-Q), and hospital days, over 12 months with intent-to-treat, mixed models repeated measures analyses covering post-intervention usual care until 12 months. RESULTS: Taking into account that 8 FBT patients (25.8%) crossed over to IMT due to lack of weight gain or psychiatric concerns, FBT and IMT were similarly feasible, acceptable, and safe, apart from more physical antagonism toward others in FBT (p = .010). FBT lasted longer (median [interquartile range, IQR]; 33.6 [17.4, 49.9] vs. 17.3 [14.4, 24] weeks, p < .001), but required fewer hospital days than IMT (median, [IQR], FBT = 1 [0, 16] vs. IMT = 123 [101, 180], p < .001). Baseline comorbidity-adjusted changes over 12 months did not differ between groups in %mBMI (FBT = 12.6 ± 11.9 vs. IMT = 13.7 ± 9.1; p = .702) and EDE-Q global score (median, [IQR]; FBT = -1.2 [-2.3, 0.2] vs. IMT = -1.3 [-2.8, -0.4]; p = .733). DISCUSSION: Implementing FBT in this pilot study was feasible, acceptable, and safe for youth eligible for IMT according to German S3 guidelines. Non-inferiority of FBT versus IMT requires confirmation in a sufficiently large multicenter RCT. PUBLIC SIGNIFICANCE: This pilot study with 62 adolescent patients with anorexia nervosa demonstrated that for 2/3rd of patients eligible for a long hospitalization in the German health care system, outpatient, Family-based treatment (FBT) was a safe and feasible treatment alternative. Over 12 months, FBT lead to similar weight gain and reduction in eating disorder cognitions as inpatient treatment with fewer hospital days. This pilot study needs to be followed up by a larger, multicenter trial.

Geographic location determines beta-cell autoimmunity among adult Ghanaians: Findings from the RODAM study.

INTRODUCTION: Beta-cell autoantibodies are established markers of autoimmunity, which we compared between Ghanaian adults with or without diabetes, living in rural and urban Ghana and in three European cities. METHODS: In the multicenter cross-sectional Research on Obesity and Diabetes among African Migrants (RODAM) study (N = 5898), we quantified autoantibodies against glutamic acid decarboxylase (GAD65Ab) by radioligand binding assay (RBA) and established cut-offs for positivity by displacement analysis. In a subsample, we performed RBA for zinc transporter-8 autoantibodies (ZnT8Ab). Associations of environmental, sociodemographic, and clinical factors with GAD65Ab were calculated. RESULTS: In this study population (age: 46.1 ± 11.9 years; female: 62%; Ghana-rural: 1111; Ghana-urban: 1455; Europe: 3332), 9.2% had diabetes with adult-onset. GAD65Ab concentrations were the highest in Ghana-rural (32.4; 10.8-71.3 U/mL), followed by Ghana-urban (26.0; 12.3-49.1 U/mL) and Europe (11.9; 3.0-22.8 U/mL) with no differences between European cities. These distributions were similar for ZnT8Ab. Current fever, history of fever, and higher concentrations of liver enzymes marginally explained site-specific GAD65Ab concentrations. GAD65Ab positivity was as frequent in diabetes as in nondiabetes (5.4% vs 6.1%; P = .25). This was also true for ZnT8Ab positivity. CONCLUSION: Geographic location determines the occurrence of GAD65Ab and ZnT8Ab more than the diabetes status. Beta-cell autoimmunity may not be feasible to differentiate diabetes subgroups in this population.