Related Experiment Video
Updated: Jan 12, 2026

Assessing Activity-based Anorexia in Mice
Published on: May 14, 2018
Maximal Weight Loss Is Strongly Correlated With Maximal BMI in Adolescents With Anorexia Nervosa and Atypical
Abigail Matthews1, Norbert Quadflieg2, Johannes Hebebrand3
1Department of Psychiatry and Psychology, Mayo Clinic, Rochester, Minnesota, USA.
Introduction:
In hospitalised patients with both anorexia nervosa (AN) and atypical AN, premorbid body mass index (BMI; kg/m2), estimated from premorbid weight (i.e., weight at the onset of weight loss preceding eating disorder development) and admission height, strongly predicts admission weight loss (i.e., premorbid weight minus admission weight). We hypothesised that maximal weight loss (i.e., maximal historical weight minus minimal weight since eating disorder onset) is also correlated with maximal BMI.
Methods:
In 520 female adolescents with AN (n = 503) and atypical AN (n = 17), relationships between admission age, admission and maximal weight loss, and maximal and minimal BMI were analysed with non-parametric Spearman's correlation coefficients.
Results:
Like premorbid BMI in previous studies, maximal BMI was strongly correlated with maximal weight loss.
Discussion:
Maximal BMI appears to be a sound substitute for premorbid BMI in correlations with admission weight loss in hospitalised adolescents with AN and atypical AN. Consistent use of maximal (vs. premorbid) BMI and admission (vs. maximal) weight loss in research would align with the metric weight suppression (derived from both maximal and current weight), thus facilitating cross study comparisons. Additional studies should assess correlations between both premorbid and maximal BMI and admission or maximal weight loss to attempt to identify the variables with the highest predictive value.
Related Concept Videos
Anorexia Nervosa
Symptoms and Physical Effects
Individuals with anorexia nervosa commonly exhibit extreme...
Bulimia Nervosa
Binge Eating Disorders
Coefficient of Correlation
If you suspect a linear relationship between x and y, then r can measure how strong the linear relationship is.
What the VALUE of r tells us:
The value of r is always between –1 and +1: –1 ≤ r ≤ 1.
The size of the correlation r indicates the...
Obesity
Regression Toward the Mean

