Here we revisit and reinterpret the original study in which the so-called‘Maudsley (London) model’ of family therapy was compared with individualtherapy for anorexia nervosa. Family therapy was more effective inadolescents with a short duration of illness. However, this is only part ofthe story. A later study describing the 5-year outcome contains importantinformation. Those adolescents randomised to family therapy achieved abetter outcome 5 years later. Moreover, the group with an onset inadolescence but who had been ill for over 3 years had a poor response toboth family and individual therapy, suggesting that unless effectivetreatment is given within the first 3 years of illness onset, the outcome ispoor. We examine other evidence supporting this conclusion and consider thedevelopmental and neurobiological factors that can account for this.