Lack of efficacy of psychological and pharmacological treatments of disorders of eating behavior: neurobiological background

BMC Psychiatry. 2014 Dec 24:14:376. doi: 10.1186/s12888-014-0376-7.


Background: Treatments of eating disorders result too often in partial psychological and physical remission, chronicization, dropout, relapse and death, with no fully known explanations for this failure. In order to clarify this problem, we conducted three studies to identify the biochemical background of cognitive-behavioural psychotherapy (CBT), individual psychology brief psychotherapy (IBPP), and psychotherapy-pharmacotherapy with CBT + olanzapine in anorexics (AN) and bulimics (BN) by measuring the levels of plasma homovanillic acid (HVA) for dopamine secretion, plasma 3-methoxy-4-hydroxy-phenylglycol (MHPG) for noradrenalin secretion, and platelet [3H]-Paroxetin-binding Bmax and Kd for serotonin transporter function. The data were then compared with psychopathological and physical alterations.

Methods: Study 1 investigated the effects of 4 months of CBT on plasma HVA, MHPG and [3H]-Par-binding in 14 AN-restricted, 14 AN-bingeing/purging, and 22 BN inpatients. Study 2 investigated the effects of 4 months of IBPP on plasma HVA in 15 AN and 17 BN outpatients. Study 3 investigated the effect of 3 months of CBT + olanzapine (5 mg/day) in 30 AN outpatients. The data were analyzed using one-way ANOVA for repeated measures for the changes between basal and post-treatment biological and psychological parameters, two-way ANOVA for repeated measures for the differences in the psychobiological data in the 3 groups, Spearman's test for the correlations between basal and final changes in the psychological and biological scores.

Results: Study 1 revealed significant amelioration of the psychopathology in the AN and BN patients, no effects on HVA, MHPG or Paroxetin binding Kd, and a significant increase in Par-binding Bmax only in the BN patients. Study 2 revealed a significant effect of IBPP on psychopathology in the AN and BN patients, and a significant increase in HVA only in the BN patients. Study 3 revealed a significant positive effect of CBT + olanzapine therapy on the psychopathology and increased HVA values. No correlations were observed in the 3 groups between biological and psychological effects of the three treatments.

Conclusions: Our data advance suggestions on the mechanism of action of the three therapies; however, the lack of correlations between biochemical and psychological effects casts doubt on their significance. Clinical Identifier NCT01990755 .

Publication types

  • Randomized Controlled Trial

MeSH terms

  • Administration, Oral
  • Adult
  • Analysis of Variance
  • Anorexia Nervosa / blood
  • Anorexia Nervosa / therapy*
  • Antipsychotic Agents / therapeutic use*
  • Benzodiazepines / therapeutic use*
  • Bulimia Nervosa / blood
  • Bulimia Nervosa / therapy*
  • Cognitive Behavioral Therapy / methods*
  • Combined Modality Therapy
  • Double-Blind Method
  • Female
  • Homovanillic Acid / metabolism
  • Humans
  • Male
  • Methoxyhydroxyphenylglycol / metabolism
  • Olanzapine
  • Paroxetine / metabolism
  • Psychotherapy, Brief / methods*
  • Serotonin Plasma Membrane Transport Proteins / metabolism
  • Treatment Failure


  • Antipsychotic Agents
  • Serotonin Plasma Membrane Transport Proteins
  • Benzodiazepines
  • Paroxetine
  • Methoxyhydroxyphenylglycol
  • Olanzapine
  • Homovanillic Acid

Associated data