A significant proportion of patients suffering from Eating Disorders (ED) present a comorbidity with anxiety disorders. Among the anxiety disorders, Obsessive Compulsive Disorders (OCD) are the third most frequently diagnosis observed in ED. However, prevalence rates from the literature are contradictory depending on the diagnostic criteria and evaluation tools used. Studies concerning the chronology of appearance of OCD and ED and the role played by denutrition are even rarer and equally contradictory.
Objective: The aim of this study is to bring new empirical data to the study of the correlations between OCD and ED by exploring, in a significant clinical sample, the current and lifetime OCD comorbidity in the diagnostic sub-groups and sub-types of ED as defined by the DSM IV and to study the chronology of appearance of these disorders taking into account the role played by denutrition. We make the assumption that there should be a difference in the prevalence of obsessive compulsive disorders in the various ED sub-groups and sub-types and that purging anorexics should, at equivalent levels of denutrition, exhibit higher OCD prevalence rates than the other sub-types as a result of their more severe general psychopathology.
Methods: Current and lifetime prevalence were investigated using the Mini International Neuropsychiatric Interview (MINI) and the Yale-Brown Obsessive Compulsive Scale in a sample of 89 DSM IV eating disorders in and out-patients aged between 15 and 30 (58 AN and 31 BN) and 89 matched controls. In order to increase the validity of the current diagnosis of OCD only patients presenting an OCD diagnosis on the MINI (excluding obsessions and compulsions related to food and body image) and a score of 16 or more on the Y-BOCS were included in the study.
Results: Current and lifetime prevalence of OCD in ED is significantly higher than in general population (15.7% and 19% vs 0% and 1.1%, p < 0.05). Anorexic and bulimic patients do not show any difference in the current and lifetime comorbidity (19% and 22.4% vs 9.7% and 12.9%, ns). Concerning the diagnostic sub-types, following our assumption purging anorexics present the higher current and lifetime prevalences (29% and 43%), followed by restrictive anorexia (16%) and purging bulimia (13%). The only significant difference is found in the lifetime prevalence of OCD between purging and non purging anorexics. In the great majority of cases (65%) OCD diagnosis precedes ED diagnosis and OCD current prevalence and Y-BOCS scores of underweight patients are not significantly higher than normal weight patients, suggesting that there are only limited links between denutrition and obsessionality.
Limits: Some methodological limitations must be considered. First of all, the small sample size of the diagnostic sub-types of ED do not enable us to draw definitive conclusions concerning the prevalence of OCD in the clinical forms of ED. Some of the diagnostic sub-types of ED, as BN-NP, appear at different ages: in order to better investigate the prevalence of OCD in all the diagnostic sub-types, larger age ranges should be considered. Secondly, our sample of anorexics is almost made of hospitalized inpatients recrutated in a specialized university unit, whereas bulimics are all consulting outpatients. It is possible that the higher than normal rates of OCD in anorexics could be related to the severity of this sample. Comparisons with samples of consulting anorexics should be performed. Lastly, it is necessary to evoke the limitation represented by the choice of a healthy control group. OCD are rare in the general population and the differences found between the clinical and the control groups do not offer strong arguments to support the specificity of the association between OCD and ED, which would be better explored by using a pathological control group.
Conclusions: So far, none of the various models proposed to explain the comorbidity observed between ED and OCD appears completely satisfactory. ED and OCD would share a genetic vulnerability to a dysregulation of serotoninergic functioning which would predispose these subjects, depending on specific clinical and biological conditions, to develop an obsessional and compulsive symptomatology more or less focused on food and body image. The knowledge of the clinical and biological specificities of the ED diagnostic sub-types presenting an OCD comorbidity could point the way towards specific therapeutic interventions in these patients. Our study of the comorbidity of OCD in ED makes a further contribution to the identification of specific links between the OCD and the various clinical forms of ED. More clinical and biological studies are needed to further explore these relationships.