Comorbidity in trichotillomania (hair-pulling disorder): A cluster analytical approach.

BACKGROUND

A promising approach to reducing the phenotypic heterogeneity of psychiatric disorders involves the identification of homogeneous subtypes. Careful study of comorbidity in obsessive-compulsive disorder (OCD) contributed to the identification of the DSM-5 subtype of OCD with tics. Here we investigated one of the largest available cohorts of clinically diagnosed trichotillomania (TTM) to determine whether subtyping TTM based on comorbidity would help delineate clinically meaningful subgroups.

METHODS

As part of an ongoing international collaboration, lifetime comorbidity data were collated from 304 adults with pathological hair-pulling who fulfilled criteria for DSM-IV-TR or DSM-5 TTM. Cluster analysis (Ward's method) based on comorbidities was undertaken.

RESULTS

Three clusters were identified, namely Cluster 1: cases without any comorbidities (n = 63, 20.7%) labeled "simple TTM," Cluster 2: cases with comorbid major depressive disorder only (N = 49, 16.12%) labeled "depressive TTM," and Cluster 3: cases presenting with combinations of the investigated comorbidities (N = 192, 63.16%) labeled "complex TTM." The clusters differed in terms of hair-pulling severity (F = 3.75, p = .02; Kruskal-Wallis [KW] p < .01) and depression symptom severity (F = 5.07, p = <.01; KW p < .01), with cases with any comorbidity presenting with increased severity. Analysis of the temporal nature of these conditions in a subset suggested that TTM onset generally preceded major depressive disorder in (subsets of) Clusters 2 and 3.

CONCLUSIONS

The findings here are useful in emphasizing that while many TTM patients present without comorbidity, depression is present in a substantial proportion of cases. In clinical practice, it is crucial to assess comorbidity, given the links demonstrated here between comorbidity and symptom severity. Additional research is needed to replicate these findings and to determine whether cluster membership based on comorbidity predicts response to treatment.