Does Comorbidity Matter in Body Focused Repetitive Behavior Disorders?

Abstract

Background

Trichotillomania (TTM) and skin picking disorder (SPD) have been characterized as Body Focused Repetitive Behavior Disorders (BFRBs). Because BFRBs frequently co-occur, we sought to discover the similarities and differences between having both TTM and SPD as opposed to just one.

Methods

421 participants with primary TTM were evaluated regarding the comorbidity of SPD and 124 were participants with primary SPD were evaluated regarding the comorbidity of TTM. The effects of comorbid overlap on demographic and clinical measures were evaluated.

Results

Of the 421 participants with primary TTM, 61 (14.5%) had co-occurring SPD. Of 124 adults with primary SPD, 21 (16.9%) had comorbid TTM. Those with primary TTM and comorbid SPD had significantly more severe trichotillomania symptoms and were more likely to have major depressive disorder than those with TTM alone. Those with primary of SPD and comorbid TTM reported significantly more severe skin picking symptoms than those who only had SPD.

Conclusions

Individuals with co-occurring TTM and SPD may have more problematic hair pulling or skin picking symptoms. Hair pullers with comorbid SPD were more likely to have comorbid depression. Evaluating people for multiple BFRBs may be important to assess severity of symptoms and may have treatment implications.

Introduction

Trichotillomania (TTM) and skin picking disorder (SPD) have both been characterized as body focused repetitive behavior disorders (BFRBs) (1–4). Their suggested grouping as BFRBs arises from parallel/common features: both disorders are more common in women (5), involve an uncontrolled repetitive behavior of excessively grooming oneself (2), and appear to share some genetic and neurobiological overlap (6–8).

TTM and SPD frequently co-occur (3, 9–10). In terms of clinical presentation, some preliminary research suggests that individuals with co-occurring TTM and SPD spend more time engaging in pulling or picking behavior each day compared to patients with TTM alone (11). In addition, self-reported data from TTM patients with comorbid BFRBs also illustrate greater “focused” pulling and perceived disability, as compared to non-comorbid cases (3).

While our clinical understanding of the impact of TTM and SPD comorbidity remains limited, more extensive research is available exploring relationships between other so-termed obsessive-compulsive and related disorders (OCRDs). The category of OCRDs, now recognized in DSM-5, includes obsessive compulsive disorder, body dysmorphic disorder, and hoarding disorder, alongside TTM and SPD. For example, adults with OCD and comorbid BDD demonstrate greater OCD symptom severity, earlier age at onset of OCD, as well as higher rates of co-occurring depression and anxiety compared to those with OCD alone (12). Similarly, BDD patients with comorbid OCD demonstrate greater severity of BDD (13), but no significant differences in age at onset compared to those with BDD alone (14). Hoarding Disorder, being a new disorder for DSM-5, has received less attention. Collectively, the available studies – mostly in OCD – indicate a possible synergistic effect when someone has multiple OCRDs, in terms of symptom severity (1).

Given these previous findings regarding the clinical significance of recognizing the comorbidity among OCRDs, the aim of the current study was to examine whether having co-occurring TTM and SPD is associated with greater symptom severity, rates of psychiatric comorbidities, and psychosocial dysfunction than having either TTM or SPD alone.

Methods

Subjects

Data from 421 participants with primary TTM (398 [94.5%] females; mean age 28.9 ± 11.6 years) from both treatment and non-treatment studies from the University of Chicago, University of Minnesota, and the Massachusetts General Hospital were included in this study. Participants with primary TTM were recruited from 2006 to 2016. Additionally, 124 participants with primary SPD (109 [87.9%] females; mean age 33.9 ± 11.7 years) from the University of Chicago and the University of Minnesota who participated in treatment and non-treatment studies between the years 2005 to 2016 were included. Study procedures were carried out in accordance with guidelines detailed in the latest version of the Declaration of Helsinki. The Institutional Review Boards at the University of Chicago, the University of Minnesota, and the Massachusetts General Hospital approved the studies and sharing of data, and all participants provided written, voluntary informed consent. Data were de-identified according to the Safe Harbor method for de-identification prior to data sharing (§164.514(b)) (15).

All participants had a current DSM-IV or DSM-5 primary diagnosis of TTM or SPD (16–17) and were between the ages of 13 and 65 years. Exclusion criteria at the University of Chicago and the University of Minnesota included current pregnancy, diagnosis of bipolar disorder or psychosis, and inability to provide written consent for participation. Exclusion criteria at the Massachusetts General Hospital included a lifetime diagnosis of psychosis, autism, or mental retardation. All participants underwent a detailed psychiatric evaluation using the Structured Clinical Interview for DSM-IV (SCID) (18). Current ADHD was assessed at the University of Chicago and the University of Minnesota by the participant reporting a previous diagnosis by a psychologist or psychiatrist and assessed at the Massachusetts General Hospital using the K-SADS ADHD module.

Assessments

All participants were asked about their hair pulling and skin picking behaviors and this included questions regarding age at onset and duration of illness. For participants seen prior to the introduction of DSM-5 in 2013, SPD was assessed using the Skin Picking Diagnostic Inventory (SPDI; Keuthen and Stewart, unpublished) or by the study doctor in a direct evaluation using SPD criteria as suggested by Arnold and colleagues (19). These criteria largely mirror those formalized in DSM-5 (17). SPD was diagnosed only if the subject displayed notable picking resulting in skin lesions. TTM was diagnosed using the Trichotillomania Diagnostic Interview-Revised (TDI-R; 20) or by the study doctor using DSM-IV or DSM-5 criteria, as appropriate based on the year in which they were assessed.

The following disorder-specific measures were used: Massachusetts General Hospital Hair Pulling Scale (MGH-HPS) (21). The MGH-HPS is a valid and reliable seven-item scale assessing hair pulling behaviors, urges, control, and distress. Each item is scored on a 5-point scale from 0= no symptoms to 4= severe symptoms. Total scores range from 0 to 28.

National Institute of Mental Health Trichotillomania Symptom Severity Scale (NIMH-TSS) (22). The NIMH-TSS is a valid and reliable scale that asks about hair pulling behaviors in the past week and the previous day, ability to resist the urge to pull hair, distress associated with hair pulling, and interference with daily life due to hair pulling.

Skin Picking Symptom Assessment Scale (SP-SAS) (9). The SP-SAS is a valid and reliable twelve question scale measuring skin picking behaviors, thoughts, urges, control, and distress in the past week. The questions are assessed using a five point scale with 0=none and 4=extreme.

Yale-Brown Obsessive Compulsive Scale Modified for Neurotic Excoriation (NE-YBOCS) (9,19). The NE-YBOCS is a valid and reliable modified version of the YBOCS for SPD assessing skin picking in the past week and has two subsections: urges and behaviors. Scores range from 0 to 40 with higher scores reflective of more severe skin picking symptoms.

General measures of symptom severity and psychosocial functioning were also used: Clinical Global Impressions- Severity Scale (CGI-S) (23). The CGI-S assesses overall symptom severity using a scale that ranges from 1 (not ill at all) to 7 (among the most severe of cases). Sheehan Disability Scale (SDS) (24). The SDS evaluates psychosocial dysfunction in three domains: work/school, social life, and home/family life.

Data Analysis

For purposes of this analysis, participants were grouped into one of the following four groups based on their mental health evaluation: (i) primary TTM, (ii) primary TTM with comorbid SPD, (iii) primary SPD, and (iv) primary SPD with comorbid TTM. In order to determine which variables were normally distributed, we used the Kolmogorov-Smirnov Tests. For variables that were not normally distributed, Mann-Whitney Tests were used to determine differences between the groups. For normally distributed variables, an Analysis of Variance (ANOVA) was used. Chi-square tests were used for non-parametric categorical variables. Significance was defined as p<.05. Effect sizes were also calculated. Effect sizes for the equality of sets of mean differences between groups are reported in terms of Cohen effect size index (“d”) or for chi-square analyses, the effect sizes are phi (Φ). A d of .2 was considered a small effect size, .5 medium, and .8 large. Phi (Φ) of .1 was considered a small effect size, .2 medium, and .3 large. SPSS Software Version 22 was used for all analyses.

Results

In participants with primary TTM, 61 out of 421 (14.5%) of cases had comorbid SPD. In participants with primary SPD, 21 out of 124 (16.9%) of cases had comorbid TTM.

The participants with primary TTM were significantly younger than those with primary TTM and comorbid SPD. There were no other significant demographic findings (Table 1).

Table 1. Primary BFRB Alone Compared to Co-Occurring BFRBs on Demographic Variables.

Primary TTM Alone Compared to Primary TTM and Comorbid SPD
Variable	TTM + SPD (n=61)	TTM (n=360)	Statistic	P-value	Cohen’s d
Sex, n (% female)	58 (95.2)	340 (94.4)	.041#	.839	-
Age, years	31.79 (11.5)	28.45 (11.6)	5376.5	.011	.281
Marital (% single) n=57, 353	30 (52.6)	249 (70.5)	8.506#	.075	-
Education, n (%), n=53, 300
High School or less	10 (18.9)	81 (27.0)	1.943#	.746	-
Some College	6 (11.3)	33 (11.0)
College Degree or more	37 (69.8)	186 (62.0)
Race (%Caucasian), n=57, 355	54 (94.7)	325 (91.6)	3.346#	.502	-
Primary SPD Alone Compared to Primary SPD and Comorbid TTM
Variable	SPD+TTM (n=21)	SPD (n=103)	Statistic	P-value	Cohen’s d
Sex, n (% female)	91 (88.4)	18 (85.7)	.114#	.736	-
Age, years	38.0 (12.2)	33.1 (11.4)	816.5	.077	-
Marital, n (% single)	10 (47.6)	55 (53.4)	.761#	.859	-
Education, n (%) n=20, 101
High School or less	0 (0)	6 (5.9)	2.566#	.767	-
Some College	7 (35.0)	30 (29.7)
College Degree or more	13 (65.0)	65 (64.4)
Race (% Caucasian), n=19, 101	18 (94.7)	85 (84.2)	4.782#	.443	-

BFRB=body focused repetitive behavior disorder

TTM=trichotillomania

SPD=skin picking disorder

Mean (SD) unless indicated otherwise; Ns indicate adjusted sample sizes based on availability of data; all statistics are Mann-Whitney U unless indicated otherwise. #Chi-square.

In terms of clinical variables, those with primary TTM and comorbid SPD versus stand-alone primary TTM had a longer duration of illness (defined as time between first TTM symptoms and assessment for this study; p=.004), greater TTM symptom severity (reflected by higher total scores on the MGH-HPS p=.004 and higher NIMH-TSS scores p=.036), and were more likely to have current comorbid depression (34.4% versus 17.5% p=.002) (Table 2).

Table 2. Primary Trichotillomania (TTM) Compared with Primary Trichotillomania with Comorbid Skin Picking Disorder (SPD) - Clinical Variables and Comorbidities.

Variable	TTM + SPD (n=61)	TTM (n=360)	Statistic	p-value	Cohen’s d
Age at onset n=60, 350	11.7 (5.4)	12.5 (5.8)	9387.5	.188	-
Duration of TTM, years, n=60, 350	20.1 (11.5)	16.0 (11.96)	8041.5	.004	.353
MGH-HPS total, n=57, 306	17.4 (4.2)	15.6 (4.8)	6659.0	.004	.393
NIMH-TSS total, n=40, 337	12.1 (4.36)	10.7 (4.0)	5376.5	.036	.359
SDS total, n=53, 270	10.8 (7.5)	9.9 (5.7)	6835.0	.606	-
Current Depression, n (%)	21 (34.4)	63 (17.5)	9.357#	.002	.305 Phi (Φ)
Current Anxiety, n (%), n=60, 358	19 (31.7)	92 (25.7)	.939#	.333	-
Current ADHD, n (%), n=60, 358	5 (8.3	27 (7.5)	.150#	.698	-
Current BDD, n (%), n=56, 311	3 (5.4)	5 (1.6)	3.129#	.077	-
Any Current Comorbidity Other than a BFRB, n (%)	38 (62.3)	185 (51.4)	2.491#	.115	-

BFRB=body focused repetitive behavior disorder

TTM=trichotillomania

SPD=skin picking disorder

Ns indicate adjusted sample sizes based on availability of data

Mean (SD) unless indicated otherwise

All statistics are Mann-Whitney U unless indicated otherwise. #Chi-square.

MGH-HPS=Massachusetts General Hospital Hair Pulling Scale

NIMH-TSS=National Institute of Mental Health- Trichotillomania Symptom Severity Scale

SDS=Sheehan Disability Scale

BDD=Body Dysmorphic Disorder

Those participants with primary SPD and comorbid TTM, versus stand-alone primary SPD, reported significantly later age of SPD onset (p=.027) and greater skin picking symptom severity (reflected by the total SP-SAS score p=.047 and the CGI-S p=.026) (Table 3).

Table 3. Primary Skin Picking Disorder (SPD) Compared to Primary Skin Picking Disorder with Comorbid Trichotillomania (TTM) - Clinical Variables and Comorbidities.

Variable	SPD+TTM (n=21)	SPD (n=103)	Statistic	p-value	Cohen’s d
Age at onset, years	18.6 (14.8)	11.8 (7.0)	743.0	.027	.585
Duration of TTM, years	19.43 (13.6)	21.29 (12.2)	986.5	.570	-
SP-SAS Total, n=21,102	31.38 (6.8)	28.20 (6.6)	4.03+	.047	.477
NE-YBOCS Total	20.19 (5.9)	18.40 (4.9)	899.0	.223	-
CGI-S	4.62 (0.7)	4.22 (0.9)	789.5	.026	.510
SDS Total	11.81 (6.7)	11.32 (6.7)	.094	.760	-
Current Depression, n (%)	8 (38.1)	37 (35.9)	.036#	.850	-
Current Anxiety, n (%)	5 (23.8)	18 (17.5)	.463#	.496	-
Current ADHD, n (%)	3 (14.3)	6 (5.8)	1.855#	.173	-
Current BDD, n (%), n=54, 9	4 (44.4)	12 (22.2)	2.011#	.156	-
Any Current Comorbidity Other than a BFRB, n (%)	13 (61.9)	55 (53.4)	.510#	.475	-

BFRB=body focused repetitive behavior disorder

TTM=trichotillomania

SPD=skin picking disorder

Ns indicate adjusted sample sizes based on availability of data

Mean (SD) unless indicated otherwise

All statistics are Mann-Whitney U unless indicated otherwise. #Chi-square. ⁺ANOVA

SP-SAS=Skin Picking-Symptom Assessment Scale

NE-YBOCS=Yale-Brown Obsessive Compulsive Scale Modified for Neurotic Excoriation

CGI-S=Clinical Global Impressions - Severity of Illness Scale

SDS=Sheehan Disability Scale

Discussion

This study examined the clinical significance of having co-occurring BFRBs in the largest sample of clinically-diagnosed adults with these disorders available to date. As predicted, BFRB symptom severity was greater in those with co-occurring TTM and SPD compared to those with either TTM or SPD alone. This finding is consistent with similar analyses in related disorders of OCD and BDD (12–13), supporting the importance of careful screening for comorbidity in clinical practice.

The comorbid BFRBs groups evidenced greater symptom severity than the other two groups (small effect size), and comorbidity of BFRBs may suggest greater dysfunction of underlying neurocircuitry. Neuroimaging research in both TTM and SPD suggest abnormally reduced integrity of white-matter tracts and structural abnormalities in neural regions involved in motor generation and suppression (such as the anterior cingulate and frontal cortices), while functional imaging has illustrated executive planning and motor inhibitory deficits, compared to healthy controls (6–7, 25–28). It is possible therefore that the combination of two similar disorders may result in additive abnormalities, or even disproportionately greater than expected structural abnormalities in regions involved in habit generation, action monitoring and top–down inhibitory control processes. Future neuroimaging research should consider comparing individuals with a single BFRB to those with co-occurring BFRBs to fully understand the biological substrates of this comorbidity. It would also be valuable to compare cognitive functioning between such groups in future work.

From a clinical perspective, the finding of greater BFRB symptom severity in the comorbid groups also may necessitate changes to current treatment approaches. Given that cognitive behavioral strategies, specifically variants of habit reversal therapy, have shown benefit for both TTM and SPD separately (29–30), the question remains as to whether interventions need to be more frequent, longer in duration, or different, when approaching individuals with comorbid TTM and SPD. Also relevant to clinical practice is that people with primary TTM, who had comorbid SPD, showed significantly higher rates of major depressive disorder, than people with TTM alone (medium effect size). While this cross-sectional study cannot assess causality, it appears likely that having these two BRFBs predisposes towards high risk of later developing depression, rather than vice versa.

There are several limitations to the current study. First, the comorbid BFRB was not examined with its own severity measure. It could be potentially informative for future research to better understand how the level of severity of the comorbid BFRB impacts the primary BFRB. Second, the samples, while large, lacked ethnic and gender diversity. Third, it should be noted that the primary TTM with co-occurring SPD group, who were more likely to have depression, were also older and so this finding needs to be interpreted with this limitation. How these findings might generalize to other populations remains to be explored. Lastly, it would be valuable in future work to examine cognitive and neurobiological correlates of comorbid BFRBs as opposed to stand-alone disorders, as the current study did not collect such data.

Conclusions

This study suggests that the co-occurrence of TTM and SPD results in a worse problem with the primary repetitive behavior. Whether and to what extent this co-occurrence necessitates novel treatment interventions await future research.