Elsevier

Psychiatry Research

Volume 239, 30 May 2016, Pages 196-203
Psychiatry Research

NEO-FFI personality clusters in trichotillomania

https://doi.org/10.1016/j.psychres.2016.03.011Get rights and content

Highlights

  • We examined whether hair pullers could be grouped based on personality traits.

  • Two personality clusters were identified.

  • These clusters differed on HP style, affective States, and quality of life.

  • The two prototypes may represent groups of more and less focused pullers.

Abstract

The purpose of this study was to determine whether personality prototypes exist among hair pullers and if these groups differ in hair pulling (HP) characteristics, clinical correlates, and quality of life. 164 adult hair pullers completed the NEO-Five Factor Inventory (NEO-FFI; Costa and McCrae, 1992) and self-report measures of HP severity, HP style, affective state, and quality of life. A latent class cluster analysis using NEO-FFI scores was performed to separate participants into clusters. Bonferroni-corrected t-tests were used to compare clusters on HP, affective, and quality of life variables. Multiple regression was used to determine which variables significantly predicted quality of life. Two distinct personality prototypes were identified. Cluster 1 (n=96) had higher neuroticism and lower extraversion, agreeableness, and conscientiousness when compared to cluster 2 (n=68). No significant differences in demographics were reported for the two personality clusters. The clusters differed on extent of focused HP, severity of depression, anxiety, and stress, as well as quality of life. Those in cluster 1 endorsed greater depression, anxiety, and stress, and worse quality of life. Additionally, only depression and cluster membership (based on NEO scores) significantly predicted quality of life.

Introduction

Limited success has been reported to date for existing treatment interventions for trichotillomania (TTM). One might argue that current treatment approaches fail to adequately tailor treatment strategies to varying symptom presentations in different pullers. Furthermore, there has been little investigation of personality variables in TTM. The examination of personality prototypes in hair pullers may elucidate phenotypes that advance our knowledge of etiology and promote the development of more effective treatments for differing presentations of TTM.

Our group has previously explored personality traits in TTM including their predictive validity for TTM diagnosis, hair pulling (HP) severity and control, and HP style (Keuthen et al., 2015). Scores on the NEO-Five Factory Inventory (FFI; Costa and McCrae, 1992) domain of neuroticism predicted TTM case vs. control status with every 1-point increase in neuroticism scores resulting in a 10% greater chance of TTM diagnosis. Higher neuroticism was associated with both greater pulling severity and less pulling control. Higher openness and lower agreeableness were also associated with greater pulling severity. Higher neuroticism and lower openness were associated with greater focused pulling while none of the NEO-FFI domains predicted automatic pulling.

The current study extends our previous research by examining personality prototypes or profiles in hair pullers, rather than isolated personality dimensions. Our prior trait analyses failed to shed light on the dynamic organization of multiple personality traits within the individual. Identification of personality patterns provides a multivariate picture of personality and illuminates how traits co-occur within the individual. It is arguably the case that identification of personality prototypes in hair pulling may be more useful in identifying genetic and environmental etiologies and in predicting disorder presentation, course and treatment outcome, than trait dimensions alone.

Asendorpf et al. (2001) previously identified a three-prototype model of personality consisting of undercontrollers (characterized by high scores in neuroticism and low scores in agreeableness and conscientiousness), overcontrollers (characterized by high scores in neuroticism and low scores in extraversion and openness to experience) and a resilient type (characterized by below average scores in neuroticism and above average scores in all other NEO-FFI dimensions). Thus, the undercontrollers have limited control over impulses and display externalizing tendencies. The overcontrollers display excessive modulation of impulses and internalizing or avoidance tendencies. Finally, those with the resilient prototype exhibit social and cognitive flexibility that allows them to optimally adapt to life challenges. These prototypes have been replicable in different populations and at different development stages. There is an extensive literature that has explored these prototypes in psychiatric (e.g., Mueller et al., 2010, Sanchez-Guarnido et al., 2015) and medical (e.g., Berry et al., 2007) populations. Mueller et al. (2010) performed a cluster analysis of NEO-FFI scales in compulsive buying, an impulse control disorder purportedly on a spectrum with TTM. Two personality clusters were identified. One cluster was similar to the undercontrolled prototype while a second cluster shared features of both the resilient and overcontrolled cluster.

It is also of note that earlier research (Flessner et al., 2008) reported two common styles of pulling. The automatic style is characterized by unaware pulling, often while engaged in sedentary activities. The focused style is associated with uncomfortable affect or sensations and pulling occurs with greater awareness. This latter style of pulling shares some features with the undercontrolled personality prototype while the automatic style of pulling, in contrast, does not appear similar to any of the personality prototypes. Given that Mueller and colleagues identified two personality clusters in another obsessive-compulsive spectrum disorder, we hypothesized that there would similarly be two personality clusters in TTM.

Section snippets

Participants

Participants were 164 adult hair pullers between the ages of 18 and 65 who had participated in two consecutive TTM studies at Massachusetts General Hospital between 2006 and 2012. All subjects either met criteria for DSM-IV-TR TTM or chronic HP (all TTM criteria satisfied except tension before pulling or when attempting to resist, formerly criterion B, and/or pleasure, relief, or gratification when pulling, formerly criterion C). Additionally, participants did not have lifetime diagnoses of

Results

Latent class cluster analysis was first performed using MCLUST to identify existing personality prototypes within the participants. The best solution for each number of clusters (1 through 9) regardless of shape and volume were identified. The BIC values as well as the log-likelihood and AIC values that were manually calculated using the appropriate formulas can be seen in Table 1. The BIC values for all solutions (including all shapes, volumes, and number of clusters) are graphed in Fig. 1. In

Discussion

Our investigation into the personality traits of individuals suffering from TTM has identified two prototypes. The first prototype contains individuals who have greater neuroticism but less extraversion, agreeableness, and conscientiousness than those in the second prototype. While these conceptions of the nature and personality of two clusters of hair pullers are compelling, they require replication in future studies. It is also unclear if these results in our largely Caucasian study would

Disclosures

Nancy Keuthen and Christopher Flessner are members of the scientific advisory board of the Trichotillomania Learning Center. Nancy Keuthen is also a member of the scientific advisory board of the International Obsessive Compulsive Disorder Foundation and receives royalties from New Harbinger, Inc. She also has equity interest in Johnson & Johnson, Merck & Co., Inc., Pfizer, Inc., and Procter & Gamble Co. The other authors report no conflicts of interest.

Acknowledgments

We gratefully recognize financial support for this project from the Trichotillomania Learning Center, Inc. (TLC) and its BFRB Precision Medicine Initiative, as well as the Greater Kansas City Foundation. We also wish to express gratitude to Saraj Mothi for his statistical consultation and Amanda Falcon, Miriam Frank, and Zachary Madison who performed data input and cleaning.

References (25)

  • R. De Bin et al.

    Subsampling versus bootstrapping in resampling-based model selection for mulivariable regression

    Biometrics

    (2015)
  • M.B. First et al.

    Structured Clinical Interview for DSM-IV Axis I Disorders, Research Version, Non-Patient Edition. (SCID-I/NP). Biometrics Research

    (2002)
  • Cited by (7)

    • The relationship between emotion regulation strategies, personality traits and skin picking behaviours in a non-clinical sample of Polish adults

      2018, Psychiatry Research
      Citation Excerpt :

      Regarding psychological factors, personality structure has also been consistently shown to be an important predictor of stress reactivity, resilience and well-being (Amstadter et al., 2016; Hengartner, 2015; Roberts et al., 2007). Moreover, dysfunctional personality patterns have consistently been confirmed to constitute an early risk factor of various psychiatric conditions including obsessive-compulsive disorder (Rector et al., 2002; Rees et al., 2006, 2014; Samuels et al., 2000; Wu et al., 2006) and trichotillomania (Chamberlain and Odlaug, 2014; Hagh-Shenas et al., 2015; Keuthen et al., 2015, 2016) which share notable similarities to SPD. Nevertheless, the link between personality traits and SPD is still understudied.

    • Clinical characteristics of trichotillomania

      2023, Annals of Clinical Psychiatry
    • Psychological factors and genetic characteristics of rural cannabis users

      2020, Annals of Agricultural and Environmental Medicine
    View all citing articles on Scopus
    1

    These authors are co-first authors as they contributed equally to this work.

    View full text