Psychology and Neuroscience Faculty Database
Psychology and Neuroscience
Arts & Sciences
Duke University

 HOME > Arts & Sciences > pn > Faculty    Search Help Login pdf version printable version 

Publications [#302616] of Scott N. Compton

search PubMed.

Journal Articles

  1. Houghton, DC; Capriotti, MR; De Nadai, AS; Compton, SN; Twohig, MP; Neal-Barnett, AM; Saunders, SM; Franklin, ME; Woods, DW (2015). Defining treatment response in trichotillomania: a signal detection analysis.. Journal of Anxiety Disorders, 36, 44-51. [doi]
    (last updated on 2019/02/16)

    Abstract:
    The Massachusetts General Hospital Hairpulling Scale (MGH-HPS) and the NIMH Trichotillomania Severity Scale (NIMH-TSS) are two widely used measures of trichotillomania severity. Despite their popular use, currently no empirically-supported guidelines exist to determine the degrees of change on these scales that best indicate treatment response. Determination of such criteria could aid in clinical decision-making by defining clinically significant treatment response/recovery and producing accurate power analyses for use in clinical trials research. Adults with trichotillomania (N=69) participated in a randomized controlled trial of psychotherapy and were assessed before and after treatment. Response status was measured via the Clinical Global Impressions-Improvement Scale, and remission status was measured via the Clinical Global Impressions-Severity Scale. For treatment response, a 45% reduction or 7-point raw score change on the MGH-HPS was the best indicator of clinically significant treatment response, and on the NIMH-TSS, a 30-40% reduction or 6-point raw score difference was most effective cutoff. For disorder remission, a 55-60% reduction or 7-point raw score change on the MGH-HPS was the best predictor, and on the NIMH-TSS, a 65% reduction or 6-point raw score change was the best indicator of disorder remission. Implications of these findings are discussed.


Duke University * Arts & Sciences * Faculty * Staff * Grad * Postdocs * Reload * Login