The aim of this study was to assess interrater reliability and provide initial data bearing on the validity of a method of assessing personality disorders (PDs) that does not presume that patients can accurately self-report personality pathology. In a sample of 24 outpatients, two clinician-judges independently applied the Shedler-Westen Assessment Procedure-200 (SWAP-200; Westen & Shedler, 1999a, 2000), a 200-item Q-sort procedure for assessing personality pathology, to data from the Clinical Diagnostic Interview (Westen, 2002), a systematic clinical interview that mirrors and standardizes methods used by experienced clinicians to diagnose personality. In 16 of the 24 cases, the treating clinician also independently described the patient using the SWAP-200 Q-sort, based on longitudinal knowledge of the patient over the course of treatment, blind to the interview data. Interrater reliability was uniformly high, with median correlations between interviewers at r > .80. Interviewer-treating clinician correlations were also high, with median convergent validity coefficients at r > .80. Diagnostic overlap (discriminant validity) was moderate for dimensional DSM-IV diagnoses, reflecting extensive comorbidity among disorders, but minimal for empirically derived diagnoses identified in prior research. Treating clinicians' dimensional PD diagnoses using this method also strongly predicted interviewer-rated measures of adaptive functioning. The findings provide preliminary support for the reliability and validity of an alternative to structured interviews for diagnosing personality pathology, and suggest that the way to improve validity of personality diagnosis may not be to minimize clinical inference but to quantify it using psychometric instruments.

This article describes the development of, and preliminary findings with, the Affect Regulation and Experience Q-Sort (the AREQ), an observer-based assessment of affect regulation and experience. In Study 1, 31 clinicians provided Q-sort descriptions of 90 patients. Factor scores correlated in predicted ways with criteria such as suicide attempts and hospitalizations, as well as with clinicians' ratings of functioning in a variety of domains. Correlations between prototype Q-sorts and actual Q-sort profiles for patients sharing a diagnosis (dysthymia, borderline personality disorder, and narcissistic personality disorder) also provided evidence for convergent and discriminant validity. The data also suggested the importance of distinguishing 2 kinds of negative affect that have very different correlates. Study 2 showed that the AREQ can be applied reliably using an interview that avoids many of the problems of self-report.

BackgroundIn 2016, 44,965 people in the United States died by suicide. It is common to see people with suicidal ideation seek help or leave suicide notes on social media before attempting suicide. Many prefer to express their feelings with longer passages on forums such as Reddit and blogs. Because these expressive posts follow regular language patterns, potential suicide attempts can be prevented by detecting suicidal posts as they are written.ObjectiveThis study aims to build a classifier that differentiates suicidal and nonsuicidal forum posts via text mining methods applied on post titles and bodies.MethodsA total of 508,398 Reddit posts longer than 100 characters and posted between 2008 and 2016 on SuicideWatch, Depression, Anxiety, and ShowerThoughts subreddits were downloaded from the publicly available Reddit dataset. Of these, 10,785 posts were randomly selected and 785 were manually annotated as suicidal or nonsuicidal. Features were extracted using term frequency-inverse document frequency, linguistic inquiry and word count, and sentiment analysis on post titles and bodies. Logistic regression, random forest, and support vector machine (SVM) classification algorithms were applied on resulting corpus and prediction performance is evaluated.ResultsThe logistic regression and SVM classifiers correctly identified suicidality of posts with 80% to 92% accuracy and F1 score, respectively, depending on different data compositions closely followed by random forest, compared to baseline ZeroR algorithm achieving 50% accuracy and 66% F1 score.ConclusionsThis study demonstrated that it is possible to detect people with suicidal ideation on online forums with high accuracy. The logistic regression classifier in this study can potentially be embedded on blogs and forums to make the decision to offer real-time online counseling in case a suicidal post is being written.

Clinicians and independent interviewers can reliably assess complex personality traits associated with personality pathology using the SWAP-200.

Clinicians and independent interviewers can reliably assess complex personality traits associated with personality pathology using the SWAP-200.