The impact of extended longitudinal observation on the assessment of personality disorders

Abstract

Multiple sources of information are necessary for a valid assessment of personality disorders (PDs). This study investigates the impact of longitudinal observation. The sample comprised 1217 patients from 15 different treatment units. PDs were assessed at admission to treatment using the Structured Clinical Interview for DSM-IV Axis II Personality Disorders (SCID-II) and additional clinical information (best estimate diagnosis). After approximately 18 weeks of treatment, the SCID-II protocols were re-examined at clinical conferences and the diagnostic status reassessed on the basis of longitudinal observations in multiple group situations (longitudinal, expert, all data principle). Using this procedure, 78% of the patients' diagnostic criteria sets were changed, and 32% of patients' diagnostic statuses were changed. Many (32%) patients who were evaluated initially as not having a PD received a PD diagnosis after re-examination. The information provided by customary clinical assessment has important limitations, and longitudinal observation provides additional information that may change the diagnostic status in approximately one-third of PD cases.

The Alcohol Use Disorder and Associated Disabilities Interview Schedule-5 (AUDADIS-5): Reliability of substance use and psychiatric disorder modules in a general population sample

“…This result is consistent with clinical literature recommending that diagnostic assessment of PDs take a longitudinal perspective and utilize multiple sources of data: interviews with target individuals and collateral informants, and extant records (Clark, 2007; Pedersen et al, 2013; Zimmerman, 1994). While unavoidable in a large national epidemiologic survey, PD diagnoses requiring impairment or distress that relied solely on respondent self-report may have suffered from distortions or instabilities in respondents’ self-appraisals, including lack of insight into the effects of symptomatic behaviors on role functioning (Clark, 2007; Pedersen et al, 2013; Zimmerman, 1994). In the case of conduct disorder, respondents also needed to recall and report impairment or distress from symptoms that may have occurred many years before interview (Rueter et al, 2000).…”

Sociodemographic and psychiatric diagnostic predictors of 3-year incidence of DSM–IV substance use disorders among men and women in the National Epidemiologic Survey on Alcohol and Related Conditions.

“…Collateral data sources are particularly important for PD assessment to mitigate potential distortions in respondents’ self-appraisals, including lack of insight into the effects of symptomatic behaviors on role functioning (Clark, 2007; Pedersen, Karterud, Hummelen, & Wilberg, 2013; Zimmerman, 1994). Additionally, the 3-year follow-up yielded relatively few incident cases of AUDs and DUDs, particularly among women.…”

Poor Validity of theDSM-IVSchizoid Personality Disorder Construct as a Diagnostic Category

“…Staff members were trained in PD diagnostics through attendance at local courses and Network conferences. DSM-IV diagnoses were established by way of the "Longitudinal, Expert, All Data" standard (LEAD standard; Pedersen et al, 2013). Tentative diagnoses were established at admission on the basis of referral letters, self-reported history and complaints, and two clinical structured diagnostic interviews: (a) the Mini-International Neuropsychiatric Interview for Axis I diagnoses (MINI; Sheehan et al, 1994), and (b) the Structured Clinical Interview for DSM-IV Personality Disorders (SCID-II; First, 1994).…”

“…A final PD diagnosis required that the criteria from the original SCID-II protocol were confirmed by clinical observations. It is assumed that the LEAD procedure resulted in more valid diagnoses (Pedersen et al, 2013).…”

“…The large range in prevalence estimates is indicative of reliability problems. Zanarini et al (2000) found that SZPD had the poorest reliability estimates of all PDs, and a recent clinical study found evidence for substantial false-positive rates of SZPD if only assessed at one point in time (Pedersen, Karterud, Hummelen, & Wilberg, 2013). Moreover, the low prevalence of SZPD in clinical samples suggests poor clinical utility as a psychiatric diagnosis.…”