Chapter 3 88 3. Results 3.1 Selection and description of included studies Figure 1 presents the study selection process. The systematic search yielded 1,012 hits (PsycInfo 227, Web of Science 481, and Scopus 304). After removal of duplicates, 659 articles remained. Screening of title and abstract resulted in the exclusion of 490 articles. The remaining 169 articles were assessed on the basis of the full texts. In total, 132 of the 169 articles were excluded (see Figure 1 for the reasons). Therefore, 37 articles were included, which in total reported 39 studies. Table 1 summarizes the 39 selected studies and their most important characteristics: year of publication, study design, sample size, measurement of emotional awareness or alexithymia, measurement of borderline personality pathology, the use of control variables, and positive or negative results. Publication year varied between 1994 and 2014, with a median of 2012, showing that most studies were conducted recently. Seventeen studies had a comparative design: 14 compared patients with BPD to a (matched) healthy control group; two compared patients with BPD to patients with another psychiatric condition; one compared BPD patients to both healthy and psychiatric controls. Twenty-two studies had a correlational design: 14 focused on the relationship between emotional awareness or alexithymia and borderline personality traits in healthy groups (10 of them on students), eight in psychiatric patients (i.e., mixed or unspecified out- or inpatient groups). The 39 studies included in this meta-analysis had 8,321 subjects. Study size ranged from 30 to 1,418 participants, with a median of 101. With regard to emotional awareness, 18 studies measured emotional awareness, with subscales of instruments addressing emotional regulation (10 of them used the DERS) and 21 studies used measures of alexithymia. All instruments used in the selected studies had adequate to good psychometric properties. With regard to measurement of BPP, 17 studies assessed BPD and 22 examined borderline personality traits. The Structured Clinical Interview (SCID), the Structured Interview for DSM Personality (SIDP), and the Personality Assessment Inventory-Borderline Features Scale (PAI-BOR) were the most often used instruments to assess BPP. Twenty-five studies did not control for possible confounding variables, whereas the remaining 14 did, mostly for gender, age, and symptoms of comorbid psychopathology. Of those studies, 7 (from five articles) had included a measure of NA as a covariate in their analyses. 3.2 Average association and heterogeneity Figure 2 presents the results of the overall meta-analysis. Seven studies did not find significant relations (p > .05), 31 studies found that BPP was positively related to lack of emotional awareness, and one study found an inverse relationship. Across the 39 studies, a significant positive association between borderline pathology and lack of
RkJQdWJsaXNoZXIy MjY0ODMw