Correction in the title (Jan 2016) https://doi.org/10.1371/journal.pone.0147095

Background: Little is known about the association between psychotic-like experiences (PLEs) and nonsuicidal self-injury (NSSI) in the general adult population. Thus, the aim of this study was to examine the association using nationally-representative data from England. Methods: Data from the 2007 Adult Psychiatric Morbidity Survey was analyzed. The sample consisted of 7403 adults aged ≥16 years. Five forms of PLEs (mania/hypomania, thought control, paranoia, strange experience, auditory hallucination) were assessed with the Psychosis Screening Questionnaire. The association between PLEs and NSSI was assessed by multivariable logistic regression. Hierarchical models were constructed to evaluate the influence of alcohol and drug dependence, common mental disorders, and borderline personality disorder symptoms on this association. Results: The prevalence of NSSI was 4.7% (female 5.2% and male 4.2%), while the figures among those with and without any PLEs were 19.2% and 3.9% respectively. In a regression model adjusted for sociodemographic factors and stressful life events, most types of PLE were significantly associated with NSSI: paranoia (OR 3.57; 95%CI 1.96-6.52), thought control (OR 2.45; 95%CI 1.05-5.74), strange experience (OR 3.13; 95%CI 1.99-4.93), auditory hallucination (OR 4.03; 95%CI 1.56-10.42), and any PLE (OR 2.78; 95%CI 1.88-4.11). The inclusion of borderline personality disorder symptoms in the models had a strong influence on the association between PLEs and NSSI as evidenced by a large attenuation in the ORs for PLEs, with only paranoia continuing to be significantly associated with NSSI. Substance dependence and common mental disorders had little influence on the association between PLEs and NSSI. Conclusions: Borderline personality disorder symptoms may be an important factor in the link between PLEs and NSSI. Future studies on PLEs and NSSI should take these symptoms into account.