Association between psychotic experiences and non-accidental self-injury: results from a nationally representative survey of adolescents.

Abstract

The association between psychotic experiences (PEs) and non-accidental self-injury (NASI; including self-harm and suicide attempts) is well established, although variables influencing this relationship have not been comprehensively examined. This study aimed to investigate (1) the cross-sectional PE-NASI association before and after adjustment for confounders, and (2) the individual contribution of each confounding and potentially mediating variable to the association. A random sample of Australian adolescents aged 14-17years (n = 1998) completed self-report questions regarding any self-harm, suicidality or PEs experienced in the past 12months in 2013-2014 as part of the Young Minds Matter Survey, a national household survey. We conducted logistic regression analyses to investigate the association between NASI and PEs, after controlling for confounders (sociodemographics, substance use, and parental mental illness) as well as the influence of potential mediators (major depression, bullying, psychological distress, sleep, self-esteem, disordered eating behaviour, social isolation, and intervention factors). Except for special messages, all PE subtypes (auditory and visual hallucinatory experiences [HEs], and two of the three delusional experiences [DEs]) were associated with NASI after adjustment for confounders (OR range: 2.60-5.21). Depression and psychological distress significantly influenced all PE-NASI associations, where depression appeared to fully explain the DE-NASI association, and partially attenuate the HE-NASI association. Variables such as parental mental illness, disordered eating behaviour, and social isolation had negligible effects in nearly all self-harm and attempted suicide models. Adolescents reporting any PE in the past 12 months reported increased likelihood of NASI in the same time period and, auditory HEs in particular, were strongly and independently associated with self-harm and suicide attempts. These results highlight the importance of PEs as indicators of risk of self-injurious behaviour among Australian youth.