Loneliness as a predictor of suicidal ideation and behaviour: a systematic review and meta-analysis of prospective studies

Background: Suicide and suicidal behaviour are global health concerns with complex aetiologies. Given the recent research and policy focus on loneliness, this systematic review aimed to determine the extent to which loneliness predicts suicidal ideation and/or behaviour (SIB) over time. Methods: A keyword search of five major databases (CINHAL, Medline, PsychArticles, PsychInfo and Web of Knowledge) was conducted. Papers for inclusion were limited to those using a prospective longitudinal design, written in English and which measured loneliness at baseline and SIB at a later time-point. Results: After duplicates were removed, 947 original potential papers were identified, with 22 studies meeting the review criteria. Meta-analysis revealed loneliness was a significant predictor of both suicidal ideation and behaviour and there was evidence that depression acted as a mediator. Furthermore, studies which consisted of predominantly female participants were more likely to report a significant relationship, as were studies where participants were aged 16-20 or >55 years at baseline. Limitations: There was considerable variability in measures, samples and methodologies used across the studies. Middle-aged adults were under-represented, as were individuals from minority ethnic backgrounds. All studies were conducted in countries where self-reliance and independence (i.e. individualism) are the cultural norm. Conclusions: Loneliness predicts later SIB in select populations. However, due to the heterogeneity of the studies further research is needed to draw more robust conclusions. Suicide death also needs to be included as an outcome measure. A focus on more collectivist countries is also required.