Abstract Purpose Loneliness is associated with poor health including premature mortality. There are cross-sectional associations with depression, anxiety, psychosis, and other mental health outcomes. However, it is not known whether loneliness is causally linked with the new onset of mental health problems in the general population. Longitudinal studies are key to understanding this relationship. We synthesized evidence from longitudinal studies investigating the relationship between loneliness and new onset of mental health problems, in the general population. Method We systematically searched six electronic databases, unpublished sources, and hand-searched references, up to August 2021. We conducted a meta-analysis of eight independent cohorts and narrative synthesis of the remaining studies. Results We included 32 studies, of which the majority focused on depression. Our narrative synthesis found most studies show loneliness at baseline which is associated with the subsequent new onset of depression. The few studies on anxiety and self-harm also showed a positive association. Our meta-analysis found a pooled adjusted odds ratio of 2.33 (95% CI 1.62–3.34) for risk of new onset depression in adults who were often lonely compared with people who were not often lonely. This should be interpreted with caution given evidence of heterogeneity. Conclusion Loneliness is a public mental health issue. There is growing evidence it is associated with the onset of depression and other common mental health problems. Future studies should explore its impact across the age range and in more diverse populations, look beyond depression, and explore the mechanisms involved with a view to better informing appropriate interventions.