Limited longitudinal population-based research exists on the bidirectional association between loneliness and common mental disorders (CMDs). Using 3-year follow-up data, this study examined whether loneliness among adults increases the risk for onset and persistence of mild-moderate or severe CMD; and whether mild-moderate or severe CMD is a risk factor for onset and persistence of loneliness. Data were used from the second ('baseline') and third (3-year follow-up) waves of the Netherlands Mental Health Survey and Incidence Study-2, a prospective study of a representative cohort of adults aged 18-64 years. Twelve-month CMDs and their severity were assessed with the Composite International Diagnostic Interview 3.0, and current loneliness using the De Jong Gierveld Loneliness Scale. Multivariate analyses were controlled for several potential confounders. Loneliness predicted onset of severe CMD at follow-up in adults without CMDs at baseline, and increased risk for persistent severe CMD at follow-up in those with CMD at baseline. Conversely, severe CMD predicted onset of loneliness at follow-up in non-lonely adults at baseline, but was not associated with persistent loneliness at follow-up in lonely adults at baseline. Observed associations remained significant after controlling for perceived social support at baseline, except for the relationship between loneliness and persistent severe CMD. No longitudinal relationships were observed between loneliness and mild-moderate CMD. Attention should be paid to loneliness, both in adults with and without CMD. Further research is needed to better understand the mechanisms underlying the observed associations between loneliness and CMDs to develop successful interventions.