Search for a command to run...
<b>Objective </b>To examine the bidirectional relationships between depression, anxiety, neuroticism, and urinary incontinence in women. <br/><br/><b>Design </b>A prospective time-to-event and two-sample Mendelian randomisation (MR) study.<br/><br/><b>Setting </b>Individual participant data from the UK Biobank and summary genome-wide association (GWAS) study data from international consortia. <br/><br/><b>Participants </b>Up to 118 526 UK Biobank women with linked health records and up to 1.6 million participants with GWAS summary data. <br/><br/><b>Main outcome measures</b> Urinary incontinence (UI) and its subtypes (stress, urge, mixed), urinary urgency (irrespective of leakage), depression, anxiety, and neuroticism. <br/><br/><b>Results </b>We triangulated evidence to demonstrate bidirectional relationships between depression/anxiety and UI. In prospective analyses adjusted for confounders, depression was associated with a higher rate of new onset UI (any UI: Hazard Ratio (HR) 1.67; 95% Confidence Intervals (CI) 1.55 to 1.81) and its subtypes, with the strongest associations observed for mixed UI (HR 1.91; 95%CI 1.59 to 2.31). Similarly, anxiety and higher neuroticism scores were prospectively associated with UI and its subtypes. In the reverse direction, all UI subtypes were associated with a higher rate of new onset depression (e.g. any UI: HR 1.40; 95%CI 1.27 to 1.54) and anxiety (e.g. any UI: HR 1.28; 95%CI 1.17 to 1.39). Two-sample MR provided evidence for a causal effect of genetic liability to depression and neuroticism on UI and its subtypes (e.g. depression on any UI: ORivw; 1.25 95%CI 1.16 to 1.35). Evidence for a causal effect in the reverse direction was weaker, with modest effects of genetic liability to any UI on depression. Little evidence was found for causal effects of anxiety with UI subtypes in either direction. Results were largely robust to sensitivity analyses. <br/><br/><b>Conclusion </b>We find evidence of bidirectional relationships between depression/anxiety and UI. Evidence that depression, anxiety and neuroticism are predictors of UI onset has implications for treatment. Research is needed to examine if treatments for depression/anxiety could be effective in alleviating UI.