Predicting urinary incontinence in women in later life: A systematic review

Eighteen prospective longitudinal studies fulfilled the search criteria. These were analysed systematically (as per the PRISMA checklist) and bias risk through study design was minimised where possible upon data analysis. One paper proposed a predictive assessment tool called the 'continence index'. It was derived following secondary analysis of a cohort study and its predictive threshold had suboptimal sensitivity (79%) and specificity (65%) rates. Seventeen studies identified multiple strong risk factors for UI but despite a large selection of papers on the topic, no robust risk assessment tool prospectively identified patients at risk of UI in later life. Thus more research in this field is required. Clinicians should be aware particularly of modifiable UI risk factors to help reduce the clinical burden of UI in the long term.