Objectives: To investigate the association of urinary incontinence (UI) and double incontinence (DI, concurrent UI and fecal incontinence) with one-year mortality among older female hip fracture patients and to identify predictors of incident UI and DI.

Design: A prospective cohort study SETTING AND SUBJECTS: 1,468 female patients aged ≥ 65 treated for their first hip fracture during the period 2007-2019 METHODS: Continence status was elicited at baseline and one-year post-fracture. Age- and multivariable-adjusted Cox proportional hazards and multinomial logistic regression models were used to determine the associations of incontinence with one-year mortality and to examine the associations of baseline predictors with incident UI and DI respectively.

Results: Of the women with no incontinence, UI and DI, 78 (13%), 159 (23%) and 60 (34%), died during follow-up. UI (HR 1.72, 95% CI 1.31-2.26) and DI (HR 2.61, 95% CI 1.86-3.66) were associated with mortality after adjusting for age. These associations lost their predictive power in multivariable analysis while age over 90, living in an institution, impaired mobility, poor nutrition, polypharmacy, and late removal of urinary catheter remained associated with mortality. Of continent women, 128 (21%) developed UI and 23 (4%) DI during follow-up. In multivariable analysis, impaired mobility was associated with incident UI (OR 2.56, 95% CI 1.48-4.44) and DI (OR 4.82, 95% CI 1.70-13.7), as well as living in an institution (OR 3.44, 95% CI 1.56-7.61 and OR 3.90, 95% CI 1.17-13.0).

Conclusions and implications: Underlying vulnerability likely explains differences in mortality between continence groups and development of incident UI and DI.