Background Urinary incontinence affects one in three women worldwide. Pelvic floor muscle training is an effective treatment. Electromyography biofeedback (providing visual or auditory feedback of internal muscle movement) is an adjunct that may improve outcomes.
Objectives To determine the clinical effectiveness and cost-effectiveness of biofeedback-mediated intensive pelvic floor muscle training (biofeedback pelvic floor muscle training) compared with basic pelvic floor muscle training for treating female stress urinary incontinence or mixed urinary incontinence.
Design A multicentre, parallel-group randomised controlled trial of the clinical effectiveness and costeffectiveness of biofeedback pelvic floor muscle training compared with basic pelvic floor muscle training, with a mixed-methods process evaluation and a longitudinal qualitative case study. Group allocation was by web-based application, with minimisation by urinary incontinence type, centre, age and baseline urinary incontinence severity. Participants, therapy providers and researchers were not blinded to group allocation. Six-month pelvic floor muscle assessments were conducted by a blinded assessor.
Setting: This trial was set in UK community and outpatient care settings.
Participants Women aged ≥ 18 years, with new stress urinary incontinence or mixed urinary incontinence. The following women were excluded: Those with urgency urinary incontinence alone, those who had received formal instruction in pelvic floor muscle training in the previous year, those unable to contract their pelvic floor muscles, those pregnant or < 6 months postnatal, those with prolapse greater than stage II, those currently having treatment for pelvic cancer, those with cognitive impairment affecting capacity to give informed consent, those with neurological disease, those with a known nickel allergy or sensitivity and those currently participating in other research relating to their urinary incontinence.
Interventions Both groups were offered six appointments over 16 weeks to receive biofeedback pelvic floor muscle training or basic pelvic floor muscle training. Home biofeedback units were provided to the biofeedback pelvic floor muscle training group. Behaviour change techniques were built in to both interventions. Main outcome measures The primary outcome was urinary incontinence severity at 24 months (measured using the International Consultation on Incontinence Questionnaire Urinary Incontinence Short Form score, range 0-21, with a higher score indicating greater severity). The secondary outcomes were urinary incontinence cure/improvement, other urinary and pelvic floor symptoms, urinary incontinence-specific quality of life, self-efficacy for pelvic floor muscle training, global impression of improvement in urinary incontinence, adherence to the exercise, uptake of other urinary incontinence treatment and pelvic floor muscle function. The primary health economic outcome was incremental cost per quality-adjusted-life-year gained at 24 months.
Results A total of 300 participants were randomised per group. The primary analysis included 225 and 235 participants (biofeedback and basic pelvic floor muscle training, respectively). The mean 24-month International Consultation on Incontinence Questionnaire Urinary Incontinence Short Form score was 8.2 (standard deviation 5.1) for biofeedback pelvic floor muscle training and 8.5 (standard deviation 4.9) for basic pelvic floor muscle training (adjusted mean difference -0.09, 95% confidence interval -0.92 to 0.75; p = 0.84). A total of 48 participants had a non-serious adverse event (34 in the biofeedback pelvic floor muscle training group and 14 in the basic pelvic floor muscle training group), of whom 23 (21 in the biofeedback pelvic floor muscle training group and 2 in the basic pelvic floor muscle training group) had an event related/possibly related to the interventions. In addition, there were eight serious adverse events (six in the biofeedback pelvic floor muscle training group and two in the basic pelvic floor muscle training group), all unrelated to the interventions. At 24 months, biofeedback pelvic floor muscle training was not significantly more expensive than basic pelvic floor muscle training, but neither was it associated with significantly more quality-adjusted life-years. The probability that biofeedback pelvic floor muscle training would be cost-effective was 48% at a £20,000 willingness to pay for a quality-adjusted life-year threshold. The process evaluation confirmed that the biofeedback pelvic floor muscle training group received an intensified intervention and both groups received basic pelvic floor muscle training core components. Women were positive about both interventions, adherence to both interventions was similar and both interventions were facilitated by desire to improve their urinary incontinence and hindered by lack of time. Limitations Women unable to contract their muscles were excluded, as biofeedback is recommended for these women. Conclusions There was no evidence of a difference bet ween biofeedback pelvic floor muscle training and basic pelvic floor muscle training.
Bibliographical noteFunding Information:
This project was funded by the NIHR Health Technology Assessment programme and will be published in full in Health Technology Assessment.
We would like to thank the external members of the TSC for their advice and support for the project: Mr Chris Mayne (University Hospitals of Leicester NHS Trust), Dr Thomas Chadwick (University of Newcastle), Ms Teresa Cook (Independent Consultant, Women’s Health Physiotherapist) and Ms Veronica
Haggar (Homerton University Hospital NHS Foundation Trust). Our thanks go also to the DMEC, comprising Dr Steff Lewis (University of Edinburgh), Dr Sue Hallam (Tameside and Glossop Integrated Care NHS Foundation Trust) and Mr Simon Emery (Abertawe Bro Morgannwg University Health Board). We are grateful to the participants who supported the study, giving so generously of their time and sharing their experiences with us. Likewise, we are grateful to the therapists and the research and administrative staff at all 23 collaborating centres who provided invaluable assistance to us throughout the study. Thank you also to Gladys McPherson and Mark Forrest for their contributions to the development and
ongoing support of the trial database and for input to the PMG; to Graeme McLennan, CHaRT Director, for his contribution to the PMG; and to Lyndsay Wilson who provided advice from a service user perspective on the preparation of the protocol and on the PMG. We would also like to thank Lorna Kerr and Kim Stewart who provided administrative and budget support for the trial.
Support Glasgow Caledonian University agreed to act as sponsor for the research and the study was adopted by the NIHR Clinical Research Network Portfolio (number 15841). The NIHR Collaboration for Leadership in Applied Health Research and Care South West Peninsula at the Royal Devon and Exeter NHS Foundation Trust also supported Sarah Dean’s position at Exeter during this work.