Background Pelvic floor muscle (PFM) training (PFMT) is a common conservative first-line treatment for SUI in women. However, the success of PFMT alone, without adjunctive therapies, has been hampered by the lack of training adherence among patients, who report the inability to contract the correct PFMs and limited motivation to perform PFMT exercises. Therefore, PFMT is often combined with adjunctive therapies, such as electromyography (EMG) biofeedback via an intravaginal probe (referred to as conventional biofeedback), transvaginal electrical stimulation (TVES), or weighted vaginal cones. However, PFMT adjunctive therapies that involve intravaginal probes/ cones are not preferred by women because they are invasive, and the vaginal insertion of probes/ cones can cause pain and discomfort. In addition, older women with atrophic vaginas have difficulty retaining probe/cones within the vaginal canal due to weak PFM tone.
To meet the existing need for a non-invasive PFMT adjunctive therapy, the research team developed the PelviSense. The PelviSense is an innovative, non-invasive biofeedback device with wearable EMG sensors, designed to assist individuals in isolating and strengthening PFMs by helping them distinguish correct PFM contractions (lifting or squeezing the muscles) from incorrect attempts (bearing down). A mixed-methods study, including a full-scale RCT and semi-structured focus groups, alongside an economic evaluation, will be conducted to evaluate both the effectiveness and cost-effectiveness of the newly developed PelviSense device for the treatment of SUI in women.
Aims:
Aim 1: Investigate the efficacy of PelviSense-assisted PFMT compared with PFMT alone for improving SUI symptom severity in women.
Aim 2: Compare the cost-effectiveness of the PelviSense device against PFMT alone for treating SUI in women.
Aim 3: Explore the views of women regarding the use of the PelviSense device as an adjunct to PFMT and the impacts of the PelviSense device on their lives and well-being.
Methods Recruitment: Women with SUI who attend the outpatient Obstetrics and Gynecology Department of Kwong Wah Hospital (KWH) will be recruited for the study. In addition, potential participants will be recruited using a non-probability convenience sampling technique through public, KWH, elderly care centers, and HK Polytechnic University (PolyU) campus advertising. Advertising will consist of study flyers posted at KWH, around the university campus, in waiting rooms of university health services and physiotherapy clinics, in community churches, and in university-owned amenities, including sports centres and gyms.
Randomisation and blinding: Computer-generated randomisation will be utilised to ensure that both study groups are assigned an equal number of participants at a 1:1 ratio. Sequentially numbered, sealed, opaque envelopes of similar size and shape will be utilised to conceal group allocation and avoid selection bias. Due to the nature of the intervention, it is not possible to blind the therapists or participants to the allocation of treatment; therefore, only the outcome assessor will be blinded. To minimise detection bias, the outcome assessors will not participate in the treatment, and all study outcomes will be assessed in a separate room without any knowledge of participant allocation. Statistical analysis will be performed in a blinded manner, with each group assigned a code (group A and group B). The group identities will be revealed after all analyses are completed.
Procedure and baseline assessment: Potentially eligible women who contact research personnel will be invited to attend an in-person screening session. On this first day of contact, all women will undergo BMI measurement; complete the screening questionnaire, MMSE, and 1-h pad test to confirm eligibility; and provide written informed consent. Baseline assessments of study outcomes will be performed, and eligible and consenting participants will be randomised to intervention groups.
Outcome measures Outcomes assessments will be completed at baseline (time point 1 [T1]), 1-month (T2) and 6-months (T3).
Intervention. The interventions will be provided for 6 months, beginning with a 1-month supervised training period at KWH and PolyU. Following supervised training, women will be instructed to perform unsupervised home exercises for 5 months. Unsupervised home exercises will consist of at least eight contractions, performed 3 times per day, for 5 days of each week, with or without the biofeedback device according to group assignment. Each contraction will last for 10 seconds, followed by relaxation for 10 seconds. The PFMT exercise parameters are based on the National Institute for Health and Care Excellence PFMT guidelines for UI.
Qualitative study (semi-structured focus groups). A computer-generated randomization schedule will be utilized to identify 20 participants (from the 71 participants included in the PelviSense-assisted PFMT group) to participate in the focus groups. Each focus group will last for 45-60 minutes. Focus groups will continue to be conducted until data saturation occurs. Focus groups will be scheduled for 8-months post-randomisation. A semi-structured focus group approach will be used, during which discussion will be guided by but not limited to a set of pre-determined, open-ended questions.
Data processing and analysis: Statistical analysis will be performed on an intention-to-treat basis using the Statistical Package for the Social Sciences (v.25; Armonk, NY).
Treatment effects for both primary and secondary outcomes between T1, T2, and T3 and across the intervention groups will be evaluated using a two-way analysis of covariance (ANCOVA) with age as the covariate. T1 and T2 will be compared to establish immediate treatment effects, and T1 and T3 will be compared to evaluate the retention of treatment effects. For significant group by time interaction effects, post-hoc analysis using Bonferroni correction will be performed.
The unadjusted mean total and disaggregated costs and differences in costs between the PelviSense and PFMT alone groups will be calculated. Seemingly unrelated regression (SUR) analyses will be used to estimate total cost differences (ΔC) and effect differences (ΔE).
The EQ-5D-5L responses will be converted into utility scores to estimate the gain or loss of quality-adjusted life-years (QUALYs). The incremental cost-effectiveness ratio (ICER) will be calculated using the formula ICER = ΔC / ΔE. Uncertainty surrounding ICERs and 95% confidence intervals for cost differences will be estimated using bias-corrected and accelerated bootstrapping with 5,000 replications.
Qualitative data from the focus groups will be analysed using the six phases of thematic analysis recommended by Braun and Clarke. The identified themes will then be defined and named according to their contents. The final set of themes will be confirmed as coherent and comprehensive by review of the team.
