Background and purpose: The research team developed the PelviSense, a non-invasive biofeedback device paired with wearable EMG sensors for use as a PFMT adjunct. The proposed study aims to (1) investigate the efficacy of PelviSense-assisted PFMT compared with unassisted PFMT on SUI in women, and (2) identify the mechanisms underlying the beneficial effects of PFMT for the treatment of SUI. Methods: A two-arm, parallel-group RCT will be conducted using 132 non-pregnant women with SUI, aged 18-60 years. The study participants will be randomised into two study groups: PelviSense-assisted PFMT or unassisted PFMT. Women will be supervised in the performance of PelviSense-assisted and unassisted PFMT on a 1:1 basis for 4 weeks and instructed to perform unsupervised home exercises for 24 weeks after the completion of supervised training. The primary outcome will be the severity of urine loss, as measured using the one-hour pad test. Secondary outcomes will include quality of life, SUI symptoms severity, and PFM strength, as measured using the incontinence impact questionnaire, short-form, international consultation on incontinence questionnaire, short-form, and modified Oxford scale, respectively. Mediator variables will include the following: bladder neck elevation and levator hiatus area. Outcomes and mediator variables will be assessed at baseline, 4, and 28 weeks. Statistical analysis: Treatment and mediation effects will be evaluated using analysis of covariance and the Hayes' PROCESS macro, respectively.

Background and purpose: Stress urinary incontinence (SUI) is a common condition among women, resulting in urine leakage during physical exertion, sneezing, or coughing. Due to the risks associated with surgical and pharmacological treatments, women with SUI often prefer conservative treatments, such as pelvic floor muscle (PFM) training (PFMT). PFMT has been shown to provide increased beneficial effects when combined with electromyography (EMG) biofeedback (referred to as conventional biofeedback [CB]) via intravaginal probe than when performed alone. However, the CB device applied in clinical practice is not preferred by women because it is invasive, and involves the insertion of a probe into the vagina, causing pain/ discomfort. Therefore, we developed the PelviSense, a non-invasive biofeedback device paired with wearable EMG sensors for use as a PFMT adjunct.

The proposed study aims to (1) investigate the efficacy of PelviSense-assisted PFMT compared with unassisted PFMT on SUI in women, and (2) identify the mechanisms underlying the beneficial effects of PFMT for the treatment of SUI.

Hypotheses: (1) women assigned to the PelviSense-assisted PFMT group will report a greater reduction in the severity of urine loss compared with women assigned to the unassisted PFMT group; and (2) improvements in the severity of urine loss will be mediated by increased PFM strength, causing an increase in bladder neck elevation and a reduction in levator hiatus (LH) area during the Valsalva manoeuvre.

Methods: A two-arm, parallel-group RCT will be conducted using 132 non-pregnant women with SUI, aged 18-60 years. The study participants will be randomised into two study groups: PelviSense-assisted PFMT or unassisted PFMT. Women will be supervised in the performance of PelviSense-assisted and unassisted PFMT on a 1:1 basis for 4 weeks and instructed to perform unsupervised home exercises for 24 weeks after the completion of supervised training.

Recruitment. Potential participants will be recruited using a non-probability convenience sampling technique through public and university campus advertising.

Randomisation and blinding. An individual who will not be involved in study recruitment will randomly assign participants to one of the two study groups (PelviSense assisted PFMT or unassisted PFMT) at a 1:1 ratio. Another individual will assign treatments according to a computer-generated random schedule in permuted blocks of two within age strata. The allocation sequence will be concealed using sealed, opaque, sequentially numbered envelopes containing the group name and a personal identification number. To eliminate expectation effects and biases, a research assistant (RA) will complete the assessment of primary outcomes, and intervention training and supervision at HK PolyU will be conducted by a physiotherapy research postgraduate (RPg) student. Raw data will be double entered into a spreadsheet by a student assistant. Data analysis will be completed in a blinded manner, with code names (Groups A and B) used for each group, and group identities will only be revealed after the analysis has been completed.

Procedure and baseline assessment. Potential participants will be invited to an in-person visit at HK PolyU. On the first day of contact, the participants will complete a screening questionnaire and a 1-h pad test to determine eligibility. Written informed consent and body mass index (BMI) measurements will also be obtained. The 1-h pad test, International Consultation on Incontinence Questionnaire, Short Form (ICIQ-SF), the Incontinence Impact Questionnaire, Short Form (IIQ-7), and the MOS will be used to establish baseline severity of urine loss, SUI symptoms severity, QoL, and PFM strength, respectively. Participants will undergo a transabdominal ultrasound (TAU) to measure baseline bladder neck elevation and LH area during a Valsalva manoeuvre (cough). Following baseline assessments, participants will select an envelope containing treatment allocation details.

Outcome measures: The primary outcome will be the severity of urine loss, as measured using the one-hour pad test. Secondary outcomes will include quality of life, SUI symptoms severity, and PFM strength, as measured using the incontinence impact questionnaire, short-form, international consultation on incontinence questionnaire, short-form, and modified Oxford scale, respectively. Mediator variables will include the following: bladder neck elevation and levator hiatus area. Outcomes and mediator variables will be assessed at baseline, 4, and 24 weeks. Statistical analysis: Treatment and mediation effects will be evaluated using analysis of covariance and the Hayes' PROCESS macro, respectively.

Statistical analysis. Hypothesis 1: Treatment effects on both primary and secondary outcomes between T1, T2, and T3 and across the intervention groups will be evaluated using analysis of covariance (ANCOVA).

Hypothesis 2: Mediation analyses will be conducted using the Hayes PROCESS SPSS macro (v2.13), Model 4.

Inclusion Criteria:

• non-pregnant women with SUI and
• aged 18-60 years; and (2) women who experience mild-to-moderate SUI, as indicated by a pad weight gain of 1-50 g during the 1-h pad test and stress test (referred to as the 1-h pad test)

Exclusion Criteria:

• non-pregnant women with SUI and
• aged 18-60 years; and (2) women who experience mild-to-moderate SUI, as indicated by a pad weight gain of 1-50 g during the 1-h pad test and stress test (referred to as the 1-h pad test)
• women with severe pelvic organ prolapse (stages 3 and 4 according to the Baden-Walker Halfway Scoring System)
• women with complicated urinary incontinence due to pelvic radiation; (3) women with severe psychological problems that would impair study participation; and
• women with urge-predominant mixed urinary incontinence or urge incontinence.

Data will shared upon request