Optimizing Remote Access to Urinary Incontinence Treatments for Women Veterans (PRACTICAL): Study Protocol for a Pragmatic Clinical Trial Comparing Two Virtual Care Options

Abstract

Objectives:

In this pragmatic clinical trial, the primary objective is to increase access to behavioral treatment of urinary incontinence (UI) for women Veterans by comparing the effectiveness of two virtual care delivery modalities.

Methods:

Veterans Affairs (VA) clinical sites in AL, GA, NC will virtually randomize 286 women Veterans with UI (ie, stress, urge, or mixed). We will compare the effectiveness of our mHealth UI application (MyHealtheBladder) to a single VA Video Connect (VVC) session delivered by trained UI providers. Women without improvement after 8 weeks will receive an optimization VVC visit using a sequential, multiple assignment, randomized trial (SMART) design. The primary outcome is UI symptom improvement at 12-weeks with or without optimization; secondary outcomes include improvements in lower urinary tract symptoms, adherence, retention rates, perceptions of improvement, and visit-related miles saved. Sample size needed to identify a 2.5-point change (range 0-21) in the International Consultation on Incontinence Questionnaire – Urinary Incontinence Short Form (ICIQ-UI SF) from baseline to 12-weeks post-randomization is 200 participants. Allowing for an attrition rate of 25%, 286 participants are required.

Key Results:

Study team initiated remote recruitment on April 2020. Recruitment is on target with a 75% retention rate. We expect completion in fall of 2023 (clinicaltrials.gov NCT04237753).

Discussion/Conclusion:

Engaging women Veterans with virtual modalities for initial UI treatment may increase access to UI care while also improving symptoms. After assessing efficacy, adherence, and retention, the next step is to implement the most effective option for remote delivery of evidence-based behavioral UI treatment for women Veterans.

BACKGROUND

Urinary incontinence (UI) is a prevalent condition that impacts quality of life for millions of women. Several evidence-based treatments are available for UI including behavioral, medical, and surgical therapies. Specifically, behavioral self-management treatments, including pelvic floor muscle training (PFMT), bladder control and voiding strategies, and fluid management, are widely recommended by consensus groups and are identified by current guidelines as first-line treatment options because of their demonstrated effectiveness and low risk of side effects.¹ Yet, 60-70% of women with UI do not seek treatment and clinicians with expertise to deliver behavioral UI treatments for women are not easily accessible across healthcare setting, especially in rural areas.^2–4 To optimize behavioral UI treatments, more than a single visit may be necessary for symptom improvement over time.⁵ To address this gap in access, we developed and tested an interactive mobile-health UI application, MyHealtheBladder, to deliver an evidence-based behavioral treatment for UI in women Veterans over an 8-week period.⁶

Many women Veterans experience UI and may be at increased risk due to exposures during military service, such as restricted toilet access and the impact on the pelvic floor of heavy protective gear and equipment. A recent study showed that 20 percent of women Veterans recently returning from active duty reported having UI.^7,8 In addition, they face other potentially contributing risk factors throughout their life course, including pregnancy, childbirth, and menopausal transitions.

Increasing numbers of women Veterans, including women aged 50 and older, are using the VA for their general and gender-specific health care, representing a doubling in the past decade. Now, women represent 7% of all Veterans seen in the VA.⁹ Primary care and prevention research among women Veterans is an area of increasing focus within VA.⁹

The purpose of this trial is to determine the optimal method for remote delivery of behavioral therapy for UI to women Veterans. We will compare MyHealtheBladder to a best usual care model delivered by a clinician with UI expertise via virtual video visits delivered on the VHA system called VA Video Connect (VVC). We hypothesized that behavioral treatment for UI delivered via the mHealth UI application for 8-weeks would be superior to delivery through a one-time VVC visit with a continence specialist. In addition, we will utilize a sequential, multiple assignment, randomized trial (SMART) design to assess whether outcomes of the programs can be optimized with the sequential addition of a VVC visit for non-responders.¹⁰ In addition to examining multiple quantitative outcomes, we will also conduct a qualitative evaluation of participants’ views of the program and its key features. This report outlines our study protocol per the Enhancing the Quality and Transparency of Health Research (EQUATOR) methodological framework, for the SPIRIT-Outcomes 2022 extension of the SPIRIT 2013 statement.¹¹

METHODS

Design Overview:

With this pragmatic clinical trial, we will compare the effectiveness of our mHealth UI application, MyHealtheBladder, to VVC visits administered by trained UI providers (best usual care) and evaluate the sequential addition of a 1:1 VVC booster visit for non-responders in both randomization groups. This study builds on our single site pilot study that evaluated the feasibility of delivering UI treatment with MyHealtheBladder.⁶ This SMART design (Figure 1) will include two randomization stages and compare 1 optimization factor at 8 weeks (defined as a VVC visit with a trained UI provider) for both groups over a 3-month intervention period with outcomes assessed at baseline, 4-weeks, 8-weeks, 12-weeks (primary outcome), and 24-weeks for durability. With the SMART design, we will investigate the effectiveness of two remote methods for the clinical delivery of a behavioral self-management program that may reduce UI severity and improve access to care for women Veterans (n=286).

Figure 1.

Study Flow Diagram

Ethical Considerations:

IRB approval for this multisite study will occur at each individual site and informed consent procedures followed per local policies as the VA office of human research protections determined the single IRB would not be required for this protocol. Research coordinators obtained signed informed consent documents via mail and with VA-approved electronic signatures.

Study Setting:

This multi-site clinical trial will be conducted at 3 VA sites: Birmingham, AL, Atlanta, GA, and Durham, NC.

Recruitment:

Women Veterans will be identified through 3 sources: 1) referrals from primary care clinic providers for women Veterans; 2) targeted letters for women Veterans (as identified by administrative records of those receiving primary care through the VA); and 3) flyers/advertisements posted in the medical centers and clinics. Women who respond will undergo a telephone screening assessment by trained research staff for eligibility (Table 1). We will track the number of screening telephone calls completed at each site. Trained staff will complete the informed consent process by phone. Signed consent forms can be received by mail or via electronic signature.

Table 1.

Eligibility Criteria

Inclusion criteria:

• Female Veterans

• Urinary incontinence (not specific to type) occurring at least monthly for 3 months

• Able to access daily internet via computer or mobile device

• Access to personal email for MyHealtheBladder and VVC visit initiation and reminders

Exclusion criteria:

• Unstable medical conditions that could contribute to incontinence (e.g., recent major hospitalization, planned major surgery, conditions that affect urine volume - hemoglobin A1c of ≥ 9.0, chronic kidney disease with planned dialysis within 3 months, as assessed by PI or Site PI)

• Unstable psychiatric conditions (e.g., psychosis, suicidal, active alcohol/substance abuse based on history and medical records)

• Unstable housing situation

• Genitourinary cancer undergoing active treatment with chemotherapy or radiation

• Neurologic conditions known to contribute to incontinence (Multiple Sclerosis, Parkinson’s disease, TBI, Dementia, and Stroke survivors with limited mobility)

• New treatments for incontinence started in the prior 3 months or planned during the 6-month study duration (includes medications and/or surgery)

• Currently pregnant

• < 12 weeks postpartum

• Current or prior surgically implanted sacral nerve stimulation device or botulinum toxin bladder injections for UI

Eligibility criteria:

Participants will be women Veterans without any age limits (by definition Veterans are 18 years and older) who meet self-reported criteria for UI for at least 3 months. See Table 1 for inclusion/exclusion criteria. No in-person evaluation for eligibility will be required for this study.

Interventions:

Participants in both randomization groups will receive evidenced-based behavioral treatment either through a mHealth UI application (Figure 1) with daily content over 8-weeks or the same standardized content delivered through a single VVC visit with a clinician experienced in continence care (Figure 2). Both groups will have access to printed materials in an evidence-based booklet form adapted for women Veterans (Supplemental material).¹² Certified Registered Nurse Practitioners (CRNPs) who deliver UI care in Birmingham and Atlanta will implement the intervention for all sites. Centralized training was conducted to standardize the VVC content across all sites.

Figure 2.

MyHealtheBladder Platform Information

The major components of the behavioral intervention for MyHealtheBladder and the VVC visit has been published previously and include: bladder education on anatomy and function, UI risk factor reduction, pelvic floor muscle exercises with behavioral strategies, toileting strategies, and self-monitoring.^6,12,13 MyHealtheBladder also has stories/quotes to reinforce behavior change, acknowledge the experience of women Veterans, as well as reminder features for adherence. Built-in reminders are a core component of MyHealtheBladder.⁶ Women will complete daily modules and have the opportunity to make up any missed modules within 2 weeks from the assigned day.

Randomization:

Participants will be randomized remotely by a computer-generated algorithm and given access to start MyHealtheBladder with email instruction (written and video) or an appointment will be given for the 1:1 VVC visit through usual care clinical pathways. We will stratify eligible women according to site and by UI symptom severity (mild, moderate, severe) using the same validated UI questionnaire used for outcome measurement. Given the randomization schema, study team members will not be blinded to intervention arm. Site investigative teams will be blinded to all outcomes; whereas study coordinators will determine responders and nonresponders for the second randomization.

Primary Outcome, Secondary Outcomes, and Key Variables (Figure 3, Table 2):

Figure 3.

VA Video Connect (VVC) Platform Information – Handout Accessed from the public VA website at VA Video Connect | VA Mobile

Table 2.

Schedule of Measures

CONSTRUCTS	MEASURES AND KEY VARIABLES	Baseline	4-wks	8-wks	12-wks	24-wks
PRIMARY OUTCOME
Urinary Incontinence Severity/Symptom burden and Bother	International Consultation on Incontinence Modular Questionnaire – Urinary Incontinence Short Form,14 ICIQ-UI SF, is a 3-item instrument that measures UI frequency, volume loss, and 1-item for bother measured through the MyHealtheBladder program or VVC administration	X	X	X	X	X
SECONDARY OUTCOMES
Other lower urinary tract symptoms: urgency, frequency, nocturia	International Consultation on Incontinence Modular Questionnaire (ICIQ)-Overactive Bladder (OAB) is a 4-item instrument that measures nocturia, urgency, frequency, urinary incontinence symptoms and bother14	X			X	X
Global Perceptions and Satisfaction with Treatment	Validated tools developed for use in clinical trials of incontinence treatments17				X	X
Pelvic floor muscle exercise adherence	Self-reported adherence to behavioral therapy - collected weekly through the mHealth platform5		X	X	X	X
Direct and indirect costs for UI	Incontinence Resource Utilization Questionnaire: IRUQ18	X			X
Miles saved	Map function with data to/from living location to the clinical sites	X			X
Usability	The Health Information Technology Usability Evaluation Scale (Health-ITUES) has demonstrated reliability and validity for use in assessing the usability of mHealth technologies in community-dwelling adults living with a chronic illness.19				X
Adaptive Behavior Index	Validated tool –UI module20	X			X

The primary outcome is the validated International Consultation on Incontinence Questionnaire – Urinary Incontinence Short Form (ICIQ-UI SF) and was selected based on psychometric properties and use in other studies using similar interventions.^14,15 The 3 item ICIQ-UI will be completed at baseline, 4-week, 8-weeks, and 12-weeks (primary outcome timepoint), with durability at 24-weeks. (Figure 1). We will determine responder or non-responder status based on the ICIQ-UI SF responses at 8-weeks to enable a second randomization for the addition of a booster VVC visit with the trained UI clinician (optimization factor). The ICIQ-UI minimal clinically important difference (MCID) threshold will be used to define a responder or a nonresponder (±2.52 points, range 0-21).

Secondary outcomes include other lower urinary tract symptoms,¹⁴ impact on quality of life,¹⁶ adherence to the pelvic floor muscle exercises,⁵ retention rates for program completion, satisfaction with treatment,¹⁷ perceptions of improvement,¹⁷ costs of incontinence care,¹⁸ miles saved, usability of the mHealth app,¹⁹ and adaptive behaviors related to UI (Table 2).²⁰ Other key variables include demographic variables, such as age, education, marital status, and race/ethnicity, medical (obesity and other co-morbidities) and mental health conditions (depression), obstetrical history (parity and hysterectomy), perceived stress, sleep characteristics, and medications.^21–23 At 12-weeks and 24-weeks, the final assessments will capture any changes in medications, additional treatments, symptom burden, and impact on quality of life.

Sample Size:

We powered this study using the minimally important clinical difference of the ICIQ-UI Short Form (Range:0-21; MICD: 2.52 (±2.56)).¹⁵ Power and sample size calculations for SMART designs have more than one power function and follow a protocol detailed in Collins et al. (2014).²⁴ Initially, the sample size is based on the power to detect an effect size at the first stage of the trial. Given the UI severity reduction at 4-weeks cited above, the standardized effect size ranges from 0.40 to 0.45, resulting in total sample sizes ranging from 200 to 158. The power to detect effects at the second stage of the SMART trial depends on the projected percentage of responders in each arm of the trial. Based on a 40% response rate (achievement of the MICD at 8 weeks), 70% of patients in each arm would remain on their initially randomized treatment program and 30% in each arm would be re-randomized to the booster VVC visit or continued intervention. With an initial sample size of 200 patients completing the first stage, 140 (70 in each arm) would be available for comparison at the 12-week endpoint. This sample size would yield over 99% power to detect between-group differences at a 5% significance level based on the UI severity reduction at 8-weeks cited above. We anticipate a drop-out rate of 25% given our pilot data. To account for a 75% retention rate (25% drop-out rate), we will recruit women at each site with a total sample size of 286 women.

Statistical Analysis:

The primary endpoint will be assessed at 12 weeks post-randomization with durability data collected at 24-weeks post-randomization. Point estimates of mean ICIQ-UI SF change and standard deviation of change, as well as 95% CI will be performed using SAS statistical software (SAS Institute). Kolmogorov–Smirnov tests will be used to test normality of distributions. Comparability of groups at baseline will be analyzed using Chi-square test for categorical variables and two-sample t-test for continuous variables (if normally distributed) for demographics, medical history, and obstetrical history.

To test the overall differences in outcomes at 12-weeks, paired t-tests of differences will be used by summing results over both groups. Repeated measures ANOVA will be used to test outcome time differences from baseline between both groups, at 8-weeks, 12-weeks, and 24-weeks. Repeated measures ANCOVA will be used to adjust for baseline outcome measurements, UI type, age, obesity, parity, and depression. Despite best efforts, there will inevitably be missing data to contend with in the analyses. Techniques we propose to use assume that missing data will be missing at random (MAR), meaning that the missingness mechanism does not depend on treatment success but may depend on baseline characteristics. In general, we plan to include baseline factors that are associated with loss-to-follow-up in the analysis of outcomes. We also plan to evaluate the MAR assumption using sensitivity analysis. The final protocol, statistical code, and de-identified data set will be available upon request after study completion.

Monitoring:

A VA central data safety and monitoring board (DSMB) will have oversight of the safety and enrollment of VA-funded multi-site research. Monitoring begins with a review of the Data Analysis Plan (DAP) prior to study funding and continues with annual evaluation of study progress, including patient accrual and retention, and adverse events. Each local VA will perform annual study audits.

Results:

The clinical trial started recruitment at all participating sites, Birmingham, Atlanta, and Durham, on 04/20/2020 during the COVID-19 pandemic. Recruitment goals are on target to date, and the trial recruitment is projected to be completed in second quarter, 2023. Current retention rates are 75% for the primary outcome. The trial is registered at clinicaltrials.gov NCT04237753.

Table 3 shows the baseline demographic and selected key variables for the women Veterans participating by randomization group. Despite having a wide age range among the participating women Veterans (23-83 years of age), our mean age was 53 ± 11.4 years with 58% self-reporting African American/Black race. The majority of women Veterans self-reported being in the Army with an average of 9.1 ± 7.9 years of service. Women Veterans also had a wide variability in miles saved by not having to travel to the closest VA medical center for UI treatments, 55.3 ± 54.9 miles saved (range 0 to 362 miles). Overall, no major differences were noted for these variables by randomization group.

Table 3.

Baseline Demographics and Selected Variables of Women Veterans Participating in the PRACTICAL Clinical Trial

	Aggregate N = 286	MyHealtheBladder N = 147	VA Video Connect N = 139
Socio-demographics
Age, Mean ± SD	53 ± 11.4	53.2 ± 11.2	52.8 ± 11.6
Age, range in years	23 to 83	27 to 75	23 to 83
Ethnicity, N (%)
Hispanic	18 (6.3)	7 (4.8)	11 (7.9)
Not Hispanic	265 (92.7)	138 (93.9)	127 (91.4)
Not Reported	3 (1.1)	2 (1.4)	1 (0.7)
Race, N (%)
White	101 (35.3)	48 (32.7)	53 (38.1)
Black/African American	166 (58.0)	92 (62.6)	74 (53.2)
More than one race	8 (2.8)	4 (2.7)	4 (2.9)
Other	11 (3.9)	3 (2.0)	8 (5.8)
Education, N (%)
HS graduate	16 (5.6)	8 (5.4)	8 (5.8)
Some college, no degree	54 (18.9)	26 (17.7)	28 (20.1)
Associate degree	64 (22.4)	39 (26.5)	25 (18.0)
Bachelor’s degree	84 (29.4)	40 (27.2)	44 (31.7)
Master’s degree	64 (22.4)	32 (21.8)	32 (23.0)
Doctoral degree	4 (1.4)	2 (1.4)	2 (1.4)
Marital status N (%)
Married	113 (39.5)	62 (42.2)	51 (36.7)
Divorced	93 (32.5)	45 (30.6)	48 (34.5)
Widowed	16 (5.6)	5 (3.4)	11 (7.9)
Separated	12 (4.2)	9 (6.1)	3 (2.2)
Never married	48 (16.8)	25 (17.0)	23 (16.6)
Unmarried couple	4 (1.4)	1 (0.7)	3 (2.2)
Branch of service N (%)
Army	157 (54.9)	84 (57.1)	73 (52.5)
Navy	39 (13.6)	20 (13.6)	19 (13.7)
Air Force	45 (15.7)	19 (12.9)	26 (18.7)
Reserve	24 (8.4)	16 (10.9)	8 (5.8)
Marine Corps	10 (3.5)	3 (2.0)	7 (5.0)
National Guard	7 (2.5)	3 (2.0)	4 (2.9)
Coast Guard	3 (1.1)	2 (1.4)	1 (0.7)
NOAA	1 (0.4)	0	1 (0.7)
Years of service, Mean ± SD	9.1 ± 7.9	8.9 ± 8.0	9.3 ± 7.8
Other Self-Reported Data
Body mass index (kg/m2): Mean ± SD	32.0 ± 6.6	32.2 ± 6.4	31.7 ± 6.9
Device types - Not mutually exclusive
Cell phone	236 (82.5)	119 (81.0)	117 (84.2)
Tablet	100 (35.0)	49 (33.3)	51 (36.7)
Laptop/PC	194 (67.8)	97 (66.0)	97 (69.8)
Other Data (not self-reported)
Miles saved from clinic visit (calculated by address): Mean + SD	55.3 ± 54.9	51.1 ± 52.1	59.8 ± 57.5
Median (Range)	37 (0, 362)	32 (2, 362)	46 (0, 346)
Site, N (%)
Alabama	106 (37.0)	51 (34.7)	55 (39.6)
Georgia	80 (28.0)	43 (29.3)	37 (26.6)
North Carolina	100 (35.0)	53 (36.1)	47 (33.8)

Discussion:

There are multiple issues of interest encountered both prior to and during early implementation that influenced trial design and protocol.

Choice of study design:

Our team designed the current clinical trial to answer two key questions related to the SMART design: 1) Can we improve remote access to effective evidence-based behavioral UI treatments for women Veterans via mHealth over best usual care delivered via VVC? and 2) Will a 1:1 video visit improve outcomes for nonresponders? We chose the SMART design given that women Veterans with UI may need an optimization or more intensive strategy to improve UI symptoms with virtual care after 8-weeks. Thus, meeting patient needs with a sequential treatment strategy with a VVC visit is an important consideration for UI treatment.¹⁰ At the time of the grant award in January of 2020, VVC was considered the ‘best usual care’ approach to remote care within the VA. With the COVID-19 pandemic in March of 2020, our team was able to maintain our study design without any protocol changes and without any in-person visits. This clinical trial design enabled remote recruitment efforts. Similarly, we designed the clinical trial to use remote methods to collect consent forms and all study outcomes. The wide use of digital signing software adopted during our study further facilitated participation in this trial. Lastly, we were able to secure a VA-approved contract with a company that we used for other clinical trials that utilized mobile health technology. We designed this trial to collect baseline data and self-reported outcomes for both intervention groups via a mobile phone, tablet, or desktop computer. For participants randomized to the MyHealtheBladder application, the daily content was available to them via the same electronic platform. The VVC group only had access to our educational booklet (supplemental material) on this platform.

Recruitment:

Our team utilized several unique recruitment strategies for this trial. We were able to send letters to women Veterans receiving primary care in our regions to inform them about this study. If interested, women Veterans could contact us for information about the study and our team would screen them for eligibility. Given that the majority of women do not discuss UI with their providers, this was a key recruitment strategy.^2–4 In Birmingham and Atlanta, we were also able to recruit through existing relationships with primary care providers, especially through primary care clinics with designated care teams serving the needs of women Veterans. Unique to our two sites, we have existing clinics that deliver behavioral care for UI. At the Durham VA, they did not have an existing clinic at the start of the study, but a pelvic floor physical therapy clinic opened during the trial. While this may have allowed increased access to behavioral UI care for women living in and around the Durham, NC area, many women who seek care may be living in more rural areas and would not be likely to attend these in person appointments. Lastly, we were able to deliver remote care to participants in NC with VVC visits with continence care experts located in Birmingham and Atlanta due to VA policies that allow care to be provided across state lines.

Pragmatic trial design to enhance potential for implementation and dissemination:

Since our goals were to assess effectiveness and implementation (e.g. hybrid design), the pragmatic design was informed by VA national priority goals and operational partner priorities to enhance the potential for broad dissemination at the conclusion of the study.²⁵ Specifically, this proposal targets performance metrics related to timeliness in getting appointments, care, and information, as well as timeliness in getting appointments for UI treatment. The proposal may also address facility-level performance goals by improving outpatient self-management support for the mHealth application (Patient Experiences Dimensions of Care) and Access to Care for the mHealth program and VVC visits. This proposal is aligned with two VA operational partners that will assist with future implementation and dissemination: the VA Offices of Women’s Health and Connected Care.

Strengths and Limitations:

There are several limitations inherent to the virtual design of this trial. As previously discussed, participants will not have in-person contact but women randomized to the VVC visits will have contact via video visits. This may result in diminished commitment to the program, resulting in a higher-than-expected drop-out rate. Given this, we designed the study to have a 25% drop out rate. Also, we want to maximize retention by making phone calls as well as intermittent contact in both arms to increase study retention. In addition, there is no physical examination, limiting the ability to assess for the presence of pelvic organ prolapse, or the ability to assess contraction of the pelvic floor muscles at baseline. Randomization may help with this limitation given the women who cannot self-perform pelvic floor muscle exercises may be equally present in either intervention group. Although these are important concerns, our aim is to align this study design with the focus on increasing access to care, especially during the COVID-19 pandemic, further justifying this decision. Another justification for remote treatment is a prior clinical trial for UI that used self-reported, questionnaire-based symptoms to enroll participants to randomized treatments without needing an in-person examination.^26,27 To facilitate symptom reporting, we include validated assessment tools such as the ICIQ-UI, as well as other self-reported and validated questionnaires for other lower urinary tract symptoms.

An additional limitation is the reliance of this study on access to the internet. This may limit the generalizability of the study to the broader population, some of whom do not possess the access, desire, or skills to use internet-based technologies. This may be a concern, particularly in the older population. To address this, our study will evaluate a subgroup of older women who do not meet study criteria and reasons for ineligibility. The VA does provide tablets with internet capabilities for Veterans who may not have the ability to buy or maintain technologies in their current living situations. We will assess the use of technology provided by the VA for participants.

Strengths of the study include the pragmatic randomized trial design, use of validated measures, and the study powered to measure clinically relevant outcomes that can be compared across other trials in the literature using mobile health technologies. To ensure fidelity of the VVC visits, the study launched after having an online 2-day training session for the continence care providers and fidelity will be monitored throughout the trial.

Conclusions and Next Steps:

In conclusion, this trial will inform whether the use of a mobile health application for evidence-based behavioral UI treatment offers higher efficacy compared to the best usual care delivered remotely via video visits. The novel trial design may address an important efficacy consideration prior to conducting future studies aimed at effectiveness with hybrid-type 2 implementation science designs.