Cost-effectiveness of behavioral and pelvic floor muscle therapy combined with midurethral sling surgery vs surgery alone among women with mixed urinary incontinence: results of the Effects of Surgical Treatment Enhanced With Exercise for Mixed Urinary Incontinence randomized trial

Abstract

Background: Urinary incontinence is prevalent among women, and it has a substantial economic impact. Mixed urinary incontinence, with both stress and urgency urinary incontinence symptoms, has a greater adverse impact on quality of life and is more complex to treat than either stress or urgency urinary incontinence alone. Studies evaluating the cost-effectiveness of treating both the stress and urgency urinary incontinence components simultaneously are lacking.

Objective: Cost-effectiveness was assessed between perioperative behavioral and pelvic floor muscle therapies combined with midurethral sling surgery and midurethral sling surgery alone for the treatment of women with mixed urinary incontinence. The impact of baseline severe urgency urinary incontinence symptoms on cost-effectiveness was assessed.

Study design: This prospective economic evaluation was performed concurrently with the Effects of Surgical Treatment Enhanced with Exercise for Mixed Urinary Incontinence randomized trial that was conducted from October 2013 to April 2016. Participants included 480 women with moderate-to-severe stress and urgency urinary incontinence symptoms and at least 1 stress urinary incontinence episode and 1 urgency urinary incontinence episode on a 3-day bladder diary. The primary within-trial analysis was from the healthcare sector and societal perspectives, with a 1-year time horizon. Costs were in 2019 US dollars. Effectiveness was measured in quality-adjusted life-years and reductions in urinary incontinence episodes per day. Incremental cost-effectiveness ratios of combined treatment vs midurethral sling surgery alone were calculated, and cost-effectiveness acceptability curves were generated. Analysis was performed for the overall study population and subgroup of women with Urogenital Distress Inventory irritative scores of ≥50th percentile.

Results: The costs for combined treatment were higher than the cost for midurethral sling surgery alone from both the healthcare sector perspective ($5100 [95% confidence interval, $5000-$5190] vs $4470 [95% confidence interval, $4330-$4620]; P<.01) and the societal perspective ($9260 [95% confidence interval, $8590-$9940] vs $8090 [95% confidence interval, $7630-$8560]; P<.01). There was no difference between combined treatment and midurethral sling surgery alone in quality-adjusted life-years (0.87 [95% confidence interval, 0.86-0.89] vs 0.87 [95% confidence interval, 0.86-0.89]; P=.90) or mean reduction in urinary incontinence episodes per day (-4.76 [95% confidence interval, -4.51 to 5.00] vs -4.50 [95% confidence interval, -4.25 to 4.75]; P=.13). When evaluating the overall study population, from both the healthcare sector and societal perspectives, midurethral sling surgery alone was superior to combined treatment. The probability that combined treatment is cost-effective compared with midurethral sling surgery alone is ≤28% from the healthcare sector and ≤19% from the societal perspectives for a willingness-to-pay value of ≤$150,000 per quality-adjusted life-years. For women with baseline Urogenital Distress Inventory irritative scores of ≥50th percentile, combined treatment was cost-effective compared with midurethral sling surgery alone from both the healthcare sector and societal perspectives. The probability that combined treatment is cost-effective compared with midurethral sling surgery alone for this subgroup is ≥90% from both the healthcare sector and societal perspectives, at a willingness-to-pay value of ≥$150,000 per quality-adjusted life-years.

Conclusion: Overall, perioperative behavioral and pelvic floor muscle therapies combined with midurethral sling surgery was not cost-effective compared with midurethral sling surgery alone for the treatment of women with mixed urinary incontinence. However, combined treatment was of good value compared with midurethral sling surgery alone for women with baseline severe urgency urinary incontinence symptoms.

Trial registration: ClinicalTrials.gov NCT04171531.

Keywords: EuroQol-5 Dimension; behavioral therapy; midurethral sling; mixed urinary incontinence; pelvic floor muscle therapy; quality-adjusted life-years; stress urinary incontinence; urgency urinary incontinence.

Copyright © 2021 Elsevier Inc. All rights reserved.

Figures

Figure 1.. Incremental Cost-effectiveness Ratio Scatterplot for Combined Treatment versus Sling Only for Health Care and Societal Perspectives (base case).

Figure 1 shows scatterplots of points representing pairs of mean differences in cost and mean differences in QALYs for combined treatment versus sling only from 5,000 bootstrapped replications with replacement for both the healthcare sector (orange) and the societal (blue) perspectives. The difference in mean cost is on the y-axis and difference in mean QALY is on the x-axis. Points that lie above the horizontal axis represent replications in which combined treatment costs more than sling only, while points below indicate replications in which sling only costs more than combined treatment. Points to the right of the vertical axis represent replications in which combined treatment was more effective than sling only, while points on the left indicate replications in which sling only was more effective than combined treatment. Estimates primarily fell in two quadrants: the upper right quadrant shows pairs with higher mean costs and QALYs for combined treatment than sling only, and the upper left quadrant shows pairs with higher mean costs but lower mean QALYs for combined treatment than sling only. The 95% confidence intervals (CI) are indicated, with points not lying between the upper and lower bound CI lines falling outside the 95% CI. The “X” points indicate the ICER point estimates. $/QALY lines are indicated; points below and to the right of each line are considered cost effective at this willingness-to-pay level.

Figure 2.. Cost-effectiveness Acceptability Curve (CEAC): Base case versus UDI-Irritative Subsamples ( ≥75th percentile at baseline and ≥50th percentile at baseline)

Figure 2 shows the cost-effectiveness acceptability curve for combined treatment versus sling only. The 5,000 bootstrapped ICER replications with replacement were used to derive cost-effectiveness acceptability frontiers, which plot the probability of the combined treatment being cost-effective across a range of values of willingness-to-pay in U.S. dollars per QALY; willingness-to-pay values ranging from zero to $250,000 per QALY gained were evaluated. Probabilities representing the percentage of replications that were cost-effective at each WTP value are plotted to create cost effectiveness acceptability curves for combined treatment versus sling only. The range of common maximum willingness-to-pay per QALY gained thresholds ($50,000 - $150,000) is shaded. For the overall study population, combined treatment had a probability of being cost effective compared with sling only of 28% from the health care sector and 19% from the societal perspective at the generally accepted maximum willingness-to-pay per QALY gained threshold of $150,000. For the subgroup of women with Urogenital Distress Inventory-irritative scores ≥50th and ≥75th percentiles at baseline, the probability that combined treatment is cost effective relative to midurethral sling surgery alone is ≥90% from both the health care sector and societal perspectives, at a willingness-to-pay value of $150,000/QALY.

Figure 3

a. Incremental Cost-effectiveness Ratio Scatterplot for Combined Treatment versus Sling Only for Health Care and Societal Perspectives (UDI-Irritative ≥50th percentile at baseline) b. Incremental Cost-effectiveness Ratio Scatterplot for Combined Treatment versus Sling Only for Health Care and Societal Perspectives (UDI-Irritative ≥75th percentile at baseline) Figure 3 shows scatterplots of points representing pairs of mean differences in cost and mean differences in QALYs for combined treatment versus sling only from 5,000 bootstrapped replications with replacement for both the healthcare sector (orange) and the societal (blue) perspectives. The sub-group of women with Urogenital Distress Inventory-irritative scores ≥50th percentile at baseline is in Panel A and ≥75th percentile at baseline is in Panel B. The difference in mean cost is on the y-axis and difference in mean QALY is on the x-axis. Points that lie above the horizontal axis represent replications in which combined treatment costs more than sling only, while points below indicate replications in which sling only costs more than combined treatment. Points to the right of the vertical axis represent replications in which combined treatment was more effective than sling only, while points on the left indicate replications in which sling only was more effective than combined treatment. Estimates primarily fell in two quadrants: the upper right quadrant shows pairs with higher mean costs and QALYs for combined treatment than sling only, and the lower right quadrant shows combined treatment as “dominant,” with lower mean costs and higher mean QALYs than sling only. The 95% confidence intervals (CI) are indicated, with points not lying between the lower and upper bound CI lines falling outside the 95% CI. The “X” points indicate the ICER point estimates. $/QALY lines are indicated; points below and to the right of each line are considered cost-effective at this willingness-to-pay level.

Figure 3

a. Incremental Cost-effectiveness Ratio Scatterplot for Combined Treatment versus Sling Only for Health Care and Societal Perspectives (UDI-Irritative ≥50th percentile at baseline) b. Incremental Cost-effectiveness Ratio Scatterplot for Combined Treatment versus Sling Only for Health Care and Societal Perspectives (UDI-Irritative ≥75th percentile at baseline) Figure 3 shows scatterplots of points representing pairs of mean differences in cost and mean differences in QALYs for combined treatment versus sling only from 5,000 bootstrapped replications with replacement for both the healthcare sector (orange) and the societal (blue) perspectives. The sub-group of women with Urogenital Distress Inventory-irritative scores ≥50th percentile at baseline is in Panel A and ≥75th percentile at baseline is in Panel B. The difference in mean cost is on the y-axis and difference in mean QALY is on the x-axis. Points that lie above the horizontal axis represent replications in which combined treatment costs more than sling only, while points below indicate replications in which sling only costs more than combined treatment. Points to the right of the vertical axis represent replications in which combined treatment was more effective than sling only, while points on the left indicate replications in which sling only was more effective than combined treatment. Estimates primarily fell in two quadrants: the upper right quadrant shows pairs with higher mean costs and QALYs for combined treatment than sling only, and the lower right quadrant shows combined treatment as “dominant,” with lower mean costs and higher mean QALYs than sling only. The 95% confidence intervals (CI) are indicated, with points not lying between the lower and upper bound CI lines falling outside the 95% CI. The “X” points indicate the ICER point estimates. $/QALY lines are indicated; points below and to the right of each line are considered cost-effective at this willingness-to-pay level.