Economic impact of using fesoterodine for the treatment of overactive bladder with urge urinary incontinence in a vulnerable elderly population in the United States

Abstract

Objectives:

To assess the costs of treating overactive bladder (OAB) with fesoterodine compared to no OAB pharmacotherapy among vulnerable elderly from the US payer perspective.

Methods:

A decision analytic cost model was developed to estimate the 52-week costs of a cohort of vulnerable elderly with OAB initiating treatment with fesoterodine or no OAB pharmacotherapy. Vulnerable elderly OAB patients were defined as those aged ≥65 years with self-reported urge urinary incontinence (UUI) symptoms for ≥3 months, 2–15 UUI episodes/day, and at risk of deteriorating health by a score of ≥3 on the Vulnerable Elders Survey (VES)-13. Patients were evaluated for fesoterodine treatment response (defined as no UUI episodes) and persistence at weeks 12, 26, and 52. The model included a hypothetical health plan with 100,000 elderly members. A total of 7096 vulnerable elderly subjects were identified as the model target population based on the percentage of vulnerable elderly and annual prevalence of OAB among vulnerable elderly. OAB-related costs included fesoterodine drug acquisition costs, healthcare resource use (inpatient hospitalization, outpatient visits, and physician office visits), and OAB-related co-morbidities (falls/fractures, urinary tract infections, depression, and nursing home admissions). All costs were inflated to 2013 US$ using the medical care component of the consumer price index (CPI).

Results:

When 7096 vulnerable elderly OAB patients were treated with fesoterodine, US healthcare payers could save $11,463,981 per year, or $1616 per patient vs no OAB pharmacotherapy. Univariate one-way sensitivity analyses supported the robustness of the findings and showed results were most sensitive to changes in fesoterodine efficacy followed by annual costs of inpatient hospitalization.

Conclusions:

From a US payer perspective, treating vulnerable elderly OAB patients with fesoterodine was cost-saving compared to no OAB pharmacotherapy.

Keywords::

• Cost analysis
• Overactive bladder
• Urgency urinary incontinence
• Fesoterodine

Introduction

Overactive bladder (OAB) is a symptom-driven disease defined by the International Continence Society as urinary urgency, with or without urgency urinary incontinence (UUI), usually accompanied by increased daytime urinary frequency and nocturia1. OAB is a chronic condition adversely affecting a patient’s quality-of-life and social well-being. OAB is associated with higher risk of co-morbidities such as falls/fractures, urinary tract infections (UTI), depression, and nursing home admissions2. The reported prevalence of UUI, a common symptom of OAB, ranged from 2.6–14.2% in men and 8.9–36.3% in women in the US, with prevalence increasing with age3. The future prevalence of OAB is expected to increase in the US as the population of elderly individuals continuously grows.

The main goal of OAB treatment is to manage symptoms and OAB-related co-morbidities. First-line treatments are behavioral therapies such as bladder training, bladder control strategies, pelvic floor muscle training, or fluid management4. Anti-muscarinic agents are the primary pharmacologic treatment for OAB, as recommended by the clinical guideline from American Urological Association (AUA)4. Clinical trials demonstrate that pharmacologic treatments for OAB are generally effective and safe in elderly patients5–9. However, the frail or vulnerable elderly population may pose a particular challenge for OAB management. In these patients, OAB can be difficult to manage due to their higher likelihood of impaired mobility, increased cognitive deficiencies, declining physical function, and more complex comorbidities10^,11.

The negative impact and economic burden of OAB was estimated to be $1925 per capita in 20071 2. The estimated total cost of OAB with UUI for adults ≥25 years in the US was $65.9 billion in 2007, and projected to be $76.2 billion in 2015 and $82.6 billion in 20203^,12. Data describing the economic burden of OAB, specific to the vulnerable elderly population, are rare. One recent study showed that US vulnerable elderly with OAB had significantly higher health resource utilization and costs compared with those without OAB, with a per-capita annual incremental cost of $7188 in 20131 3. Although limited, the published evidence suggests that the economic impact of OAB in vulnerable elderly is more substantial than in the general OAB population.

Previous economic analyses have demonstrated the relative value of treating OAB with pharmacotherapy in the general population2^,14^,15, but such data are lacking for vulnerable elderly. The objective of this study was to assess the economic value of treating vulnerable elderly patients with OAB with fesoterodine compared to no OAB pharmacotherapy from the US payer perspective.

Methods

Model design

A decision-analytic cost model was developed to estimate the 52-week costs of a cohort of vulnerable elderly with OAB initiating treatment with fesoterodine or managing the condition without OAB pharmacologic therapy (Figure 1). The vulnerable elderly OAB population was defined as in a recent clinical trial8: patients ≥65 years with self-reported UUI symptoms for ≥3 months and 2–15 UUI episodes/day, and at risk of deteriorating health by a score of ≥3 on the Vulnerable Elders Survey-13. The model represented typical clinical treatment patterns of OAB patients receiving anti-muscarinic therapy where they were evaluated for treatment response and persistence at weeks 12, 26, and 52. The model included a hypothetical health plan with 100,000 elderly members. The percentages of vulnerable elderly and annual prevalence of OAB among vulnerable elderly were calculated from an analysis of the Medicare Current Beneficiary Survey (MCBS) from 2001 to 201016. In the MCBS, 52.4% of the participants ≥65 years were vulnerable elderly (data on file), and the annual prevalence of OAB among the vulnerable elderly was 13.5%11. In the hypothetical cohort of 100,000 elderly patients, this results in a total of 7096 vulnerable elderly subjects as the model target population.

Figure 1. Decision analytic model design diagram.

Treatment efficacy

Treatment responders were defined as no UUI episodes based on self-reported 3-day bladder diaries after 12 weeks of treatment. Responders were assumed to maintain treatment efficacy until discontinuation or the end of the model horizon. This assumption was supported by findings from previous long-term trials of fesoterodine and other anti-muscarinic drugs17–19. Patients not responding to treatment at week 12 were assumed to permanently discontinue. Switching to a different OAB medication was not considered within this model.

The proportion of responders at week 12 (50.8%, Table 1) was taken from a randomized, double-blind, placebo-controlled trial of vulnerable elderly subjects with UUI8. In this trial, patients were randomized to fesoterodine received 4 mg once daily for 4 weeks with a possible dose increase to 8 mg once daily.

Table 1. Model inputs.

	Base-case value^a
Treatment efficacy, %
Fesoterodine responders at week 128	50.8
Remission over 2 years among  untreated patients^b20	2.0
Patients remaining on treatment  at weeks 12, 26, and 5221	92.4, 58.6, 37.1
Model costs,^c $
Fesoterodine, per day22	7.27
Inpatient hospitalization13
Responder	4576
Non-responder	6228
Outpatient hospitalization13
Responder	1321
Non-responder	2406
Physician office visits13
Responder/Remission	2991
Non-responder13^,23	7333
Non-remission	5518
Falls/fracture, per event30	13,563
UTI, per event31	1225
Depression, per event32	19,915
Nursing facilities, per 12 weeks25	9878

^aUnivariate sensitivity analysis varied model inputs by ±25% unless otherwise specified.

^bAlternative values analyzed within scenario analyses were: 0%, 10.3%, and 36%.

^cCosts presented in 2013 US dollars.

Spontaneous remission has been shown to occur in OAB patients who are not on treatment20. We incorporated this phenomenon into the model as a proportion of untreated and treatment-discontinued patients assumed to have spontaneous remission from OAB symptoms (i.e., no UUI episodes), and was maintained until week 52. Scenario analyses were also conducted with alternative yearly remission probabilities. Concomitant behavioral therapies are commonly used instead of OAB medications and may contribute to spontaneous remission without treatment. Behavioral modifications were assumed to occur in both treatment arms, and thus were not explicitly modeled.

Treatment persistence

Treatment discontinuation was estimated from a published retrospective study of three primary healthcare centers in Spain including OAB patients aged ≥18 years, which reports the proportion of patients continuing fesoterodine therapy to be 92.4%, 58.6%, and 37.1% at weeks 12, 26, and 5221. Although not US-specific, these discontinuation data were similar to the values obtained with the 12-week discontinuation probability observed in the fesoterodine clinical trial8 extrapolated to 52 weeks assuming an exponential survival distribution. The current analysis did not model specific reasons for discontinuation, but multiple factors, such as adverse events, lack of efficacy, loss of follow-up, etc. were expected to impact treatment persistence. Patients discontinuing fesoterodine treatment were assumed to return to untreated status because a switch to a different OAB medication was not considered.

Model costs

This analysis was conducted from the US payer perspective; therefore, only direct medical costs were considered. Direct medical costs consisted of fesoterodine costs, costs of healthcare resource utilization (inpatient hospitalization, ER visits, outpatient hospitalization, and physician office visits), and costs of OAB-related comorbidities (falls/fractures, urinary tract infections, depression, and nursing home admissions). All costs were inflated to 2013 US$ using the medical care component of the consumer price index.

Drug costs

The average wholesale price of fesoterodine included daily drug cost of $7.27 for 4 mg and 8 mg dosages22. Although 52.7% of patients titrated to 8 mg within the fesoterodine clinical trial, drug prices after week 4 remained the same, due to pricing parity between the two doses.

Healthcare resource use (HCRU)

The annual costs of inpatient hospitalization, outpatient visits, and physician office visits were also estimated from the MCBS13. Costs of ER visits were not separately assessed and were included in either inpatient hospitalization or outpatient visits costs, depending on where the services were provided. In our model, responders as well as untreated or discontinued patients with spontaneous remission were assumed not to have UUI episodes and were, therefore, considered to be similar to the vulnerable elderly population without OAB. Consequently, the HCRU costs of the responders and those with remission were assigned the HCRU costs of the vulnerable elderly without OAB. Conversely, costs of non-responders and untreated patients without remission were assumed equal to those with OAB because they were assumed to continue to experience OAB symptoms.

It has been reported that UUI patients treated with pharmacotherapy have a significantly higher frequency of physician visits (rate ratio = 1.329, p = 0.0001) than those without treatment23. To take this difference into consideration, the costs of physician office visits for treated (non-responders) were 1.329-times that of untreated patients who did not have remission. The difference in the frequency of physician visits between those treated and not treated with pharmacotherapy was only observed among the UUI patients. Thus, non-UUI patients, including those who were untreated with remission as well as treatment responders, were assumed to have an equal frequency of physician visits and were assigned equal costs.

OAB-related comorbidities costs

The costs of OAB-related comorbidities were obtained from published literature (Table 2). The annual probabilities of experiencing an OAB-related comorbidity were obtained from the MCBS database11, with the same assumptions that the probabilities of treatment responders and untreated patients with spontaneous remission were equal to those of non-OAB patients and probabilities of treatment non-responders and non-remitters were equal to those of OAB patients. UTIs may occur more than once per year and the number of events per patient were obtained from a previously published cost study24. The unit cost for nursing home was estimated for the first 12 weeks, which was estimated as the difference in average Medicare expenditures between long-term care residents and Medicare beneficiaries who never lived in a long-term care facility25.

Table 2. OAB-related comorbidities.

OAB-related comorbidities	Treatment responders		Treatment non-responders		No treatment, OAB remission2		No treatment, without OAB remission3
	Annual probability	No. per patient	Annual probability	No. per patient	Annual probability	No. per patient	Annual probability	No. per patient
Falls/fracture11	42.3%	–	55.4%	–	42.3%	–	55.4%	–
Urinary tract infection11^,24	16.3%	1.58	45.7%	2.28	16.3%	1.58	45.7%	2.28
Depression11	11.7%	–	23.9%	–	11.7%	–	23.9%	–
Nursing home33	2.9%	–	7.5%	–	2.9%	–	7.5%	–

Vulnerable elderly may have decreased cognitive function, but there is conflicting published evidence on whether anti-muscarinic treatments exacerbate this condition26^,27. Due to the lack of prevalence and cost data associated with cognitive impairment in the OAB population and the uncertainty surrounding the impact of treatment, this factor was not considered within the model.

Sensitivity and scenario analyses

In the base case scenario, the proportion of patients with spontaneous OAB remission was based on a published study conducted in the US, reporting 2% of women aged 54–79 with ≥1 urinary incontinence episode per week had complete OAB remission within 2 years. This study also reported 10.3% of patients with any urine leakage/month had spontaneous remission over 2 years, which was used as an alternative value. Additionally, we assessed probabilities of 0% and 36% (the placebo response observed in the clinical trial of vulnerable elderly) as alternative scenarios.

Univariate one-way sensitivity analysis was also performed to assess how variation of individual inputs affected the model results. Parameters included in the univariate sensitivity analysis were varied by ±25% of the base case estimates.

Results

Within the cohort of 7096 vulnerable elderly patients with OAB, total medical costs were estimated to be $219,847,373 in patients without OAB pharmacotherapy and $208,383,392 in fesoterodine-treated patients over 52 weeks (Table 3). This results in a savings of $11,463,981 for the entire cohort, or $1616 per patient. Fesoterodine-treated patients incurred $7.8 million in drug costs, but were associated with a reduction of $7.2 million in HCRU costs and a reduction of 12 million in OAB-related comorbidity costs.

Table 3. Estimated yearly medical costs in vulnerable elderly OAB patients (n = 7096) in 2013 US$.

Cost component	Fesoterodine	No treatment	Cost difference
Total anti-muscarinic drug, $	7,832,919	0	7,832,919
Total HCRU, $	92,889,606	100,162,050	−7,272,445
Inpatient hospitalization	41,407,439	44,111,891	−2,704,452
Outpatient visits	15,242,498	17,019,337	−1,776,839
Physician office visits	36,239,668	39,030,822	−2,791,154
Total OAB-related comorbidities, $	107,660,868	119,685,323	−12,024,455
Falls/fracture	62,308,176	66,983,236	−4,675,059
Urinary Tract Infection	8,759,786	10,763,598	−2,003,812
Depression	31,982,765	36,551,387	−4,568,622
Nursing Homes	4,610,141	5,387,103	−776,962
Total medical costs, $	208,383,392	219,847,373	−11,463,981
Total medical cost per patient, $	29,367	30,982	−1616

HCRU, healthcare resource use; OAB, overactive bladder.

With zero probability of spontaneous remission, the cost savings of fesoterodine treatment increased to $1643 per patient. Assuming probabilities of 10.3% and 36%, the cost savings were reduced to $1500 and $664. Univariate sensitivity analysis of other variables indicated that the incremental cost savings per patient were most sensitive to changes in fesoterodine efficacy, followed by costs of inpatient hospitalization and physician office visits (Figure 2). Within ±25% ranges of each tested parameter, the fesoterodine-treated arm remained cost-saving compared to the no medication treatment arm. Given the reduced medical costs associated with treatment, our model demonstrates the price of fesoterodine would need to be $17.91 to be cost neutral.

Figure 2. Univariate sensitivity analysis on varying input parameters by ±25% centered on the base case estimates. Vertical dotted lines represent ±25% change from the base case result.

Discussion

Overactive bladder is a chronic and age-related disorder, imposing a substantial economic burden in the US. OAB is particularly impactful in vulnerable elderly, as this population has increased cognitive deficiencies, declining physical function, and more complex comorbidities7. This analysis found that overall medical costs from the US payer perspective were lower when a patient was treated with fesoterodine rather than no OAB medication at all. When treated, fesoterodine drug costs accounted for only 3.8% of total medical costs among the vulnerable elderly as the target population. Further, the additional drug costs were offset by decreases in costs for HCRU and OAB-related comorbidities. Fesoterodine remained cost-saving among all alternative scenario analyses and sensitivity analyses varying a number of data inputs tested, supporting the robustness of these findings. As would be expected, the model results were most sensitive to the fesoterodine efficacy assumption. However, fesoterodine was still associated with a lower cost vs no OAB pharmacotherapy when its efficacy was reduced by 25%.

The presence or absence of the UUI symptom was used as the sole definition of treatment response. Although clinical trial data specific to the vulnerable elderly were available to determine model efficacy, corresponding cost data were not. We, thus, imputed cost data for UUI responders and non-responders with MCBS data of non-OAB and general OAB patients, respectively. Since UUI is a common and bothersome among the patients with OAB with UUI sub-population28^,29, our use of cost data from general OAB patients (with and without UUI) may be a conservative estimate of costs, as the presence of urge incontinence has been shown to directly correlate with higher incidence of falls/fractures and depression10.

It is also important to note that reduction (not necessarily eradication) of UUI or other OAB symptoms may also be considered by patients and physicians as a satisfactory treatment response, and that some patients may be undiagnosed or elect not to receive treatment for their OAB. Sufficient data to relate costs to other OAB symptoms and costs associated with increased diagnosis and treatment awareness were beyond the scope of this analysis. Nevertheless, these factors are important for understanding patient satisfaction and determining accurate estimates of budget impact from the payer perspective.

Our model includes real-world persistence data from a general adult OAB population, who may remain on treatment longer than the vulnerable elderly population. Further, once patients do discontinue fesoterodine, we do not include an option to switch to another therapy or augment fesoterodine with additional drugs. Medication switch or addition is common for OAB management and recommended by clinical guidelines4, but lack of clinical data prohibited consideration of these scenarios.

This analysis was conducted to estimate the economic costs of treatment with fesoterodine, but the results observed should not be interpreted to be unique to fesoterodine. Other available OAB medications may also produce overall cost savings, depending on their efficacy and price. Currently, clinical trial evidence in the vulnerable elderly population, which is necessary for economic analysis of other treatments, is lacking.

Important long-term treatment outcomes and cost considerations beyond the payer perspectives were not included within the model. These include any expected improvements in patients’ quality-of-life and social well-being by alleviation of OAB symptoms, economic values of reduction in long-term mortality7, and the costs of incontinence containment products that are not covered by the payer in the most-common US healthcare systems.

A final limitation is that our analysis only modeled a 1-year time horizon. Some of the costs of OAB-related comorbidities could be considered long-term benefits, such as depression and nursing-home admissions, that may not be manifested within the first year of treatment. Excluding the cost savings from these two comorbidities nevertheless results in overall savings of $862 per patient for the fesoterodine group (Table 3).

Conclusion

Results from our model suggest that treating vulnerable elderly OAB patients with fesoterodine can be considered a cost-saving option compared with no OAB pharmacotherapy, due to the cost offsets from reduced HCRU and avoided OAB-related co-morbid conditions. Clinical trial evidence for other OAB medications in the vulnerable elderly population is warranted. This model can be updated to incorporate additional treatment options when these data become available.

Transparency

Declaration of funding

This study was funded by Pfizer Inc, the manufacturer of fesoterodine.

Declaration of financial/other relationships

XL and KHZ are employees and shareholders of Pfizer Inc. LQ and SJS are employees of Pharmerit International, who were paid consultants to Pfizer in connection with the development of this study. JME peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Acknowledgments

The authors thank Mei Xue and Beth Lesher of Pharmerit International for their editorial support for this manuscript.