Work Productivity Associated With Treated Versus Never-Treated Overactive Bladder Symptoms

Overactive bladder (OAB) has been defined by the International Continence Society as urinary urgency, with or without urgency urinary incontinence, usually associated with increased daytime frequency and nocturia, in the absence of urinary tract infection or other obvious pathology.^1,2 In turn, urgency has been defined as a sudden, compelling desire to pass urine without delay.¹ The overall prevalence of OAB in adults 18 years or older has been estimated at 16% to 17% among men and women in the United States, with prevalence increasing with advancing age.³ The effect of age can be seen in the higher prevalence reported among 20,000 US adults 40 years or older who participated in the Epidemiology of Lower Urinary Tract Symptoms (Epi- LUTS) study, in which OAB symptoms were reported at least “sometimes” by 27.2% of men and 43.1% of women.⁴ By 2018, it is projected that 546 million adults 20 years or older throughout the world will be affected by OAB.⁵

Current treatment rates for OAB are low,⁶ which is a concern because OAB symptoms can have a negative impact on health-related quality of life, including physical, psychological, vocational, and social functioning, and the performance of routine daily activities.^3,7,8 The functional impairment associated with OAB contributes to its economic burden.^9,10 Although effective treatments (ie, antimuscarinic agents, behavioral therapies) are available to manage OAB symptoms, ^11,12 many individuals do not discuss their symptoms with a healthcare provider and therefore are not prescribed treatment for OAB.^6,13

The present study evaluated OAB-related impairment in work productivity and daily activity among treated and never-treated US subjects with OAB symptoms using data from a National Health and Wellness Survey (NHWS) subpopulation. From these analyses, annual indirect costs associated with OAB-related work productivity impairment were estimated and factors that predicted impairment in work productivity and daily activity were identified.

METHODSStudy Design and Subjects

The NHWS was a self-administered, Internet-based questionnaire given in 2009 to a nationwide sample of 75,000 adults in the United States. The NHWS respondents were recruited via stratifi ed random sampling on the basis of demographics reported by the US Census Bureau to represent the total US adult population. A total of 24,866 respondents from the 2009 NHWS data set were screened for OAB symptoms with the OAB Awareness Tool¹⁴ and prespecifi ed inclusion and exclusion criteria (listed below); 2887 respondents reported a history of OAB symptoms and qualifi ed for a follow-up OAB survey; 2750 qualifi ed respondents completed the follow-up OAB survey.

For inclusion in the survey, subjects had to be 18 years or older, able to read and write English, and willing to provide personal health information. The presence of OAB symptoms was documented by current use of medication to control OAB symptoms and by scores higher than 14 (for men) or higher than 16 (for women) on the 8-item OAB Awareness Tool (score range, 0 to 40),¹⁴ with screening scores of 8 or higher indicating possible OAB. Exclusion criteria included current pregnancy, selfreported hematuria or pink-tinged urine, pain or burning sensation on urination due to urinary tract infection, use of a catheter, diagnosis of benign prostatic hyperplasia (BPH) or prostate cancer, or current use of medication for BPH (eg, tamsulosin, dutasteride).

The 2750 qualified respondents who completed the follow-up OAB survey were classifi ed into 3 groups: 549 (20%) who were currently taking a prescription OAB medication (OAB-treated); 622 (23%) who had never taken a prescription OAB medication, although they may have been receiving nonpharmacologic treatment, such as behavioral modification (OAB never-treated); and 1579 (57%) who had ceased taking prescription OAB medication (OAB past treatment). The never-treated subjects reported that they believed that their bladder control condition would probably get worse over time and require treatment with medications. Subjects who were not currently being treated but had been treated at some time in the past were excluded from further analyses to avoid any confounding effects associated with differences in the reasons for discontinuing previous treatment. Of 1171 OAB-treated and never-treated subjects, 476 (41%) reported that they were employed (full-time, part-time, or self-employment).

OAB Survey Assessments

The Work Productivity and Activity Impairment (WPAI) questionnaire was used to assess impaired work productivity and daily activity function due to OAB symptoms over the past 7 days.¹⁵ Employed subjects indicated the number of hours they missed from work due to OAB symptoms, the number of hours they missed for any other reason, and the total number of hours they worked. The degree of impairment in work productivity or daily activities is rated on a scale on which 0 equals no impairment and 10 equals total impairment. From these data, 4 subscales (ie, absenteeism, presenteeism, percentage of overall work impairment, and percentage of overall activity impairment) were generated. Absenteeism refers to the percentage of hours missed at work in the previous week. Presenteeism refers to the percentage of impairment in productivity during the hours worked. Overall work impairment is the total impact of absenteeism and presenteeism. Activity impairment is the percentage of impairment during nonwork daily activities. Only subjects who were currently employed were included in the assessments of absenteeism, presenteeism, and overall work impairment, whereas all subjects were included in assessments of daily activity.

Subjects provided socioeconomic and demographic information and completed items following up on symptoms identifi ed in the OAB Awareness Tool. Overactive bladder symptoms experienced in the past month, regardless of whether they were rated as bothersome on the OAB Awareness Tool, were summed (symptom count).¹⁴ Subjects also rated the degree to which bothersome symptoms changed in severity over the past year (scores from 1 for a substantial decrease to 7 for a substantial increase), and symptom scores were averaged to calculate the overall change (symptom bother).¹⁶ Generic health-related quality of life was assessed with the 12-item Short-Form Health Survey (SF-12, version 2) with Physical Component Summary and Mental Component Summary scores, each ranging from 0 (poorest state of health) to 100 (highest state of health), normalized to the US population (mean [standard deviation] score = 50 [10]).¹⁷ Responses to the SF-12 v2 were used to generate the Short-Form Six-Dimension Health Utility score, a preference-based single index of health status (physical functioning, role participation, social functioning, pain, mental health, and vitality) using general population values; scores range from 0 to 1, with 1 representing optimal health.¹⁸

Finally, the Charlson Comorbidity Index assessed the 10-year risk of mortality associated with the presence of any of 19 conditions, each of which is assigned a numeric score (1, 2, 3, or 6; higher scores indicate more lethal conditions), with the index score equaling the sum of the scores.¹⁹

Statistical Analysis

As the NHWS is stratified by sex, age, and race/ethnicity, all bivariate results (comparing OAB-treated and never-treated populations) were weighted to reflect adult population values from the US Census. Demographic and socioeconomic characteristics, quality of life, and assessments of work productivity and daily activity impairment due to OAB symptoms were compared between OAB-treated and never-treated subjects using either a 2-sample t test for continuous data or a _χ² test for categorical data (significance level, 2-sided P <.05).

Indirect costs associated with work productivity impairment were calculated using the US Department of Labor’s 2009 Bureau of Labor Statistics average wages (2009 US dollars), with adjustments for sex and age. The adjusted wages were multiplied by the percentage of impairment in work productivity and then annualized to provide an estimate of the projected annual per capita costs associated with lost productivity due to OAB.²⁰ Annual costs related to work productivity impairment due to OAB symptoms were compared between OAB-treated and never-treated subjects, using 2-sample t tests.

Generalized linear models were used in multivariate analyses to predict impairment in work productivity and daily activity for OAB-treated versus never-treated subjects, with adjustments for the following covariates: age, sex, ethnicity, employment, marital status, household income, educational status, health insurance status, Charlson Comorbidity Index score, and the number of OAB symptoms (symptom count) and symptom bother. Work productivity and daily activity impairment were continuous variables (percent impairment ranging from 0% to 100%) but were often highly skewed. Therefore, the generalized linear models specifi ed a negative binomial distribution to provide the best fit to the data. Additional corrections to the standard errors were implemented automatically to adjust for model underdispersion. Given that the traditional use of negative binomial models is for count distributions and the WPAI questionnaire yields a summary score, we also applied a model used for continuous outcomes as a sensitivity analysis in which the 4 WPAI metrics (scores) were modeled as continuous variables using linear regression analysis. All statistical analyses were performed by using SAS 9.1 software (SAS Institute Inc, Cary, North Carolina).

RESULTSSubject Characteristics

Two-sample bivariate analyses of respondent characteristics showed several statistically significant differences between OAB-treated and never-treated subjects. Compared with never-treated subjects, OAB-treated subjects typically were older, more likely to be white and less likely to be Hispanic, more likely to have health insurance, less likely to be married or living with a partner, and less likely to be employed (

Table 1

).

Daily Activity and Work Productivity

In multivariate analyses, younger age, male sex, unemployment, nonsingle marital status, lack of OAB treatment, increased OAB symptom count, and increased OAB symptom bother were statistically signifi cant predictors of daily activity impairment (all P <.03;

Table 2)

.

Among 476 employed respondents, 193 (41%) were OAB-treated and 283 (59%) were never treated. Bivariate analyses of respondent characteristics showed some statistically significant differences between OAB-treated versus never-treated subjects. Compared with nevertreated subjects, OAB-treated subjects had less OAB-related impairment at work and in daily activities, and fewer days absent from work or present at work but with productivity reduced 50% or more over the previous 3 months, but a smaller total number of hours worked in the previous week (Table 3). In multivariate analyses (

Table 4

), various factors were identified as signifi cant predictors of absenteeism (younger age, male sex, and black ethnicity; all P <.04) or of presenteeism and overall work productivity impairment (ie, OAB-untreated, younger age, male sex, increased OAB symptom count, and increased OAB symptom bother; all P <.01). Goodness-of-fit data indicated that the model reflected the WPAI outcome data (Table 4). The pattern of effects and their statistical significance levels in the linear regression sensitivity analysis were similar to those in the negative binomial model.

The estimated total annual costs (2009 dollars) associated with OAB-related absenteeism and impaired work productivity were $9670 for OAB-treated subjects versus $17,477 for never-treated subjects; this statistically significant difference was largely due to impairment while at work (ie, presenteeism) rather than to absenteeism from work (

Table 5

).

DISCUSSION

This study screened respondents from the 2009 NHWS to identify subjects with OAB symptoms for a follow-up OAB survey used to assess the impact of OAB on daily activity and work productivity. A clinically meaningful threshold for work productivity impairment has not been established. Our data indicated a total work impairment rate of 33.6% for treated subjects and 49.8% for never-treated subjects with OAB symptoms (P <.001). Data from 4 trials of subjects with Crohn’s disease indicated total work productivity impairment rates of 49% to 71% across the trials at baseline (pre-treatment).²¹ A study of work productivity in subjects with osteoarthritis reported total impairment rates of 21% to 47% for mild to severe disease severity.²² These previously reported results in other chronic diseases provide an external benchmark for total work productivity impairment.

The focus of our study was on identifying factors associated with daily activity and work productivity impairment in OAB-treated and never-treated subjects and estimating the annual indirect costs of work productivity impairment. A lack of OAB prescription treatment could reflect subjects’ reluctance to seek care for OAB symptoms, lack of access to care, or both. Impairments in daily activity and work productivity were associated with lack of OAB prescription treatment, younger age, male sex, and as would be expected, increased OAB symptom count and symptom bother. The estimated annual costs of impaired work productivity were nearly twice as high in never-treated subjects ($17,477) as in OAB-treated subjects ($9670).

The present findings are consistent with those of previous studies that assessed the negative impact of OAB symptoms on daily activities and work productivity. A 2004 report concluded that urinary incontinence in women adversely affects their involvement in routine activities related to house cleaning and shopping, attendance at religious services, social life, employment, and personal hygiene.²³ Similarly, a large-scale, populationbased survey in 5 European countries found that 32% of subjects with OAB symptoms reported symptom-related depression and 28% reported stress; depression and stress were significantly worse when OAB was associated with incontinence, which had a significant negative impact on social situations and work.²⁴ An evaluation of the impact of lower urinary tract symptoms (LUTS), including OAB, on work productivity among 5696 employed US men and women in the EpiLUTS study demonstrated that all types of LUTS interfered with work productivity, with the greatest degree of interference with work productivity reported by subjects with incontinent OAB.⁸ In another survey of work-related costs attributable to OAB that used an administrative database of 1.2 million beneficiaries, including those with medical and disability claims from 1999 to 2002 and matched controls, employees with OAB had signifi cantly (P <.01) higher rates of work loss (2.2 additional days due to medically related absenteeism and 3.4 additional days because of disability) than employees without OAB.²⁵

The economic burden of a disease is the total cost of all resources used or lost by individuals and society as a result of the disease. The total cost is derived from direct costs (costs of diagnosis, treatment, and patient care), indirect costs (lost productivity to society and lost income to patient and caregivers), and intangible costs (pain, suffering, and decreased quality of life). Estimates of the indirect costs associated with OAB can vary widely depending on the study design and methodology, and the population surveyed. The present analysis included an estimation of the mean annual per capita indirect costs due to OAB symptoms of employed US subjects, based on subject data from the comprehensive, validated WPAI questionnaire, which assesses both absenteeism and presenteeism.

Previous studies of the economic burden of OAB in the United States generally have focused on the total per capita costs of OAB. A comprehensive cost analysis based on age- and sex-specific OAB prevalence rates, practice guidelines, Medicare and managed care fee schedules, and published data estimated that the average total annual per capita cost of OAB in 2007 was $1925 for communitybased and institutionalized adults, which included $1433 for direct medical costs and $426 for indirect (absenteeism only) costs; the authors noted that the total annual per capita cost was substantially higher than previous estimates.⁹ For example, a 2003 report from the National Overactive Bladder Evaluation program estimated the annual costs per community-based individual with OAB in the United States at only $267,¹⁰ and a 2009 analysis estimated total annual per capita costs in 2007 at $590 for community-based US adults with frequent symptoms and $561 for those with occasional symptoms.²⁶ Although these estimated total per capita costs are lower than the annual per capita indirect costs we report here ($17,477 for never- treated subjects and $9670 for OAB-treated subjects), these previous analyses were based on the overall OAB population in the United States, whereas our cost analysis included only employed subjects with OAB symptoms and assessed the costs of both absenteeism and presenteeism.

Limitations

Potential limitations of the present study are that subject classifi cation was based on self-reported medication use and OAB symptoms, work productivity assessments were based on subject recall, and men with BPH and subjects taking medication for BPH were excluded from the analyses. Many men with LUTS are diagnosed with and treated for BPH, with the clinical distinction between LUTS suggestive of BPH and OAB symptoms not always clearly established. Additionally, we did not collect data on reasons for treatment discontinuation among respondents who had previously received treatment for OAB symptoms but were not currently receiving treatment. Previous research suggests respondents may have ceased taking medication for a variety of reasons, including poor efficacy and/or tolerability, a general dislike of taking chronic medication, cost, their symptoms resolved or stopped being bothersome, and so forth.²⁷ The effect of OAB symptoms on work productivity may have differed among respondents who discontinued for different reasons; therefore, previously treated respondents were excluded from this analysis. Future research that includes men with OAB symptoms and BPH and subjects who previously received treatment for OAB symptoms (stratified by reason for discontinuation) may provide additional information on work productivity and daily activity impairment.

CONCLUSIONS

Overactive bladder symptoms can cause impairment in daily activity and work productivity, especially among younger subjects who remain untreated. The annual costs of OAB-related work productivity impairment are nearly twice as high for untreated versus treated subjects. The present findings support the need to change the perception of OAB from a lifestyle issue to a highly bothersome and costly medical condition, not only for the benefit of patients with OAB but also because improved awareness, diagnosis, and treatment of OAB may reduce its negative impact on daily activity and work productivity and thereby reduce its economic burden on society.