The cost of clinically significant urinary storage symptoms for community dwelling adults in the UK

Authors


D. Turner, Department of Health Sciences, University of Leicester, 22–28 Princess Road West, Leicester, UK.
e-mail: dat3@leicester.ac.uk

Abstract

OBJECTIVES

To estimate the cost of clinically significant urinary storage symptoms (CSUSS), including costs borne by the National Health Service (NHS) and individuals, in terms of the use of goods and services in community-dwelling adults.

SUBJECTS AND METHODS

The subjects were community-dwelling adults aged ≥ 40 years and living in Leicestershire. The prevalence of CSUSS was estimated using a postal questionnaire with a randomly selected sample of 23 182 respondents. The costs associated with CSUSS were estimated using home interviews with 613 cases with and 523 subjects without CSUSS. Cases were defined on the basis of urinary symptoms of leakage, urgency, frequency and nocturia. Willingness-to-pay was used to measure intangible costs as an indicator of the value of alleviating symptoms.

RESULTS

The estimated total annual cost to the NHS for treating CSUSS cases in community-dwelling adults was £536 million at 1999/2000 prices (£303 million and £233 million for men and women, respectively). The total value of costs borne by individuals was estimated to be £207 million (£29 million and £178 million for men and women, respectively). This gives total annual costs related to the use of services of £743 million. There were large intangible costs borne by individuals estimated to be £669 million (£301 and £368 million for men and women) for the UK in terms of willingness-to-pay.

CONCLUSIONS

The costs of CSUSS in the community amounted to ≈ 1.1% of overall NHS spending for 1999/2000. Personally borne and intangible costs are also large and important components of the costs of CSUSS. There are large gender differences in the proportion of costs borne by the NHS, i.e. 91% of male and 57% of female costs.

Abbreviations
NHS

National Health Service

MRC

Medical Research Council

CSUSS

clinically significant urinary storage symptoms

ICS

International Continence Society

WTP

willingness-to-pay.

INTRODUCTION

About 30% of community dwelling individuals aged ≥ 40 years have some form of significant urinary incontinence or related storage symptoms, and this prevalence is higher for women and increases with age [1,2]. This condition has implications for the UK National Health Service (NHS) in terms of use of services; it also has cost implications for individuals, in that they may need to purchase services and consumables themselves. In addition to these costs there may also be substantial costs associated with urinary symptoms that are related to changes in lifestyle and negative feelings (e.g. embarrassment) resulting from the condition. Although these may not necessarily be accompanied by resource implications it is likely that alleviation is of substantial value to individuals. These costs will be referred to as ‘intangible costs’; their existence means that a considerable burden associated with urinary symptoms may be unrecognized by healthcare decision makers.

Several studies have quantified the costs associated with urinary incontinence; one recent UK study estimated the annual cost to be £353.6 million for the UK (1998 UK pounds) [3]. This study used evidence from national datasets combined with assumptions about how much of the work of particular staff types would be concerned with incontinence; e.g. 10% of all district nurse time was assumed to be devoted to urinary incontinence. However, this study only focused on NHS costs and did not include costs borne by individuals. There have also been UK estimates published that focus on particular aspects of costs, such as equipment, pads and appliances [4,5] or the resources used by a sample of individuals [6,7].

Evidence from the USA suggested that costs associated with urinary incontinence can be substantial, e.g. two studies estimated the annual cost of urinary incontinence in the USA for 1995 to be US $26.3 billion [8] and US $16.3 billion [9] (these values equate to £16.5 billion and £10.2 billion using an exchange rate of 1.59 dollars to 1 UK pound sterling). Both studies used estimates of the prevalence of urinary incontinence combined with evidence from the literature and national datasets on the costs of urinary incontinence. However, they had a USA perspective so may not be appropriate to UK-based individuals because of differences in the healthcare system between the countries. Also, none of these studies addressed the issue of intangible costs and all measured the costs associated with urinary incontinence; there may be costs associated with other urinary storage symptoms in addition to those associated with the symptom of leakage alone.

The present study aimed to estimate the cost of illness, to both the NHS and individuals, associated with clinically significant urinary storage symptoms (CSUSS) in community-dwelling adults aged ≥ 40 years, and to estimate costs for the UK as a whole. A further aim was to investigate the contribution of intangible costs. Finally, for comparison, we aimed to identify all costs associated with urinary symptoms for individuals who were not classified as having CSUSS.

SUBJECTS AND METHODS

This work was carried out as part of the Medical Research Council (MRC)-funded incontinence study based in Leicestershire between 1995 and 2002 [2]. The approach used was to estimate the annual cost of a CSUSS case from interview data. The number of cases in the UK was estimated by multiplying estimates of the prevalence by population estimates for the UK. The total cost of CSUSS for the UK was derived by multiplying the average cost of a case by the estimated number of cases. Several different types of cost are used in this report; the first is associated with the use of goods and services, e.g. visits to practitioners or purchasing padding. This type of cost applies to NHS and personal costs. The second type of cost used is the value of time spent by respondents in travelling and receiving care, and the third type measured were intangible costs. All costs identified here are those that arise because of the urinary symptoms of the individuals surveyed.

DEFINITION OF CASES AND SELECTION FOR HOME INTERVIEW

Detailed information on resource use was collected in a home-interview study of cases and non-cases identified from a postal survey. Participants in the postal survey were women and men aged ≥ 40 years who were living in the community. The questionnaire was mailed to 4996 individuals and was completed by 3015 responders (response rate 60%). The questionnaire included several validated questions on urinary symptoms and the results were used to define individuals as cases or non-cases. The definition of urinary symptoms conforms to the standards recommended by the International Continence Society (ICS) [10] except where specifically noted (Table 1). Cases were defined as above using clinically significant severity thresholds based on earlier pilot work in relation to the uptake of a nurse-led service [11]. This included reporting of one or more of the symptoms of leakage of urine (monthly or more), urgency (monthly or more), frequency (hourly or more), or nocturia (three times or more). Non-cases had no incontinence, urgency, frequency, or nocturia above these thresholds. Invitations to home interview were sent to all 961 cases identified in the postal survey and detailed interviews were carried out on 612 cases (response rate 64%). A random sample of 933 non-cases was also invited for interview, and 524 of these completed the home interview (response rate 56%). Sufficient resource-use information was available to estimate annual costs for 602 (98.2%) cases (233 men and 369 women) and 510 (95.7%) non-cases (250 men and 260 women). Most of these interviews were carried out between November 1999 and April 2000.

Table 1.  Definitions and thresholds used for storage symptoms and caseness
ICS storage symptomQuestion used in present studyResponse
  • *

    This was consistent with ICS definition of incontinence in 1997 [24] operative at the time of sampling.

Increased daytime frequencyDuring the daytime how often do you usually go to the toilet to pass urinePositive: frequency every hour or more
Negative: 2–4 hourly
NocturiaHow often do you usually feel the need to get up at night to pass urine?Positive: ≥ 3 times
UrgencyHow often is the urge to pass urine very strong, making it difficult to delay urination?Positive: Daily
Urinary incontinenceDo you ever leak urine when don’t mean to (i.e. anything from a few drops to a flood during the day or night)?Positive: several times a month or more
Stress urinary incontinenceDoes any urine leak when you laugh, cough or exercise?Positive: several times a month or more
Urge urinary incontinenceDo you have such a strong desire to pass urine that you leak before reaching the toilet?Positive: several times a month or more
Other urinary incontinenceDo you leak urine when awake without feeling it happen or realizing it?Positive: several times a month or more
* Do you leak urine when you are asleep?Positive: several times a month or more
Do you leak urine because you have difficulties going to, or getting on or off a toilet or commode?Positive: several times a month or more
* Do you dribble or leak urine immediately after you feel you have finished urinating?Positive: several times a month or more
Bladder sensationWhen you need to pass urine, how strong is the urge usually?Positive: very strong or overwhelming
Negative: strong, normal, weak, or absent

COST DATA COLLECTED IN THE HOME INTERVIEW

The resource-use questions asked covered a range of services provided by the NHS and purchased by respondents themselves (personal costs). Details of the methods used for estimating the costs of individual items are given in Table 2. Respondents were asked to give details of all services received and these were costed using appropriate local and national data. Where no reliable cost was available an estimate was made from local data sources. This was conducted by itemizing all ‘components’ necessary to provide a service and combining them with estimates of the cost of these ‘components’. Costs are expressed in UK sterling for the year 1999/2000. Discounting was not used as all costs are assumed to accrue in the same year. However, where equipment was used that could have a lifespan of > 1 year, a ‘net present value’ was calculated using a discount rate of 6%[12].

Table 2.  Resources costed and sources of data used
Category of costSource of unit cost data used
NHS-borne costs
CathetersTime, consumables and equipment needed were obtained from expert opinion of two local providers of continence services. Relevant local unit cost data was used to cost the resources used.
Pads obtained from the NHSThe cost of NHS pads was obtained from a local continence nurse.
Clinical InvestigationsEquipment and consumables necessary to carry out these were obtained from expert opinion. Costs associated with health care provider time needed to provide interventions was valued elsewhere (costs of visits to healthcare providers) Costs were obtained from local nurse-led intervention to treat incontinence.
Contacts with health services provider in respondent's homeCost per minute and travel costs [21].
Respondents visits to GPs surgeryCost per minute values for GPs and practice nurses were obtained from a published national source [21].
Visits to outpatient departmentFor outpatient visits a cost was obtained from the relevant NHS outpatients (Leicester Hospitals NHS Trust)
Inpatient staysCosts for inpatient stays were estimated from several sources.
HospitalHospital Episode Statistics [25] for OPCS codes related to UI were obtained for 1999/2000. This was used to obtain the number of contacts for each code for males and females for the UK. Healthcare resource group codes were obtained for these OPCS codes from a local NHS trust. National reference costs [26]were then used to cost all activity for these OPCS codes for the UK. This was used to generate average costs for inpatient and day case admissions for both genders
Aids and appliancesCosts of items obtained from local continence nurse. Net present value calculated using a discount rate of 6%[12].
Medicines obtained on prescriptionCost obtained from the British National Formulary [27].
Personal-borne costs
Pads paid for by respondentMany respondents were able to give costs for purchase of panty liners and sanitary towels. These were used to estimate a mean cost per pad, this was used for all respondents reporting use of these products
Travel costsFor travel costs a cost per mile for car transport was obtained from the AA website [28]. Costs for public transport and taxis were obtained from the respondents, for a one-way trip to the contact. Total costs were assumed to be twice the one-way cost.
Replacing expensive itemsRespondents own valuations were used for expensive item replacement
LaundryCost only included electricity costs [29].
Over-the-counter medicinesCosts obtained via the website of the proprietary Association of Great Britain [30].
Time costsTaken to be the mean cost per hour for all workers in the East Midlands for those who stated they would otherwise be working [31]. For those not otherwise working a cost per hour was taken as that of a homecare worker [21]. Costs of any individuals who needed to accompany the respondent to ensure the contact was made were also costed using the hourly cost of a home care worker

ESTIMATE OF PREVALENCE

The estimate of prevalence used was obtained from a postal survey also conducted as part of the Leicestershire MRC Incontinence Programme. The focus of this questionnaire was also LUTS and it included all the same symptom questions used to define CSUSS [13]. The questionnaire was mailed to 39 602 individuals in October 1998, and after two reminders to those not responding, was completed by 23 182. After adjustment for the 2799 identified as not eligible (deceased or moved from address) a response rate of 63% was obtained. An in-depth study into those not responding showed little indication of nonresponse bias in the reporting of urinary symptoms, except for a slight under-reporting in the very elderly [14].

EXTRAPOLATION OF COSTS TO THE UK

The resource-use data from the home interview were used to estimate all costs associated with an individual's urinary symptoms. The costs estimated in the home interview were combined with the estimate of prevalence and the population estimates for the UK as a whole [15] to derive estimates of the total cost of urinary symptoms for individuals defined as CSUSS cases for both Leicestershire and the UK.

WILLINGNESS-TO-PAY (WTP)

To investigate the value of intangible costs the present study used a WTP instrument (see Appendix). This technique asks respondents to assign a monetary value to changes in health. A monetary value would correspond to the amount of spending on other goods and services that an individual would be prepared to forego to receive the health change. The more a health change is valued the more individuals would be prepared to forego and hence the higher a monetary value they would be willing to pay. The technique of WTP has been widely used to value changes in health and the benefits of healthcare interventions, e.g. a recent review contained 71 WTP studies [16]. WTP has also been used to estimate respondents values of their urinary symptoms [17–19].

In the present study respondents were asked if they would wish to change their urinary pattern in any way. If ‘yes’, they were asked if they would be willing to pay for this change. Individuals were then asked if they would pay any one of an ascending series of values; the interviewer would continue until they reached a value the respondent was unwilling to pay.

RESULTS

Reported use of NHS services were much higher for men than for women, with average annual NHS costs per person estimated to be £92.74 and £46.71, respectively (Table 3). The biggest component of NHS costs was inpatient stays, which comprise about half of NHS costs for both males and females. Other important contributors to total NHS costs included GP visits and prescribed medications. For personal costs the average annual costs per person for women were substantially greater than those for men (Table 3). The most important personal cost is the use of pads and padding, which was over £20 for women. Replacing expensive items is also costlier for women, and women tend to spend more on over-the-counter medications.

Table 3.  Mean annual NHS costs per person, personal-borne costs, annual time costs and intangible costs associated with CSUSS cases and non-cases
Mean cost £ sterling (95% CI)MaleFemale
CaseNon-caseCaseNon-case
N233250369260
NHS costs
Catheters    6.46 (−1.81–14.73)    2.98 (−1.34–7.30)    1.76 (−0.73–4.26)    1.37 (−1.31–4.05)
NHS pads    3.10 (−1.37–7.58)    0 (0–0)    1.23 (0.09–2.37)    0.62 (−0.60–1.84)
Investigations    1.95 (1.0–2.90)    0.31 (−0.01–0.63)    1.57 (0.77–2.37)    0.28 (0.02–0.55)
Practitioner visits to own home    1.81 (−0.70–4.31)    0 (0–0)    3.10 (0–6.20)    0.17 (−0.16–0.50)
Visits to GP surgery  12.53 (0.72–24.33)    0.57 (−0.38–1.51)    6.66 (0.04–13.28)    0.98 (−0.01–1.97)
Outpatients    6.28 (2.0–10.57)    3.15 (−1.52–7.82)    4.61 (0.94–8.29)    1.88 (−1.80–5.56)
Inpatients  44.63 (8.41–80.84)    0 (0–0)  23.26 (−1.85–48.37)    0 (0–0)
NHS aids    0.20 (0.04–0.35)    0.06 (−0.01–0.13)    0.48 (0.28–0.68)    0.13 (−0.01–0.27)
Prescription medicines  15.79 (6.46–25.11)    0 (0–0)    4.02 (0.81–7.24)    0 (0–0)
Total NHS costs  92.74 (41–144)    7.07 (0.47–13.66)  46.71 (8.31–75.1)    5.43 (0.33–10.52)
Personal costs
Pads paid for by respondent    2.05 (−1.59–5.69)    0 (0–0)  20.79 (15.60–25.98)    1.33 (0.43–2.22)
Travel costs of GP surgery visits    0.69 (−0.02–1.41)    0.03 (−0.02–0.07)    0.23 (0.08–0.37)    0.10 (−0.07–0.27)
Travel costs of outpatient visits    2.12 (−0.31–4.56)    0.25 (−0.16–0.66)    0.51 (−0.03–1.05)    0.07 (−0.07–0.22)
Replacement of expensive items    0.99 (−0.39–2.36)    0 (0–0)    7.23 (−2.17–16.63)    0 (0–0)
Extra laundry costs    1.47 (0.01–2.93)    0.01 (−0.01–0.02)    1.69 (0.86–2.53)    0.08 (−0.01–0.17)
Aids purchased by respondent    1.36 (−0.05–2.78)    0 (0–0)    0.22 (0.06–0.38)    0.02 (−0.02–0.06)
Over the counter medicines    0.21 (−0.21–0.63)    0 (0–0)    5.06 (2.5–7.63)    0.43 (−0.41–1.28)
Respondent borne costs total    8.90 (3.5–14.3)    0.29 (−0.13–0.7)  35.74 (24.3–47.17)    2.03 (0.76–3.29)
Time costs and intangible costs
Time spent visiting GP    2.85 (1.06–4.64)    0.13 (−0.05–0.32)    1.70 (0.56–2.84)    0.42 (0–0.84)
Time spent visiting outpatients    2.98 (0.78–5.19)    0.92 (−0.65–2.49)    1.17 (0.4–1.94)    0.64 (−0.62–1.9)
Total time costs    5.84 (2.46–9.21)    1.05 (−0.53–2.63)    2.87 (1.28–4.46)    1.06 (−0.26–2.39)
WTP  92.08 (65.58–118.59)  21.28 (12.83–29.72)  73.82 (61.65–85.99)  13.62 (7.3–19.93)

Other costs measured were the time costs of travelling to, and attending GP and outpatient appointments, and the WTP values for alleviating symptoms (Table 3). Annual WTP estimates are large, at a mean of £92.08 and £73.82 for men and women, respectively. These values are similar in magnitude to the value of NHS costs in the men, and greater in the case of women.

There are positive costs associated with non-cases for NHS, personal costs, and time and intangible costs: although these are substantially lower than those for cases (Table 3). Important costs associated with non-cases include the cost of outpatient visits, pads paid for by respondents (women only), and willingness to pay for alleviating symptoms.

Figure 1 shows that estimates of the mean costs per person vary between individuals who are aged < and > 65 years. There was a different pattern between men and women, with mean costs per person for men appearing to increase while those for women tended to decrease with age.

Figure 1.

Annual individual costs for women and men for CSUSS divided into < (green bars) and > (red bars) 65 years old.

Estimates of prevalence from the postal questionnaire are given in Fig. 2, by age and gender. The prevalence of CSUSS for women was estimated at 33.5%. Prevalence increased with age, at 30.1% for those aged 40–49 years to 49.4% for those aged > 80 years. The prevalence of CSUSS in men was 25.8% and again was age-related, from 16.5% for those aged 40–49 to 45% for those aged > 80 years. Overall, prevalence was slightly lower in men, with a steeper gradient with age than women. Total prevalence, for both men and women, was 30%.

Figure 2.

Prevalence of storage symptoms by age group. Males, green bars; females, red bars.

Table 4 shows estimates of the total costs extrapolated to all cases in the UK. Four categories of cost are shown; NHS, personal, time costs and WTP. Total resource-based costs are slightly higher in women, in line with gender differences in prevalence. NHS costs are higher for men and personal costs are substantially higher for women. The latter costs are equal to 9% of all male and 43% of all female resource use-related costs.

Table 4.  Total annual costs the UK for urinary symptoms associated with CSUSS cases
Costs, millions of UK £Storage symptoms
MenWomenAll
NHS costs302.8232.7  535.6
Personal costs  29.1178.1  207.2
Total resource331.9410.8  742.7
Time costs  19.1  14.3    33.4
WTP/intangible300.7367.9  668.6
Total costs651.6793.01444.6

WTP forms a substantial component of all costs; the overall value attributed to the symptoms by the WTP instrument was similar to the total resource-based costs, with a similar gender differential. This suggests that costs estimated by measuring direct resource use represent only about half of total costs for this condition if intangible costs are considered.

DISCUSSION

The costs presented here are those associated with CSUSS cases in community-dwelling individuals aged ≥ 40 years. The present estimates for NHS spending in the UK represent ≈ 1.1% of the NHS budget for the year 1999/2000. At £536 million the NHS costs presented here are higher than those estimated by the Continence Foundation [3]. This estimate was £354 million for England in 1998 prices, which approximates to £415 million for the UK as a whole (using 1999/2000 prices). These costs can be compared with cost estimates for other conditions. For example, heart failure has been estimated to cost £799 million as NHS costs in the UK [20] inflated to 1999/2000 prices using a hospital- and community health service-relevant inflator [21].

In the data used here there were clear differences in the cost associated with cases between women and men, and in how these costs varied between those < and > 65 years old. Total NHS costs were higher for men than for women despite the prevalence being slightly higher in women. Differences in costs were evident in more visits to the GP, more prescriptions and more inpatient care for men. Some difference may be a result of the difference in apparent treatability of different underlying conditions between men and women (e.g. between stress incontinence and prostate symptoms in older cases). However, it may also reflect differences in attitude, including investment in the development of appropriate services.

The costs given relate to individuals who are defined as being CSUSS cases. This does not include the costs associated with urinary symptoms for individuals who are below this clinical threshold. If we include costs associated with non-cases then the estimate of total cost for the UK increases to £656 million for NHS costs and £239 million for personal costs. Although the bulk of costs for urinary symptoms are connected with individuals who are defined as cases there are important costs associated with individuals who are not at these clinically significant thresholds.

The focus of this study was individuals aged > 40 years and living in the community. These results therefore underestimate the total population costs by excluding younger people and older people living in residential care. Data from the 2001 census indicates that 3.8% of people aged > 65 were resident in ‘medical and care establishments’[22]. In Leicestershire it has been estimated that 40% of such residents had weekly wetting, two-thirds of whom used continence aids [23]. This residential care population are therefore likely to have significant associated costs. The present estimate for the costs associated with CSUSS in community-dwelling adults is therefore an important part of the total for all costs associated with urinary symptoms, but there are likely to be important costs not considered here. A similar detailed approach considering costs in residential care is, we think, an important subject for future research.

The purpose of the WTP question was to estimate the value that an individual would place upon improvements in their symptoms, and this value is taken here to be the costs of these symptoms to the individual. However, the values used here are intended to indicate the potential size of these costs unrelated to resources rather than as a precise estimate. Potentially there are problems associated with this approach, notably in relation to individuals’ unfamiliarity with the procedure and the hypothetical nature of the question. There may also be questions relating to what exactly is being valued by individuals. Although the WTP values should be interpreted cautiously, the magnitude of these intangible costs is informative. For both genders WTP indicates that these intangible costs appear to be at least as great as resource use-related costs. In considering the overall burden of the condition and the potential for interventions to alleviate urinary symptoms the potential costs of urinary symptoms may be substantially greater than those generated by measurable resource use alone.

This is the first study to report NHS, personal and WTP costs for CSUSS. Future research should focus on subgroups excluded in this analysis, particularly elderly people in residential care. Future studies should also take account of costs associated with different types of disorder, such as stress or urge incontinence.

ACKNOWLEDGEMENTS

DAT was funded by the Trent Institute for Health Services Research. The authors thank the MRC for funding the Leicestershire MRC incontinence project. The authors also thank Dr Kate Williams, Madeleine Donaldson and Dr Gurminder Matharu for helpful comments on draft manuscripts. We also thank all staff who carried out home interviews for this study and all members of the public who completed interviews and postal questionnaires.

CONFLICT OF INTEREST

None declared. Source of funding: D. Turner was funded by the Trent Institute for Health Services Research, N. Cooper was funded by the University Hospitals of Leicester (UHL) NHS Trust, the remaining authors were funded by the MRC.

Appendix

The willingness to pay question used in the study

We are interested in how people value the impact of their urinary pattern on their normal activities.

(a) Would you wish to change your urinary pattern in any way?

Yes 1

No 2

The purpose of this question is to find out how much you would be willing to pay for a change in your urinary pattern. There are no plans to charge you for any of the services currently provided by the NHS or any future NHS services.

Imagine that there was a new pill that you could take that would give you your desired urinary pattern. There are no side-effects from taking this pill and taking it is not unpleasant in any way. Imagine also that you would have to pay for this pill (PROMPT: remind respondent that they will not be asked to pay for any real services provided).

So, if this pill existed, would you be willing to pay for it, i.e. to change your urinary pattern?

Yes 1

No 2

IF YES:

How much money, per month, would you be prepared to pay to change your urinary pattern? We would like you to think of the extra amount that you would be willing to pay over-and-above any current expenses that you already have that are related to your pattern. Remember, also, that any amount that you would pay would be taken out of your income, so that you would have less money to spend on other things.

How much would you be willing to pay? (PROMPT IF NO RESPONSE: would you be willing to pay £5, £10, £15 . . . per month? continuing until respondent says no, record last yes response).

£ PER MONTH

Note: This question was asked as part of a face-to-face interview where the term ‘urinary pattern’ had been used throughout. Individuals could ask the interviewer if they did not understand this term. Pilot work had revealed that this phrase was generally well understood and accepted. The term was used as many individuals would either not have clinically significant symptoms, or if they did may not consider that they had an illness or problem as such.

Ancillary