Validation of the Urgency Perception Scale

Authors


Linda Cardozo, Kings College Hospital, 8 Devonshire Place, London W1G 6HP, UK.
e-mail: coyne@medtap.com

Abstract

OBJECTIVE

To ascertain the validity and responsiveness of the Urgency Perception Scale (UPS) as an indicator of perceived urinary urgency.

PATIENTS AND METHODS

The UPS was developed to assess perceived urinary urgency in clinical studies evaluating the efficacy of the antimuscarinic drug, tolterodine, for treating the overactive bladder (OAB) syndrome. Secondary analyses of clinical and patient assessment data from three clinical studies of tolterodine were conducted to evaluate the UPS. Construct validity was assessed by correlations between the UPS and patient voiding diary variables and other patient assessments, including: perception of bladder condition; perception of treatment benefit; the Medical Outcomes Study Short-Form 36 health survey; the King's Health Questionnaire; the OAB Questionnaire; and the Overall Treatment Effect scale.

RESULTS

The UPS correlated well with the patients’ perception of bladder condition (correlation coefficients − 0.29 to − 0.46; all P < 0.001) and with the voiding diary variables, especially incontinence episodes (−0.30 to − 0.41) and pad usage (−0.25 to − 0.38; all P < 0.001). An improvement in UPS score at the end of treatment was consistently reflected by significant improvements among all voiding variables.

CONCLUSION

Urinary urgency is a central symptom of OAB and is particularly bothersome because of its unpredictability and consequent impact on daily life. The UPS is a valid scale by which urinary urgency can be assessed subjectively, and provides a useful tool for research into urinary urgency.

Abbreviations
OAB

overactive bladder

UPS

Urgency Perception Scale

RCT

randomized clinical trial

SB

single-blind, dose-escalation study

DECO

Detrol in the Community Study

SF-36

Medical Outcomes Study Short-Form 36

HRQL

health-related quality of life

KHQ

King's Health Questionnaire

OAB-q

Overactive bladder questionnaire

OTE

Overall treatment effect

INTRODUCTION

Symptoms of urinary urgency and frequency, with or without urge incontinence, are collectively referred to as the overactive bladder (OAB) syndrome, and affect ≈ 33 million people in the USA [1,2]. Of these symptoms, incontinence has received most attention, but urgency and frequency are far more common and troublesome [2–4]. Urgency is defined as a strong, sudden desire to urinate, typically at low bladder volumes (<150 mL), and frequency is defined as passing urine ≥ 8 times/day [1]. In a recent population-based survey of subjects with symptoms of OAB from six European countries, 85% and 54% of respondents reported frequency and urgency, respectively [4]. Urgency is a central and particularly bothersome hallmark symptom of OAB because it is unpredictable and consequently affects daily living.

The efficacy of treating OAB has traditionally been measured using voiding diaries or urodynamic variables. However, rigid voiding diaries might fail to highlight the most bothersome symptoms (e.g. urgency), and improvements in diary variables need to be viewed as clinically meaningful from the patient's perspective. Therefore, to determine the overall benefit of therapy, subjective patient evaluations are needed to complement objective voiding diaries. A specific measure of urinary urgency would allow the patient to focus on a specific complaint that they might find most bothersome, but although there are standardized measures of urge incontinence and frequency, there are currently none for urgency. Thus, to more fully assess the impact of urgency, a standardized, reliable instrument is required [5,6]. The Urgency Perception Scale (UPS) was developed to assess perceived urinary urgency in clinical studies evaluating the efficacy of an antimuscarinic drug, tolterodine, in the treatment of OAB. The purpose of the present research was to ascertain the validity and responsiveness of the UPS as an indicator of perceived urinary urgency.

PATIENTS AND METHODS

Content validity of the UPS was evaluated by face-to-face cognitive debriefing interviews with individuals who had symptoms of urinary urgency. Respondents to a newspaper advertisement were selected for interview via a telephone-based screening procedure. Sixteen participants aged ≥ 18 years and with urgency symptoms were selected; respondents with prostate cancer and other comorbidities that might have an effect on urinary urgency and frequency were excluded. Eight of the 16 participants were receiving pharmacological treatment for their urinary urgency. An Institutional Review Board approved the interviews and all participants gave written informed consent. Participants were questioned about the ease of self-administration of the questionnaire and the response options. Seven clinical experts who were unfamiliar with the tolterodine clinical studies, in which this instrument was used, were asked to review the UPS to ascertain the appropriateness of the questionnaire. All clinical experts responded to a series of standardized questions on the content, format, and validity of the measure.

An instrument's empirical performance is based on its distributional characteristics, evidence of reliability and validity, and its sensitivity or responsiveness to change [7]. Data on, and confidence in, an instrument's reliability and validity accrue over time, as it is used in repeated investigations in diverse conditions and settings. Thus, to evaluate the psychometric properties of the UPS, secondary analyses were conducted on clinical and patient data gathered during three separate clinical studies examining the efficacy and tolerability of tolterodine. While not ideal, the evaluation of construct validity and responsiveness provides useful information as to the assessment of urinary urgency.

PATIENT SAMPLE

The three diverse studies used for this psychometric evaluation were: (i) a randomized clinical trial (RCT) examining the clinical efficacy and tolerability of tolterodine prolonged-release capsules and tolterodine immediate-release tablets vs placebo for a treatment duration of 12 weeks (1169 patients) [8]; (ii) a single-blind, dose-escalation study (SB) with tolterodine (1–2 mg twice daily) in 978 patients with symptoms of OAB, including urinary urgency and frequency with or without urge incontinence, for a treatment duration of 16 weeks [9]. (iii) An open-label, observational study of 539 patients initiating therapy for OAB or changing it to tolterodine (2 mg twice daily). The study duration was 12 weeks (Detrol in the Community Study, DECO (data on file)).

All studies were performed in accordance with the Good Clinical Practice Guidelines and the Declaration of Helsinki. The ethics committees approved the study protocols, and all patients provided written informed consent before the start of the study. The per-protocol study populations were used from each dataset. All three studies were multicentre but only one (the RCT) was international, while the others were USA-based. When needed, questionnaires were translated on-site for non-English speaking countries. For two of the studies (SB and DECO), patients did not have to have urinary incontinence to be eligible, just symptoms of urgency and frequency, but they had to have both incontinence and urinary frequency to be eligible for enrolment in the RCT.

The UPS is a single-item, three-response question that has been both self- and interviewer-administered amongst patients with symptoms of OAB. The patient is asked to describe his/her typical experience when he/she feels the desire to urinate. The response options are: 1, ‘I am usually not able to hold urine’; 2, ‘I am usually able to hold urine until I reach the toilet if I go immediately’; 3, ‘I am usually able to finish what I am doing before going to the toilet’. This question was administered at baseline and at the end of treatment.

CLINICAL ASSESSMENTS

Voiding diaries, similar to those used by Groutz et al.[10], were kept of all incontinence episodes, all voids, and all voids associated with the sensation of urgency, before the study baseline visit and before the end-of-treatment visit. For at least 2 days of the diary period, every incontinence pad used was counted and the volume of urine per void recorded using a standardized measuring jar. Diaries were collected for a 7-day period in the RCT study and a 3-day period in the SB study. No diary data were collected in the DECO study.

All patient assessments except the perceived treatment benefit and the overall treatment effect were assessed at baseline and the end of the study. These assessments were in addition to standard patient outcome variables and were used to examine the validity of the UPS.

The Medical Outcomes Study Short-Form 36 (SF-36), a generic 36-item measure of health-related quality of life (HRQL), consists of eight subscales: physical functioning, role physical, bodily pain, general health, vitality, social functioning, role-emotional and mental health; it has been widely used in a variety of clinical conditions [11,12]. The SF-36 was used in the RCT and DECO studies.

The King's Health Questionnaire (KHQ, version 7) developed by Kelleher et al.[13] is a self-administered incontinence-specific HRQL instrument that has been validated in patients with OAB [14]. There are seven subscales: role limitations, physical limitations, social limitations, personal relationships, emotions, sleep and energy, and severity measures. Two single-item questions addressing incontinence impact and general health perceptions are also included. The KHQ was used in the RCT study.

The OAB Questionnaire (OAB-q) is a newly developed disease-specific questionnaire that assesses symptom bother and HRQL in patients with OAB, and consists of an eight-item symptom bother scale and a 25-item HRQL scale divided into four domains: coping, concern/worry, sleep, and social interaction [15]. The OAB-q was used in the DECO study.

The ‘Perception of bladder condition’ is a single-item self-assessment of perceived bladder condition severity used in patients at baseline and the end of the study in the RCT and DECO studies. Patients rated their perceived bladder condition severity on a six-point rating scale (1 = no problems; 2 = very minor problems; 3 = minor problems; 4 = moderate problems; 5 = severe problems; and 6 = many severe problems).

To assess the perception of treatment benefit, patients were asked: ‘Have you had any benefit from your treatment?’ and could give the following response options: ‘No’, ‘Yes, little benefit’, ‘Yes, much benefit’ in all three studies.

The Overall Treatment Effect (OTE) scale was used to assess patient-rated change in bladder symptoms [16]. The first question asked respondents to indicate whether their bladder symptoms had improved, remained the same, or worsened since the last evaluation. If participants indicated that their symptoms had improved, they were asked to rate the degree of improvement on a seven-point scale from ‘Almost the same, hardly better at all’ (1), to ‘A very great deal better’ (7). If subjects indicated that their symptoms have worsened, they were asked to rate the degree of worsening on a seven-point scale from ‘Almost the same, hardly worse at all’ (−1) to ‘A very great deal worse’ (−7). For the OTE scale, a score between − 1 and +1 was considered as no change in bladder symptoms. This was only assessed in the DECO study.

All statistical analyses were considered exploratory and no adjustments were made for multiple-test procedures. The a priori significance level was set at 0.05 unless otherwise indicated. Because the efficacy of tolterodine in the relief of urinary urgency symptoms in patients with OAB was not known (i.e. was being tested in the trials from which the data for the current study were drawn), treatment groups were not used to test the instrument's sensitivity.

For the UPS, change scores (baseline minus end of study value), were calculated for all patient-reported measures and patient diary variables. For the UPS, positive changes were coded as ‘Improved’; ‘0’ was coded as ‘Unchanged’; negative scores as ‘Worsened’. Analyses for the UPS were conducted on both the raw change scores (−2 to + 2) and the coded scores. No data were imputed; if data were missing, the patient was excluded from the analysis.

Construct validity was assessed by correlations between the UPS and the patient diary variables (voiding frequency, incontinence episodes, urge-related voids and continence pad usage) and other patient assessments at baseline and end of study or interim visit. Cohen's [17] guidelines for interpreting correlation coefficients were used, with a correlation of 0.1 being small, of 0.3 moderate and of 0.5 being large. In the behavioural sciences, validity coefficients (correlation with a criterion) are generally 0–0.6 [17].

When applicable, anova was used to assess the responsiveness and discriminant validity of the UPS among varying patient groups and disease severity. Responsiveness was examined by comparing the change in UPS score with the change in the patient's perception of bladder condition/treatment benefit and changes in the clinical voiding variables. To identify data trends, post hoc pairwise comparisons were conducted using the Student-Newman-Keuls’post hoc test with a significance level of 0.05. As the UPS response options are ordinal level data, categorical data analyses (chi-square and contingency table analyses) were conducted when needed.

RESULTS

All patient and physician reviewers commented that the concept of the UPS was valid and rational. Concerns about the response gradations were noted, as well as the need to evaluate tasks and time, but overall the UPS was viewed as an appropriate and reasonable assessment of perceived urinary urgency.

The baseline characteristics of each trial sample are presented in Table 1. Each study population was similar in terms of age, race and gender. A higher percentage of patients in the RCT and DECO studies reported being unable to hold their urine at baseline than in the SB study (30.4%, 29.5% and 18.0%, respectively) and relatively few patients were able to hold their urine until task completion in the RCT study, which reflects the enrolment criteria of each study.

Table 1.  Demographic and clinical characteristics, and the frequency of the UPS score change from baseline, with an assessment of treatment benefits of the study groups
VariableRCTSBDECO
N1169978271
Mean (sd) age, years  60.3 (13.8) 65.5 (13.3) 61.8 (15.3)
Gender, % female  83.1 79.8 78.2
Race, n (%)
White1111 (95.0)889 (90.9)246 (90.8)
Black  39 (3.3) 59 (6.0) 17 (6.3)
Asian  12 (1.0) 9 (0.9) 3 (1.1)
Other/mixed   7 (0.6) 21 (2.1) 2 (0.7)
Missing   0 0 3 (1.1)
UPS score, n (%)
1 = Unable to hold urine 356 (30.4)176 (18.0) 80 (29.5)
2 = Hold, but hurry 757 (64.8)672 (68.7)147 (54.2)
3 = Hold until finish task  55 (4.7)127 (13.0) 36 (13.3)
Missing   1 (0.09) 3 (0.3) 8 (3.0)
Perception of bladder condition, n (%)
No problems   3 (0.3) 4 (1.4)
Very minor problem  29 (2.5) 14 (5.2)
Minor problem 134 (11.5) 57 (21.0)
Moderate problem 521 (44.6)123 (45.4)
Severe problems 372 (31.8) 52 (19.2)
Many severe problems 109 (9.3) 14 (5.2)
Missing   0 7 (2.6)
UPS score change, n(%)
Improved 469 (40.2)422 (43.2)103 (38.0)
1–2 174 (14.9) 98 (10.1) 35 (12.9)
2–3 225 (19.3)276 (28.3) 52 (19.2)
1–3  70 (6.0) 48 (4.9) 16 (5.9)
Unchanged 615 (52.7)499 (51.2)137 (50.6)
Worsened  84 (7.2) 54 (5.5) 16 (5.9)
3–2  23 (2.0) 24 (2.5) 4 (1.5)
2–1  57 (4.9) 27 (2.8) 9 (3.3)
3–1   4 (0.3) 3 (0.3) 3 (1.1)
Missing   0 0 15 (5.5)
Perception of treatment benefits
Yes, very much 434 (37.3)500 (51.1)138 (50.9)
Yes, a little 396 (34.0)390 (39.9)113 (41.7)
No 335 (28.8) 88 (9.0) 18 (6.7)
Missing   0 0 2 (0.7)
OTE
Better (>+1)151 (55.7)
Same (−1,0,1) 94 (34.7)
Worse (<−1) 12 (4.4)
Missing 14 (5.2)

Table 1 also shows the change in UPS score and for the perceived treatment benefit and OTE. Across the three studies, 38–43% of patients reported an improvement in perceived urgency, while 50–53% reported no change; relatively few (6–7%) worsened. However, 61–93% of patients reported some treatment benefit (73% by OTE). Figure 1 compares the UPS change score and perceived treatment benefit for patients in the RCT study; of those improving by two response options on the UPS, 93% noted a treatment benefit, and 88% of those improving by one response option noted a treatment benefit. There was a consistent response trend between the measures, with similar trends in the SB and DECO study groups.

Figure 1.

Change in the UPS score by perceived treatment benefit (patients who reported ‘much’ or ‘little’ benefit).

The UPS correlated well with the perception of bladder condition, with correlations of − 0.3 and − 0.46 at baseline and 12 weeks in the RCT trial, and − 0.34 and − 0.41 in the DECO study (all P < 0.001). Of the voiding diary variables, the UPS correlated most strongly with incontinent episodes/day and number of incontinence pads used (Table 2). Correlations of the UPS with the incontinence-specific and OAB-specific questionnaires, the KHQ and OAB-q, were small to moderate (r = − 0.09 to 0.37, all P < 0.01, and 0.19–0.42, all P < 0.001, except the sleep subscale, where r = 0, P > 0.05) but were much stronger than correlations with the SF-36 (0–0.10, all P < 0.01).

Table 2.  Correlations of the UPS with clinical variables
Voiding diary variableRCTSB
Baseline12 weekBaseline16 week
  1. Correlations of RCT and SB baseline and end of treatment variables with UPS responses at the same period; *P < 0.001.

Incontinence episodes−0.30*−0.33*−0.41*−0.38*
Micturition frequency 0.0−0.19* 0.06−0.20*
Proportion associated  with urgency 0.15*−0.17*
Volume voided/24 h,  mL (n = 1155) 0.05 0.13* 0.02 0.14*
Number pads/24 h−0.25*−0.30*−0.38*−0.37*

UPS response options were also differentiated by voiding diary variables at baseline and at the end of treatment in both the RCT and SB studies. Table 3 shows the baseline results for the RCT study and compares voiding diary change scores with UPS change scores, showing that improved patients had a marked decrease in incontinence episodes and urge-related voids, and a decrease in incontinence pad use in both the RCT and SB studies.

Table 3.  Voiding diary variables compared with UPS response at baseline for the RCT group, and the change in the variables vs the change in the UPS score (baseline to 12 weeks)
UPSNMean (sd) or change in voiding diary variable [n]
Incontinence episodes/weekVoids/24 hProportion assoc. with urgencyVolume voided, mLPads/24 h
  1. Post hoc pairwise comparisons were among the UPS change groups using Student-Newman-Keuls test with P < 0.05. At baseline, there were no differences in voiding frequency or volume voided based on UPS response. For the remaining diary variables, the ‘unable to hold’ group was significantly worse than the other two groups. There was a significantly greater change in all voiding variables for the improved group than for the unchanged or worsened groups except voiding frequency, where only the improved group vs the unchanged group was significant.

Unable to hold356 34.2 (25.1)11.1 (4.4) 0.6 (0.4)133.9 (43.8) [349] 
Hold, but hurry757 18.2 (16.2)11.1 (3.7) 0.4 (0.3)137.9 (39.1) [751] 
Hold, finish task 55 21.7 (24.8)11.3 (4.4) 0.4 (0.4)139.2 (41.7) [55] 
Changes
Improved469−16.5 (18.9) 2.5 (3.4)−0.0 (0.5) 38.1 (52.2) [460]−0.9 (1.7)
Unchanged615 −8.5 (15.4)−1.4 (3.4)−0.1 (0.4) 27.8 (49.3) [612]−0.3 (1.7)
Worsened 84 −8.5 (20.2)−2.0 (3.2)−0.2 (0.4) 24.6 (45.0) [83]−0.1 (1.3)

DISCUSSION

Therapeutic trials for OAB have traditionally used objective measures of efficacy, e.g. voiding diaries, but such outcome measures might not adequately reflect the heterogeneous nature of OAB symptoms (e.g. incontinence only occurs in about a third of such patients [4]). Urinary urgency occurs in more than half of patients with OAB [4] and markedly compromises HRQL [18]. Nevertheless, although it is a core symptom of the ICS definition of OAB [19], urgency has rarely been included as a specific endpoint in clinical studies. However, given the effect that this symptom has on HRQL, urinary urgency is an important outcome to assess in treatments of OAB.

Currently, there are no specific quantifiable measures of urinary urgency. The current study describes an attempt to validate a standardized patient-reported subjective assessment of urinary urgency in clinical studies of OAB treatments. The results were consistent in three diverse patient samples, providing strong evidence of the construct validity of the UPS. While urinary urgency could be assessed subjectively, the UPS is not a diagnostic tool and cannot distinguish urge incontinence from stress incontinence.

The moderate correlations between diary variables and the changes in patient OAB symptoms of urgency after treatment with tolterodine suggest that the UPS is sensitive to incontinence, which reflects the focus of the scale, i.e. the ability to hold urine when ‘the urge’ to urinate occurs. This is supported by the correlations between UPS scores and reduction in incontinence episodes and number of incontinence pads used. As the SF-36 is a generic measure of HRQL (while the KHQ and OAB-q are disease-specific), the smaller correlation between the UPS and the SF-36 than for the KHQ and OAB-q was not unexpected. Domains of the disease-specific HRQL assessments that had the highest correlations in the disease-specific questionnaires were: incontinence impact, severity, and role and physical limitations in the KHQ; and symptom bother, concern/worry and social interaction of the OAB-q. These domains are often affected by incontinence, thus providing further evidence of the construct validity of the UPS.

The UPS is a valid scale in patients with OAB symptoms, but its responsiveness is uncertain, and this is probably related to its limited response options. Indeed, several patients interviewed for the content validity evaluation suggested that including additional response options would be useful, as they had perceived changes when on drug therapy, but not enough to merit a change in response option on the UPS. When comparing the UPS with the OTE and perceived treatment benefit, more patients perceive favourable treatment effects than report positive changes in the UPS, suggesting less sensitivity in the UPS responses. This was shown in the DECO study, where 58% claimed to be unchanged using the UPS, but 73% reported improvements using the OTE scale (with a 14-point gradation). Patients also noted that specific periods or tasks should be provided in the response options (e.g. they can hold their urine for 5, 10, 15 min). These issues are being addressed in the further refinement of the UPS that includes additional questions, response options, and overall visual analogue scales.

The UPS shows that urinary urgency can be assessed, and that patients can perceive changes in urinary urgency with treatment. Importantly, these changes were associated with a decrease in the number of incontinence episodes, voids associated with urgency, and incontinence pad use, as well as an increase in urine volume voided. This work provides a valid useful tool for research into urinary urgency, and lays a strong foundation for refining and developing a more comprehensive and responsive UPS.

ACKNOWLEDGEMENTS

This study was supported by a grant from Pharmacia Corporation. The authors acknowledge and thank Jerry G. Blaivas, MD, for his careful and thoughtful review of this manuscript. Material in this paper was presented at the International Society for Pharmacoeconomics and Outcomes Research 7th Annual Meeting; May 19–21, 2002; Arlington, Virginia, USA.

CONFLICT OF INTEREST

Karin Coyne is a paid consultant to Pharmacia. Source of funding: research funded by Pharmacia Corporation (now Pfizer).

Ancillary