Jeanette S. Brown, MD; Catherine S. Bradley, MD, MSCE; Leslee L. Subak, MD; Holly E. Richter, MD, PhD; Stephen R. Kraus, MD; Linda Brubaker, MD, MS; Feng Lin, MS; Eric Vittinghoff, PhD; Deborah Grady, MD, MPH; Diagnostic Aspects of Incontinence Study (DAISy) Research Group
Grant Support: By a UCSF research contract from Astellas Pharma US (formerly Yamanouchi Pharma America).
Potential Financial Conflicts of Interest: Honoraria: C.S. Bradley (Pfizer Inc.), S.R. Kraus (Pfizer Inc., Astellas, Novartis); Grants received: C.S. Bradley (Yamanouchi), S.R. Kraus (National Institutes of Health, Astellas, GlaxoSmithKline).
Requests for Single Reprints: Jeanette S. Brown, MD, University of California, San Francisco, 1635 Divisadero Street, Suite 600, San Francisco, CA 94115; e-mail, email@example.com.
Brown J., Bradley C., Subak L., Richter H., Kraus S., Brubaker L., Lin F., Vittinghoff E., Grady D., ; The Sensitivity and Specificity of a Simple Test To Distinguish between Urge and Stress Urinary Incontinence. Ann Intern Med. 2006;144:715-723. doi: 10.7326/0003-4819-144-10-200605160-00005
Download citation file:
Published: Ann Intern Med. 2006;144(10):715-723.
Nearly 35% of women older than 40 years of age have urinary incontinence, which is associated with increased social isolation, falls, fractures, and admissions to specialized nursing units (1-3). Diagnosis and treatment of this common condition by primary care physicians would be optimal.
Incontinence is generally classified as urge (urine leakage with the urge to urinate), stress (urine leakage when straining, coughing, or exercising), mixed (both types), or other uncommon types of incontinence (such as neurogenic and overflow). While both urge and stress incontinence may improve with behavioral interventions, such as bladder training, urge incontinence is effectively treated with antimuscarinic or anticholinergic medications (4, 5) and stress incontinence is treated with pelvic muscle exercises and surgery (6, 7). Because treatment differs, it is important to distinguish urge incontinence from stress incontinence.
Karel G.M. Moons
Julius Center for Health Sciences and Primary Care
May 19, 2006
Improving the applicability of a test to distinguish between stress and urge urinary incontinence
With great interest I have read the paper by Brown et al.(1) The proposed test will certainly aid physicians in their work-up of women with urinary incontinence. I have two questions or rather suggestions to further improve the accuracy of their tool and its applicability in practice.
My first issue involves the intended use of their test, the 3 incontinence questionnaire (3IQ). The study population included women with at least 3 episodes of urinary incontinence per week for 3 months or more. Accordingly, the authors repeatedly state that the 3IQ test is to be applied as a diagnostic tool in women with urinary incontinence to distinguish between urge and stress incontinence. Given this, I do not understand the first question of their questionnaire. This question rather serves the purpose of screening or case finding of women with urinary incontinence in the general population. Is the 3IQ meant to be a diagnostic test to be applied in women with urinary incontinence (than the first question is redundant and even misleading) or a screening tool for case finding (which actually can not be inferred from their study)?
My second issue concerns the value of the 3IQ test in combination with other tests. The authors focused on the accuracy of the 3IQ test in isolation. This single test approach is indeed prevailing in the medical literature. But there is an increasing interest for a multivariable approach in diagnostic research.(2,3) Such multivariable approach allows for the estimation of the added value of a test beyond other tests, provides results that are commonly more relevant to practice (hardly any diagnosis is based on a single test), and may also improve the accuracy and applicability of a test. The authors documented various other patient characteristics that are known - based on their previous research(4) - to be related to the presence of either stress or urge urinary incontinence. These include, for example, race, body mass index, diabetes, parity and hysterectomy. The authors could easily undertake a multivariable (e.g. logistic regression) analysis or other type of subgroup analysis, to provide insight in the diagnostic accuracy of the 3IQ test in relation to these other easy to obtain characteristics. I do not think it would compromise the applicability of the 3IQ test. On the contrary, it would result in a more targeted use of their questionnaire if its accuracy indeed differs across these relevant patient subgroups.
1. Brown JS, Bradley CS, Subak LL, Richter HE, Kraus SR, Brubaker L, Lin F, Vittinghoff E, Grady D. The sensitivity and specificity of a simple test to distinguish between urge and stress urinary incontinence. Ann Intern Med 2006;144:715-23.
2. Moons KG, van Es GA, Michel BC, Buller HR, Habbema JD, Grobbee DE. Redundancy of single diagnostic test evaluation. Epidemiology 1999;10:276- 81.
3. Moons KG, Biesheuvel CJ, Grobbee DE. Test research versus diagnostic research. Clin Chem 2004;59:213-5.
4. Brown JS, Grady D, Ouslander JG, Herzog AR, Varner RE, Posner SF. Prevalence of urinary incontinence and associated risk factors in postmenopausal women. Heart & Estrogen/Progestin Replacement Study (HERS) Research Group. Obstet Gynecol 1999;94:66-70.
Karel G.M. Moons Professor of Clinical Epidemiology
Mahatma Gandhi Institute of Medical Sciences, Sevagram, India
May 21, 2006
Re: Improving the applicability of a test to distinguish between stress and urge urinary incontinenc
Brown and colleagues (16 May, p 715) (1) evaluate the diagnostic accuracy of the 3 Incontinence Questionnaire (3IQ) for discriminating types of urinary incontinence and suggest that in primary care setting the questionnaire may be used to classify and treat women with incontinence. The data from the study, however, indicate that the authors' confidence on the 3IQ is misplaced. We can interpret the results of the study by using likelihood ratios to compute post-test probabilities (predictive values). With the extended evaluation as the reference standard, the prevalence (pretest probability) of urge incontinence in the study was 39% (119/301). For classifying urge incontinence, the 3IQ generated a positive likelihood ratio (LR+) of 3.29 (95% CI, 2.39 to 4.51) and a negative likelihood ratio (LR-) of 0.32 (CI, 0.23 to 0.43). Thus when a woman's response to the 3IQ was positive, the probability of her actually having urge incontinence is 68% (CI, 59 % to 76%) [computed using the formula: pretest oddsÃ—LR+ = post -test odds]; when her response to the 3IQ was negative, the probability that she has urge incontinence drops to 17% (CI, 12% to 24%).
Similarly, the prevalence of stress incontinence in the study was 44% (132/301). For classifying stress incontinence, the 3IQ generated a LR+ of 2.13 (95% CI, 1.71 to 2.66) and LR- of 0.24 (CI, 0.15 to 0.36). A woman with a positive response to the 3IQ would have 62% chance (CI, 55% to 69%) of truly having stress incontinence; a woman with a negative response to the 3IQ would have a 16% chance (CI, 10% to 23%) of having stress incontinence. Clearly a positive test which misclassifies two-thirds of women with urge incontinence and three-fifths of women with stress incontinence would have limited application in clinical practice. This is not surprising because the likelihood ratios of 2-5 and 0.5-0.2 have been shown to generate small changes in probability of the target disorder (2). The test properties (LR+) of the 3IQ are unable to make large and clinically meaningful shifts between pretest and post-test probabilities, and may not help one in deciding the type of urinary incontinence. I would urge caution in using the 3IQ for ruling in or ruling out types of urinary incontinence in primary care setting.
1. Brown JS, Bradley CS, Subak LL, Richter HE, Kraus SR, Brubaker L, Lin F, Vittinghoff E, Grady D.The sensitivity and specificity of a simple test to distinguish between urge and stress urinary incontinence. Ann Intern Med 2006; 144(10):715-23.
2. Jaeschke R, Guyatt G, Lijmer J. Diagnostic tests. In: Guyatt G, Rennie D (eds). Users' guide to the medical literature. A manual for evidence-based clinical practice. Chicago: AMA Press, 2002:121"“40.
Steven A. Rich
Geriatrics/Lifetime Health Medical Group
June 5, 2006
Questinaires are inadequate to prescribe drug therapy for urinary incontinence
To the Editor: I disagree with the conclusions reached by Dr. Brown and colleagues in their study of the use of a three item questionnaire (3IQ) to diagnose and treat urinary incontinence (1). I recommend against the implementation of this technique in primary care.
The author's use of an "extended evaluation" as the gold standard for determining the type of urinary incontinence is not the accepted standard for such research. While urodynamic studies are not justifiable as routine test in all urinary incontinence, it is the accepted gold standard for research (2). Therefore, the estimates of sensitivity and specificity cannot be considered accurate.
The generalizabilty of the 3IQ test to internal medicine practice is not assured because of the exclusion of patients of more advance age and those with any neurological disorder.
The recommendation that the 3IQ can be used to direct non-specialist care of urinary incontinence underemphasizes the potential harm of anti-muscarinic agents used for urge incontinence. While mentioning the risk of urinary retention as a minor concern in the potential treatment of misclassified patients, the authors do not consider the central nervous system adverse effects of cognitive impairment and delirum. Gastrointestinal adverse experiences such as reflex esophagitis and constipation are also not considered.
The outcome of diagnosing and treating urinary incontinence based upon the 3IQ would include the prescribing of potentially harmful medication to patients without an indication up to 23% of the time. While this may serve the financial interest of the sponsor of this study (manufacturers of solifenacin succinate), it cannot be considered evidence-based practice.
An evidenced-based primary care approach to diagnosing urinary incontinence is presented in the ACP's Diagnostic Strategies for Common Medical Problems, second edition. The third edition (in press) will further emphasize the need, and means, for an acceptable diagnosis so that treatment can be based upon the best evidence and the appropriate thresholds of certainty and risk. Questionnaires can establish the presence of incontinence but, when compared to an appropriate "gold standard" test, they can not differentiate the type with enough accuracy to direct drug treatment.
Steven A. Rich MD Geriatrics/Lifetime Health Medical Group Rochester, NY Steven.Rich@lifetimehealth.org
2. Glazener CM. Lapitan MC Urodynamic investigations for management of urinary incontinence children and adults. Cochrane Database of Systematic Reviews. (4):CD003195, 2004.
3. Rich S, Pannill F, Urinary Incontinence. In:Black E, Bordley D, Tape T, Panzer R, editors. Diagnostic Strategies for Common Medical Problems, Second Edition. Philadelphia:American College of Physicians-American Society of Internal Medicine; 1999:527-539
Jeanette S. Brown
July 6, 2006
We appreciate the opportunity to answer Drs. Kalantri, Moons, and Rich's questions. We agree that the accuracy of the 3 IQ is modest. However, Dr. Kalantri considerably overstates the proportions of women with urge (43%, not two-thirds) and stress (23% not three-fifths) incontinence who are misclassified by the 3IQ (Table 3 of paper). As we stated in our conclusions, the modest accuracy of the 3 IQ is acceptable given that the risk of misclassification and inappropriate treatment by primary care is low.
The 3 IQ is meant to be a diagnostic test to be applied in women with urinary incontinence and not as a screening tool for case finding. The initial question was included because women who have incontinence may have less than monthly incontinence and current incontinence in the last 3 months seemed a reasonable threshold for continuing on to the next 2 questions to determine the type of incontinence.
There is not an agreed upon "gold standard". To determine the "gold standard", we consulted with international experts, our investigators, and did a literature review. We respectfully disagree with Dr. Rich's conclusion that urodynamics are required for the "gold standard".
For clarification, we did not exclude older participants (age range for the DAISy was 40 to 94) and only excluded major neurologic disease. We contend our cohort was, in fact, very generalizable to patients that should be treated by primary care.
We agree with Dr. Moon that prediction based on multiple risk factors might in principle outperform our assessment essentially based on two questions. However, the risk factors he cites from our earlier work are considerably more useful for predicting weekly incontinence rather than for classifying it by type among women known to have it. In addition, more complicated diagnostic algorithms are less useful in primary care practice.
We suggest use of a self-help booklet as the first line of treatment 1. We agree with Dr. Rich that further studies of adverse effects of medications for urge incontinence should be conducted to evaluate the effect on cognitive functioning. We also recommend further study to determine the clinical outcomes that would result from using the 3 IQ. Prior to that study, it seems reasonable to use the 3 IQ in a primary care setting.
1. Burgio, K. L., P. S. Goode, et al. (2002). "Behavioral training with and without biofeedback in the treatment of urge incontinence in older women: a randomized controlled trial." JAMA 288(18): 2293-9.
to gain full access to the content and tools.
Learn more about subscription options.
Register Now for a free account.
Nephrology, Urological Disorders.
Results provided by:
Copyright © 2016 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use
This PDF is available to Subscribers Only