Ryan C. Hedgepeth,* Jessica Labo, LingLing Zhang and David P. Wood, Jr.

Transcription

1 Expanded Prostate Cancer Index Composite Versus Incontinence Symptom Index and Sexual Health Inventory for Men to Measure Functional Outcomes After Prostatectomy Ryan C. Hedgepeth,* Jessica Labo, LingLing Zhang and David P. Wood, Jr. From the Division of Urologic Oncology, Department of Urology, University of Michigan, Ann Arbor, Michigan Purpose: Evaluating quality of life outcomes following prostate cancer treatment is important because different treatments provide similar survival outcomes. A wide variety of quality of life surveys are used with an unknown correlation between domain specific and broad domain instruments. We compared the urinary and sexual outcome measures of the Expanded Prostate Cancer Index Composite, a broad domain instrument, to those of the Incontinence Symptom Index and the Sexual Health Inventory for Men, which are domain specific instruments. Materials and Methods: A total of 640 patients undergoing radical prostatectomy at our institution completed a combination of the Expanded Prostate Cancer Index Composite, Incontinence Symptom Index and Sexual Health Inventory for Men questionnaires. Matching functional domains were compared and correlation coefficients were calculated. Subgroup analysis was performed to compare specific information pertinent to recovery from prostatectomy. Results: Correlations between measures of incontinence were 0.26 to 0.84, while indicators of sexual function were 0.70 to Subgroup analysis comparing measures of irritative symptoms demonstrated weaker correlations. Analysis in patients reporting no sexual activity also showed a significantly lower correlation of scores than that in patients reporting sexual activity. Conclusions: Expanded Prostate Cancer Index Composite scores generally demonstrate strong correlations with corresponding Incontinence Symptom Index and Sexual Health Inventory for Men scores, indicating similar measurements of information. Divergent correlations between irritative scores as well as scores in men who are not sexually active may indicate that the Expanded Prostate Cancer Index Composite has more descriptive validity in this population. Wider use of a single broad domain instrument such as the Expanded Prostate Cancer Index Composite to assess outcomes after prostate cancer treatment may improve clinical efficiency and allow comparative quality of life research across treatment types in the future. Abbreviations and Acronyms ED erectile dysfunction EPIC Expanded Prostate Cancer Index Composite EPIC-S EPIC-sexual EPIC-UIN EPIC-urinary incontinence EPIC-UIR EPIC-urinary irritative HRQOL health related quality of life IIEF International Index of Erectile Function ISI Incontinence Symptom Index SHIM Sexual Health Inventory for Men Submitted for publication November 5, Study received University of Michigan institutional review board approval. * Correspondence: Department of Urology, University of Michigan Health System, 3875 Taubman Center, 1500 East Medical Center Dr., Ann Arbor, Michigan Financial interest and/or other relationship with Intuitive Surgical. Key Words: prostate, prostatic neoplasms, prostatectomy, quality of life, questionnaires CLINICIANS treating patients with prostate cancer use various physiological and functional parameters to define success. Questionnaires that measure functional outcomes must accurately reflect patient symptoms. They can be broad domain instruments that assess several symptom areas or domain specific instruments that examine only 1 symptom area. The use of multiple patient assessment questionnaires before and after prostate cancer treatment /09/ /0 Vol. 182, , July 2009 THE JOURNAL OF UROLOGY Printed in U.S.A. Copyright 2009 by AMERICAN UROLOGICAL ASSOCIATION DOI: /j.juro

2 222 ASSESSMENT TOOLS TO COMPARE OUTCOMES AFTER PROSTATECTOMY commonly occurs because correlations between similar symptom areas in broad domain and domain specific instruments are largely unknown. The classic 3 outcomes used to evaluate therapeutic success following prostatectomy are cancer control, continence and potency. 1 HRQOL issues, including urinary control and sexual ability, are common after treatment and they may be as important to patient satisfaction as cancer control. 2 A large, population based longitudinal cohort study in men who underwent prostatectomy showed that more than 8% were incontinent and 60% were impotent 18 months after surgery. 3 Satisfaction with the treatment outcome in patients with prostate cancer has been shown to be directly impacted by changes in these as well as other quality of life domains. 4 Therefore, measuring such domains accurately and reliably is important for assessing and improving patient satisfaction after prostate cancer treatment. No consensus exists regarding the best way to measure functional outcomes after prostate cancer treatment. Many commonly used symptom questionnaires are site or study specific. Understanding the relative level of information contained in HRQOL instruments offers the opportunity to streamline the functional assessment process and it may enable standardized comparisons among treatment groups and sites. We established the correlation between the validated broad domain EPIC instrument and 2 domain specific instruments, that is ISI and SHIM. MATERIALS AND METHODS Study Population All patients with prostate cancer undergoing radical prostatectomy at our institution beginning in January 1994 who provided consent to the prospective prostate cancer database were eligible for study. EPIC, ISI and SHIM were routinely administered to all patients beginning in Patients who completed all 3 questionnaires at the initial preoperative visit were included in the study. Instrument Descriptions EPIC is a validated HRQOL instrument based on the UCLA Prostate Cancer Index that was created in response to an evolving need for quality of life information as surgical, radiation and hormonal treatments expanded. 5 The EPIC short form is intended to require less than 10 minutes to complete. It consists of 26 questions evaluating 4 domains, including urinary, sexual, bowel and hormonal domains. The urinary domain is further subdivided to allow for discrimination between incontinence (EPIC- UIN) and irritative (EPIC-UIR) symptoms. Development and validation of the EPIC instrument has been previously described in detail. 6 Scores are standardized and they range from 0 to 100 with higher scores indicating better perceived functional outcomes and satisfaction in patients. ISI was developed to quantify symptoms of stress and urge incontinence across many patient populations presenting with various etiologies. ISI is divided into severity and impairment (bother) scores with the severity domain subdivided into 3 subdomains focusing on stress incontinence, urge incontinence and pad use. Severity scores range from 0 to 32 and bother scores range from 0 to 8. Higher ISI scores indicate worse outcomes. 7 SHIM is an abbreviated version of IIEF. IIEF was created to assess variations in erectile function, particularly in response to oral phosphodiesterase inhibitor therapy. 8 SHIM consists of 5 IIEF questions that have the highest diagnostic predictive value for ED. 9 SHIM scores range from 1 to 25. Higher SHIM scores indicate better erectile function, including a total score of greater than 21 good function, 17 to 21 mild ED, 8 to 16 moderate ED and less than 8 severe ED. Outcome Variable and Statistical Analysis Questionnaires were administered preoperatively, postoperatively at 3, 6, 9 and 12 months, and at any additional followup visits. Three comparisons of EPIC with ISI and SHIM were performed. 1) The EPIC urinary domains, EPIC- UIN and EPIC-UIR, were compared with ISI severity and bother scores to examine the incontinence and urgency components of urinary distress. 2) The EPIC urinary domains Table 1. Patient demographics for EPIC, ISI and SHIM comparisons No. pts 640 Mean SD age (range) (38 77) No. surgery type (%): Converted from da Vinci to radical 9 (1.4) retropubic prostatectomy da Vinci radical retropubic prostatectomy 332 (51.9) Radical perineal prostatectomy 2 (0.3) Radical retropubic prostatectomy 297 (46.4) No. nerve sparing (%): Bilat 466 (72.9) Unilat 74 (11.6) Not performed 99 (15.4) No. neoadjuvant/adjuvant hormonal therapy (%): Androgen deprivation 33 (5.2) Chemotherapy 2 (0.3) Research protocol 19 (3.0) Median ng/ml pretreatment prostate specific 6.0 (0 50.8) antigen (range) No. pathological Gleason score (%): (26.6) (66.9) (6.3) No. pathological T stage (%): T2a 103 (16.5) T2b 411 (65.7) T3a 80 (12.8) T3b 23 (3.7) T4 8 (1.3) Mos since primary intervention: Mean SD (range) ( ) Median 4.8 No. race (%): Black 34 (7.1) White 539 (86.5) Other 54 (8.6)

3 ASSESSMENT TOOLS TO COMPARE OUTCOMES AFTER PROSTATECTOMY 223 were compared with ISI urgency incontinence subdomain questions, including question 4 could not wait to void, question 5 sudden urge to void and question 6 could not get to bathroom before voiding. Their sum emerged after differences in correlation scores between ISI and the 2 EPIC urinary domains. For ISI negative correlation values indicated positive associations due to inversely oriented outcome scales. 3) EPIC-S and SHIM score were compared. Subanalysis stratified by sexual activity was performed using a zero response on the 4 SHIM functional questions, including penetration, maintain, completion and satisfaction, as a surrogate for no sexual activity. All analysis was performed in a chronologically dependent manner (preoperatively vs postoperatively) as well as in a chronologically independent manner (combined). Correlation coefficients were calculated using Pearson s r correlation test, including an absolute value of less than 0.3 weak, 0.3 to 0.6 moderate and greater than 0.6 strong correlation. Normality is not required for Pearson s correlation test as long as the relationship is linear. An additional 2 tests, that is the Kendall b and Spearman tests, were also performed to substantiate Pearson values. These tests have decreased power and, therefore, they are less descriptive, although they are readily applied to skewed populations. They confirmed that Pearson s r test values were valid. Longitudinal analysis and linear mixed models were also performed when appropriate to account for repeated measures. All statistical analysis was performed using SAS, version with p 0.05 considered statistically significant. This research was approved by the University of Michigan institutional review board. RESULTS Demographics and Score Distribution Table 1 lists study population characteristics. Prostatectomy was performed in 640 patients who participated in the study. Data were obtained from EPIC in 554 patients, from ISI in 429 and from SHIM in 402. Median followup was 4.8 months and some patients were followed for a substantially longer time. Mean patient age was 59.6 years. Population characteristics were comparable to those in previous EPIC validation studies. Table 2 lists the preoperative and postoperative distribution of scores. EPIC-UIN and EPIC-UIR vs ISI EPIC-UIN and EPIC-UIR scores were compared to ISI scores using 611 and 583 pairs of complete questionnaires, respectively. Table 3 lists longitudinal correlation coefficients of EPIC urinary scores with ISI scores and EPIC-S with SHIM scores. Calculated coefficients between EPIC-UIN, and ISI severity and bother scores showed predominantly strong correlations (range 0.53 to 0.87) with negative values indicating appropriate symptom correlations. Histograms of EPIC-UIN and ISI severity revealed distributions that were similarly skewed (fig. 1), indicating that most patients who underwent prostatectomy had a successful urinary outcome. However, correlations between EPIC-UIR and ISI were much more moderate in strength, although they remained relatively constant across periods. Figure 2 shows these longitudinal comparisons graphically, allowing clear appreciation of the relative differences in the strength of the correlations. EPIC-UIR and EPIC-UIN vs ISI 4 to 6 and Sum The observed difference in the strength of the correlations between EPIC-UIR and EPIC-UIN with ISI prompted us to explore urgency symptoms more specifically. Thus, EPIC-UIN and EPIC-UIR scores Table 2. Preoperative and postoperative scores by instrument Instrument No. Pts Mean SD Minimum 25th Percentile Median 75th Percentile Max Minimum % Score Max EPIC-UIN: Preop Postop EPIC-UIR: Preop Postop ISI severity: Preop Postop ISI bother: Preop Postop EPIC-S: Preop Postop SHIM: Preop Postop

4 224 ASSESSMENT TOOLS TO COMPARE OUTCOMES AFTER PROSTATECTOMY Table 3. Longitudinal correlations between EPIC-UIN and ISI, EPIC-UIR and ISI, and EPIC-S and SHIM Score Comparison Preop 3 1 mos 6 1 mos 9 1 mos 12 1 mos 18 1 mos 24 1 mos UIN: vs ISI severity* vs ISI bother 0.57* 0.58* 0.68* 0.58* 0.66* * No. questionnaires No. pts UIR: vs ISI severity 0.45* 0.52* 0.52* 0.35* vs ISI bother 0.37* 0.48* 0.50* * No. questionnaires No. pts EPIC-S vs SHIM: 0.85* 0.70* 0.78* 0.79* 0.80* * No. questionnaires No. pts * Significant to p Significant to p were compared to ISI questions 4 (could not wait), 5 (sudden urge) and 6 (could not get to bathroom), and to the sum score of these 3 questions, representing the ISI urge incontinence component. Table 4 shows the results of this subgroup analysis using a linear mixed model to account for repeated measures. EPIC-UIR and the ISI urge questions demonstrated only weak to moderate correlations (range 0.25 to 0.36). These data indicate that the information gathered by EPIC-UIR and ISI is somewhat different. However, comparing EPIC-UIN to the same ISI urge incontinence scores showed fairly strong correlations (range 0.56 to 0.82), indicating a significant overlap of information. Thus, the correlation differences between ISI compared to EPIC-UIR vs EPIC-UIN seem to indicate that EPIC-UIR may capture unique information about urgency without incontinence that ISI and EPIC-UIN are not able to gather in this population. EPIC-S vs SHIM Finally, EPIC-S scores were compared to SHIM scores in 605 pairs of questionnaires. Longitudinal scores demonstrated strong correlations (range 0.7 Figure 1. Histogram distribution of postoperative EPIC-UIN, ISI, EPIC-S and SHIM scores

5 ASSESSMENT TOOLS TO COMPARE OUTCOMES AFTER PROSTATECTOMY 225 Figure 2. Longitudinal correlation between EPIC-UIN and EPIC-UIR with ISI scores to 0.85) (table 3). However, further analysis of these patients grouped by those reporting sexual activity and those reporting no sexual activity revealed a startling difference in correlations. Patients reporting sexual activity showed continued strong correlations between EPIC-S and SHIM scores but those reporting no sexual activity showed a decreased correlation (0.46). Furthermore, after again adjusting for repeated observations this correlation was decreased to 0.27, which was much weaker than that in the overall group. While the overall agreement Table 4. EPIC-UIR and EPIC-UIN correlations with ISI questions 4 to 6 and sum scores ISI Linear Mixed Model Coefficients 95% CI EPIC-UIR EPIC-UIN EPIC-UIR EPIC-UIN Question: 4 (couldn t wait) 0.36* 0.56 ( 0.43, 0.30) ( 0.62, 0.50) 5 (sudden urge) ( 0.35, 0.21) ( 0.85, 0.76) 6 (couldn t get 0.25* 0.68 ( 0.31, 0.18) ( 0.73, 0.63) to bathroom) Sum score 0.34* 0.82 ( 0.41, 0.27) ( 0.86, 0.78) *p p between instruments appeared good, this finding suggests that SHIM may underestimate erectile function in men who do not attempt to achieve sexual intercourse. Figure 3 shows comparative scatterplots of these groups with best fit lines representing the difference in the correlations. DISCUSSION EPIC urinary and sexual domain scores demonstrated good overall correlations with ISI and SHIM scores, indicating that the broad domain instrument obtains the same quality of life information as the 2 domain specific instruments. Strong correlations between EPIC-UIN scores, and ISI severity and bother scores confirmed that the postoperative incontinence assessment between the instruments is similar. Strong correlations between EPIC-S scores and SHIM scores demonstrated that information on sexual function is similar. However, subgroup analysis revealed an important divergence between EPIC- UIR and ISI urgency scores, and between EPIC-S and SHIM scores in patients with no attempted sexual activity. These decreased correlations imply that EPIC captures specific additional information

6 226 ASSESSMENT TOOLS TO COMPARE OUTCOMES AFTER PROSTATECTOMY Figure 3. Scatterplots show EPIC-S and SHIM scores grouped by patient report of sexual activity pertinent to patients with prostate cancer after treatment. EPIC may have been more comprehensive for measuring quality of life outcomes in this study because EPIC was specifically developed for the prostate cancer treatment population. As noted, ISI and SHIM were developed to evaluate many other types of patients. For urinary function the diverging correlations observed for urgency symptoms may have been due to the fact that EPIC detected urge symptoms without incontinence or ISI assigned an urgency component to symptoms that were predominately due to stress incontinence. Since previous studies have shown that urgency symptoms improve in most men after prostatectomy, it is unlikely that ISI detected a subtle worsening of symptoms. 10 For sexual function we dichotomized men by sexual activity using SHIM scores to illustrate that EPIC-S evaluates erectile function separately from sexual intercourse components. This likely provides a better description of anatomical erectile function, especially in men who are not experiencing intercourse with a partner but still have some measurable degree of erection. This difference may represent a key advantage of EPIC-S, in that identifying such men provides an opportunity to target them early with therapy and possibly improve long-term outcomes. Furthermore, the bother component of sexual function in EPIC-S may serve as an indirect measure of patient valuation and tradeoffs in the treatment process, which is an important part of understanding its impact on quality of life and an avenue for further research in this area. A problem with comparing quality of life outcomes after prostate cancer treatment is that clinicians use a diverse range of symptom questionnaires and instruments to evaluate patients. For example, the Prostate Cancer Outcomes Study reported urinary and sexual function after radical prostatectomy using survey questions adapted from several validated instruments. 3,11 The 2007 American Urological Association guidelines on prostate cancer noted that a main reason that quantitative synthesis of quality of life data after treatment is impossible is that a wide variation of assessment methodology is used in the literature. 12 Inconsistent reporting formats and followup further complicate the possibility of comparing quality of life outcomes. Multiple instruments may even be used at a single institution, including our institution. A finding that surprised us was that instrument response rates were different. Whether this was due to surgeon bias toward a survey that they believed was more specific to prostate cancer is not known. However, because most questionnaires were mailed to us by patients, it may have been that patients believed that the comprehensive nature of EPIC obviated the need to always complete other surveys. Regardless of the reason, the higher response rate of EPIC may prove advantageous for monitoring outcomes after treatment. EPIC offers an opportunity to compare HRQOL outcomes among treatment groups using only 1 instrument that was developed and validated in the prostate cancer treatment population. The advantages of using a single, descriptive quality of life instrument are important at the clinical, administrative and academic levels. Domain specific HRQOL instruments such as ISI and SHIM have had broad appeal because they include easily understood scoring systems and can be applied to various urological disease processes. Although EPIC is not as intuitive, its 26 questions on patient reported symptoms can still be easily completed by patients before the clinic appointment. Scoring and interpreting EPIC should become more accessible as computers continue to refine the clinical encounter. A single, comprehensive instrument can decrease the burden that multiple questionnaires place on patients, ancillary staff and

7 ASSESSMENT TOOLS TO COMPARE OUTCOMES AFTER PROSTATECTOMY 227 treating physicians, thereby improving office and clinical efficiency. Furthermore, using such an instrument may facilitate a direct comparison of prostate cancer outcomes across patient groups, between competing and novel treatments, and after defined recovery regimens. EPIC can detect meaningful differences in domain scores between treatment types such as prostatectomy, external radiation and brachytherapy on short-term evaluation and long-term followup. 13 Our study demonstrates that EPIC provides HRQOL information that is equivalent to or better than that of domain specific instruments and using it alone to measure outcomes in this population is appropriate. Several limitations of our study must be addressed. The study was performed at a single institution where the EPIC instrument was developed in a population similar to the validation cohort. This likely has little impact on its generalizability since our institution is a large referral center and external validation has been done at other institutions. 4 Another limitation is that the number of individuals with long-term followup is still relatively limited in our cohort, so that continued followup and correlational analysis are ongoing. CONCLUSIONS EPIC urinary and sexual function scores in patients with prostate cancer treated with radical prostatectomy show good correlation with ISI and SHIM scores. EPIC may obtain additional HRQOL data on these patients because it was specifically developed for this patient population. Using a well developed, broad domain instrument such as EPIC has advantages at the clinical, administrative and research levels. The widespread use of such an instrument could facilitate large-scale quality of life comparisons among treatment groups in the future. REFERENCES 1. Bianco FJ Jr, Scardino PT and Eastham JA: Radical prostatectomy: long-term cancer control and recovery of sexual and urinary function ( trifecta ). Urology 2005; 66: Fulmer BR, Bissonette EA, Petroni GR and Theodorescu D: Prospective assessment of voiding and sexual function after treatment for localized prostate carcinoma: comparison of radical prostatectomy to hormonobrachytherapy with and without external beam radiotherapy. Cancer 2001; 91: Stanford JL, Feng Z, Hamilton AS, Gilliland FD, Stephenson RA, Eley JW et al: Urinary and sexual function after radical prostatectomy for clinically localized prostate cancer: the Prostate Cancer Outcomes Study. JAMA 2000; 283: Sanda MG, Dunn RL, Michalski J, Sandler HM, Northouse L, Hembroff L et al: Quality of life and satisfaction with outcome among prostate-cancer survivors. N Engl J Med 2008; 358: EDITORIAL COMMENTS 5. Litwin MS, Hays RD, Fink A, Ganz PA, Leake B and Brook RH: The UCLA Prostate Cancer Index: development, reliability, and validity of a health-related quality of life measure. Med Care 1998; 36: Wei JT, Dunn RL, Litwin MS, Sandler HM and Sanda MG: Development and validation of the Expanded Prostate Cancer Index Composite (EPIC) for comprehensive assessment of health-related quality of life in men with prostate cancer. Urology 2000; 56: Wei JT, Dunn RL, Hoag L, Faerber GJ, Dorr R and McGuire EJ: The Incontinence Symptom Index (ISI): a novel and practical symptom score for the evaluation of urinary incontinence severity. J Urol, suppl., 2003; 169: 33, abstract Rosen RC, Riley A, Wagner G, Osterloh IH, Kirkpatrick J and Mishra A: The International Index of Erectile Function (IIEF): a multidimensional scale for assessment of erectile dysfunction. Urology 1997; 49: Rosen RC, Cappelleri JC, Smith MD, Lipsky J and Pena BM: Development and evaluation of an abridged, 5-item version of the International Index of Erectile Function (IIEF-5) as a diagnostic tool for erectile dysfunction. Int J Impot Res 1999; 11: Namiki S, Ishidoya S, Saito S, Satoh M, Tochigi T, Ioritani N et al: Natural history of voiding function after radical retropubic prostatectomy. Urology 2006; 68: Potosky AL, Legler J, Albertsen PC, Stanford JL, Gilliland FD, Hamilton AS et al: Health outcomes after prostatectomy or radiotherapy for prostate cancer: results from the Prostate Cancer Outcomes Study. J Natl Cancer Inst 2000; 92: Thompson I, Thrasher JB, Aus G, Burnett AL, Canby-Hagino ED, Cookson MS et al: Guideline for the management of clinically localized prostate cancer: 2007 update. J Urol 2007; 177: Miller DC, Sanda MG, Dunn RL, Montie JE, Pimentel H, Sandler HM et al: Long-term outcomes among localized prostate cancer survivors: health-related quality-of-life changes after radical prostatectomy, external radiation, and brachytherapy. J Clin Oncol 2005; 23: In postdiluvian Babylon all humankind spoke in 1 language, although the ensuing self-aggrandizement led to the construction of a monument so megalomaniacal that the Almighty is said to have issued punishment, confusing their languages so that they would not understand each other s speech. 1 Such is the case with the broad array of quality of life measures available to track men with prostate cancer. Focusing on 3 such measures, these authors argue convincingly that the hodgepodge of validated instruments creates a discordant cacophony of reported outcomes. After comparing EPIC, a widely used and comprehensive instrument developed specifically for prostate cancer, to the continence world ISI and the potency world SHIM they contend that their tool provides the most propitious data on this population. This seems correct. Their earnest plea for a uniform methodological approach to measuring quality of life outcomes in men with prostate can-

8 228 ASSESSMENT TOOLS TO COMPARE OUTCOMES AFTER PROSTATECTOMY cer is rational and worthy. However, one wonders whether such monolingualization would strip the diverse beauty from the pointillistic quality of life portrait to which we have grown accustomed. Mark S. Litwin Departments of Urology and Health Services Jonsson Comprehensive Cancer Center University of California-Los Angeles Los Angeles, California REFERENCE 1. Genesis 11: 1 7. These authors systematically compared different questionnaires that measure sexuality and urinary incontinence. A recent report showed strong correlation between SHIM and EPIC for measuring ED before prostate cancer treatment. 1 These authors now report that post-prostatectomy scores between the 2 questionnaires correlated similarly but they observed a modest difference between the 2 instruments at lower levels of sexual function, where a more compressed lower end of the SHIM scale may not be as sensitive to different ED levels that are detected by a broader distribution of EPIC-S scores. Whether these differences are clinically meaningful could be determined by testing which questionnaire score at baseline is more predictive of sexual outcome after treatment. As expected, the ISI score showed strong correlation with the EPIC-UIN score (each questionnaire measures incontinence) but only modest correlation with the EPIC-UIR score. Since the latter measures obstructive urinary concerns, comparing it to other measures of urinary obstruction such as the International Prostate Symptom Score or American Urological Association symptom index would be expected to show stronger correlation. A practical advantage of EPIC for evaluating prostate cancer treatment outcomes is that, like the UCLA-Prostate Cancer Index, from which EPIC was derived, it measures different aspects of HRQOL concurrently. Sexual and urinary outcomes are determined at once on 1 questionnaire, improving survey efficiency for the patient and the clinical or research staff alike. On the other hand, for clinicians or researchers with an interest in only 1 aspect of urological outcomes, eg those interested in erectile function but not incontinence or vice versa, using SHIM or ISI may be more practical. Martin G. Sanda Division of Urology Beth Israel Deaconess Medical Center Boston, Massachusetts REFERENCE 1. Schroeck FR, Donatucci CF, Smathers EC, Sun L, Albala DM, Polascik TJ et al: Defining potency: a comparison of the International Index of Erectile Function short version and the Expanded Prostate Cancer Index Composite. Cancer 2008; 113: 2687.