Clinically meaningful improvement on the Self-Esteem And Relationship questionnaire in men with erectile dysfunction

Transcription

1 Qual Life Res (2007) 16: DOI /s Clinically meaningful improvement on the Self-Esteem And Relationship questionnaire in men with erectile dysfunction Joseph C. Cappelleri Æ Stanley E. Althof Æ Michael P. O Leary Æ Sidney Glina Æ Rosie King Æ Vera J. Stecher Æ Martin Carlsson Æ Richard L. Siegel Received: 14 December 2006 / Accepted: 4 June 2007 / Published online: 7 July 2007 Ó Springer Science+Business Media B.V Abstract Purpose To estimate the minimal clinically meaningful improvement (MCMI) on the Self-Esteem And Relationship (SEAR) questionnaire. Methods Using combined data from the 2 pivotal SEAR trials of men treated with sildenafil for erectile dysfunction (ED), MCMIs were estimated as the lower limit of the 2-sided 95% confidence intervals of SEAR mean change scores (from baseline to end of study) for 2 anchor groups: (1) men who improved 1 ED severity category on the Erectile Function domain of the International Index of Erectile Function, and (2) men who improved 5 60% (inclusive) on these erectile function scores. Results Both anchors gave comparable results. A 10- point MCMI was proposed because the estimates of 95% lower bounds centered on around 10 points for most SEAR J. C. Cappelleri (&) Pfizer Inc, Global Research & Development, MS , Eastern Point Road, Groton, CT , USA joseph.c.cappelleri@pfizer.com S. E. Althof Case Western Reserve University School of Medicine, Cleveland, OH, USA M. P. O Leary Harvard Medical School, Brigham and Women s Hospital, Boston, MA, USA S. Glina Instituto H. Ellis, Sao Paulo, Brazil R. King Sydney Centre for Sex and Relationship Therapy, Sydney, NSW, Australia V. J. Stecher M. Carlsson R. L. Siegel Pfizer Inc, New York, NY, USA components (Sexual Relationship, Confidence, Self-Esteem, Overall Score). An exception was the Overall Relationship whose 95% lower bounds were too low to recommend them as an MCMI. Conclusions Two anchor-based approaches suggest that a change of about 10 points represents an MCMI on most components of the SEAR questionnaire. Keywords Erectile dysfunction Impotence Psychometrics Quality of life Self-concept Introduction The Self-Esteem And Relationship (SEAR) questionnaire is a 14-item patient-reported outcome that possesses strong psychometric properties that support its validity and reliability for measuring the impact of erectile dysfunction (ED) on men s self-esteem, confidence, sexual relationship, and relationship satisfaction (Table 1) [1, 2]. The SEAR contains 5 components: a Sexual Relationship Satisfaction domain, a Confidence domain, a Self-Esteem subscale, an Overall Relationship Satisfaction subscale, and an Overall Score. Extensive research documented in several publications [1 8] has confirmed that the SEAR questionnaire is multidimensional. The 4 concepts (Sexual Relationship, Confidence, Self-Esteem, Overall Relationship), along with their overall score, were based on focus groups (35 men with ED [aged years], 34 female partners of these men, and 27 physicians [10 internal medicine, 11 general or family practice, 6 urology]), factor analysis, knowledge of the subject matter, and comprehensive validation detailed in 2 published reports [1, 2]. These 5 components were then prespecified in 2 subsequent placebo-controlled

2 1204 Qual Life Res (2007) 16: Table 1 Self-Esteem And Relationship questionnaire During the past 4 weeks 1. Sexual relationship domain 1. I felt relaxed about initiating sex with my partner 2. I felt confident that during sex my erection would last long enough 3. I was satisfied with my sexual performance 4. I felt that sex could be spontaneous 5. I was likely to initiate sex 6. I felt confident about performing sexually 7. I was satisfied with our sex life 8. My partner was unhappy with the quality of our sexual relations 2. Confidence domain 2a. Self-esteem subscale 9. I had good self-esteem 10. I felt like a whole man 11. I was inclined to feel that I am a failure 12. I felt confident 2b. Overall relationship subscale 13. My partner was satisfied with our relationship in general 14. I was satisfied with our relationship in general Response options 1 = Almost never/never 2 = A few times (much less than half the time) 3 = Sometimes (about half the time) 4 = Most times (much more than half the time) 5 = Almost always/always Note: Questions 8 and 11 are reverse-scored so that a higher score indicates a more favorable response for all 14 items randomized trials of sildenafil [4, 5], which attested to the validity of the 5 components (including their responsiveness and sensitivity). In the sildenafil treatment studies, the Self-Esteem subscale was prospectively defined as the primary endpoint from baseline to week 12. Several clinical trials of sildenafil for the treatment of ED have incorporated the SEAR questionnaire [2 8]. Combined [6] and individual data analyses of the 2 pivotal double-blind, placebo-controlled SEAR trials conducted in and outside the United States [4, 5] demonstrated that men randomized to treatment with flexible-dose sildenafil (50- mg initial dose adjustable to 25 or 100 mg) for ED reported statistically significant (P < versus placebo) and substantially higher change scores from baseline on all 5 SEAR components, whereas scores in men randomized to placebo remained virtually unchanged. Moreover, changes in SEAR component scores correlated positively and strongly with changes in scores on the Erectile Function domain of the International Index of Erectile Function (IIEF). During a 36-week open-label extension of the US trial in which all men received sildenafil, mean Erectile Function domain scores and SEAR scores remained high in men who had received sildenafil in the randomized, doubleblind phase and increased (P < ) in men who had received placebo in the randomized, double-blind phase and then crossed over to sildenafil. In addition, positive and substantial correlations between changes in Erectile Function domain scores and changes in SEAR scores were maintained (P < ), suggesting that improved erectile quality is associated with improvements in long-term psychosocial well-being and functioning [8]. Quality-of-life measures, like the SEAR questionnaire, have become a standard and essential part of patient-reported outcomes across different types of disease. In addition to standard psychometric evaluations on validity and reliability, interpreting scores and establishing criteria for meaningful change in scores are crucial to understanding the relevance of outcomes on a health-status measure. Such an understanding is important not only in helping to complete a psychometric evaluation but also in designing trials, evaluating interventions, informing consumers and health policy makers, and providing information for formulary and reimbursement decisions [9 11]. Importance and relevance are highlighted by inclusion of a discussion of minimum important difference in the US Food and Drug Administration s recently published draft guidance for industry, Patient-Reported Outcome Measures: Use in Medical Product Development to Support Labeling Claims [12]. Clinically meaningful change in the outcome on a health-status measure can be estimated by comparing its change with change in a known external disease-related measure that has familiar clinical relevance and has an appreciable association with the targeted measure an anchor-based approach [10, 11, 13]. Anchor-based

3 Qual Life Res (2007) 16: approaches can be cross-sectional (e.g., difference in outcome on the health-status measure of interest between patients with different severity levels at the same time according to the external anchor measure) or longitudinal (e.g., change in outcome on the health-status measure of interest over time or with treatment that results in change in response on the external anchor) [10, 11]. In this article, we estimate a clinically meaningful improvement, and the minimal such improvement, on the SEAR questionnaire using a longitudinal anchor-based approach. Methods Data source The analyses were based on results from 2 randomized, double-blind, placebo-controlled trials of flexible-dose sildenafil treatment, which had identical protocols and were conducted inside [4] and outside (Australia, Brazil, Japan, and Mexico) the United States [5], and for which combined results have been published [6]. Assignment was in a 1:1 ratio by computer-generated randomization code to treatment with sildenafil or identical placebo that was to be taken approximately 1 h before anticipated sexual activity and not more than once daily. Patients were men 18 years or older who were in a stable sexual relationship, had a clinical diagnosis of ED confirmed by a score of 21 on the Sexual Health Inventory for Men, and had reduced self-esteem documented by a score of 75 out of a maximum of 100 on the Self-Esteem subscale of the SEAR questionnaire. Exclusion criteria were reported previously [4 6]. Efficacy assessments used for the current analyses were the SEAR questionnaire components (with scores transformed linearly to a 0 [worst] to 100 [best] scale), with the Self-Esteem subscale being the primary endpoint (Table 1). Also assessed were the Erectile Function domain of the IIEF, which categorizes erectile function by severity category as no ED (score 26 30), mild ED (22 25), mildto-moderate ED (17 21), moderate ED (11 16), and severe ED (1 10) [14 16]. Measurements on the SEAR questionnaire and the IIEF were also assessed at week 8, and missing values at week 12 or end of study were imputed with the last observation carried forward method. Statistical model for treatment differences An analysis of covariance model using treatment group as the key explanatory variable and study center and baseline scores as covariates was used to determine the difference in mean change from baseline to week 12 (or end of treatment) between groups in SEAR scores and in IIEF Erectile Function domain scores [4 6]. Corresponding 2-sided 95% confidence intervals (CIs) were obtained as well. Clinically meaningful improvement Analyses were based on observed mean (standard error [SE], 95% CI) SEAR scores within the entire intent-to-treat (ITT) population (sildenafil group plus placebo group) at baseline, 12-week (end of treatment) follow-up, and their difference (change from baseline to week 12). Anchor method 1 Severity of ED was measured concurrently at baseline and again at week 12 (or end of treatment) for each individual. SEAR change scores were stratified by change in ED severity category of the IIEF Erectile Function domain. Improvement of 1 ED severity category was used as the anchor for estimating clinically meaningful improvement in SEAR scores. Mean value and the lower limit of the 2- sided 95% CI were derived from the change scores for each SEAR component in men who had improvement by 1 ED severity category. Estimates on Minimal Clinically Meaningful Improvement (MCMI) were proposed based on the estimated lower limit of the 95% CI for mean changes. Mean changes were considered as estimates of clinically meaningful improvement (not necessarily minimal values). The relationship between change in the outcome of interest (SEAR score) and the anchor outcome (improvement of 1 ED severity category) was substantiated by broadening the anchor to include improvement of 1 ED severity category. If the relationship was valid, broadening to include men with a greater degree of improvement in the anchor should result in greater improvement in the outcome of interest. Anchor method 2 Another anchor-based approach was sought, one based on a percentage improvement instead of a 1-category improvement, that was also based on ED severity category levels: severe, 1 10; moderate, 11 16; mild-to-moderate, 17 21; mild, 22 25; no ED, Three formulations of it were initially considered: (1) the percent change from the first score of a category level with the first score of the next category level of improvement, (2) the percent change from a midpoint of a category level with the midpoint of the next category of improvement, and (3) the percent change from the last score of a category level with the last score of the next category level of improvement.

4 1206 Qual Life Res (2007) 16: For each formulation, the maximum percent change was obtained (1,000%, 145%, and 60%, respectively) so as to encompass all percentage scores for a 1-category improvement. Formulation 3 (based on the last score of a category level) was chosen because its highest percentage (60% from 10 to 16, or from severe to moderate) resulted in the most reasonable value and the most stable set of 1- category improvements (SE of 19.1% vs. 484% for formulation 1 and 59.2% for formulation 2). Therefore, with a 5% improvement in EF domain scores from baseline to end of study as the lower bound, 60% improvement from baseline to end of study in EF domain scores became the upper bound [percent change was measured as 100%* (Final EF score Baseline EF score)/(baseline EF score). Estimated MCMI values were proposed based on the lower limit of the 2-sided 95% CI for mean changes. Mean changes were considered as estimates for clinically meaningful improvement (not necessarily minimal values). The relationship between change in the outcome of interest (SEAR score) and the anchor outcome (5 60% improvement in EF domain score) was substantiated by broadening the anchor to include improvement of more than 5%, with no upper bound imposed. If the relationship was valid, broadening to include men with higher percentages of improvement in the anchor should result in greater improvement in the outcome of interest. Results Population In total, 553 patients (US trial, n = 253; Brazil, Mexico, Australia, and Japan trial, n = 300) were enrolled and randomized to receive placebo (n = 274) or sildenafil (n = 279) from April 2002 to February Patients in each treatment group were well balanced on age, weight, duration of ED, etiology of ED, and SEAR scores (Table 2). The ITT population consisted of 538 men with a completed baseline and postbaseline SEAR measurement (267 randomized to placebo, 271 randomized to sildenafil). Regarding the overall attrition rates (combined over the 2 studies), a total of 79 subjects (14.7% = 79/538) discontinued before study end at 12 weeks. Of these 79 patients, 35 were in the sildenafil group and 42 in the placebo group. Efficacy outcomes The difference in mean change in SEAR scores from baseline to end of treatment between the sildenafil group and placebo group (95% CI) was 22.9 ( ) for the Sexual Relationship domain, 21.9 ( ) for the Confidence domain, 22.5 ( ) for the Self-Esteem subscale, 20.0 ( ) for the Overall Relationship subscale, and 22.4 ( ) for the Overall Score (P < ) [6]. The difference in mean change in IIEF Erectile Function domain scores from baseline to end of treatment between the sildenafil group and placebo group (95% CI) was 6.2 ( ). At end of treatment, 184 of 256 (72%) men with ED who received sildenafil versus 91 of 254 (36%) who received placebo had erectile function that was categorized as mild ED or no ED (P < , chi-square test; Fig. 1). Clinically meaningful improvement Overall, a total of 134 men stayed in the same ED severity category from baseline to 12-week follow-up, 335 im- Table 2 Patient characteristics Placebo (n = 274) Sildenafil (n = 279) Mean ± SD age, year (range) 55 ± 12 (23 81) 56 ± 11 (25 83) Mean ± SD weight, kg (range) 85 ± 16 (53 148) 86 ± 16 (47 142) Mean ± SD duration of ED, year (range) 4.3 ± 4.5 ( ) 4.4 ± 4.4 ( ) Primary ED etiology, n (%) Organic 113 (41) 119 (43) Psychogenic 55 (20) 44 (16) Mixed 106 (39) 116 (41) Baseline mean ± SD SEAR scores (range) 1. Sexual relationship domain 39 ± 20 (0 91) 38 ± 20 (0 86) 2. Confidence domain 42 ± 21 (0 96) 42 ± 21 (0 100) 2a. Self-esteem subscale 39 ± 21 (0 94) 39 ± 21 (0 100) 2b. Overall relationship subscale 46 ± 28 (0 100) 48 ± 29 (0 100) 3. Overall score 40 ± 19 (0 87) 40 ± 18 (0 82) ED = erectile dysfunction; SD = standard deviation; SEAR = Self-Esteem And Relationship

5 Qual Life Res (2007) 16: Fig 1 Percentage of men in each severity category of the Erectile Function domain of the International Index of Erectile Function, at baseline compared with end of treatment, by treatment group. ED = erectile dysfunction proved by 1 ED severity category (of whom 88 improved by 1 ED severity category), and the remaining 69 men fell to a worse ED severity category. Anchor method 1 was based on men who improved exactly 1 ED severity category (n = 88), while anchor method 2 was based on men who improved between 5% and 60% (n = 154). MCMI, based on the lower limit of the 95% CI for mean change scores, was estimated as 13.0 (anchor method 1) and 11.3 (anchor method 2) for the SEAR Self-Esteem subscale and as (anchor 1) and (anchor 2) for the other SEAR components (Table 3). Nonminimal clinically meaningful improvement, based on mean change scores, was estimated as 17.8 (anchor 1) and 15.2 (anchor 2) for the SEAR Self-Esteem subscale and as (anchor 1) and (anchor 2) for the other SEAR components (Table 3). These values, calculated using improvement of 1 ED severity category as the anchor (anchor method 1) and 5 60% improvement in EF domain scores as the anchor (anchor method 2), were half or less than half of those calculated using mean improvement of 1 ED severity category as the anchor or using mean improvement of more than 5% in EF domain scores as the anchor (Table 4). Thus, the greater improvement in SEAR scores associated with greater improvement in the ED severity scores substantiated the relationship between these 2 measures. Furthermore, there was no real statistical shift in SEAR scores among men who stayed in the same ED severity category or among men who changed between minus 5% and 5% (inclusive) in EF domain scores (Table 5). When the MCMI estimated with both anchor-based methods were averaged, outcomes were similar for all SEAR components (range around 9 12 points) except for the Overall Relationship subscale. For simplicity, a value of 10 points was proposed for use as the MCMI for all SEAR components except for the Overall Relationship subscale whose MCMI, unlike the other components, was not satisfactorily provided by its lower 95% confidence limit. Discussion Anchor-based approaches are quite useful [10, 11, 13]. However, the validity of an estimate obtained with an anchor-based approach depends on the strength of the correlation between the disease-related outcome measure (the anchor) and the patient-reported outcome measure [17]. As reported previously in these data, improvement in IIEF Erectile Function domain scores showed moderate to high correlations (range, ; P < ) with improvement in SEAR component scores [6]. The strength of the relationship between the anchor and the SEAR questionnaire was also substantiated in the current analyses by the increase in the estimated clinically meaningful improvement (in line with expectations) when the anchor was broadened to include improvement of 1 ED severity category (instead of exactly 1 ED category improvement) of the IIEF Erectile Function domain and when the anchor was expanded to include improvement of more than 5% with no upper bound imposed (instead of an upper limit of 60%). The MCMI estimates obtained with the anchor-based approaches in the current analyses are further validated by their similarity and consistency. Estimates on the 95% lower bound for the mean change, as well as the mean change itself, for the SEAR components varied by fewer than 2 points between the 2 anchor-based approaches.

6 1208 Qual Life Res (2007) 16: Table 3 Anchor-based approaches: difference in mean change scores on the Self-Esteem And Relationship questionnaire among men who improved by 1 erectile dysfunction severity category and, separately, among men who improved from 5% to 60% in erectile function score SEAR component Change in ED severity Improved by 1 category a 5 60% improvement in EF scores b n SEAR change score, mean ± SE (95% CI) n SEAR change score, mean ± SE (95% CI) Sexual relationship domain ± 2.1 ( ) ± 1.7 ( ) Confidence domain ± 2.3 ( ) ± 2.0 ( ) Self-esteem subscale ± 2.4 ( ) ± 2.0 ( ) Overall relationship subscale ± 3.1 ( ) ± 2.6 ( ) Overall score ± 2.0 ( ) ± 1.7 ( ) CI = confidence interval; ED = erectile dysfunction; EF = erectile function; SE = standard error; SEAR = Self-Esteem And Relationship a Change from baseline to 12-week follow-up (end of treatment) in ED severity categorized by score on the Erectile Function domain of the International Index of Erectile Function (IIEF) b Percentage improvement from baseline to 12-week follow-up in Erectile Function domain scores of the IIEF Table 4 Anchor-based approaches: difference in mean change scores on the Self-Esteem And Relationship questionnaire among men who improved by 1 or more erectile dysfunction severity category and, separately, among men who improved by more than 5% in erectile function scores SEAR component Change in ED Severity a Improved by 1 category Improved by >5% in EF scores b n SEAR change score, mean ± SE (95% CI) n SEAR change score, mean ± SE (95% CI) Sexual relationship domain ± 1.3 ( ) ± 1.4 ( ) Confidence domain ± 1.4 ( ) ± 1.5 ( ) Self-esteem subscale ± 1.4 ( ) ± 1.5 ( ) Overall relationship subscale ± 1.8 ( ) ± 1.9 ( ) Overall score ± 1.3 ( ) ± 1.3 ( ) CI = confidence interval; ED = erectile dysfunction; SE = standard error; SEAR = Self-Esteem And Relationship a Change from baseline to 12-week follow-up (end of treatment) in ED severity categorized by score on the EF domain of the International Index of Erectile Function (IIEF) b Percentage improvement from baseline to 12-week follow-up in Erectile Function domain scores of the IIEF Table 5 Anchor-based approaches: difference in mean change scores on the Self-Esteem And Relationship questionnaire among men who stayed in the same erectile dysfunction severity category of the EF domain and, separately, among men who changed between 5% and 5% in EF domain scores SEAR component No change in EF category a 5% to 5% change in EF scores n SEAR change score, mean ± SE (95% CI) n SEAR change score, mean ± SE (95% CI) Sexual relationship domain ± 1.7 ( 6.0 to 0.9) ± 3.0 ( 9.7 to 2.6) Confidence domain ± 2.0 ( 6.8 to 1.3) ± 3.4 ( 7.2 to 6.5) Self-esteem subscale ± 2.2 ( 5.2 to 3.6) ± 3.9 ( 6.9 to 9.0) Overall relationship subscale ± 2.9 ( 11.2 to 0.4) ± 4.9 ( 14.5 to 5.5) Overall score ± 1.6 ( 5.8 to 0.5) ± 2.6 ( 7.6 to 3.1) CI = confidence interval; EF = erectile function; SE = standard error; SEAR = Self-Esteem And Relationship a Change from baseline to 12-week follow-up (end of treatment) in erectile dysfunction severity categorized by score on the EF domain of the International Index of Erectile Function Excluding the Overall Relationship subscale, the mean estimates of the 95% lower bounds across the 4 other SEAR components were 10.7 for anchor-based method 1 and 9.7 for anchor-based method 2, suggesting that an MCMI of approximately 10 points is tenable. The merit of using the lower limit of the 95% CI for the MCMI

7 Qual Life Res (2007) 16: estimates here is supported by the large sample size and by incorporation of natural sampling variation in estimation, along with the possibility that minimal or slight improvement on the Erectile Function domain may be somewhat less than a full-category improvement or a 60% improvement on the Erectile Function domain. The Overall Relationship subscale was an outlier, for which estimates varied more and were substantially lower. Because the empirical support is lacking, we cannot confidently place an MCMI on the Overall Relationship subscale and recommend instead that its clinically meaningful improvement is around its estimated mean change of 7 points. There are at least 3 reasons why the Overall Relationship subscale is an outlier with the anchor-based approach. First, among men who improved by 1 ED severity category on the Erectile Function domain of the IIEF, the mean score on the Overall Relationship subscale was the most sensitive (than other SEAR components) to its baseline score, which has been a practical issue in estimating the MCMI [18]. Second, of the 5 SEAR components, the Overall Relationship subscale showed the most variability on its mean change scores in terms of a 1-category improvement in ED severity. Third, the 2-item Overall Relationship subscale is the least precise of the SEAR components, as measured by internal consistency and test retest reliability [2], and is the least valid because its 2 broad questions may not fully measure an area as intricate as overall relationship satisfaction in a relatively short period of 12 weeks [1, 5]. Our investigation has at least four limitations. First, because clinically meaningful improvements on the SEAR questionnaire were anchored by the Erectile Function domain of the IIEF, such improvements are wholly gauged by erectile function and do not directly consider other measures of sexual function such as premature ejaculation. Second, although some SEAR questions cover partner relationship satisfaction, there are no questions that are responded to directly by the partner of the man with ED. Third, because the SEAR questionnaire is not intended to distinguish among male sexual disorders, its clinically meaningful improvements do not either. Finally, we made no attempt to investigate clinically meaningful deterioration on the SEAR questionnaire, which may or may not be of the same magnitude as its clinically meaningful improvements. We use last observation carried forward in our analyses. Assessments on SEAR scores in the 2 studies were scheduled at baseline, week 8, and week 12. With only 2 postbaseline measurements, limited information was available to capitalize on a more complicated model (e.g., a mixed-effects model or response profile model) to address missing data. Last observation carried forward here minimizes the number of subjects who are eliminated from the analysis and allows the analysis to examine trends over time. Similar discontinuation rates and reasons for discontinuation between treatments provide evidence that missing scores were not induced by drug. Most importantly, results at week 8 were similar to the results at week 12 [4 6], and this stable set of postbaseline scores across time lends justification for using last observation carried forward. The SEAR questionnaire, which is specific to men with ED, is among the latest in an increasing list of healthrelated quality-of-life measures for which criteria for meaningful change in scores have been established. Others include the Parkinson s Disease Questionnaire (PDG-39) [19], the Functional Assessment of Cancer Therapy-Breast (FACT-B) [20], the Functional Assessment of Cancer Therapy-Lung (FACT-L) [21], the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (QLQ-C30) [22], the Overactive Bladder Questionnaire (OAB-q) [23], and the Acne-Specific Quality of Life Questionnaire (Acne-QoL) [24]. For the SEAR questionnaire, as for these other questionnaires, criteria for meaningful change in scores must be used appropriately in the clinical environment. Based on the current analyses, we believe that the proposed MCMI in SEAR scores can be used as a guide to meaningful change in SEAR scores between treatments, even accounting for measurement error in individual scores. Repeat testing or corroboration with other psychosocial endpoints (e.g., the Erectile Dysfunction Inventory of Treatment Satisfaction [25]) is encouraged to enhance decision making regarding meaningful improvement or its absence. In clinical trials, the proposed MCMI in SEAR scores could inform decisions regarding required sample size and could assist in interpretation of scores between and within treatment groups. Conclusions Interpreting SEAR scores and establishing criteria for meaningful improvement in scores is important to understanding the relevance of outcomes on this validated health-status measure. According to convergence of 2 anchor-based criteria, both based on the erectile function domain of the International Index of Erectile Function, a 10-point increase in SEAR scores is estimated to represent the MCMI for all SEAR components except for the Overall Relationship domain (for which empirical data do not confidently offer an MCMI). Therefore, for these SEAR components, we propose that, in clinical trials, the mean score of the superior treatment group has a minimal improvement of 10 points and, in addition, that it exceeds the mean score of the inferior treatment group by 10 points.

8 1210 Qual Life Res (2007) 16: Acknowledgments This manuscript was sponsored by Pfizer Inc. Editorial support was provided by Deborah Campoli-Richards, RPh, at Complete Healthcare Communications, Inc., and was funded by Pfizer Inc. References 1. Cappelleri, J. C., Althof, S. E., Siegel, R. L., Shpilsky, A., Bell, S. S., & Duttagupta, S. (2004). Development and validation of the Self-Esteem And Relationship (SEAR) questionnaire in erectile dysfunction. International Journal of Impotence Research, 16, Althof, S. E., Cappelleri, J. C., Shpilsky, A., Stecher, V., Diuguid, C., Sweeney, M., et al. (2003). Treatment responsiveness of the Self-Esteem And Relationship questionnaire in erectile dysfunction. Urology, 61, Steidle, C. P., Stecher, V. J., Pace, C., & Tseng, L. J. (2006). Correlation of improved erectile function and rate of successful intercourse with improved emotional well-being assessed with the Self-Esteem And Relationship questionnaire in men treated with sildenafil for erectile dysfunction and stratified by age. Current Medical Research and Opinion, 22, O Leary, M. P., Althof, S. E., Cappelleri, J. C., Crowley, A., Sherman, N., Duttagupta, S., et al (2006). Self-esteem, confidence, and relationship satisfaction in men with erectile dysfunction treated with sildenafil citrate: A multicenter, randomized, parallel-group, double-blind, placebo-controlled study in the United States. Journal of Urology, 175, Althof, S. E., O Leary, M. P., Cappelleri, J. C., Hvidsten, K., Stecher, V. J., Glina, S., et al. (2006). Sildenafil citrate improves self-esteem, confidence, and relationships in men with erectile dysfunction: Results from an international, multi-center, doubleblind, placebo-controlled trial. Journal of Sexual Medicine, 3, Althof, S. E., O Leary, M. P., Cappelleri, J. C., Glina, S., King, R., Tseng, L.-J., et al. (2006). Self-esteem, confidence, and relationships in men with erectile dysfunction treated with sildenafil citrate: Pooled analysis of two double-blind, placebocontrolled trials. Journal of General Internal Medicine, 21, Cappelleri, J. C., Althof, S. E., Siegel, R. L., Stecher, V. J., Tseng, L. J., & Duttagupta, S. (2005). Association between the Erectile Dysfunction Inventory of Treatment Satisfaction and the Self-Esteem and Relationship Questionnaire following treatment with sildenafil citrate for men with erectile dysfunction. Value in Health, 8(suppl 1), S54 S Althof, S. E., O Leary, M. P., Cappelleri, J. C., Crowley, A. R., Tseng, L.-J., & Collins, S. (2006). Impact of erectile dysfunction on confidence, self-esteem, and relationship satisfaction after 9 months of sildenafil citrate treatment. Journal of Urology, 176, Symonds, T., Berzon, R., Marquis, P., & Rummans, T. A. (2002). The clinical significance of quality-of-life results: Practical considerations for specific audiences. Mayo Clinic Proceedings, 77, Crosby, R. D., Kolotkin, R. L., & Williams, G. R. (2003). Defining clinically meaningful change in health-related quality of life. Journal of Clinical Epidemiology, 56, Guyatt, G. H., Osoba, D., Wu, A. W., Wyrwich, K. W., & Norman, G. R. (2002). Methods to explain the clinical significance of health status measures. Mayo Clinic Proceedings, 77, Food and Drug Administration (2006). Guidance for industry. Patient-reported outcome measures: Use in medical product development to support labeling claims. guidance/5460dft.pdf 13. Hays, R. D., Farivar, S. S., & Liu, H. (2005). Approaches and recommendations for estimating minimally important differences for health-related quality of life measures. COPD: Journal of Chronic Obstructive Lung Disease, 2, Cappelleri, J. C., Siegel, R. L., Osterloh, I. H., & Rosen, R. C. (2000). Relationship between patient self-assessment of erectile function and the erectile function domain of the International Index of Erectile Function. Urology, 56, Cappelleri, J. C., Rosen, R. C., Smith, M. D., Mishra, A., & Osterloh, I. H. (1999). Diagnostic evaluation of the erectile function domain of the International Index of Erectile Function. Urology, 54, Rosen, R. C., Riley, A., Wagner, G., Osterloh, I. H., Kirkpatrick, J., & Mishra, A. (1997). The International Index of Erectile Function (IIEF): A multidimensional scale for assessment of erectile dysfunction. Urology, 49, Sloan, J. A., Cella, D., Frost, M., Guyatt, G. H., Sprangers, M., & Symonds, T. (2002). Assessing clinical significance in measuring oncology patient quality of life: Introduction to the symposium, content overview, and definition of terms. Mayo Clinic Proceedings, 77, Hays, R. D., & Woolley, J. M. (2000). The concept of clinically meaningful difference in health-related quality-of-life research. How meaningful is it? Pharmacoeconomics, 18, Fitzpatrick, R., Norquist, J. M., & Jenkinson, C. (2004). Distribution-based criteria for change in health-related quality of life in Parkinson s disease. Journal of Clinical Epidemiology, 57, Eton, D. T., Cella, D., Yost, K. J., Yount, S. E., Peterman, A. H., Neuberg, D. S., et al. (2004). A combination of distribution- and anchor-based approaches determined minimally important differences (MIDs) for four endpoints in a breast cancer scale. Journal of Clinical Epidemiology, 57, Cella, D., Eton, D. T., Fairclough, D. L., Bonomi, P., Heyes, A. E., Silberman, C., et al. (2002). What is a clinically meaningful change on the Functional Assessment of Cancer Therapy-Lung (FACT-L) Questionnaire? Results from Eastern Cooperative Oncology Group (ECOG) Study Journal of Clinical Epidemiology, 55, Osoba, D., Rodrigues, G., Myles, J., Zee, B., & Pater, J. (1998). Interpreting the significance of changes in health-related qualityof-life scores. Journal of Clinical Oncology, 16, Coyne, K., Matza, L., Thomspon, C., Kopp, Z., & Khullar, V. (2006). Determining the importance of change in the Overactive Bladder Questionnaire. Journal of Urology, 176, McLeod, L. D., Fehnel, S. E., Brandman, J., & Symonds, T. (2003). Evaluating minimal clinically important differences for the acne-specific quality of life questionnaire. Pharmacoeconomics, 21, Althof, S. E., Corty, E. W., Levine, S. B., Levine, F., Burnett, A. L., McVary, K., et al. (1999). EDITS: Development of questionnaires for evaluating satisfaction with treatments for erectile dysfunction. Urology, 53,