Enriching the Interpretation of the Erectile Dysfunction Inventory of Treatment Satisfaction: Characterizing Success in Treatment Satisfaction

Abstract

BackgroundPatient-reported outcomes, such as the Erectile Dysfunction Inventory of Treatment Satisfaction (EDITS) index, are essential for successful evaluation and treatment of patients with erectile dysfunction. AimTo enrich interpretation of the EDITS index score and to complement the existing 0 to 100 scoring. MethodsThis supplemental analysis evaluated EDITS questionnaire data (11 items; index score range = 0–100; higher scores indicate more treatment satisfaction) after completion of an 8-week double-blinded trial of 279 men 18 to 65 years old with erectile dysfunction randomized to sildenafil 100 mg, sildenafil 50 mg, or placebo. Response options for each EDITS item were grouped into “success” (the 2 most satisfied or favorable responses) and “no success” (the remaining 3 responses). The binary response (success or no success) for each item was expressed as a function of overall EDITS score in a simple logistic regression model with all treatments combined. OutcomesOdds ratios and success probabilities (using Wald χ2 tests) were calculated for specified point differences and total EDITS index scores, respectively. ResultsEDITS index score increases corresponded with significant increases in odds of success in different EDITS aspects (P < .0001 for all comparisons). For instance, a 10-point EDITS index score difference was associated with odds ratios of 11.3, 42.0, 17.7, and 6.8 for overall treatment satisfaction, treatment meeting expectations, satisfaction with treatment quickness, and satisfaction with how long treatment lasts, respectively. For a given EDITS index score, likelihood of success was determined for different aspects of treatment satisfaction. For example, a mean EDITS index score of 78 (sildenafil 100 mg; SD = 18) corresponded to 96%, 88%, 94%, and 88% chances of success for the 4 EDITS items referenced earlier, respectively. Corresponding probabilities for a mean EDITS index score of 50 (placebo; SD = 18) were 3%, less than 0.1%, 1%, and 4%, respectively. Clinical ImplicationsInterpretation of the EDITS index score can be augmented using key aspects of treatment satisfaction as reported by the patient. Strengths and LimitationsThis analysis used a well-established anchor-based approach to interpret EDITS index scores. The methodology used and corresponding results are appropriate for clinical practice and clinical trial settings. Limitations include data evaluation only for the Patient EDITS and not the complementary Partner EDITS and use of data from a clinical trial enrolling a well-defined patient population only in stable relationships. ConclusionThese results enable a meaningful interpretation of EDITS index scores, facilitating decision making by stakeholders for better-informed health care choices.Cappelleri JC, Tseng L-J, Stecher V, Goldstein I. Enriching the Interpretation of the Erectile Dysfunction Inventory of Treatment Satisfaction: Characterizing Success in Treatment Satisfaction. J Sex Med 2018;15:732–740.