To test the reliability of the Patient Reported Scar Evaluation Questionnaire (PR-SEQ) in a general scar population at two time points and in two formats (pen/paper and electronic). The PR-SEQ, a 30-item 4-domain measure of scar experience, was administered to a total of 512 clinic- and self-recruited subjects with qualifying scars. The 101 clinic-recruited subjects were administered the PR-SEQ in a pen/paper format at Baseline and at Timepoint 1 (T1) and Timepoint 2 (T2) about 7 +/- 2 days apart using a computer-based version on the internet. The 413 self-recruited subjects were administered only the internet version at T1 and T2. Internal consistency reliability was assessed using Cronbach’s alpha for the total T1 cohort (n=512); test-retest reliability was assessed using the intraclass correlation coefficient (ICC) based on an ANOVA model for the T1-T2 retest cohort (n=359); and between forms reliability was assessed using the ICC for the clinic cohort data from baseline and T1 (n=100). Cronbach’s alpha was 0.96 for the PR-SEQ total score and 0.86 for Appearance (5 items), 0.68 for Symptoms (3 items), 0.92 for Bother (8 items), and 0.95 for Impact (14 items) domains. The ICC for the test-retest cohort was 0.85 for the total score and 0.76, 0.78, 0.80, and 0.81 for the domains of Appearance, Symptoms, Bother, and Impact, respectively. Between forms reliability was similarly acceptable with scores of 0.84, 0.90, 0.92, and 0.95 for the domains, and 0.94 for the total score. Overall, the two versions of the PR-SEQ demonstrated acceptable levels of reliability across the domains and total score, with the possible exception of the internal consistency of the 3-item Symptom domain. Further item reduction may be possible without significant risk to the reliability of the instrument.

Photo-based questionnaires have been used to evaluate aesthetic treatment outcomes such as wrinkle reduction and eyelash growth. The objective of this study was to develop a photonumeric guide of scar severity intended for use by both clinicians and patients to measure scar treatment outcomes. Nine clinicians and 43 patients with linear surgical scars participated in a photograph sorting exercise. Patients sorted approximately 50 scar photographs consistent with their own skin type (light, medium or dark skin), while clinicians sorted all three sets of photographs (n=151 photographs). All participants arranged the photographs into 5 categories of perceived scar severity (least to most severe) and 5 photographs were considered for inclusion in the photonumeric guide based on analysis of inter-rater reliability, response consistency, redundancy, and variability. When photographs yielded similar results, clinical judgment was used to select the best photo. Instruction and response anchors were developed for the 5 scar photographs and the final guide was cognitively debriefed for relevance, comprehensibility, and acceptability in 24 additional scar patients. Based on the pre-specified criteria for inclusion and exclusion, 5 light skin and 5 dark skin photographs were included in the final photonumeric guide. A “medium” guide was not developed because of significant overlap between it and the light skin guide rendering it superfluous. Inter-rater reliability of the 52 subject cohort was strong (0.95-0.96) across all skin types. The 5 photographs included in the photonumeric guide demonstrated goodness-of-fit (infit mean-square < 1.4 and > 0.6), low variance in severity ratings (SD < one category change), and strong agreement between patients and clinicians. The Patient and Clinician Reported Scar Severity Scales were systematically developed to easily assess scar severity outcomes in clinical trials. Continued psychometric evaluation of the guide is planned to ensure the scales meet regulatory standards for labeling purposes.

Photoguides are used as measures of treatment effect. To interpret the results, an understanding of what change is “enough” is required. The purpose of this study was to determine the Patient Detectable Difference (PDD) and Patient Relevant Difference (PRD) in scar severity using a discrete choice experiment with paired comparisons of scar photos via the internet. Patients were asked to select the scar most like their own from a 5-photo photoguide as the referent. Each patient was then presented with a randomly selected scar from a scar library previously scored on severity by clinicians. For the PDD, patients were asked if the random scar was better, worse, or about the same. For the PRD, patients were asked if they would undergo treatment to “get” the random scar as opposed to keeping the referent scar. Each exercise was repeated five times. Optimal cutoff scores for PDD and PRD were based on a random intercept logistic model (RILM). Using a series of RILM estimations across 514 participants, the PDD was calculated as 15.4 points on a 100-point scale (area under curve = 0.85, sensitivity = 0.83, specificity = 0.73). In subgroup analyses by gender and scar source, the PDD ranged from 9.5 to 22.0 points. The replicate analysis adjusting for directionality of preference (improvement only) showed a PRD of 26.9 points (area under curve = 0.97, sensitivity = 0.90, specificity = 0.93), which ranged from 19.5 to 32.0 in the subgroups. Results show that, as expected, scar patients report detectable differences in scar severity that are smaller than what they report as relevant differences necessary to choose treatment. The Internet-based approach using a discrete choice experiment is a novel method to quantify the threshold of detectable and relevant differences in scar severity using a photoguide.

Currently available scar assessment tools focus on the appearance and symptoms of scars but avoid the complete scar experience. Additionally, existing instruments lack evidence or documentation (or both) of their development (e.g., content validity) and performance (e.g., reliability, construct-related validity, and responsiveness), and it is unlikely they would meet current regulatory requirements for labeling. The objective of the present work was to conduct qualitative research with patients to better understand scar appearance, symptoms, and impacts, and to use those results to inform the development of a content-valid PRO instrument for use in clinical trials. Eight physicians (3 dermatologists and 5 plastic surgeons) recruited subjects to participate in the CE research activities. Each subject (aged 18 to 65) was required to have a linear surgical (cosmetic or non-cosmetic) scar below the neck. Subjects with burn scars were excluded as were subjects with significant medical comorbidities. The interviews were conducted in multiple cities in the United States, transcribed, and used to derive the scar questionnaire items. A total of 43 interviews across five surgery types were conducted to elicit information regarding their experiences with their disfiguring scar. During the CE interviews, the most important and relevant concepts patients used to describe scar appearance were color, size (height, width, thickness), and texture. The most important and relevant scar symptom concepts reported were itchiness and pain. The most important, relevant, and bothersome scar impact concepts were limitations of wearing certain clothing and feeling self-conscious, sad, and less attractive because of the scar. The CE interviews provided rich information about how patients perceive and experience their scars which extends beyond the typical appearance dimensions. This work is fundamental in providing the basis for the conceptual framework of the scar experience and the first steps in the development of the PR-SEQ.