Figure. Association between the site of vitiligo lesions and specific quality-of-life impairment. Ordinal logistic regression models were constructed with the total Dermatology Life Quality Index (DLQI) score categorized into 3 levels (0-5, no effect or small effect; 6-10, moderate effect; and >10, very or extremely large effect) (A) and with self-consciousness (B) and sexual dysfunction (C) divided into 4 responses (not relevant/not at all, a little, a lot, and very much) as the dependent (outcome) variables. The independent (explanatory) variables were sites of vitiligo lesions, including the face, chest, abdomen, back, arms, hands, legs, feet, and genitals. Data are presented as the percentage of patients with a DLQI score of at least 6 (A), a lot or very much self-consciousness about their vitiligo (B), and a little, a lot, or very much sexual dysfunction (C) among patients who reported vitiligo lesions at the respective site on the diagram. In addition, odds ratios (ORs) (with parenthetical 95% confidence intervals) and P values are given.
Silverberg JI, Silverberg NB. Association between vitiligo extent and distribution and quality-of-life impairment. JAMA Dematol.
Published online November 1, 2012. doi:10.1001/jamadermatol.2013.927.
eTable 1. Questions used in the study.
eTable 2. Validation of vitiligo affected BSA from questionnaire
eTable 3. Effect of vitiligo on individual aspects of lifestyle and quality of life.
eTable 4. Ordinal logistic regression analysis of effect of number of body parts with vitiligo lesions on individual aspects of the DLQI
Silverberg JI, Silverberg NB. Association Between Vitiligo Extent and Distribution and Quality-of-Life Impairment. JAMA Dermatol. 2013;149(2):159-164. doi:10.1001/jamadermatol.2013.927
Author Affiliations: Departments of Dermatology, Beth Israel Medical Center and St Luke’s–Roosevelt Hospital Center, New York, New York.
Although vitiligo has been reported to cause significant impairment of quality of life (QOL),1- 4 specific areas of QOL impairment, such as functional limitations and impaired activities of daily living, have not been well characterized in large studies. In addition, the relationship of vitiligo extent, distribution, and duration with subsequent QOL impairment has not been examined. We hypothesized that greater extent of vitiligo would be associated with worse QOL impairment and that different distributions of vitiligo lesions would be associated with different aspects of QOL impairment. Vitiligo is thought of as a silent pigmentary disorder with few or no symptoms. Anecdotally, we observed that some patients with vitiligo complain of skin symptoms, such as pruritus or burning, associated with their disease. Few studies have described the burden of skin and other associated symptoms in vitiligo or their determinants. We conducted a prospective questionnaire-based study of vitiligo patients to elucidate these points.
This prospective questionnaire-based study was approved by the institutional review board at St Luke’s–Roosevelt Hospital Center for men and women (age, ≥18 years) with vitiligo. The survey was validated in paper format by adult vitiligo patients (n = 45) at the hospital to verify the ease and understandability of questions. Survey responses were compared with clinical information to determine the validity and interpretability of the questions, including physician diagnosis, affected body surface area (BSA), and number and distribution of body parts affected by vitiligo.
The questionnaire was uploaded to http://www.surveymonkey.com/, and members of nonprofit support groups for vitiligo vulgaris were invited to participate by responding to the survey. Responses from the initiation of the study in June 2010 through July 2011 were reviewed. The questionnaire took an average of 16 (range, 10-25) minutes to complete. Data were de-identified and confidential and posed no risk to participants; the institutional review board waived the informed consent requirement.
Questionnaire items were developed to determine novel risk and prognostic factors for vitiligo. The a priori aims of the questionnaire included identification of psychological sequelae of vitiligo. The questionnaire consisted of questions pertaining to demographics/phenotype of vitiligo and the standard 10-question Dermatology Life Quality Index (DLQI),5,6 for 77 questions overall (55 closed questions and 22 open questions). The questions related to this study and the response rates are listed in eTable 1. Responses were verified by screening for noninteger or implausible values (eg, ages given as younger than 0 or older than 100 years).
All data processing and statistical analyses were performed with commercially available software (SAS, version 9.2; SAS Institute Inc). Age at diagnosis, duration (in years), and number of body parts affected were divided into tertiles for statistical analysis owing to wide skewing.
We used the McNemar test and the Bowker test of asymmetry to compare survey responses with clinical information. Nonsignificant P values indicate consistency between both data sources and the validity of the question wording/format. We compared survey responses with clinical information for the number of body parts and duration of vitiligo using Spearman correlations. The Cronbach α coefficient ranged from 0.74 to 0.83 for questions used in the survey.
Ordinal logistic regression models were constructed to examine the effect of vitiligo extent on QOL, using the DLQI to categorize the effect as none or mild (0-5), moderate (6-10), and very or extremely large (>10), as previously reported.7 Individual aspects of the DLQI were divided into 4 responses (not relevant/not at all, a little, a lot, and very much) as the dependent variables. The independent variables were affected BSA of at least 25%, the number of body parts affected, laterality and distribution of vitiligo lesions, and duration of vitiligo (in tertiles). Retention of covariates in multivariate models was based on consideration of confounding and effect modification. Sex, current age (continuous), and comorbid autoimmune disease (binary) were evaluated individually and in combination. Any variable that had a P value of less than .05 or changed the parameter estimate for the independent variables of interest by more than 10% was retained as a covariate in that model. Adjusted odds ratios (AORs) were calculated by including them in the full models. Linear interaction terms were tested and were included in final models if significant (P < .05). Contrast statements were used to generate estimates for AORs, 95% confidence intervals, and P values for each combination of predictors.
We constructed logistic regression models to determine whether the site of vitiligo lesions (independent variable) predicts the impaired aspects of QOL and comorbid autoimmunity in vitiligo (dependent variables). Missing values were encountered owing to nonresponse of the patients to various questions (eTable 1 provides overall response rates). Complete case analysis was performed (ie, missing values were ignored throughout the study). Data imputation was considered by multiple imputations. However, few or no differences were found between the estimates from the 2 approaches. Therefore, final models did not involve data imputation.
P < .05 indicated statistical significance for all estimates. However, the multiple dependent tests performed in this study increase the risk of falsely rejecting the null hypothesis. Therefore, P values near .05 should be interpreted with caution.
The survey was piloted in 45 adult patients at St Luke’s–Roosevelt Hospital Center in paper-based format, including 30 with vitiligo and 15 with dermatitis. Clinical characteristics of these participants were collected by means of medical record review and compared with survey responses. All 30 vitiligo patients reported having physician-diagnosed vitiligo; none of the patients with dermatitis reported vitiligo (McNemar test, P > .99). The affected BSA of the 30 vitiligo patients was very similar to that of survey responses (Bowker test, P = .92) (eTable 2). Duration of vitiligo and the number of body parts affected by vitiligo lesions that were documented clinically correlated significantly with survey responses (Spearman correlation, ρ = 0.91 and ρ = 0.96, respectively [P < .001]). No subjects reported difficulty understanding or interpreting any questions. After validation, the survey was posted online and invitations were sent via e-mail.
One thousand seven hundred subjects started the survey, of whom 1632 completed it (96.0% completion rate), and 1553 were diagnosed as having vitiligo by a physician. Twelve patients were excluded for being younger than 18 years. Sex, age, duration of disease, affected BSA, face and/or body involvement, and presence of bilateral lesions for respondents are listed in Table 1. Of 1525 subjects with available and usable data, 541 (35.48%) had comorbid autoimmune disorders, most commonly hyperthyroidism or hypothyroidism (Table 2).
Overall, the total mean (SD) DLQI score in subjects with vitiligo was 5.9 (5.5). We found no significant difference between DLQI scores in subjects with and without comorbid autoimmune disorders (5.8 [5.4] vs 6.2 [5.5]; Kruskal-Wallis test, P = .08) or men vs women (5.7 [5.4] vs 6.0 [5.5]; P = .28). However, we found an inverse association between age and DLQI score (Spearman correlation, P = .004).
Total DLQI score was categorized into a combination of no effect or small effect (0-5), moderate effect (6-10), and very or extremely large effect (>10) and modeled using ordinal logistic regression. The DLQI was associated with an affected BSA of more than 25% (AOR, 2.17 [95% CI, 1.71-2.75; P < .001]) and the number of body parts affected (AOR for tertile 2, 1.47 [95% CI, 1.10-1.96; P = .009]; for tertile 3, 2.18 [95% CI, 1.64-2.89; P < .001]) but not with bilaterality (1.59 [0.98-2.57; P = .06]) or duration (for tertile 2, 0.97 [0.73-1.29; P = .84]; for tertile 3, 1.09 [0.82-1.45; P = .54]) of vitiligo lesions (Table 3).
Vitiligo had a negative effect on all aspects of the DLQI (eTable 3). Five hundred ten subjects (35.1%) reported itching and/or burning of their skin within the past week, which was associated with an affected BSA greater than 25% (AOR, 1.59 [95% CI, 1.26-2.01; P < .001]). An affected BSA greater than 25% was also associated with self-consciousness about lesions, impaired ability to go shopping and take care of their home or garden, influence on clothing decisions, effect on socializing or leisure activities, effect on participation in sports activities, and impaired relationships and sexual function (P < .001 for all) (Table 4). The odds of sexual dysfunction and other DLQI outcomes were similar between men and women. However, an affected BSA greater than 25% was not associated with impaired QOL secondary to treatment or loss of time owing to treatment in the past week (P = .53).
Similar results were found when the independent variable was the number of body parts affected by vitiligo lesions (eTable 4). However, the bilaterality of lesions was not a consistent predictor of impaired QOL in vitiligo (data not shown).
Distribution of vitiligo lesions was associated with impairment of different aspects of the DLQI. Lesions located on all areas of the body were associated with significantly increased total DLQI scores (Figure, A). However, only lesions on the face, arms, hands, legs, and feet were significantly associated with being self-conscious about vitiligo (Figure, B). In contrast, only lesions on the genital area and chest were significantly associated with sexual dysfunction (Figure, C).
The present study found that vitiligo had a negative effect on numerous QOL factors, including self-consciousness; clothing decisions; socializing and leisure activities; working or studying; relationships with partners, close friends, or relatives; and sexual function. Impaired QOL in vitiligo was associated with increased vitiligo extent and with the distribution of vitiligo lesions. Itching and burning skin occurs in about one-third of patients and is predicted by an affected BSA greater than 25%.
The mean (SD) DLQI scores found in the present study (5.9 [5.5]) are comparable to scores for other skin disorders, including psoriasis and atopic dermatitis.3,7,8 A wide range of DLQI scores measuring the effect of vitiligo has been reported in the literature (mean DLQI, 1 to >10).2,9- 13 These estimates likely vary with disparate study designs, small sample sizes, and selection bias.
The highly significant association between the percentage of affected BSA and DLQI scores observed in this study is consistent with prior studies. A prospective study of 78 Belgian adolescents or adults with vitiligo found that DLQI was associated with disease extent and improved with the use of camoflouge2; however, the study size was small and subjects were selected from an academic dermatology clinic. Our study focused on vitiligo patients and, therefore, we are unable to determine whether the association between the percentage of affected BSA and DLQI scores is specific to vitiligo vulgaris or perhaps is applicable to other dermatologic disorders (eg, psoriasis). A case-control study of 60 Tunisian vitiligo patients found that depression, anxiety, and QOL scores were worse than those of 60 other dermatologic disease controls.1 A questionnaire-based study of 119 Belgian patients found that vitiligo was associated with increased DLQI scores similar to those of psoriasis.3 In contrast, a prospective case-control study of 33 Serbian adolescents with vitiligo found no differences in anxiety or depression compared with 60 age-matched healthy subjects.4
The association between vitiligo and impaired QOL underscores the clinical importance of conducting psychological screening in vitiligo patients (eg, with the DLQI). Patients with a very large (DLQI score, 11-20) or an extremely large (DLQI score, >20) effect of vitiligo on QOL may benefit from more aggressive intervention, such as phototherapy or systemic therapy (eg, oral corticosteroids). Furthermore, psychological intervention may be warranted in such cases. Climatotherapy was found to improve QOL, coping with the disease, general well-being, and individual stress levels in 71 vitiligo patients compared with 42 matched controls.14 A nonrandomized study of 45 patients with vitiligo found that cognitive behavioral therapy improved patients' ability to cope with the disease as judged by the General Health Questionnaire, although the DLQI score did not change.15 Disease extent and duration were not examined in either study. Early psychological intervention might improve QOL and perhaps even vitiligo severity.
Vitiligo is classically viewed as a localized destruction of melanocytes and not as being associated with any skin symptoms. A recent study found that 20% of patients with vitiligo reported itching of the vitiligo lesions.16 In the present study, 35.1% of patients in this study reported that itching and/or burning of their skin affected their QOL, which was associated with increased vitiligo extent. Some potential mechanisms for skin symptoms may be (1) higher-grade inflammation with more extensive disease, (2) increased pruritus and/or heightened sensitivity secondary to anxiety, (3) koebnerization of lesions due to manipulation, and/or (4) symptoms secondary to comorbid autoimmune disease (eg, pruritus in autoimmune thyroid disease). Regardless, consideration of pruritus and other skin symptoms appears to be warranted in the initial management of vitiligo.
Different distributions of vitiligo lesions have different effects on QOL. Sexual dysfunction was associated with genital lesions, whereas self-consciousness was associated with lesions on noncovered areas, such as the face and extremities. However, lesions in all sites were associated with significantly higher DLQI scores. This association is likely a reflection of the DLQI instrument, which is a measure of functional and lifestyle impairment by a skin disease but not a measure of the emotional burden of disease. Vitiligo patients might feel limited in their daily activities by lesions anywhere on the body.
In the present study, we found no significant difference for sexual dysfunction between men and women and no significant linear interaction with sex. This finding is in contrast with a previous prospective study of 50 vitiligo patients aged 16 to 60 years from Istanbul that found increased prevalence of sexual dysfunction in female, but not male, patients compared with age- and sex-matched healthy controls.17 The results of the present study suggest that vitiligo has a negative effect on sexual dysfunction and many other aspects of QOL in men and women.
This study has several strengths, including being prospective, having a large diverse sample size, and using a validated questionnaire for QOL (the DLQI). The patient population includes a large sample of men and women and represents various adult age groups and extent of vitiligo. The DLQI is a well-established approach for the evaluation of QOL in dermatologic disease.6 However, this study has potential limitations. Measures of vitiligo extent are self-reported and not assessed clinically. To address this limitation, we validated the questionnaire before posting it online. Furthermore, we used multiple measures of disease extent, including the percentage of affected BSA, the sum of body parts reported to be affected, and the laterality and distribution of lesions. Many of the analyses found highly significant results across all or multiple measures of vitiligo extent. The invitation to participate in the survey was distributed by vitiligo support groups, which may result in a selection bias toward subjects with greater severity or with poorer QOL owing to vitiligo. Nevertheless, the study included a large number of subjects with lower vitiligo extent and DLQI scores. Finally, Fitzpatrick skin type was not assessed and should be considered in future studies.
In conclusion, increased extent of vitiligo is associated with poorer QOL, including pruritus and/or burning of skin and sexual dysfunction. These associations indicate that screening for QOL impairment in vitiligo patients should be included in the routine assessment and incorporated into therapeutic decision making.
Correspondence: Nanette B. Silverberg, MD, Department of Dermatology, St Luke’s–Roosevelt Hospital and Beth Israel Medical Centers, Ste 11D, 1090 Amsterdam Ave, New York, NY 10025 (firstname.lastname@example.org).
Accepted for Publication: August 26, 2012.
Published Online: November 19, 2012. doi:10.1001/jamadermatol.2013.927
Author Contributions: Both authors had full access to all the data in the study and take responsibility for the integrity of the data and accuracy of the data analysis. Study concept and design: J. I. Silverberg and N. B. Silverberg. Acquisition of data: J. I. Silverberg and N B. Silverberg. Analysis and interpretation of data: J. I. Silverberg. Drafting of the manuscript: J. I. Silverberg and N. B. Silverberg. Critical revision of the manuscript for important intellectual content: J. I. Silverberg and N. B. Silverberg. Statistical analysis: J. I. Silverberg. Administrative, technical, or material support: N. B. Silverberg.
Conflict of Interest Disclosures: None reported.