Table II.
Test–retest reliability and internal consistency reliability studies
| Reference | Country | DLQI completed | Disease | Method used | Results | COSMIN |
|---|---|---|---|---|---|---|
| Test–retest reliability | ||||||
| Finlay 1994(1) | United Kingdom | 100 | Any skin disease | Spearman’s rank correlation | Test–retest reliability correlation coefficients Spearman rank gamma = 0-99, p < 0.0001); test–retest reliability of individual question scores (gamma =0.95–0.98, p <0.001) | ? |
| Badia 1999(34) | Spain | 246 | Eczema and psoriasis | ICC | ICC eczema 0.77, psoriasis 0.90 | + |
| Jobanputra 2000(35) | South Africa | 660 | 84 different diagnoses were made during the study. Dermatitis, including atopic and contact dermatitis (26%), psoriasis (18%), and acne (10%) were the most common disorders | Spearman’s rank correlation | r=0.97; p < 0.0001 (n = 65) | |
| Ferraz 2006(36) | Brazil | 115 | Multiple for reliability incl. onychomycosis and psoriasis (6 patients each), Contact dermatitis 4, and solar keratosis, viral warts, vitiligo (3 patients each). Lupus Erthematous for validity | ICR,Pearson’s | Pearson correlation coefficient for inter-observer reliability was 0.96 (p <0.001), n = 44 | ? |
| Takahashi 2006(37) | Japan | 197 | Acne | ICC | Test–retest reliability of the DLQI-J was slightly less than that of the original English version. n = 44 ICC=0.90) | + |
| Baranzoni 2007(38) | Italy | 22 | Any skin disease | ICC, Wilcoxon’s signed rank test | Retest at 1–2 weeks. n = 19 ICC = 0.983, p <0.001. Weighted kappa between 0.644 and 0.984 for items. No statistically significant difference found in total score between 1st and 2nd assessments (p = 0.016). p > 0.15 for all but 2 questions: symptoms p = 0.083 and clothes p = 0.096. mean DLQI 1st assessment = 9.14 ± 5.50, and 2nd assessment = 9.45 ± 5.86 | + |
| Mackenzie, 2011(39) | Canada | 60 | Psoriasis and Psoriatic arthritis | ICC | 0.96 (0.93, 0.97) (n = 60) | + |
| Madarasingha 2011(40) | Sri Lanka | 200 | Eczema (24.5%), Psoriasis (23.0%), Acne (10.0%), Vitiligo (14.5%), Infections (10.5%), Other (17.5%) | Cohen’s Kappa | Kappa test–retest reliability coefficient of 0.83 | + |
| Khoudri 2013(41) | Morocco | 244 | Psoriasis | ICC | ICC of the test–retest reliability was 0.97 for the overall DLQI and exceeded 0.70 in all scales. | + |
| Liu 2012(42) | China | 131 | Urticaria | ANOVA | ANOVA with Friedman’s test chi2 = 320.61 (p <0.001) indicated good repeatability using the DLQI of Chinese version. | ? |
| Ali 2017(43) | United Kingdom | 104 | Any skin disease | ICC, Wilcoxon’s signed rank test | ICC = 0,98; 95% confidence interval (CI) 0.97–0.99 | + |
| Jesmin 2021(44) | Bangladesh | 80 | Psoriasis | ICC | ICC= 0.97 | + |
| Meneguin 2021(45) | Brazil | 188 | psoriasis, cellulitis/erysipelas, chronic ulcers and eczematous dermatosis, other dermatoses | ICC | For cases that did not show any clinical change in their disease status (n = 44), first interview media n = 9 (4.5–11), second interview after 7 to 14 days media n = 10 (5.5–11.5). ICC 0.95 (CI 0.88–0.98) | + |
| Schwartzman 2021(46) | United States | 994 | Atopic dermatitis | ICC | 0.81 (95% CI 0.76–0.85) | + |
| Internal consistency reliability studies | ||||||
| Finlay 1994(1) | United Kingdom | 100 | Any skin disease | Consistency between all questions when paired was found to be statistically significant (p = 0.002) ranging from Rank correlations of 0-23-0-70 | ? | |
| Badia 1999(34) | Spain | 246 | Eczema and psoriasis | α = 0·83 | + | |
| Jobanputra 2000(35) | South Africa | 660 | 84 different diagnoses were made during the study. Dermatitis, including atopic and contact dermatitis (26%), psoriasis (18%), and acne (10%) were the most common disorders | α = 0.83. The inter-item rank correlation coefficients ranged from 0.04 to 0.54 | + | |
| Zachariae 2000(47) | Denmark | 400 | Psoriasis, Atopic eczema, Other eczema, Urticaria, Bullous disease, Erythroderma, Hyperhidrosis, Collagenosis, Pruritus, Acne, Viral warts, Miscellaneous | α = 0.88 | + | |
| Shikiar 2003(48) | United States | 1095 | Psoriasis | Study A baseline α =0.871, week12 α =0.921; Study B baseline α =0.869, week12 α =0.919 | + | |
| Aghaei 2004(49) | Iran | 70 | Vitiligo | α = 0.77 | Cronbach’s alpha by domain, and by gender, marital status, severity, and extension of disease, and | + |
| Ilgen 2005(50) | Türkiye | 108 | Acne | α = 0.87 | + | |
| Mazzotti 2005(51) | Italy | 900 | Psoriasis | a = 0.83; item-total correlation = 0.40–0.70 | + | |
| Ozturkcan 2006(52) | Türkiye | 79 | Eczema-contact dermatitis, Psoriasis, Urticaria, chronic urticaria, Tinea, Alopecia areata, Acne | Cronbach’s α = 0.87. The item versus total (overall). Spearman’s correlation coefficients ranged from 0.48–0.81 with a median of 0.66, and the subscales versus total (overall) ranged from 0.71–0.83 with a median of 0.77. The α value was 0.84 for the age groups under 20 years and 0.89 for the 21+ years age group | males 0.83 and females 0.88; outpatients 0.86 and inpatients 0.87; eczema/acne 0.90 and other dermatological disorders 0.84). | + |
| Shikiar 2006(53) | United States | 147 | Psoriasis | α was 0.89 at baseline, 0.92 at Week 12 | + | |
| Takahashi 2006(37) | Japan | 197 | Acne | α = 0.83. Exclusion of any one of the 10 items did not increase α by > than 0.01. | + | |
| Baranzoni 2007(38) | Italy | 22 | Any skin disease | α = 0.787 for 1st assessment, 0.828 for 2nd assessment | + | |
| Mazharinia 2007(54) | Iran | 109 | Burns | Cronbach’s α for physical Q1,3,5,7,10, psychological Q2,4,6,8, and sexual domains Q9 and for total DLQI were 0.78, 0.77, 0.72, and 0.75, respectively. | + | |
| Henok 2008(55) | Ethiopia | 74 | Podoconiosis | Overall α value was 0.90, standardized item alpha 0.89. Average inter-item correlation was 0.44, item total correlation ranged from 0.15 to 0.81. Only item 6 (about sport) had a value of <0.2. The average item total correlation was 0.64. | + | |
| Aghaei 2009(56) | Iran | 125 | Psoriasis | α = 0.79 | + | |
| An 2010 (57) | China | 128 | Leprosy | Cronbach’s α = 0.765, standardized item α =0.759, | Average inter-item correlation was 0.240 (> 0.2), Item total correlation ranged from 0.212 to 0.596. Average item total correlation was 0.427. | + |
| Madarasingha 2011(40) | Sri Lanka | 200 | Eczema (24.5%), Psoriasis (23.0%), Acne (10.0%), Vitiligo (14.5%), Infections (10.5%), Other (17.5%) | Cronbach’s α 0.561 to 0.741 (except for the personal relationship domain). Healthy volunteers (n = 40): Symptoms and feelings (0.598), Daily activities (0.654), Leisure (0.569). Personal relationships (0.498). Patients (n = 200): Symptoms and feelings (0.561), Daily activities (0.741, Leisure (0.687). Personal relationships (0.442). | – | |
| Liu 2012(42) | China | 131 | Urticaria | α was 0.82, and it became 0.84 when item 1 was deleted. The α value reached 0.85 after standardization | + | |
| Maksimovic 2012(58) | Serbia | 66 | Atopic dermatitis | α = 0.84 | + | |
| Twiss 2012(59) | United Kingdom | 292 | Psoriasis and Atopic dermatitis | The Person Separation Index (PSI) indicated that the DLQI had adequate internal reliability. | + | |
| An 2013(60) | China | 395 | Neurodermatitis or psoriasis vulgaris | Cronbach’s α =0.889. Average inter-item correlation = 0.415, item-total correlation ranged from 0.483 to 0.711, average item-total correlation was 0.628. | + | |
| He 2013(61) | China | 851 | Psoriasis | α = 0.91. Exclusion of any one of the 10 items did not increase a by more than 0.01. Corrected item-total correlations ranged from 0.51 to 0.79 | + | |
| Khoudri 2013(41) | Morocco | 244 | Psoriasis | Overall 0.70 (α = 0.84) and ranged in all scales from 0.33 to 0.75. Item internal congruency 0.82– 0.90. ICC 0.85–0.97 | + | |
| Lilly 2013(62) | United States | 90 | Vitiligo | Cronbach α = 0.935. Item-total correlations ranged between 0.56 and 0.84 except for VitiQoL question 13 (‘’Has your skin condition affected your sun protection efforts during recreation?’’) with a correlation of 0.36. | + | |
| Liu 2013(63) | China | 106 | Pruritic papular eruption | α = 0.673 for the six dimensions (Symptoms and feelings, Daily activities, Leisure, Work and School, Personal relationships and Treatment) were 0.633, 0.777, 0.771, 0.785, 0.772 and 0.684 respectively. | – | |
| Lockhart 2013(64) | United Kingdom | 85 | Vulval intraepithelial neoplasia | α = 0.93 | + | |
| Thomas 2014(65) | India | 38 | Lymphatic Filariasis | α = 0.73 | + | |
| Wachholz 2014(66) | Brazil | 41 | Leg ulcers | α = 0.729 | + | |
| Qi 2015(67) | China | 698 | Alopecia | α = 0.887, standardized item alpha was 0.881, The average inter-item correlation was 0.425 (> .2), suggesting good reliability. The item total correlation ranged from 0.180 to 0.797. The average item total correlation was 0.617 | + | |
| Chernyshov 2016(68) | Ukraine | 126 | Psoriasis and atopic dermatitis | α =0.81 for AD and 0.86 for psoriasis | + | |
| Solgajová 2016(69) | Slovakia | 104 | Acne or atopic dermatitis | α = 0.82 | Note: Aghaei et al., 2004; (Liu) Zhibin et al., 2013 don’t give this value, it must be from this study! | + |
| Kirby 2017(70) | United States | 154 | Hidradenitis Suppuratvia | α = 0.90 | R, version 3.3.2 | + |
| Cozzani 2018(71) | Italy | 50 | Psoriasis and psoriatic arthritis | α = 0.90 (0.88–0.92 for items). Highest value for the item-test correlation (r = 0.89) was for item 9 (interpersonal relationships), while the lowest corresponded to item 6 (leisure; r = 0.31). α increased to 0.90 only with the deletion of item 5 (sociability) | + | |
| Hunt 2018(72) | Vietnam | 102 | Leprosy | α = 0.78 | + | |
| Shimizu 2018 (73) | Brazil | 116 | Alopecia | α = 0.87 | + | |
| Xiao 2018(74) | China | 465 | Arsenic-related skin lesions and symptoms | Cronbach’s α was 0.79, and the split-half reliability was 0.77 | + | |
| Beamer 2019(75) | United States | 40 | Radio-dermatitis | α = 0.69 with work and study item was removed from analysis because the variance was zero. Inter-item correlation from 0.10 to 0.66 | Removal of treatment subscale (item) would improve alpha by .15. | – |
| Patel 2019(33) | United States | 340 | Atopic dermatitis | Cronbach’s α = 0.89. Spearman rho Interitem correlations 0.30 to 0.62 | + | |
| Satti 2019(76) | Pakistan | 173 | Uremic pruritus | α = 0.71 | + | |
| Storck 2018(77) | Germany | 79 | Pruritus | α Paper based 0.80, iPAD electronic1 0.81, iPAD electronic2 0.81 | + | |
| Temel 2019(78) | Türkiye | 150 | Acne vulgaris (AV) or vitiligo, or alopecia areata (AA) | α acne vulgaris 0.812, vitiligo 0.329, alopecia areata 0.915 | + | |
| Demirci 2020(79) | Türkiye | 100 | Psoriasis | α =0.82 (SPSS 20.0) | + | |
| Jorge 2020(80) | Brazil | 1286 | 14 dermatoses. (Basal cell carcinoma Bullous disorders, Female alopecia Genital warts, Hidradenitis suppurativa, Leprosy, Melasma, Onychocriptosis, Photoaging, Psoriasis, Rosacea, Uremic pruritus, Urticaria, Vitiligo) | Total Cronbach’s α (CI 95%) 0.90 (0.89–0.91); 0.72–0.91 for individual diseases. If any item was excluded, Cronbach’s α for the total sample ranged from 0.87 to 0.89 . | Highlighed cultural difficulty of q9 (sexual life) within the population. IRT analysis indicates that q9 is most affected with severe HRQOL impact. | + |
| Paudel 2020(81) | Nepal | 149 | Urticaria | α = 0.88, standardised 0.89, and did not change with the deletion of any of the items. The interitem correlation matrix revealed that the Pearson’s correlation coefficients (r) ranged from 0.097 to 0.730. All items had a satisfactory correlation with each other. Items 1–4 α = 0.79, items 5–10 α =0.86 | + | |
| Meneguin 2021(45) | Brazil | 188 | Psoriasis, cellulitis/erysipelas, chronic ulcers and eczematous dermatosis, other dermatoses | α = 0.85 (CI 0.82–0.88) | + | |
| Pollo 2021(82) | Brazil | 281 | Psoriasis | α = 0.87 | + | |
| Kolokotsa 2022(83) | Greece | 150 | Acne | α = 0.80 | + | |
Data was extracted from referenced publications.
For test–retest reliability COSMIN: “+” ICC or weighted Kappa ≥ 0.70; “?” ICC or weighted Kappa not reported; “–” ICC or weighted Kappa < 0.70. The criteria are based on Prinsen et al.(17).
For internal consistency reliability COSMIN: “+” At least low evidence for sufficient structural validitya AND Cronbach’s alpha(s) ≥ 0.70 for each unidimensional scale or subscaleb; “?” Criteria for “At least low evidence for sufficient structural validitya” not met; “–” At least low evidence for sufficient structural validitya AND Cronbach’s alpha(s) < 0.70 for each unidimensional scale or subscaleb.
aThis evidence may come from different studies.
bThe criteria ‘Cronbach alpha < 0.95’ was deleted, as this is relevant in the development phase of a PROM and not when evaluating an existing PROM. The criteria are based on Prinsen et al. (17)