Cross validation of the Turkish version of dermatology life quality index


Serap Öztürkcan Celal Bayar Üniversitesi Tıp Fakültesi Dermatoloji AD, 45020 Manisa, 45020 Turkey E-mail:


Background  The aim of this study was to test the linguistic validation of the Turkish version of the Dermatology Life Quality Index (DLQI) for Turkish speaking dermatology patients.

Methods  The DLQI is a 10-item dermatology specific index developed originally in English. The methodology of this study consists of four consecutive sections: Translation, cognitive debriefing, field testing and statistical analysis. Translation steps: (a) Two forward independent translations into Turkish, (b) reconciliation of these translations by a dermatologist, (c) backward translation of the consensus Turkish version by a bilingual person into its original language (English), (d) and comparing the original questionnaire with the backward translated one. Cognitive debriefing: Sessions were performed on five patients from each of the seven different dermatological diagnosis groups. Field testing: The final Turkish version on which the face validity was approved by specialists on a total of 79 inpatients/outpatients with various dermatological diagnoses treated at Celal Bayar University Hospital. Statistical analysis: Internal consistency (using Cronbach a) and item–total score correlations (Pearson correlation) were used for reliability analysis. Validity analysis was carried out by construct testing (principal components factor analysis), convergent (Pearson correlation) and (discriminate Student's t-test and Mann–Whitney U-test) validity, and SF-36 was used in parallel with DLQI in order to test convergent validity. The data were analyzed by the SPSS version 10.0 (SPSS Inc., Chicago, IL, USA) statistical package.

Results  The mean age of the patients in the study was 30.77 ± 15.91 years; the mean score of DLQI was 7.61 ± 6.12. The median of item–total correlation coefficient was found to be 0.66, within a range of 0.48–0.81. The internal consistency of the index was found to be highly sufficient (α = 0.85). The DLQI was found to be highly related to the physical domain of SF-36. Life quality score was found to be significantly low for the inpatients compared with outpatients (differential validity).

Conclusion  It was found that the Turkish version of the DLQI was an acceptable index for dermatologists and dermatology patients and, moreover, to be valid and reliable in a cross-sectional level. The responsiveness of the Turkish version of the DLQI needs to be tested further on a variety of dermatological conditions with different severities.