This study showed that Hindi version of IRLS has good internal consistency, is clinically valid, and can be used with ease in the Hindi speaking population of India. This scale is also available in English, and has been translated in Brazilian Portuguese.[3
] Original version has shown good reliability of 0.95.[3
] It has been found to have good internal consistency and reliability (alpha = 0.76) in clinical trial.[12
] Brazilian Portuguese translation has shown a good reliability with the Cronbach's alpha score of 0.83.[11
] Our study shows that Hindi version is also having a good reliability (Cronbach's alpha score of 0.86). Also, the scale is clinically valid and has a high discriminative ability.[3
] Brazilian translation has not been tested on the control group.[11
] Our results have shown that when the standard procedure is followed to administer the scale, the scores differ significantly between RLS and non-RLS subjects.[10
] Non-RLS subjects score zero on all items because of the inclusion of the phrase “because of the RLS…” in all questions. However, when the scale is given to sleep disorder patients not having RLS, they marked some of the responses positively. This happened even when we explained the symptoms of RLS to all participants before asking them to fill the scale themselves.
During the translation procedure, we have changed some of the words according to the needs and understanding of the patients. Cha et al
] have suggested that decentring should be applied instead of direct translation of the scale for a cross - cultural population. These translations face few common problems like vocabulary equivalence, idiomatic equivalence, grammatical - syntactical equivalence, experiential equivalence and lastly conceptual equivalence.[13
] We have faced the problems related to vocabulary equivalence and conceptual equivalence. Cha et al
] recommend that vocabulary problems can be solved by finding closely similar words of the target language. Similarly, for better understanding, syntax of the language can be modified by deleting or adding words or phrases according to the need of the target language. Conceptual equivalence was dealt with testing the scale on ten patients (version 5) and changing the items according to local needs. Masuko et al
] had also followed the same approach while translating the scale in Brazilian - Portuguese language.
We have used the scale to sequentially assess the improvement in symptoms, and the scores were consistent with the clinical symptoms (data will be published separately). However, there was a poor correlation between test and retest scores. This type of reliability testing might sometimes give spurious results.[14
] There are many reasons for this problem - first, the scale measures problem in past week only; secondly, scores are affected by clinical improvement as it is a severity rating scale, rather than a diagnostic tool; thirdly, symptoms of RLS vary on day-to-day basis; and, lastly, RLS usually responds dramatically to the therapy. Translating a linguistic instrument is difficult for those places where a segment of population is illiterate. We have shown that the instrument was applicable in all strata of patients- illiterate, literate, rural as well as urban. Minor grammatical errors may be ignored for the better acceptability of the instrument.[11
] Hindi version was applicable in all subjects without any additional information being given to them at the time of administration of scale.
The strength of this study is the exclusion of subjects with conditions that may mimic RLS.[9
] Hence, the data reflects the applicability of scale in true RLS cases. In conclusion, we were able to translate the IRLS in Hindi successfully and the translated version has shown good reliability and clinical validity, when the instructions were followed.