Translation and validation of restless leg syndrome quality of life questionnaire in Hindi language
Correspondence Address: DOI: 10.4103/0028-3886.103188 PMID: 23135023
Background: Restless legs syndrome (RLS) itself as well as the problems that are secondary to RLS may worsen the quality of life. Aim: The aim of this study is to translate and validate the Restless Legs Syndrome Quality of Life (RLS-QOL) questionnaire in Hindi language. Settings and Designs: Patients attending psychiatry OPD and sleep clinic in a tertiary care teaching hospital. Materials and Methods: Thirty four consecutive patients of RLS and twenty nine control subjects were included in the study. Permission for translation and validation of RLS-QOL questionnaire scale was obtained. Translation was done according to the guidelines provided by the publisher. After translation, the final version of the scale was applied in both the groups to find the reliability and validity. Statistical Analysis: SPSS, version 17.0 was used for the analysis. Independent sample t test was used to compare age. Chi-square test was applied to compare non-parametric variables. Pearson's and Spearman's correlations were used to find out the correlation between parametric and non-parametric items, respectively. Reliability analysis was done by using Cronbach's alpha. Results: Among the RLS subjects, mean Insomnia Severity Index (ISI) score, International Restless Legs Syndrome Severity Rating Scale (IRLS) score and Computed Score of RLS-QOL questionnaire were 25.43 (7.39), 12.7 (8.34) and 29.8 (8.39), respectively. A statistically significant difference was observed between both the groups on all these scores (ISI: t= −11.2, P<0.001, 95% CI= −22.62 to −15.76; IRLS: t = −8.1, P< 0.001, 95% CI =−15.81 to −9.58; RLS-QOL: t =−19.07, P<0.001 with 95% CI = −32.69,- 28.83). A significant correlation between ISI and RLS-QOL (r 2 = 0.59; P< 0.001) was seen. Most of the items also showed good correlation with each other. Internal consistency done by Cronbach's alpha showed good correlation (0.85). Conclusion: Hindi version of the Restless Legs Syndrome Quality of Life (RLS-QOL) questionnaire is a valid and reliable tool for the assessment of the quality of life in patients with RLS.
Keywords: Insomnia severity index, restless legs syndrome, restless legs syndrome quality of life, translation, validation
Restless legs syndrome (RLS) is characterized by an irresistible urge to move the legs, which usually begins or worsens during periods of rest. Physical activity such as walking often relieves this urge. ,, The symptoms of RLS mostly begin in the late evening and continue throughout night. Resultant problems such as difficulty in initiating or maintaining sleep worsen sleep efficiency and cause daytime fatigue. Reduced concentration and poor motivation often accompanies, which may mimic depression or anxiety. , RLS itself, as well as problems that are secondary to RLS cumulatively worsen the quality of life. For this reason, the Restless Legs Syndrome Quality of Life questionnaire (RLS--QOL), a patient-reported measure of quality of life specific to RLS, was developed and validated.  This questionnaire measures four domains including the overall life impact score, employment, sexual interest and work. This has been translated in many languages including Dutch, Finnish, French, German, Greek, Hungarian, Italian, Japanese etc. 
The exact prevalence of RLS in India is not known however, hospital based studies have reported the incidence to be 6.25% in non-anaemic subjects and 34.37% in anaemic patients.  Taken together, these facts suggest a large proportion of Indian population might be suffering from RLS. Disease specific and validated questionnaires are required for the clinical assessment of illness, to keep a track of improvement after treatment and for research purpose. As already mentioned above, RLS-QOL questionnaire is not available in Hindi which is the national language in India. Hence, we planned to translate and validate this scale in Hindi language.
Patients attending outpatients departments in Psychiatry, Neurology and sleep clinic at a tertiary care teaching institute were included in the study after an approval by the institutional ethics committee. Thirty four consecutive subjects with RLS were Included and diagnosed as per the criteria proposed by International-RLS-Research-Group.  Control group consisted of 29 subjects suffering from conversion disorder or transient insomnia or recent onset unspecified headache.Informed consent was obtained from all the study subjects. However, subjects suffering from co-morbid severe depression, opioid/opiate withdrawal, sciatica, osteoarthritis of knee, rheumatoid arthritis affecting leg joints, myelopathies, neuropathies, uremia and end-stage-renal-disease were excluded. Similarly subjects taking neuroleptics, antidepressants, anti-parkinsonian drugs or opioids were also not included. In addition, pregnant female and those not willing to participate in the study were also excluded.Subjects in both the groups were included irrespective of their gender, educational status and domicile (urban or rural). Each of the question (From the Hindi version of RLS-QOL questionnaire) was read aloud (to control for the illiteracy) by the investigator and response of the patient was recorded as mentioned in the questionnaire.
International Restless Legs Syndrome Severity Rating Scale- Hindi version (IRLS) , Insomnia Severity Index-Hindi Version (ISI) , and Hindi translation of Restless Legs Syndrome-Quality of Life (RLS-QOL) questionnaire  were used to judge the severity of RLS, sleep disruption and quality of life affected by RLS respectively after seeking the permission from the authors/distributors.
Restless leg syndrome quality of life: Original instrument
Restless Leg Syndrome Quality of Life (RLS-QOL) questionnaire is a copy-right to the MAPI Research Trust, France.  After obtaining written permission, translation was done as per the instructions provided by the publisher.
The original scale was translated independently by two persons proficient in both English and Hindi. This resulted in two Hindi versions. Both the translators discussed the language of their respective versions, to keep it closest to the original version and came out with a third common Hindi version. This version was discussed with the clinicians having experience in dealing with RLS patients. The requisite changes were then made and was back-translated to the English (version 4) by another translator well-versed with both the languages. This was matched with the original scale to check for the consistency of the items. Wherever, there was a gross discrepancy issue was discussed and a change was made in the third version so that it communicated what was mentioned in original version. This brought the Hindi version (version 5) ready for validation.
Analysis was done using SPSS version 17.0 (SPSS Inc. Chicago, IL, USA). Descriptive statistics was calculated for the categorical variables. Independent sample t test was used to compare age. Chi-square test was applied to compare non parametric variables. Pearson's correlation was used to find out the correlation between parametric items whereas Spearman's correlation coefficient was calculated for non-parametric items of the questionnaire (Item Numbers 1, 2, 3, 4, 5, 7, 9, 10,13 and 14). Reliability analysis was done by using Cronbach's alpha.
The study included 34 cases and 29 controls. In RLS group and the control group, 64.7% and 62.1% subjects were females, respectively. Average age of subjects in the RLS group was 38.61 (14.64) years and 39.58 (9.72) years in the control group. Both the groups were similar with respects to the gender distribution and age (X 2= 0.04, P = 0.82 and t = 0.3, P = 0.76, respectively).
Among the RLS subjects, mean ISI score, IRLS score and computed score of RLS-QOL-questionnaire were 25.43 (7.39), 12.7 (8.34) and 29.8 (8.39), respectively. The control group scored zero on IRLS and RLS-QOL- questionnaire because items in both these scales use the phrase "because of RLS". However, control group scored 6.24 (5.95) on the ISI. A statistically significant difference was observed between both the groups on all these scores (ISI: t = −11.2, P < 0.001, 95% CI = −22.62 to −15.76; IRLS: t =-8.1, P < 0.001, 95% CI =−15.81 to −9.58; RLS-QOL: t =−19.07, P < 0.001, 95% CI = −32.69 to −28.83). Correlation analysis depicted a significant correlation between ISI and RLS-QOL-questionnaire scores (r 2 = 0.59; P< 0.001). However, we could not find any correlation between IRLS and RLS-QOL- questionnaire scores.
Translation of scale
A number of problems were encountered during the translation procedure. Most of them resulted from binding to keep the language simple and colloquial. We do not have any literal word in Hindi for "distressing". When used in Hindi, it denotes the disturbance that a person feels. Hence, in the questions 'distressing' was changed to disturbance (pareshani). Secondly, literal translation of 'restless legs' did not carry any meaning in Hindi. Hence, it was changed to restlessness in legs (paavon kee baichaini).
Similarly, literal Hindi translational for the word "disrupt" produces a word that is not used in day to day language. Hence, this was simplified to 'dikkat' (interference) to maintain the flow of language. "Your first appointment of the day" was omitted as it was interfering with the flow of the language and most of the female subjects were housewives. Resultantly, we had to change it to 'subah kaam shuru karne mein deri' (delay in starting the work in the morning) in item numbers 5 and 6. 'Pareshani' (problem) was changed to 'dikkat ya kami' (disturbance or reduction) in item number 12. Language was modified for the sake of simplicity in item number 18. These changes were not gold standard but most appropriate for the subjects to understand properly as discussed with the treating physician and patients while filling up the questionnaire (Appendix 1).
Hindi version of RLS-QOL questionnaire
In the Hindi version of RLS-QOL questionnaire, most of the items showed good correlation with each other [Table 1]. Other non-parametric items of the scale (items 11 and 12) also showed good correlation with each other (Spearman's rho = 0.7, P<0.001). Items 6 and, 15 to 18 were analysed separately as they depict interference with daily activities because of RLS and most of them were correlated with each other [Table 2]. The Hindi version showed good internal consistency (Cronbach's alpha= 0.85).
Our study has shown a good reliability score (Cronbach's alpha = 0.85) of the translated version. This is in concordance with the results of earlier studies. , Good correlation between different items show good internal consistency as this scale measures different aspects of RLS viz., overall life impact, employment, sexual interest and work.
Translation is the most common method for preparing study instruments. It has been observed during the process of translation that word by word translation is often not possible and may be a source of misunderstanding.  In our study we came across a number of problems during the translation procedure. So, several changes were made to maintain the flow of the language as well as make the language simple and appropriate for the subjects to understand properly. This is to mention here that this approach is not new and has been advised in the past as well during cross-cultural translation. ,
In our study we found female predominance confirming the result of earlier study.  A significant correlation was found between scores on ISI and RLS-QOL questionnaire scores. This showed that poor quality-of-life in RLS patients was mostly due to disturbance in sleep as seen in previous study.  However, we could not find any significant association between IRLS score and scores on RLS-QOL questionnaire. This could be attributed to the difference in the duration of symptoms examined by both instruments. While the RLS-QOL questionnaire assessed the symptoms during past 4 weeks from the date of interview, the IRLS assessed severity of symptoms during past one week only. Because of the variability of RLS both the scores may differ and this shows that none of the instruments in question is a substitute for the other. ,
This study had some methodological limitations. First, this questionnaire can be difficult to be used in non-Hindi speaking population. We suggest translation of this instrument in other Indian languages. Second, this questionnaire cannot be used in children as their level of understanding the items may vary. In conclusion, Hindi version of the Restless-Legs-Syndrome-Quality-of-Life-Questionnaire is a valid and reliable tool for the assessment of the quality of life for patients having RLS.
We are thankful to Dr. Charles Morin and MAPI research trust for allowing us to use the instruments used in this study.
[Table 1], [Table 2]