Background: Restless Legs Syndrome (RLS) is a common disease with an estimated overall prevalence of 3% based on findings from a recent meta-analysis. Due to the lack of a reliable Persian questionnaire for the diagnosis of RLS, this study aimed to translate and evaluate the content validity and reliability of the Persian version of the modified-Restless Legs Syndrome Diagnostic Questionnaire (m-RLS-DQ).
Methods: The m-RLS-DQ, a reliable and accurate tool for diagnosing RLS was selected. Three bilingual physicians translated it into Persian, then unified by the fourth expert. Eight specialists evaluated the content validity using a Likert scale for relevance. Reliability was tested by administering the final version to 10 RLS the patients. 
Results: The reliability index (i.e., Cronbach’s alpha coefficient) was also obtained as 0.78. The content validity was determined with the help of Microsoft Excel software, showing a Content Validity Index (CVI) of 0.98.
Conclusion: In this study, the appropriate validity and reliability of the Persian version of m-RLS-DQ was confirmed.
Keywords: Prevalence, Reproducibility of results, Restless legs syndrome, Surveys and questionnaires

Introduction 
Restless Legs Syndrome (RLS) is a common disease with an overall prevalence of 3%, according to a recent meta-analysis study and also another study reported that 34.5% of the participants aged over 40 years old were afflicted with this disorder (1). However, the prevalence of this condition is relatively higher in some population groups, including women (2). Restless leg syndrome is associated with numerous diseases, and there is evidence supporting the association of this condition with iron deficiency and renal dysfunction (3). This syndrome has also been observed in association with neurological disorders, for example, its prevalence has been reported as 27.8% in Multiple Sclerosis (MS) patients, 14% in those suffering from Parkinson’s disease (4,5), and 8% among all individuals referring to neurology clinics (6). Citing a well-designed cross-sectional study, the prevalence of RLS was reported to be 6% in Iran (7); however, a systematic review declared a prevalence of 30% in Iran (8). Restless leg syndrome refers to a condition in which the patient is urged to move or rub the foot in response to an unpleasant or painful sensation in the extremities. The symptoms are exaggerated in the evening or during rest but subside after walking or by moving the legs (9). This condition may become so severe that can trigger suicidal thoughts and behaviors in affected people (10). The correct diagnosis of RLS can be problematic due to ambiguous and non-specific symptoms, and unfortunately, many patients are misdiagnosed (11). Therefore, new diagnostic criteria emphasize ruling out other differential diagnoses such as postural pain, myalgia, etc. (9).
Various questionnaires have been developed to diagnose RLS, most of which are unable to rule out various differential diagnoses of this syndrome (12). Recently, a promising questionnaire has been developed with the ability to not only assess multiple differential diagnoses of RLS but also to identify the comorbidities associated with this syndrome (13). Compared to the conventional tools, the recent questionnaire has offered higher average sensitivity and specificity (14). As far as it is known, none of the mentioned recommended tools for RLS screening have been translated into Persian (12). Most studies conducted in Iran have used the IRLSS questionnaire (6,15), a tool that has not been modified for up-to-date diagnostic criteria and cannot rule out differential diagnoses (16). Thus, regarding the lack of a valid Persian language questionnaire for RLS diagnosis, this study aimed to translate it into Persian and validate the modified-Restless Legs Syndrome Diagnostic Questionnaire (m-RLS-DQ).

Materials and Methods 
This study took place in Iran in 2024. The most recent questionnaire (i.e., m-RLS-DQ) was selected for the diagnosis of RLS. The m-RLS-DQ is a diagnostic tool (not for severity) with item-based scoring and strong sensitivity (94.9%) and specificity (94.1%). It was redesigned to include RLS mimics, tested in English and Hindi, and validated for clinical use. It shows excellent psychometric properties and allows comorbid mimic identification. The English version of this questionnaire was independently translated into Persian by three physicians who mastered both Farsi and English. In the next step, these three translated versions were matched together and unified by the fourth person to finally compile a Persian version of the tool. To assess the content validity, the final Persian version of the questionnaire was provided to 8 medical specialists who were proficient in the diagnosis and treatment of RLS (movement disorder neurologist, psychiatrist, sleep specialist and physical medicine and rehabilitation specialists), and their comments on the questions in terms of appropriateness and relevance to the objective (i.e., the diagnosis of RLS) were received on a Likert scale. The experts assigned a score between 1 and 4 to each question based on the level of its relevance or need for modifications. The reliability of the final Persian version of the tool was assessed by providing it to 10 patients suffering from RLS. The inclusion criteria were having symptoms compatible to RLS including feeling restlessness or discomfort in the legs along with the urge to move them. 

Validity
A valid study is one that can measure the intended objective without any errors in the study design. The American Psychological Association (APA) defines the following categories for the type of validity: Construct Validity, Content Validity, Concurrent Validity and Predictive Validity. 
This research focuses on the second category. Reviewing the content of scale to determine content validity is one of the most important parts of evaluating the validity of a questionnaire. The goal of this evaluation is to determine whether the content of the questionnaire is capable of measuring the defined objective. Experts should consider three key steps when evaluating content validity:
1. Defining the domain that the tool’s content aims to measure.
2. Identifying the specific content area that each item aims to measure.
3. Comparing the tool’s content with the content of study.
Two qualitative and quantitative methods are used to determine content validity. Experts are asked to review the tool based on criteria such as grammar, wording, item allocation, and scaling for qualitative content analysis, and provided feedback for necessary revisions. Two indices are used for quantitative content analysis: CVR and CVI. Experts are asked to review each item on a three-point scale (essential, useful but not essential, and not essential) to calculate CVR.
The obtained CVR is compared to a corresponding table value for decision-making. If this value exceeds the table threshold, the item’s content validity is confirmed.
For CVI assessment, simplicity, relevance, and clarity are rated on a four-point Likert scale by 8 experts for each item (e.g., 1: irrelevant, 2: somewhat relevant, 3: relevant, and 4: highly relevant). CVI is calculated by aggregating the ratings of the items which have values of 3 or 4 and divided by the total number of experts. Items with a CVI score above 0.79 are accepted.

Internal consistency
The goal of internal consistency is to evaluate the correlation between items within a scale. This study used Cronbach’s alpha for evaluation of internal consistency. After receiving the experts’ opinions, required statistical methods were performed in Microsoft Excel software to calculate CVI. After gathering the answers from the patients, the reliability index (i.e., Cronbach’s alpha coefficient) was calculated using STATA 11.5 software (Stata Corp., College Station, TX, USA). Due to the low frequency of the disorder, the number of 10 was confirmed by the methodologist of the study for the initial reliability testing. The patients were chosen among all the psychiatry outpatients who were visited in Fajr clinic in Ilam in 3 months by a psychiatrist.

Results
The validity of the Persian Modified-Restless Legs Syndrome Diagnostic Questionnaire (m-RLS-DQ) was carefully evaluated by eight medical experts, who reviewed each item using a four-point Likert scale. CVR was calculated as shown in table 1; most items received a perfect CVR score of 1.0, indicating that all the experts agreed on their necessity. Only two items (Q21 and Q25) had a slightly lower score of 0.75, which still met the minimum required threshold, confirming their importance in the questionnaire.
CVI was also calculated to measure how relevant, clear, and simple the items were. The CVI of the questionnaire was 0.98, which is above the recommended 0.79 cutoff, indicating strong expert consensus on its effectiveness for diagnosing RLS.
Additionally, Cronbach’s alpha coefficient was calculated to assess the internal consistency of the questionnaire, yielding a value of 0.78. This result confirms that the items work well together and consistently measure what they are intended to assess.

Table 1.  Content validity measures for the persian m-RLS-DQ

Note: CVR was calculated for each item in the Persian m-RLS-DQ, along with the number of experts who rated each item as essential. CVR: Content Validity Ratio.

Discussion
The diagnosis of RLS is not always straightforward, and patients are misdiagnosed in many cases. The prevalence of this condition has been reported to be about 3% in most recent studies (2,11). A practical way to boost diagnostic accuracy for this condition is to use a validated and appropriate questionnaire with high sensitivity and specificity. Fortunately, numerous suitable questionnaires have been developed for this purpose so far (12); however, none of these possess Persian versions yet, a problem that was addressed in the present study. 
Other valid questionnaires are available for diagnosis of RLS, with acceptable psychometric properties. The validity and reliability RLS-Self-care Behaviour questionnaire (RLS-ScBq) were assessed in a study by Odzakovic et al which demonstrated that the eight-item RLS-ScBq can be used by healthcare professionals to assess the use and effectiveness of self-care activities in patients with RLS (17). The Cambridge-Hopkins diagnostic questionnaire for RLS (CH-RLSq) was another widely studied tool. In one study, the sensitivity and specificity of that were 87.2 and 94.4%, respectively, indicating that it provides a reasonable level of sensitivity and specificity for identifying RLS in population-based studies (18). While a sensitivity of 94.9% and a specificity of 94.1% for m-RLS-DQ was prepared and validated by Kumar et al in 2023 and offers a sensitivity of 94.9% and a specificity of 94.1% (13), representing particularly higher values compared to previous tools. In a systematic review and meta-analysis of 52 studies on RLS diagnostic tools by Fulda et al, the means of sensitivity and specificity were obtained as 88% and 90%, respectively (14).
The m-RLS-DQ questionnaire not only has appropriate psychometric properties but is also able to discern conditions mimicking RLS. Moreover, this tool is capable of detecting concomitant comorbidities associated with RLS in individuals. In a study by Kwatra et al, the differential diagnoses of RLS were reviewed, denoting conditions such as night-time leg cramps, postural pain, vascular disorders, vascular lameness, Akathisia, restlessness leg movement, neuropathies, and radiculopathy (11). Notably, the m-RLS-DQ contains queries related to most of the above-mentioned conditions. Considering the prevalence of RLS, it is important to pay attention to the Positive Predictive Value (PPV) of any given screening tool. In this study, the PPV of this questionnaire at a prevalence of 5% was obtained as 45.9% (95%CI=30.6, 60.8%). In the meta-analysis conducted by Fulda et al, the average value of PPV was obtained as 31% (14). These studies show that the m-RLS-DQ delivers a higher PPV compared to its previous counterparts.
In a 2016 systematic review, the prevalence of RLS was reported as 30% in Iran, which was significantly higher compared to the rate of 6% reported by a population-based study in the country conducted on 19,176 participants (7).
In the recent systematic review, the authors acknowledged that they included all the relevant studies regardless of their quality, which may be a possible explanation for this considerable gap. This discrepancy further confirms that no reliable and appropriate diagnostic tool for RLS is available in Iran, leading to the inclusion of possible differential diagnoses and the overestimation of the prevalence of this condition. 

Limitations
While m-RLS-DQ demonstrated strong content validity and reliability, it was validated using a small sample size. The study assessed content validity with eight medical specialists and tested reliability on ten patients, which may not fully represent the broader population. Additionally, as the questionnaire relies on self-reported symptoms, there is a risk of bias. Therefore, large-scale studies are necessary to confirm its generalizability, as well as its widespread applicability and reliability.

Conclusion
The result of the present study verified the applicability of the Persian version of the m-RLS-DQ questionnaire as a screening tool for RLS, and it can be used as a suitable instrument in epidemiological studies in Iran. The authors will be happy to provide this tool to Iranian researchers working on RLS.

Acknowledgement
The authors sincerely appreciate the support of the ethical committee of Ilam University of Medical Sciences for approving this study under the ethical code IR.MEDILAM.REC.1402.272. 
The authors sincerely thank Professor Gupta Ravi (MD, Ph.D., MAMS, Department of Psychiatry and Division of Sleep Medicine, All India Institute of Medical Sciences) for providing the original version of the questionnaire for translation and also for his further guidance. We are also deeply grateful to Dr. Naseh Yousefi, Dr. Elham Moinpour, Dr. Motahara Mirhaj, Dr. Parshad Pourzarabian, Dr. Amirhassan Habibi, and Dr. Masoume Zoghali for their help in preparing the Persian questionnaire.

Conflict of Interest
There was no conflict of interest in this manuscript.