Assessment of Restless Legs Syndrome Features With Standardized Methods

Published on 12/04/2015 by admin

Filed under Neurology

Last modified 12/04/2015

Print this page

rate 1 star rate 2 star rate 3 star rate 4 star rate 5 star
Your rating: none, Average: 0 (0 votes)

This article have been viewed 1672 times

Chapter 20 Assessment of Restless Legs Syndrome Features With Standardized Methods

Diagnostic assessments and evaluation of therapeutic interventions in patients with restless legs syndrome (RLS) are almost exclusively based on the patient’s subjective reports and, in part, on their clinical interpretation by experts. Similar to the evaluation of psychiatric disorders like depression or anxiety, there is a need for standardized methods that contribute to an increase in the objectivity, reliability, and validity of such judgments by patients and physicians. Surprisingly, the development of standardized methods was delayed in RLS research, with publication of the first method not occurring until 2001,1 although treatment trials had been conducted for two decades, often using polysomnographic measures as outcome variables.

There is a broad potential demand for validated methods both in research and in routine practice. The most important demand is the valid diagnosis of RLS, which needs to identify the presence of the key RLS features, urge to move with or without unpleasant sensations, including their circadian distribution. If distressing symptoms are present, a cascade of impairments of the individual’s life may be activated. The main clinical effect of RLS is disturbed sleep with prolonged sleep latency, sleep fragmentation, and nocturnal sleep deprivation,2 which may result in daytime sleepiness. RLS is usually a chronic disorder with burdening symptoms, and together with continuous sleep problems and daytime somnolence may culminate in an impaired quality of life of the patients, which may also include the development of comorbid psychiatric disorders such as depression, anxiety, or adjustment disorders.36

This chapter provides an overview on standardized methods that have been developed for the assessment of RLS. In addition, we describe some methods from other disciplines that have been used in RLS research.

Assessment of Severity of Restless Legs Syndrome

Methodological Problems

Because RLS is a condition characterized largely by subjective symptoms of the patients, it has been important to develop and validate instruments that could be used to measure the severity of the syndrome based on patient rating. It is not yet clear whether severity of RLS is better assessed with items rating the frequency or those rating the intensity of symptoms. In frequency items, patients should report if they have experienced an RLS feature on scales ranging from “never in the last 12 months” to “6 or 7 times a week.” Severity items require a rating of intensity of RLS features on scales ranging from “not present” to “very severe”—either for the current situation or averaged across periods of time, usually 7 days. The number of categories between the extremes varies from one scale to the other. Although frequency questions are preferred in United States, scales developed in Europe make use of intensity ratings. The frequency question appears to be more easily answered objectively, because counting days with symptoms should be more reliable than assessing the severity of a symptom like urge to move. Both types of scaling are highly intercorrelated.7 Frequency-based scales might be poorly sensitive for treatment differences in patients with moderate to severe forms of RLS, but they are very useful for epidemiologic research.7 Intensity ratings are more easily influenced by the specific question wording.

A major use of severity scales is the evaluation of treatments for RLS. Quantitative changes between baseline and end of treatment are analyzed. One important question is how to translate scale changes into clinically relevant changes.8 In addition, different responder criteria (e.g., 50% improvement at end of therapy compared with the baseline value) or remitter criteria (end-of-therapy values below a certain threshold, or even an end-of-treatment score of 0) have been used in clinical trials (e.g., Stiasny-Kolster and colleagues.9). These threshold criteria, however, have not been validated by comparison with investigator-based ratings.10

A related concern for RLS severity scales is determining the margins for clinically relevant differences between active treatment and placebo (superior efficacy) or between two active treatments (equivalent or noninferior efficacy).

One way to define a noninferiority margin statistically is the 90% confidence interval of the error of measurement of the severity scale. The noninferiority margin should be within the confidence interval of the error of measurement of the severity scale. For example, when using reliability (Cronbach’s α =.93) and standard deviation (8.8 points) of the IRLS validation study (International Restless Legs Study Group Severity Scale7), the confidence interval calculates to mean ±2.98; thus, a noninferiority margin of 3 IRLS points is achieved. However, statistically developed margins deserve approval from a more clinical perspective.

The definition of a clinically relevant difference between two active treatments or an active treatment and placebo is hampered by largely varying outcomes of clinical trials with drug treatments. Differences in the IRLS between active therapy and placebo were reported between approximately 3 points (ropinirole, e.g., Walters and coworkers11) and more than 8 points (cabergoline, Stiasny-Kolster and coworkers9; rotigotine: Stiasny-Kolster and coworkers12 and Oertel and coworkers13). The margin for clinically relevant differences may be derived from a meta-analysis of all double-blind, randomized drug trials in RLS, which has not yet been conducted.


International Restless Legs Syndrome Severity Scale

The IRLS (Box 20-1) is a set of 10 questions based largely on features of RLS as proposed by the IRLSSG.14,15 The goal during scale development was to establish a scale that would examine various areas of clinical impact of RLS. The items use 5-point scales that are defined according to the content of the items, with 0 = “symptom not present” to 4 = “very severe intensity.” In a large-scale validation study,7 the IRLS showed excellent psychometric properties (high inter-rater reliability, high retest reliability and internal consistency, high criterion, convergent and discriminant validity). A total score was determined as the sum of the scores of the 10 items, which ranges between 0 = “non RLS” to 40 = “very severe RLS.” The authors of the scale proposed a categorical interpretation of the total score in five classes of severity (0 = no RLS, 1-10 = mild, 11-20 = moderate, 21-30 = severe, 31-40 = very severe RLS). A single-factor solution provided sufficient evidence that a summary score was a useful measure.

BOX 20-1 International Restless Legs Syndrome Study Group Rating Scale (IRLS)

Copies of this scale and its translations into several languages can be obtained from Mapi Research Institute, 27, rue de la Villette, 69003 Lyon, France; Fax: +33 (0) 472 13 66 8; e-mail:; Website: The scale is available free to university or foundation-based researchers; a modest fee is charged for commercial use. Individuals can also contact MAPI if they wish to produce their own translation to discuss methods and to list the new translation with MAPI.

In a separate factor analysis study of the 10 items of the IRLS on 516 self-identified RLS patients, two primary factors were identified—one with six items related to symptom severity and a second factor with three items related to impact of the symptoms on life.16 Recently, a further psychometric study determined that “option characteristic curves” (the probability of scoring different options for a given item as a function of overall IRLS score) and item characteristic curves (the expected score on an item as a function of overall IRLS score) were also generated with the data of a sample of 196 RLS patients. The IRLS items demonstrated excellent item response properties, with option and item characteristic curves closely approximating those of an ideal item. Item 3 (relief of arm or leg discomfort from moving around) was the most problematic item in that a “floor” effect was evident; however, the item response characteristics for this item were still acceptable.17 An additional study indicated that the scale was responsive to changing clinical status in a therapeutic trial.18 The IRLS has also been successfully validated against the number of periodic limb movements (PLM) during sleep on a sleep study or waking during a suggested immobilization test.19

The IRLS has become the “gold standard” in severity assessment in RLS research, especially in clinical drug trials. In all published trials, a difference between active treatment and placebo in changes of the IRLS total score between baseline and end of treatment could be demonstrated that supports the sensitivity of the scale for treatment differences. Nonetheless, it has been argued that the IRLS is prone to large placebo effects.20

Johns Hopkins Restless Legs Severity Scale

The Johns Hopkins Restless Legs Severity Scale (JHRLSS) was first published clinical scale to assess severity of RLS. It consists of one item questioning for the usual (>50% of days) time of the day that symptoms started. A 4-point scale is provided for the patient’s rating (Box 20-2) depending on the time of onset of symptoms: earlier onset is more severe. It has generally been indicated that this measure should apply to subjects with at least near-daily symptoms. The JHRLSS has been validated against objective measures of RLS severity such as sleep efficiency and periodic limb movements.1 It showed high inter-rater agreement and significant correlation with the sleep and movement variables. The JHRLSS measure is a useful, easily used clinical assessment of RLS severity. Like other frequency- or time-based methods, its primary domain is screening and epidemiologic research, but it has also been used to assess severity in pathophysiologic studies.21 As in the case of other frequency or time-based methods, its primary domain is screening and epidemiologic research but not treatment evaluation, with the exception of augmentation. It does have the limitation that it was designed to evaluate RLS patients with daily or near-daily symptoms.

RLS-6 Severity Scales

The RLS-6 scales22 have been in use since the early 1990s23; however, over the years, the scales have been extended and modified. The six RLS-6 scales use 11-point scales (0 = “not present” to 10 = “very severe”) to establish a severity profile of different night and daytime periods (Box 20-3

Buy Membership for Neurology Category to continue reading. Learn more here