Juliana Meyers, MA
Associate Director, Health Economics, RTI Health Solutions, Research Triangle Park, N.C.
Sean Candrilli, PhD
Head, Health Economics Data Analytics, RTI Health Solutions
Richard Allen, PhD
Associate Professor, Department of Neurology, Johns Hopkins Bayview Medical Center, Baltimore, Md.
Ranjani Manjunath, MPH
Manager U.S. Health Outcomes, GlaxoSmithKline, Research Triangle Park, N.C.
Michael Calloway, PhD
Manager U.S. Health Outcomes, GlaxoSmithKline

The researchers find that the costs of treating restless legs syndrome are fairly low, relative to the improved symptoms and associated health care outcomes, and are mainly attributable to prescription medication. Health plans are encouraged to expand coverage to reduce the associated suffering and costs.

Juliana Meyers, MA

Associate Director, Health Economics, RTI Health Solutions, Research Triangle Park, N.C.

Sean Candrilli, PhD

Head, Health Economics Data Analytics, RTI Health Solutions

Richard Allen, PhD

Associate Professor, Department of Neurology, Johns Hopkins Bayview Medical Center, Baltimore, Md.

Ranjani Manjunath, MPH

Manager U.S. Health Outcomes, GlaxoSmithKline, Research Triangle Park, N.C.

Michael Calloway, PhD

Manager U.S. Health Outcomes, GlaxoSmithKline


Purpose: This study assessed the direct economic burden of restless legs syndrome (RLS) among patients treated with dopamine agonists (DAs) using a large United States managed care database.

Design: Retrospective database analysis.

Methodology: Patients were required to have ≥1 prescriptions for a DA (i.e., pergolide, pramipexole, ropinirole) between 1/1/2005 and 12/31/2007 (date of first DA, or “index”); continuous enrollment for ≥6 months before and ≥12 months after index; >1 diagnosis of RLS, before and after index; and no diagnosis of Parkinson’s disease. Study measures included annual all-cause and RLS-related costs by care setting (hospitalizations, emergency room, office, pharmacy, other, total) and treatment-pattern events (discontinuations, switches, adjunctive treatments, titrations).

Principal findings: A total of 7,796 patients met the inclusion criteria. About 70% of patients received ropinirole, and 30% received pramipexole at index. Approximately 91% had ≥1 RLS-related office visits, and patients filled an average of 6.5 RLS-related prescriptions (DAs, gabapentin, carbidopa/levodopa) during the 1-year follow-up period. Mean (SD) all-cause health care costs were $11,485 ($21,362) per patient, mostly due to multiple medical conditions occurring with RLS. RLS-related costs were 6.7% of total all-cause costs (mean [SD] $774 [$1,504]), consisting of office visits (16%), pharmacy (63%), and other costs (20%). Approximately 58% had a treatment-pattern event suggesting a dopamine-related side effect. Opioids were the most commonly used adjunctive therapy (13% of patients).

Conclusion: We found relatively low costs associated with RLS treatment. These findings should encourage expanding the coverage of treatment to reduce the suffering and costs associated with RLS.


Restless legs syndrome (RLS) is a chronic neurological disorder characterized by an urge to move the legs, often accompanied by unpleasant leg sensations (e.g., crawling, tingling). Prevalence estimates for RLS generally range between 2% and 15% of the general population; however, the disorder is often unrecognized or misdiagnosed (Evidente 1999; Milligan 2002; Allen 2005; Allen 2010; Allen 2011). The incidence of RLS tends to increase with age and is higher in women than in men (Allen 2005). Furthermore, several medical conditions significantly increase the risk for developing RLS (e.g., pregnancy, need for hemodialysis, iron deficiency anemia, multiple sclerosis, and Parkinson’s disease).

Numerous double-blind, placebo-controlled clinical trials have found that dopamine agonists (DAs) are effective in reducing moderate to severe RLS symptoms (Montplaisir 1999; Allen 2004; Trenkwalder 2004 a; Bogan 2006; Montplaisir 2006; Partinen 2006; Winkelman 2006b; Garcia-Borreguero 2007; Trenkwalder 2008). Additionally, numerous treatment guidelines recommend DAs as first-line therapy for the treatment of RLS (Sibler 2004; Vignatelli 2006; Hening 2007).

Symptoms of RLS are intensified by inactivity and have a circadian pattern in which they worsen in the evening. Patients with RLS often have difficulty falling and staying asleep, resulting in reduced sleep duration and quality as well as impairments in subsequent daytime performance (Phillips 2006). Long-term implications of RLS include psychological distress and diminished health-related quality of life (Allen 2001; Allen 2005). Eisensehr and colleagues (2007) estimated that RLS patients experienced a 30% impairment in productivity while working, and Allen and colleagues found that individuals with RLS symptoms severe enough to be classified as sufferers experienced an average of 1 day per week of lost work productivity (Allen 2011).

Although RLS symptoms can have a profound and negative impact on individuals whose symptoms are severe enough to warrant pharmacological treatments (i.e., moderate to severe symptoms), there are limited data about either the real-world health care resource utilization or the costs associated with treating RLS in managed care. The primary objective of this retrospective database analysis was to evaluate resource use and costs associated with RLS among managed care enrollees in the United States who have been treated with a DA. A secondary objective of this analysis was to evaluate the incidence of side effects associated with DA treatment.


Data for this analysis were taken from the LifeLink database (formerly the PharMetrics Integrated Outcomes database), a national claims database encompassing 95 U.S. managed care organizations covering over 61 million lives between 1997 and 2008. This database contains patient-level demographics; periods of health plan enrollment; primary and secondary diagnoses; detailed information about hospitalizations, diagnostic testing, and therapeutic procedures; inpatient and outpatient physician services; and prescription drug use. Additionally, cost data, in the form of managed care reimbursement rates, are provided for each service within the database. Patients are tracked longitudinally via unique de-identified patient numbers. In compliance with the Health Insurance and Portability and Accountability Act of 1996, all data in the database have been de-identified to protect the privacy of individual patients, physicians, and hospitals. RTI International’s institutional review board determined that this study met all criteria for exemption.

Patients were selected for inclusion if they had at least one prescription for a DA recognized as effective in treating RLS (i.e., pergolide, pramipexole, and ropinirole) between January 1, 2005, and December 31, 2007. To ensure that treatment was for RLS and not another underlying condition (e.g., DAs also are indicated for Parkinson’s disease), all patients also were required to have at least one RLS diagnosis on or before their first prescription for a DA and at least one RLS diagnosis in the 12-month period following their first prescription. RLS diagnosis was identified using International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) codes corresponding to RLS (333.94) or neuroleptic-induced acute akathisia (333.99). DAs were identified using generic and brand names. Patients with a diagnosis of Parkinson’s disease (ICD-9-CM codes 332.xx) at any point between Jan 1, 2005, and Dec 31, 2008, were excluded from the analysis due to overlapping drug indications for Parkinson’s disease and RLS.

The index date for each patient was defined as the date of the first observed prescription for a DA. Patients were included only if they had at least 6 months of continuous health plan enrollment pre-index and at least 12 months of post-index eligibility. These times sufficed to address the research questions and ensured the patients’ enrollments in their health plans covered the time periods of the study.

Patient characteristics measured at the index date included age, geographic region, and index DA (i.e., pergolide, ropinirole, or pramipexole), defined by the first observed DA prescription. The impact of medical conditions other than RLS (comorbid conditions) was assessed during the 6-month pre-index period, using an established algorithm, the Charlson Comorbidity Index score (Charlson 1987). This score is made up of 17 comorbidities (defined by ICD-9-CM diagnosis and procedure codes), such as myocardial infarction and chronic pulmonary disease, which are weighted to correspond to the severity of the comorbid condition of interest. A higher comorbidity score represents a higher overall comorbidity burden during the pre-index period.

Health care utilization and costs were examined over the 12-month period immediately following the index DA. For each patient, all-cause health care utilization and other associated costs during the 12-month post-index period were aggregated. Costs were assessed as the amounts paid by health plans. The following settings of overall health care utilization and costs were evaluated and reported: inpatient stays, skilled nursing facility stays, emergency room (ER) visits, office visits, pharmacy, other outpatient or ancillary care, and a grand total. For each setting, the number and percentage of patients with at least one claim, the number of unique claims, and the per-patient costs were reported. Additionally, for inpatient and skilled nursing facility stays, the total number of days in each setting was reported for each patient with at least one stay. Total health care utilization and associated costs represented the sum of the individual cost components, including each prescription medication filled, and every encounter with a health care professional for which a claim was generated. For inpatient services, each unique hospital stay and all claims generated within that stay were counted as one encounter.

The total number and associated costs of health care services specifically related to RLS during the 12 months post-index were similarly evaluated. All inpatient hospital confinements with a primary discharge diagnosis of RLS were considered RLS related. RLS-related office, ER, and other visits were identified by searching for medical claims with both a diagnosis of RLS (primary or secondary) and a relevant place-of-service code for the care setting of interest. RLS-related medications included all DAs, gabapentin (when no neuropathy diagnosis was present in the pre- and post-index periods), and carbidopa/levodopa. All RLS-related medications were identified using appropriate National Drug Codes, Healthcare Common Procedure Coding System codes, and brand and generic drug names.

DA-related side effects included augmentation (i.e., worsening of RLS symptom severity during treatment) and compulsive behaviors. Because ICD-9-CM codes do not exist for these side effects, DA treatment patterns were assessed using a proxy measure. Specific treatment-pattern events that may have indicated a medication-related side effect included DA dosage increases or decreases and switches to or adjunctive treatment with gabapentin (with no concomitant diagnosis of neuropathy at any point during pre- and post-index periods), opioids, or pregabalin. To allow for sufficient time for dose adjustments of the index DA (i.e., titration period), these treatment-pattern events were limited to those (found in patients’ claims records) that had occurred more than two months after the index date. Dose increases or decreases were identified by any change in the patient’s average daily dose at least 2 months after starting treatment. A medication switch or addition was identified when a patient received gabapentin (with no concomitant diagnosis of neuropathy), an opioid, or pregabalin following his or her index date.

Analyses were carried out using SAS (Version 9) statistical software. Descriptive analyses were conducted for all outcome measures and included means and standard deviations (SDs) for continuous variables of interest (e.g., number of inpatient admissions, costs) and for frequency distributions of categorical variables of interest (e.g., health plan type). All costs were updated to 2008 U.S. dollars using the medical care component of the Consumer Price Index.

Multivariate logistic regression analyses were conducted to assess predictors of any treatment-pattern event (i.e., any dosage increase or dosage decrease, switch, or adjunctive treatment). Independent variables included age group (i.e., 45–54 years, 55–64 years, and 65+ years vs. 18–44 years [reference category]), gender (i.e., female vs. male [reference category]), geographic region (i.e., South, Midwest, and West vs. East [reference category]), the type of DA received at index date (i.e., pramipexole vs. ropinirole [reference category]), and the Charlson Comorbidity Index score (i.e., between 1 and 2 or > 2 vs. < 1 [reference category]). Regression results were reported as odds ratio (OR) estimates.


A total of 7,796 patients met the selection criteria (Table 1). Patients were, on average, 54 years old, 67% were female, and 46% were from the Midwest. Drugs prescribed on the index date were ropinirole (71% of patients), pramipexole (29%), and pergolide (less than 1%). The mean (SD) Charlson Comorbidity Index score was 0.7 (1.3). The most common comorbidities included chronic pulmonary disease (12.96%) and diabetes without chronic complications (12.08%).

Table 1 Characteristics of the study sample
Characteristic N/mean Percentage/SD
Total sample 7,796  
Age (N, %)    
< 18 years 0 0.00%
18–25 years 65 0.83%
25–34 years 388 4.98%
35–44 years 1,222 15.67%
45–54 years 2,435 31.23%
55–64 years 2,376 30.48%
≥ 65 years 1,310 16.80%
Mean (SD) 54.05 (12.77)
Gender (N, %)    
Male 2,539 32.57%
Female 5,257 67.43%
Geographic region (N, %)    
East 1,186 15.21%
South 1,390 17.83%
Midwest 3,559 45.65%
West 1,661 21.31%
Dopamine agonists received at index date (N, %)    
Pergolide 11 0.14%
Pramipexole 2,291 29.39%
Ropinirole 5,495 70.48%
Carbidopa/levodopa1 9 0.12%
Charlson Comorbidity Index score2    
Mean (SD) 0.70 (1.32)
Common comorbidities (N, %)3    
Chronic pulmonary disease 1,010 12.96%
Diabetes without chronic complications 942 12.08%
Cancer 362 4.64%
Liver disease 357 4.58%
Cardiovascular disease 307 3.94%
SD = standard deviation.
1 Use of carbidopa/levodopa alone was not sufficient for study inclusion because it is recommended only for those patients with infrequent RLS symptoms. Patients with carbidopa/levodopa at index also received another concomitant DA.
2 Evaluated in the 6-month pre-index period.
3 Percentage reported among all patients.

Mean (SD) all-cause total health care costs per patient in the 1-year follow-up period were $11,485 ($21,362) (Table 2). The largest components of overall total health care costs were pharmacy (mean [SD], $3,546 [$6,483]), other outpatient services (mean [SD], $4,441 [$9,566]), and inpatient services (mean [SD], $2,306 [$13,248]). Approximately 16% of patients were hospitalized, and the average length of stay was 7.6 days. Furthermore, 29% of patients had an ER visit; and 99% of patients had at least one physician visit, with an average of 19 visits in the 12-month follow-up period.

Table 2 Descriptive summary of overall health care utilization and costs during the 12-month follow-up period
Component N/mean Percentage/ SD
Inpatient services    
Had ≥ 1 hospital admission (N, %) 1,242 15.93%
Mean number of unique admissions (SD) 0.23 (0.67)
Mean length of stay (SD)1 7.59 (13.14)
Mean (SD) total costs $2,306 ($13,248)
SNF stays    
Had ≥ 1 SNF admission (N, %) 73 0.94%
Mean number of unique SNF admissions (SD) 0.01 (0.16)
Mean length of stay (SD)1 28.95 (37.35)
Mean (SD) total costs $52 ($1,141)
ER visits    
Had ≥ 1 ER visit (N, %) 2,254 28.91%
Mean number of ER visits (SD) 1.57 (5.74)
Mean (SD) total costs $209 ($808)
Office visits    
Had ≥ 1 office visit (N, %) 7,705 98.83%
Mean number of office visits (SD) 18.79 (25.25)
Mean (SD) total costs $930 ($1,269)
Had ≥ 1 prescription (N, %) 7,796 100.00%
Mean number of prescriptions obtained (SD) 48.94 (38.96)
Mean (SD) total costs $3,546 ($6,483)
Other outpatient services    
Had ≥ 1 other outpatient services (N, %) 7,676 98.46%
Mean number of other outpatient services (SD) 46.44 (54.05)
Mean (SD) total costs $4,441 ($9,566)
Total health care utilization    
Had ≥ 1 medical encounter (N, %) 7,796 100.00%
Mean number of encounters (SD) 120.90 (102.00)
Mean (SD) total costs $11,485 ($21,362)
ER = emergency room; SD = standard deviation; SNF = skilled nursing facility.
1 Mean length of stay estimated among only those patients with at least one unique admission.

Mean (SD) total annual RLS-related costs per patient were $774 ($1,504), which represented approximately 7% of overall total annual health care costs (Table 3). Almost 92% of patients had an RLS-related physician office visit, with an average of 1.9 RLS-related office visits per patient in the 12-month follow-up period. Additionally, patients received, on average, 6.5 prescriptions for an RLS-related medication in the 12-month follow-up period. RLS-related costs occurred mostly from prescriptions (mean [SD], $485 [$598]), office visits (mean [SD], $123 [$135]), and other outpatient services (mean [SD], $157 [$1,354]).

Table 3 Descriptive summary of RLS-related1 health care utilization and costs during the 12-month follow-up period
Component N/mean Percentage/SD
RLS-related inpatient stays    
Had ≥ 1 hospital admission (N, %) 14 0.18%
Mean number of unique hospital admissions (SD) 0.00 (0.04)
Mean length of stay (SD)2 1.29 (0.61)
Mean costs (SD) $0 ($6)
RLS-related SNF stays    
Had ≥ 1 SNF admission (N, %) 0 0.00%
Mean number of unique SNF admissions (SD)
Mean length of stay (SD)2
Mean costs (SD)
RLS-related ER visits    
Had ≥ 1 ER visit (N, %) 159 2.04%
Mean number of ER visits (SD) 0.07 (1.19)
Mean costs (SD) $9 ($133)
RLS-related office visits    
Had ≥ 1 office visit (N, %) 7,131 91.47%
Mean number of office visits (SD) 1.89 (1.52)
Mean costs (SD) $123 ($135)
RLS-related pharmacy (N, %)    
Had ≥ 1 prescription for pramipexole 2,687 34.47%
Had ≥ 1 prescription for ropinirole 5,761 73.90%
Had ≥ 1 prescription for pergolide 18 0.23%
Had ≥ 1 prescription for carbidopa/levodopa 201 2.58%
Had ≥ 1 prescription for gabapentin 422 5.41%
Any RLS-related pharmacy    
Had ≥ 1 prescription (N, %) 7,796 100.00%
Mean number of prescriptions obtained (SD) 6.46 (4.77)
Mean costs (SD) $485 ($598)
RLS-related other outpatient services    
Had ≥ 1 other outpatient services (N, %) 3,460 44.38%
Mean number of other outpatient services (SD) 2.03 (4.06)
Mean costs (SD) $157 ($1,354)
RLS-related total health care utilization    
Had ≥ 1 medical encounter (N, %) 7,796 100.00%
Mean number of encounters (SD) 10.46 (6.93)
Mean costs (SD) $774 ($1,504)
ER = emergency room; RLS = restless legs syndrome; SD = standard deviation; SNF = skilled nursing facility.
1 Inpatient and SNF stays were RLS related if there was evidence of a primary diagnosis of RLS. ER visits, physician office visits, and other outpatient services were RLS related if there was evidence of any diagnosis of RLS. Prescriptions were considered RLS related if they were for pramipexole, ropinirole, pergolide, carbidopa/levodopa, or gabapentin (with no diagnosis of neuropathy).
2 Mean length of stay estimated among only those patients with at least one unique admission.

Treatment-pattern events, serving as a proxy measure for medication-related side effects, occurred in approximately 58% of patients (Table 4). Overall, 43% of patients had a medication use change; 13% had adjunctive treatment; 13% had a dosage increase; and 3% had a dosage decrease. The most common medication that replaced or was added to a DA was an opioid (in 88.96% of patients who discontinued the DA and in 95.7% of patients who added to the DA). The mean (SD) time to change to a non-DA was 132.60 (99.52) days, and the mean (SD) time to add to a DA was 141.80 (112.20) days. Among patients with a change in their index DA dose, the mean (SD) time to a dosage increase was 169.3 (84.4) days, and the mean (SD) time to a dosage decrease was 176.8 (87.5) days.

Table 4 Incidence of treatment pattern events during the 12-month follow-up period
Treatment-pattern event N/mean Percentage/SD
Dosage increase    
Patients with a dosage increase (N, %) 982 12.60%
Mean (SD) days to dosage increase1 169.30 (84.43)
Dosage decrease    
Patients with a dosage decrease (N, %) 255 3.27%
Mean (SD) days to dosage decrease1 176.80 (87.54)
Therapy switch    
Patients with a therapy switch (N, %) 3,359 43.09%
Mean (SD) time to switch1 132.60 (99.52)
Type of therapy switched to (N, %)    
Gabapentin 227 2.91%
Opioid 2,988 38.33%
Pregabalin 144 1.85%
Therapy addition    
Patients with a therapy addition (N, %) 1,034 13.26%
Mean (SD) time to therapy addition1 141.80 (112.20)
Type of therapy added (N, %)    
Gabapentin 23 0.30%
Opioid 990 12.70%
Pregabalin 21 0.27%
Any treatment-pattern event (N, %) 4,493 57.63%
SD = standard deviation.
1 Among patients with the event.

Patients were found to be significantly more likely to have any treatment-pattern change if they were female (OR = 1.57; P < 0.001), lived in the South (OR = 1.37; P = 0.001) or West (OR = 1.48; P < 0.001) versus East, or had a Charlson comorbidity score greater than 1 (the OR increased with increasing Charlson comorbidity score, from 1.53 for patients with a score between 1 and 2 to 2.16 for patients with a score greater than 2; both P < 0.001) (Table 5).

Table 5 Predictors of any treatment pattern event, logistic regression results1
Characteristic Any treatment-pattern event
OR Lower 95% CI Upper 95% CI P value
Received pramipexole at index (vs. ropinirole) 0.94 0.85 1.04 0.230
Age (vs. 18-44 years)        
45–54 years 0.97 0.85 1.10 0.586
55–64 years 0.92 0.81 1.05 0.197
≥ 65 years 0.91 0.78 1.06 0.222
Female (vs. male) 1.57 1.42 1.73 < 0.001
Geographic region (vs. East)        
South 1.37 1.16 1.60 0.001
Midwest 1.05 0.92 1.20 0.465
West 1.48 1.27 1.73 < 0.001
Charlson Comorbidity Index score (vs. < 1)        
Between 1 and 2 1.53 1.25 1.86 < 0.001
> 2 2.16 1.82 2.56 < 0.001
CI = confidence interval; OR = odds ratio.
1 Using a logistic regression modeling the likelihood that a treatment-pattern event occurred among patients receiving pramipexole or ropinirole.


This retrospective database analysis examined health care utilization and costs among patients with RLS who were treated with DAs. The incidence of treatment-change events as a proxy for medication-related side effects also was assessed. We found that the per-patient mean (SD) annual all-cause medical costs were $11,485 ($21,362), of which 7% could be directly attributed to RLS. Approximately 58% of patients had a treatment-change event, and the most common events were change to new medication, addition of new medication, and dose adjustments.

Limited data were available regarding the costs associated with RLS in U.S. managed care populations. Curtice and colleagues (2009) examined 2,319 patients with RLS in a managed care database and found that RLS patients’ annual all-cause medical services were $8,843, with the highest costs being for outpatient services ($4,549). That study noted that annual costs were approximately $2,642 lower than the costs found in the present analysis, possibly due to differences in study methodologies (e.g., we required that all patients have at least two RLS diagnoses and that all patients be treated for RLS with a DA).

Our study found that annual RLS-related health care costs were $774, mainly composed of pharmacy costs; this amount is a fairly small proportion of total health care costs for RLS patients and payers. Previous studies have noted that RLS leads to significant reductions in quality of life for RLS patients compared with age- and sex-adjusted populations (e.g., Rothdach 2000; Sevim 2003; Winkelman 2006a) and to significant reductions in work productivity (Reinhold 2009; Allen 2011). Therefore, effective pharmacotherapy for RLS patients should improve both patients’ quality of life as well as their work productivity.

To put our findings into perspective, we compared the results of our analysis to the results of a study examining attention deficit/hyperactivity disorder (ADHD) in a population of employed adults (Secnik 2005). We chose to compare our analysis results with those of a study of ADHD in adults because of the underlying similarities in disease burden (i.e., both diseases are treated with oral medication, and neither is considered to be a life-threatening illness). We found similar rates of hospitalization (14% in the Secnik study vs. 16% in our analysis) and ER visits (7% in the Secnik study vs. 8% in our analysis). However, the Secnik study found lower total health care costs compared with our analysis ($7,542, when adjusted to 2008 dollars, vs. $11,485 in our study), most likely due to the substantially lower mean age of the ADHD population (32 years in the Secnik study compared with 54 years in our analysis) (Secnik 2005) but also possibly due to the large number of medical conditions found to be associated with RLS, including cardiovascular diseases (Winkelman 2008).

We also compared our findings with another retrospective database analysis by Pollack and colleagues (2009), which examined the economic burden of insomnia in an employed population. The Pollack study observed that patients with insomnia had a relatively low mean Charlson Comorbidity Index score (0.28 in the Pollack study compared with 0.70 in our analysis). Again, this might be due to the slightly younger mean age observed in the Pollack study (43 years in the Pollack study vs. 54 years in our analysis), but it could also be due to the increased risk of several medical disorders occurring with RLS. Specifically, we found higher rates of chronic pulmonary disease (7.85% of patients in the Pollack study compared with 12.96% of patients in our analysis) and diabetes (4.93% of patients in the Pollack study compared with 12.08% in our analysis). Additionally, total costs in the Pollack study were noted to be lower than our findings ($7,568 in the Pollack study, when adjusted to 2008 dollars, compared with $11,485 in our analysis), which may be due in part to the higher comorbidity burden and older mean age observed in our analysis (Pollack 2009).

In October 2006, the ICD-9-CM code 333.94 was developed specifically for RLS. Prior to that time, ICD-9-CM code 333.99 was used to identify RLS and other akathisias. Because this study required both the use of a DA as well as ICD-9-CM code 333.99, patients identified prior to October 2006 were very likely RLS patients. To allow for likely lags in the uptake of new coding practices, we continued to use both ICD-9-CM codes to identify patients after October 2006 (O’Malley 2005). Because DAs are not approved for the treatment of akathisia, patients identified after October 2006 using ICD-9-CM code 333.99 are still likely RLS patients and were therefore included in the analysis.


This study has several limitations common to retrospective database analyses. Patient records were not available to confirm that the RLS diagnoses was correct or to describe symptom severity. Previous studies have suggested that costs are linearly related to symptom severity (Allen 2011). Only treated RLS patients were included in the analysis; therefore, costs may be higher than those of the entire RLS population. The number of all-cause physician office visits was found to be right-skewed. The median number of all-cause physician office visits was 12 (compared with a mean of 18.65). One patient had nearly 550 visits and about 400 patients had more than 100 visits during the follow-up period. Because ICD-9-CM diagnosis codes do not exist for the side effects associated with DAs, costs associated with these side effects were included in the calculation of RLS-related costs only if a diagnosis of RLS was also present on the medical claim. Although not approved by the Food and Drug Administration for the treatment of RLS, pergolide has been shown to effectively reduce RLS symptoms and was included as a relevant DA in the present analysis (Staedt 1997; Early 1998; Wetter 1999; Trenkwalder 2004 b). Given the relatively small number of patients receiving pergolide, the net effect of their inclusion on the study results is likely minimal.

Finally, the costs examined in this analysis represented health plan payments and reflect rates for services that are negotiated by the plans in the database. Therefore, results from this study may not be applicable to Medicaid, Medicare, or for patients paying out-of pocket.


While the literature suggests that RLS is a distressing condition and, even when moderately severe, can be disabling, it is treatable with DAs. Our study found that, relative to the improved symptoms and associated health outcomes, the costs of treating RLS are fairly small and are mainly attributable to prescription medications. These findings should encourage managed care payers to consider expanding treatment of RLS to reduce the associated suffering and costs. This study is likely to be of interest to managed care payers and clinicians treating RLS patients.


Allen R, Becker PM, Bogan R, et al. Ropinirole decreases periodic leg movements and improves sleep parameters in patients with restless legs syndrome. Sleep. 2004;27:907–914.

Allen RP, Bharmal M, Calloway M. Prevalence and disease burden of primary restless legs syndrome: results of a general population survey in the USA. Mov Disord. 2011; 26(1): 114–120, 2011.

Allen RP, Earley CJ. Restless legs syndrome: a review of clinical and pathological features. J Clin Neurophysiol. 2001;18(2):128–147.

Allen RP, Stillman P, Myers AJ. Physician-diagnosed restless legs syndrome in a large sample of primary medical care patients in Western Europe: prevalence and characteristics. Sleep Med. 2010;132(pt 9):2403–2412.

Allen RP, Walters AS, Montplaisir J, et al. Restless legs syndrome prevalence and impact: REST general population study. Arch Intern Med. 2005;165(11):1286–1292.

Bogan RK, Fry JM, Schmidt MH, Carson SW, Ritchie SY, for the TREAT RLS US Study Group. Ropinirole in the treatment of patients with restless legs syndrome: a US-based randomized, double-blind, placebo-controlled clinical trial. Mayo Clin Proc.2006;81:17–27.

Charlson ME, Pompei P, Ales KL, McKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chron Dis. 1987;40:373–383.

Curtice TG, Long SR, Baser O, Montejano LB, Lobo F. Economic burden of restless legs syndrome in a privately-insured population [abstract]. Value Health. 2009;9(3):A8.

Early CJ, Yaffee JB, Allen RP. Randomized double blind placebo controlled trial of pergolide in restless legs syndrome. Neurology. 1998;51:1599–1602.

Eisensehr I, Moller C, Koster J, Bruggenjurgen B. Effectiveness of pramipexole in the treatment of RLS and the effect on work productivity: a 12-weeks non-interventional trial in patients with primary restless legs syndrome (RLS) [abstract]. Value Health. 2007;10(6):A393.

Evidente VG, Adler CH. How to help patients with restless legs syndrome. Discerning the indescribable and relaxing the restless. Postgrad Med. 1999;105(3):59–61, 65–66, 73–74.

Garcia-Borreguero D, Grunstein R, Sridhar G, et al. A 52-week open-label study of the long-term safety of ropinirole in patients with restless legs syndrome. Sleep Med. 2007;8:742–752.

Hening WA. Current guidelines and standards of practice for restless legs syndrome. Am J Med. 2007;120(1 suppl 1):S22–S27.

Milligan SA, Chesson AL. Restless legs syndrome in the older adult: diagnosis and management. Drugs Aging. 2002;19(10):741–751.

Montplaisir J, Fantini ML, Desautels A, Michaud M, Petit D, Filipini D. Long-term treatment with pramipexole in restless legs syndrome. Eur J Neurol. 2006;13:1306–1311.

Montplaisir J, Nicolas A, Denesle R, Gomez-Mancilla B. Restless legs syndrome improved by pramipexole: a double-blind randomized trial. Neurology. 1999;52:938–943.

O’Malley KJ, Cook KF, Price MD, Wildes KR, Hurdle JF, Ashton CM. Measuring diagnoses: ICD code accuracy. Health Serv Res. 2005;40(5 pt 2):1620–1639.

Partinen M, Hirvonen K, Jama L, et al. Efficacy and safety of pramipexole in idiopathic restless legs syndrome: a polysomnographic dose-finding study—the PRELUDE study. Sleep Med. 2006;7:407–417.

Phillips B, Hening W, Britz P, Mannino D. Prevalence and correlates of restless legs syndrome: results from the 2005 National Sleep Foundation Poll. Chest. 2006;129:76–80.

Pollack M, Seal B, Joish VN, Cziraky MJ. Insomnia-related comorbidities and economic costs among a commercially insured population in the United States. Curr Med Res Opin. 2009;25(8):1901–1911.

Reinhold T, Muller-Riemenschneider F, Willich SN, Bruggenjurgen B. Economic and human costs of restless legs syndrome. Pharmacoeconomics. 2009;27(4):267–279.

Rothdach AJ, Trenkwalder C, Haberstock J, Keil U, Berger K. Prevalence and risk factors of RLS in an elderly population: the MEMO study. Neurology. 2000;54(5):1064–1068.

Secnik K, Swensen A, Lage M. Comorbidities and costs of adult patients diagnosed with attention-deficit hyperactivity disorder. Pharmacoeconomics. 2005;23(1):93–102.

Sevim S, Dogu O, Camdeviren H, et al. Unexpectedly low prevalence and unusual characteristics of RLS in Mersin, Turkey. Neurology. 2003;61(11):1562–1569.

Sibler MH, Ehrenberg BL, Allen RP, et al. An algorithm for the management of restless legs syndrome. Mayo Clin Proc. 2004;79:916–922.

Staedt J, Wabmuth F, Ziemann U, et al. Pergolide: treatment of choice in restless legs syndrome (RLS) and nocturnal myoclonus syndrome (NMS). A double-blind randomized crossover trial of pergolide versus L-Dopa. J Neural Transm. 1997;104:461–468.

Trenkwalder C, Benes H, Poewe W, et al, and the SP790 Study Group. Efficacy of rotigotine for treatment of moderate-to-severe restless legs syndrome: a randomised, double-blind, placebo-controlled trial. Lancet Neurol. 2008;7:595–604.

Trenkwalder C, Garcia-Borreguero D, Montagna P, et al, and the Therapy With Ropinirole, Efficacy and Tolerability in RLS 1 Study Group. Ropinirole in the treatment of restless legs syndrome: results from the TREAT RLS 1 study, a 12 week, randomised, placebo controlled study in 10 European countries. J Neurol Neurosurg Psychiatry. 2004a;75:92–97.

Trenkwalder C, Hundemer HP, Lledo A, et al. Efficacy of pergolide in treatment of restless legs syndrome. The PEARLS study. Neurology. 2004b;62:1391–1397.

Vignatelli L, Billiard M, Clarenbach P, et al, and the EFNS Task Force. EFNS guideline on management of restless legs syndrome and periodic movement disorder in sleep. Eur J Neurol. 2006;13(10):1049–1065.

Wetter TC, Stiasny K, Winkelmann J, et al. A randomized controlled study of pergolide in patients with restless legs syndrome. Neurology. 1999;52:944–950.

Winkelman JW, Finn L, Young T. Prevalence and correlates of restless legs syndrome symptoms in the Wisconsin Sleep Cohort. Sleep Med. 2006a;7(7):5.

Winkelman JW, Sethi KD, Kushia CA, et al. Efficacy and safety of pramipexole in restless legs syndrome. Neurology. 2006b;67:1034–1039.

Winkelman JW, Shahar E, Sharief I, Gottlieb DJ. Association of restless legs syndrome and cardiovascular disease in the Sleep Heart Health Study. Neurology. 2008;70:35–42.

Corresponding author

Juliana Meyers, MA
RTI Health Solutions
200 Park Offices Drive
Research Triangle Park, NC 27709
Fax: (919) 541–7222
E-mail: jmeyers@rti.org

Disclosures/conflicts of interest

Juliana Meyers and Sean Candrill report that GlaxoSmithKline provided funding for this study. Richard Allen has received honoraria from UCB Pharma and GlaxoSmithKline and has served as a consultant for Boehringer Ingelheim, GlaxoSmithKline, Luitpold Pharmaceuticals, Pfizer, EMD Serono, Pharmacosmos, and UCB Pharma. Ranjani Manjunath and Michael Calloway are employees of GlaxoSmithKline.

Managed Care’s Top Ten Articles of 2016

There’s a lot more going on in health care than mergers (Aetna-Humana, Anthem-Cigna) creating huge players. Hundreds of insurers operate in 50 different states. Self-insured employers, ACA public exchanges, Medicare Advantage, and Medicaid managed care plans crowd an increasingly complex market.

Major health care players are determined to make health information exchanges (HIEs) work. The push toward value-based payment alone almost guarantees that HIEs will be tweaked, poked, prodded, and overhauled until they deliver on their promise. The goal: straight talk from and among tech systems.

They bring a different mindset. They’re willing to work in teams and focus on the sort of evidence-based medicine that can guide health care’s transformation into a system based on value. One question: How well will this new generation of data-driven MDs deal with patients?

The surge of new MS treatments have been for the relapsing-remitting form of the disease. There’s hope for sufferers of a different form of MS. By homing in on CD20-positive B cells, ocrelizumab is able to knock them out and other aberrant B cells circulating in the bloodstream.

A flood of tests have insurers ramping up prior authorization and utilization review. Information overload is a problem. As doctors struggle to keep up, health plans need to get ahead of the development of the technology in order to successfully manage genetic testing appropriately.

Having the data is one thing. Knowing how to use it is another. Applying its computational power to the data, a company called RowdMap puts providers into high-, medium-, and low-value buckets compared with peers in their markets, using specific benchmarks to show why outliers differ from the norm.
Competition among manufacturers, industry consolidation, and capitalization on me-too drugs are cranking up generic and branded drug prices. This increase has compelled PBMs, health plan sponsors, and retail pharmacies to find novel ways to turn a profit, often at the expense of the consumer.
The development of recombinant DNA and other technologies has added a new dimension to care. These medications have revolutionized the treatment of rheumatoid arthritis and many of the other 80 or so autoimmune diseases. But they can be budget busters and have a tricky side effect profile.

Shelley Slade
Vogel, Slade & Goldstein

Hub programs have emerged as a profitable new line of business in the sales and distribution side of the pharmaceutical industry that has got more than its fair share of wheeling and dealing. But they spell trouble if they spark collusion, threaten patients, or waste federal dollars.

More companies are self-insuring—and it’s not just large employers that are striking out on their own. The percentage of employers who fully self-insure increased by 44% in 1999 to 63% in 2015. Self-insurance may give employers more control over benefit packages, and stop-loss protects them against uncapped liability.