Cost Implications of Early Treatment Initiation Among Patients With Newly Diagnosed Fibromyalgia

, , ,
The American Journal of Pharmacy Benefits, November/December 2017, Volume 9, Issue 6

Initiation of treatment with guideline-recommended medications within 60 days of fibromyalgia diagnosis was associated with immediate annual healthcare cost savings.

Abstract

Objectives: To evaluate treatment initiation timing and healthcare costs among newly diagnosed fibromyalgia patients receiving pain medications commonly used in fibromyalgia treatment.

Study Design: Retrospective cohort study (IMS Pharmetrics Plus).

Methods: Eligible patients had ≥2 International Classification of Diseases, Ninth Revision, Clinical Modification codes 729.1X within a year (first [index] diagnosis in 2012; none in 12 months prior to index diagnosis [baseline]), and no baseline use of medications commonly used in fibromyalgia management (pregabalin, duloxetine, milnacipran, gabapentin, tramadol, venlafaxine, amitriptyline, cyclobenzaprine). “Early-” (≤60 days after index diagnosis) and “late-” treated cohorts were propensity score (PS) matched using patient characteristics and other pain-related medications. Total healthcare costs during the 12-month follow-up period were compared for early- versus late-treated cohorts and were adjusted for baseline costs, unbalanced comorbidities, and follow-up use of other pain medications.

Results: Among 33,470 newly diagnosed fibromyalgia patients identified, 5208 (15.6%) received common fibromyalgia medications. Median age for PS matched cohorts (n = 2019 each) was 47 years, and 72.2% were female. Most frequent comorbidities were rheumatic conditions, hyperlipidemia, back and neck pain, and depression. Proportion of patients receiving other pain-related medications was similar in both cohorts, with >20% receiving opioid, anti-inflammatory, anti-anxiety, and selective serotonin reuptake inhibitor medications, respectively. Total follow-up costs were 15% lower ($1970 less) in the early-treated cohort (adjusted mean costs: $11,287 versus $13,258; P <.0001).

Conclusions: Total healthcare costs were significantly lower among early-treated patients, suggesting that prompt use of fibromyalgia-related medications is a potential cost-saving strategy. Patients with fibromyalgia had a substantial comorbidity burden.

Am J Pharm Benefits. 2017;9(6):200-207

Fibromyalgia is a chronic pain syndrome characterized by widespread neuromuscular pain and generalized tender points.1,2 It is estimated that fibromyalgia affects as many as 6% of US adults and 3% of adults worldwide.1,3,4 Symptoms of fibromyalgia may be subjective, and diagnosis may be difficult and may vary between practitioners. Frequent misdiagnosis of fibromyalgia results in underdiagnosis in the general population, since fibromyalgia may often be confused with other regional pain syndromes and systemic disorders that share pathophysiologic features with fibromyalgia.1,5 Although early treatment is optimal, current diagnosis time averages about 5 years.1 Patients with fibromyalgia have frequent comorbidities, including other pain disorders, sleep disorders, depression, and mood/anxiety disorders,1,6,7 and fibromyalgia patients average 4.2 comorbid conditions each year.8 Studies estimate that up to half of fibromyalgia patients also suffer from obesity, and an additional 21% to 30% are overweight.9-12

Fibromyalgia is associated with significant direct and indirect costs,13-16 and total annual costs among patients with fibromyalgia have been estimated at $17,165 (2012 US$), with approximately 68% of total costs attributed to indirect costs due to pain.16 High prescription costs may be due to polypharmacy, attributable not only to fibromyalgia management but also reflective of the high comorbidity burden associated with fibromyalgia.13 Fibromyalgia is associated with significant indirect costs since it has an impact on patient productivity.14-16 Furthermore, fibromyalgia has a significant impact on patient quality of life, as it can impair the patient’s ability to participate in everyday activities and affect relationships with family, friends, and employers.1

Fibromyalgia treatment recommendations have been put forth by the European League Against Rheumatism (EULAR)17 and the American Pain Society (APS)18; however, the APS cautions that its treatment guidelines are outdated and should not be used for current medical practice.19 EULAR 2016 guidelines recommend the use of amitriptyline (at low dose), duloxetine, milnacipran, tramadol, pregabalin, and cyclobenzaprine for pharmacological management of fibromyalgia, although initial non-pharmacological management strategies are encouraged. Absent from the EULAR recommendation is gabapentin, which, despite not having a specific indication for fibromyalgia, is often used in this patient population.19,20 The American College of Rheumatology summarizes medications for fibromyalgia treatment on its patient/caregiver website; this list includes antidepressants and gabapentin.19 In 2012, Canadian Guidelines for the Diagnosis and Management of Fibromyalgia were developed21; however, milnacipran, which is FDA approved for fibromyalgia treatment, is not available in Canada. Only 3 medications (pregabalin, duloxetine, and milnacipran) are approved by the FDA in the United States to treat fibromyalgia pain.22-24 The use of nonsteroidal anti-inflammatory drugs (NSAIDs) and/or opioid medications to manage fibromyalgia is common, despite a lack of evidence supporting their use in fibromyalgia.25 While there is some consensus that opioids are not recommended options for treating fibromyalgia, tramadol, a weak opioid with mild serotonin-noradrenaline reuptake inhibitor activity, is supported for short-term use if the prescriber believes that opioid treatment is needed to treat fibromyalgia pain.17,19

Previous research has suggested that fibromyalgia treatment rates are low, and that more than two-thirds of patients newly diagnosed with fibromyalgia are not treated with prescription pain medication within 1 year of diagnosis.20 Furthermore, among patients who are treated pharmacologically, discontinuation of prescribed pain medications is common,20,26 and doses of medications prescribed to patients with fibromyalgia are frequently lower than recommended.20 While underdiagnosis, undertreatment, and delayed diagnosis are described in the fibromyalgia literature,1,5,20 and early treatment is recommended,1 little is known regarding the impact of treatment initiation timing on patient outcomes or healthcare cost.

The primary objective of this study was to assess the impact of prompt fibromyalgia treatment on total healthcare costs. Secondary objectives included the evaluation of the initial pain management therapy for newly diagnosed fibromyalgia patients, and to estimate the proportion of patients treated with common fibromyalgia-related medications.

METHODS

Study Patients

This was a retrospective cohort study that used administrative medical and pharmacy claims for commercial, employer self-insured, and Medicare risk plans from the IMS Pharmetrics Plus™ database. Patients who were aged ≥18 years at the time of first diagnosis (index date) of fibromyalgia (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] 729.1) during 2012, with a second fibromyalgia diagnosis within 1 year of the index date, were identified. Patients were required to have continuous health plan enrollment for both medical and pharmacy coverage during the 12 months prior to and following the index date, with no fibromyalgia diagnosis during 1 year prior to the index date. The 12-month period preceding the index date was used as the study baseline period. Patients were excluded if they received any of the following common fibromyalgia-related prescription medications during the 12-month study baseline period: amitriptyline, cyclobenzaprine, venlafaxine, duloxetine, milnacipran, pregabalin, gabapentin, or tramadol.19,20

Measurements and Study Endpoints

Patient demographic information was obtained from claims data and subsequently assessed. Comorbid conditions were evaluated at baseline using ICD-9-CM codes from medical claims. Study endpoints included the proportion of patients treated with pain medications within 1 year of index date. These included prescription medications commonly prescribed to treat fibromyalgia in the United States (amitriptyline, cyclobenzaprine, venlafaxine, duloxetine, milnacipran, pregabalin, gabapentin, tramadol),19,20 as well as other prescription medications which may be used in fibromyalgia, including opioid (excluding tramadol) and non-opioid pain medications, muscle relaxants, antidepressants, sleep aids, attention-deficit/hyperactivity disorder (ADHD) medications, and anti-anxiety medications (eAppendix Table; eAppendices available at www.ajpb.com). Time to initial prescription fill of common fibromyalgia-related medications from the index date was evaluated, and patients were categorized and placed in an “early-treated” (≤60 days) or “late-treated” (>60 days) cohort. A sensitivity analysis was performed to evaluate the impact of using an alternate threshold for the early-treated cohort (<90 days). Total healthcare costs for all healthcare services during 1 year following index date were obtained from administrative claims data, including inpatient, outpatient, and pharmacy total costs, and these costs were compared between early- and late-treated patient groups. Pharmacy costs represent costs associated with prescriptions filled outside a hospital setting; inpatient costs represent the total cost associated with hospitalizations; and all other costs (not pharmacy or inpatient) were attributed to outpatient costs (eg, emergency department visits, physician office visits, outpatient surgical or diagnostic procedures, laboratory, pathology, or radiology encounters).

Statistical Analyses

Descriptive statistics (means, SDs, and frequencies) were calculated for patient characteristics. These characteristics and baseline healthcare costs during the follow-up period were compared using t tests for continuous variables, and χ2 and Fisher’s exact tests for categorical variables, between early- and late-treated patient groups. Early- and late-treated groups were propensity score matched to ensure comparability of baseline characteristics and to reduce the possibility of selection bias.27-29 Propensity scores were calculated using logistic regression, with patient demographics, comorbidities, health plan payer type, and other pain-related therapies at baseline as independent variables. Chi-square and Fisher’s exact tests were used to assess the balance between groups across all independent variables at P >.1. Generalized linear regression with a gamma distribution and log link function was used to compare costs between early- and late-treated patient groups, adjusting for baseline costs, unbalanced baseline comorbidities, and other pain medication use during the follow-up period.

RESULTS

Study Patients

Preliminary inclusion criteria were met by 33,470 newly diagnosed fibromyalgia patients (Figure 1). During the follow-up period, 15.6% (5208) received common fibromyalgia-related medications; 39.9% of these patients started therapy within 60 days and 47.7% within 90 days, of diagnosis. During the baseline period, 45.8% of the 33,470 newly diagnosed patients were prescribed >1 other pain medication. Almost half (49.4%) of all patients were prescribed other pain medications during follow-up.

Among all patients treated with a fibromyalgia-related medication (N = 5208), median age was 47 years and 70.0% were female (Table 1). Female patients were more likely to be early-treated than late (72.6% vs 68.3%; P <.05). The majority of patients (65.3%) were covered by a health maintenance organization health plan. The most frequent comorbid conditions were rheumatic conditions, hyperlipidemia, back/neck pain, and depression. Late-treated patients had higher rates of back/neck pain and lumbago (Table 1). Characteristics for the propensity score matched cohort are also included in Table 1 (N = 2019 for each group). The median age for the propensity score matched cohort was 47 years, and 72.2% were female. Subsequent to propensity score matching, only depression remained slightly unbalanced between groups (P = .09).

Fibromyalgia-Related Medication Use

The most commonly prescribed fibromyalgia-related treatment was cyclobenzaprine, prescribed to slightly more than 4 in 10 study patients (Table 2). Many patients who were treated with a fibromyalgia-related treatment were also treated with NSAIDs, anti-anxiety medications, and antidepressants during both baseline and during follow-up. Among all patients who were treated with a fibromyalgia-related medication, the most commonly used other pain medications at baseline were opioids (40.7%), anti-inflammatory medications (25.9%), selective serotonin reuptake inhibitors (SSRIs) (23.9%), and anti-anxiety medications (21.0%). Early- and late-treated patients had similar baseline rates of use of most other pain medications; however, early-treated patients were more likely to have used opioids and anti-inflammatory medications (P <.05 for both). Among propensity matched groups, baseline use of all other pain medications was similar. However, during follow-up, late-treated patients were more likely to be treated with SSRI antidepressants (P = .001). Use of opioids, anti-inflammatory medications, and, to a lesser extent, anti-anxiety medications, increased from baseline to follow-up. Among propensity matched groups, only SSRI use during follow-up differed between early- and late-treated patients (23.7% vs 28.5%; P <.01). Overall, about 1 in 10 patients were prescribed sleep aids.

Healthcare Costs

Figure 2 includes unadjusted baseline and follow-up healthcare costs by category for propensity matched groups. Both baseline and follow-up total healthcare costs were higher among late-treated patients. Figure 2B shows that total healthcare costs were higher during the first year following initial fibromyalgia diagnosis for late-treated patients, across all categories of healthcare spend (outpatient, inpatient, and pharmacy). Outpatient costs represented the largest driver of total healthcare costs for both treatment groups across both baseline and follow-up periods. Outpatient costs during the follow-up period represented 62.9% of total direct costs among early-treated patients, and 67.2% of total direct costs among late-treated patients. Pharmacy costs represented 17.4% and 15.4% of total healthcare costs among early- and late-treated patients, respectively. Mean total healthcare costs during follow-up, adjusted for log-transformed baseline cost, baseline depression, and follow-up SSRI antidepressant and ADHD drug use, are shown in Figure 3. Adjusted total healthcare costs during the 1-year follow-up period were estimated at $11,287 for the early-treated group and $13,258 for the late-treated group, and costs incurred by the early-treated group were 14.9% lower ($1970) than those of the late-treated group.

Sensitivity Analysis

To examine the impact of our study’s definition of “early-treated” patients, we analyzed the impact of using a 90-day threshold instead of 60 days (“early-treated,” ≤90 days; “late-treated,” >90 days). After propensity score matching, 2337 patients were included in the early- and late-treated patient groups. Adjusted total healthcare costs during the 1-year follow-up period were estimated at $11,281 for the early-treated group and $12,803 for the late-treated group (difference = $1522; P <.0001).

DISCUSSION

Our study results, in agreement with those of previously published research,20 indicated low fibromyalgia treatment rates, as about half of newly diagnosed patients with fibromyalgia (47.9%) did not receive medications commonly used in fibromyalgia management in the United States, or other pain medications which may be used to treat fibromyalgia. A high comorbidity burden was associated with fibromyalgia. Rheumatic conditions, back/neck pain, and depression were common at baseline, and many patients were prescribed opioids, pain medications, and depression/anxiety medications during the year preceding the fibromyalgia diagnosis. While we were unable to evaluate sleep-related comorbidities from medical claims data, about 10% of patients were prescribed sleep aids.

Early initiation of fibromyalgia treatment was associated with lower total healthcare costs, suggesting that prompt initiation of these medications may be a cost-saving strategy. During the first year following the initial fibromyalgia diagnosis, the early initiation of treatment was associated with 15% lower ($1970) adjusted mean total healthcare costs, despite additional medication costs related to the earlier initiation of fibromyalgia-related medications during the study follow-up period. These cost savings appeared to be largely driven by lower outpatient costs among the early-treated patient group during the follow-up period.

In our study, 16% of fibromyalgia patients received treatment with either a recommended medication17 or a medication found to be commonly used in fibromyalgia management.20 Liu and colleagues20 studied 240,144 patients who were diagnosed with fibromyalgia in the United States between 2009 and 2011, and found that more than two-thirds of patients (69%) were not pharmacologically treated within a year of diagnosis. Among patients who were treated with pain medication, about half filled an initial prescription within 1 month of diagnosis, and many patients who filled a prescription for fibromyalgia management discontinued its use shortly after initiation.20

Consistent with prior research,20,30 the use of opioids in our fibromyalgia patient population was common, despite the lack of evidence or treatment guidelines supporting their use in fibromyalgia. More than half of our fibromyalgia patient population filled at least 1 prescription for an opioid other than tramadol during the 12-month follow-up period. The proportion of patients using opioids increased after fibromyalgia diagnosis in both early- and late-treated groups. The common use of opioids in fibromyalgia may be concerning given their high potential for abuse, especially given the lack of evidence supporting their effectiveness in fibromyalgia management. However, a large proportion of our study population suffered from comorbid conditions with chronic pain (eg, rheumatic conditions, back/neck pain), and since claims data were used for this study, we cannot ascertain whether opioids were prescribed for fibromyalgia or other comorbid chronic pain conditions.

Several other studies have evaluated the direct cost of fibromyalgia. Total annual direct adjusted costs in our study were $11,287 for early-treated and $13,258 for late-treated patients during 2012-2013. A study by Schaefer and colleagues that examined the cost of fibromyalgia in 2012 US$ estimated the 3-month direct cost of fibromyalgia at $1394, which would annualize to approximately $5576.16 Schaefer’s lower direct cost estimates may be related to study design, as costs were captured via patient questionnaire, rather than the administrative claims used for the current study. Chandran et al examined the impact of fibromyalgia severity on costs, and estimated direct annualized costs that ranged from $4854 to $9318 for patients with mild to severe fibromyalgia (2009 US$).15 The current study’s estimates are also somewhat higher than the mean direct annual costs observed by White and colleagues ($7286) in 2008, but this difference may be attributable to differences in the time periods studied.14 White and colleagues partially attributed high prescription drug costs to polypharmacy, and in our study, polypharmacy was also common, as the majority of patients treated with a fibromyalgia-related medication also received other pain medications. In 2011, the observations of Sanchez et al, similar to results of the current study, were that outpatient costs were the largest driver of the total direct cost burden, accounting for about half.13 They found that the highest outpatient utilization occurred within the first 6 months of fibromyalgia diagnosis before decreasing and subsequently leveling off, with the initial utilization peak possibly associated with increased patient monitoring and with therapy changes to optimize initial management.13 Finally, though it was not assessed in the current study, it is important to recognize that fibromyalgia may impact patient productivity and the associated economic consequences. Schaefer and colleagues estimated that approximately 68% of total costs were indirect, while White and colleagues estimated the annual indirect cost burden of fibromyalgia at $2983.14 Chandran et al found that fibromyalgia severity influenced the contribution of indirect costs, and estimated that mean total indirect costs accounted for 53%-78% of mean total costs for patients with mild to severe fibromyalgia, respectively.15

Limitations

This retrospective study of a geographically diverse population allowed the assessment of “usual care” in a large cohort of patients with fibromyalgia with good generalizability of results, although our results may not be reflective of those without health insurance coverage. However, it is important to consider our findings in the context of several important limitations. This study used administrative claims to assess fibromyalgia-related prescriptions filled using a pharmacy benefit, and prescriptions obtained outside the pharmacy benefit (out-of-pocket) would not be captured. Thus, use of over-the-counter medications, particularly nonprescription NSAIDs, was not assessed due to our use of pharmacy claims. We were also unable to capture information related to use of non-pharmacologic treatments (eg acupuncture, massage) which may be used for fibromyalgia management. Some prior research suggested that the majority of people with fibromyalgia use over-the-counter medications to treat fibromyalgia pain.30 The patient diagnosis of fibromyalgia was obtained from claims with the fibromyalgia ICD-9-CM codes, and due to the nature of claims data, a fibromyalgia diagnosis could not be verified and mis- and underdiagnosis of fibromyalgia may have played roles. A sizable proportion of our study population was prescribed antidepressants, despite a relatively low prevalence of depression or anxiety as classified by medical claims history; the frequency of these conditions in medical claims data may be lower than what may be observed among the results of other studies that used patient self-report to obtain this information. We were unable to assess the severity of fibromyalgia, or provider characteristics, and both of these factors may have impacted our study findings. Obesity may play a role in fibromyalgia management, and patient body mass index was not available for inclusion in the analysis. Finally, our study only captured direct costs; it did not capture the indirect costs associated with productivity loss, which have been shown to be considerable in other studies.14

CONCLUSIONS

Patients with fibromyalgia have a substantial comorbidity burden and considerable annual healthcare costs. The majority of newly diagnosed fibromyalgia patients did not receive pharmacological treatment with a fibromyalgia-related medication. Treatment of fibromyalgia with opioids and other classes of medications not recommended for use in fibromyalgia appears to be relatively common. Total healthcare costs were significantly lower among early-treated patients, suggesting that despite the increased pharmacy costs, early use of fibromyalgia-related pharmacological therapy may be a cost-saving strategy. Future research should investigate whether early treatment of fibromyalgia may also be associated with indirect cost savings, lower patient out-of-pocket costs for alternative care (eg, massage, acupuncture), and improved patient quality of life.

Authorship Information: Concept and design (CQ, MG, QZ, FF); acquisition of data (CQ, FF); analysis and interpretation of data (CQ, MG, QZ, FF); drafting of the manuscript (CQ, FF); critical revision of the manuscript for important intellectual content (CQ, QZ); statistical analysis (CQ, MG, FF); obtaining funding (FF); administrative, technical, or logistic support (FF); and supervision (QZ, FF).

FF and QZ were employed by Daiichi Sankyo, Inc at the time of the study but are no longer employed by Daiichi Sankyo, Inc.

1Daiichi Sankyo, Inc, Basking Ridge, NJ.

2Avalon Health Solutions, Inc, Philadelphia, PA.

Corresponding author:

Chunlin Qian, PhD, Health Economics & Outcomes Research, Daiichi Sankyo, Inc, 211 Mount Airy Road, Basking Ridge, NJ 07920;

Tel: (908)992-7056

Email: cqian@dsi.com

References

1. Arnold LM, Clauw DJ, McCarberg BH; FibroCollaborative. Improving the recognition and diagnosis of fibromyalgia. Mayo Clin Proc. 2011;86(5):457-464. doi: 10.4065/mcp.2010.0738.

2. Chakrabarty S, Zoorob R. Fibromyalgia. Am Fam Physician. 2007;76(2):247-254.

3. Lawrence RC, Felson DT, Helmick CG, et al; National Arthritis Workgroup. Estimates of the prevalence of arthritis and other rheumatic conditions in the United States. part II. Arthritis Rheum. 2008;58(1):26-35. doi: 10.1002/art.23176.

4. Queiroz LP. Worldwide epidemiology of fibromyalgia. Curr Pain Headache Rep. 2013;17(8):356. doi: 10.1007/s11916-013-0356-5.

5. Clauw DJ, Crofford LJ. Chronic widespread pain and fibromyalgia: what we know, and what we need to know. Best Pract Res Clin Rheumatol. 2003;17(4):685-701.

6. Arnold LM. Management of fibromyalgia and comorbid psychiatric disorders. J Clin Psych. 2008;69(suppl 2):14-19.

7. Fietta P, Fietta P, Manganelli P. Fibromyalgia and psychiatric disorders. Acta Biomed. 2007;78(2):88-95.

8. Schaefer C, Chandran A, Hufstader M, et al. The comparative burden of mild, moderate and severe fibromyalgia: results from a cross-sectional survey in the United States. Health Qual Life Outcomes. 2011;9:71. doi: 10.1186/1477-7525-9-71.

9. Kim CH, Luedtke CA, Vincent A, Thompson JM, Oh TH. Association of body mass index with symptom severity and quality of life in patients with fibromyalgia. Arthritis Care Res (Hoboken). 2012;64(2):222-228. doi: 10.1002/acr.20653.

10. Neumann L, Lerner E, Glazer Y, Bolotin A, Shefer A, Buskila D. A cross-sectional study of the relationship between body mass index and clinical characteristics, tenderness measures, quality of life, and physical functioning in fibromyalgia patients. Clin Rheumatol. 2008;27(12):1543-1547. doi: 10.1007/s10067-008-0966-1.

11. Okifuji A, Bradshaw DH, Olson C. Evaluating obesity in fibromyalgia: neuroendocrine biomarkers, symptoms, and functions. Clin Rheumatol. 2009;28(4):475-478. doi: 10.1007/s10067-009-1094-2.

12. Okifuji A, Donaldson GW, Barck L, Fine PG. Relationship between fibromyalgia and obesity in pain, function, mood, and sleep. J Pain. 2010;11(12):1329-1337. doi: 10.1016/j.jpain.2010.03.006.

13. Sanchez RJ, Uribe C, Li H, et al. Longitudinal evaluation of health care utilization and costs during the first three years after a new diagnosis of fibromyalgia. Curr Med Res Opin. 2011;27(3):663-671. doi: 10.1185/03007995.2010.550605.

14. White LA, Birnbaum HG, Kaltenboeck A, Tang J, Mallett D, Robinson RL. Employees with fibromyalgia: medical comorbidity, healthcare costs, and work loss. J Occup Environ Med. 2008;50(1):13-24. doi: 10.1097/JOM.0b013e31815cff4b.

15. Chandran A, Schaefer C, Ryan K, Baik R, McNett M, Zlateva G. The comparative economic burden of mild, moderate, and severe fibromyalgia: results from a retrospective chart review and cross-sectional survey of working-age U.S. adults. J Manag Care Pharm. 2012;18(6):415-426.

16. Schaefer C, Mann R, Masters ET, et al. The comparative burden of chronic widespread pain and fibromyalgia in the United States. Pain Pract. 2016;16(5):565-579. doi: 10.1111/papr.12302.

17. Macfarlane GJ, Kronisch C, Dean LE, et al. EULAR revised recommendations for the management of fibromyalgia. Ann Rheum Dis. 2016;76(2):318-328. doi: 10.1136/annrheumdis-2016-209724.

18. Guideline for the Management of Fibromyalgia Syndrome Pain in Adults and Children (archival version; copyright 2005). American Pain Society website. http://apps.americanpainsociety.org/store/productdetails.aspx?ProductId=473. Accessed November 10, 2015.

19. Fibromylagia. American College of Rheumatology website. http://www.rheumatology.org/Practice/Clinical/Patients/Diseases_And_Conditions/Fibromyalgia/. Updated March 2017.

20. Liu Y, Qian C, Yang M. Treatment patterns associated with ACR-recommended medications in the management of fibromyalgia in the United States. J Manag Care Spec Pharm. 2016;22(3):263-271. doi: 10.18553/jmcp.2016.22.3.263.

21. Fitzcharles MA, Ste-Marie PA, Goldenberg DL, et al; National Fibromyalgia Guideline Advisory Panel. 2012 Canadian Guidelines for the diagnosis and management of fibromyalgia syndrome: executive summary. Pain Res Manag. 2013;18(3):119-126.

22. Cymbalta (duloxetine delayed-release capsules) for oral use [package insert]. Indianapolis, IN: Lilly USA, LLC; 2012.

23. Lyrica (pregabalin) capsules [package insert]. New York, NY: Pfizer Pharmaceuticals; 2013.

24. Savella (milnacipran HCl) tablets [package insert]. St. Louis, MO: Forest Laboratories, Inc; 2013.

25. Smith HS, Harris R, Clauw D. Fibromyalgia: an afferent processing disorder leading to a complex pain generalized syndrome. Pain Physician. 2011;14(2):E217-E245.

26. Robinson RL, Kroenke K, Williams DA, et al. Longitudinal observation of treatment patterns and outcomes for patients with fibromyalgia: 12-month findings from the REFLECTIONS study. Pain Med. 2013;14(9):1400-1415. doi: 10.1111/pme.12168.

27. Rosenbaum P, Rubin D. The central role of the propensity score in observational studies for causal effects. Biometrika. 1983;70(1):41-55.

28. Rubin DB. Estimating causal effects from large data sets using propensity scores. Ann Intern Med. 1997;127(8 Pt 2):757-763.

29. Rubin DB, Thomas N. Matching using estimated propensity scores: relating theory to practice. Biometrics. 1996;52(1):249-264.

30. Bennett RM, Jones J, Turk DC, Russell IJ, Matallana L. An Internet survey of 2,596 people with fibromyalgia. BMC Musculoskelet Disord. 2007;8:27.