Objective To describe the characteristics, treatment, and health care expenditures of Medicare Supplemental-insured patients with painful diabetic peripheral neuropathy (pDPN), post-herpetic neuralgia (PHN), or fibromyalgia.
Design Retrospective cohort study.
Setting United States clinical practice, as reflected within a database comprising administrative claims from 2.3 million older adults participating in Medicare supplemental insurance programs.
Subjects Selected patients were aged ≥65 years, continuously enrolled in medical and prescription benefits throughout years 2008 and 2009, and had ≥1 medical claim with an International Classification of Diseases, Ninth Revision, Clinical Modification diagnosis code for DPN, PHN, or fibromyalgia, followed within 60 days by a medication or pain intervention procedure used in treating pDPN, PHN, or fibromyalgia during 2008–2009.
Outcome Measures Utilization of, and expenditures on, pain-related and all-cause pharmacotherapy and medical interventions in 2009.
Results The study included 25,716 patients with pDPN (mean age 75.2 years, 51.2% female), 4,712 patients with PHN (mean age 77.7 years, 63.9% female), and 25,246 patients with fibromyalgia (mean age 74.4 years, 73.0% female). Patients typically had numerous comorbidities, and many were treated with polypharmacy. Mean annual expenditures on total pain-related health care and total all-cause health care, respectively, (in 2010 USD) were: $1,632, $24,740 for pDPN; $1,403, $16,579 for PHN; and $1,635, $18,320 for fibromyalgia. In age-stratified analyses, pain-related health care expenditures decreased as age increased.
Conclusions The numerous comorbidities, polypharmacy, and magnitude of expenditures in this sample of Medicare supplemental-insured patients with pDPN, PHN, or fibromyalgia underscore the complexity and importance of appropriate management of these chronic pain patients.
Painful Diabetic Peripheral Neuropathy
Health Care Utilization
Health Care Expenditures
Painful diabetic peripheral neuropathy (pDPN), post-herpetic neuralgia (PHN), and fibromyalgia are chronic pain conditions that share similar pharmacological treatments, including select antidepressants, anticonvulsants, and analgesics [1–3]. As with many chronic illnesses, the prevalence of these conditions increases with age and peak among older individuals [4–7]. Since older individuals tend to have more comorbidity and are often treated with polypharmacy, the management of pDPN, PHN, and fibromyalgia may be more complicated in such individuals than in younger patients.
“Real-world” illness-specific information regarding the prevalence of comorbidities, use of concomitant medications, common treatment approaches, and health care expenditures can help to identify areas in which innovation and improvement of treatment practices may result in the greatest benefits to patients. For example, knowledge of the comorbid conditions that present most commonly with a given disease have been used to design health care interventions that simultaneously address multiple aspects of a patients' condition . Furthermore, co-formulated medications, such as those used to treat both cholesterol and high blood pressure or both cholesterol and hyperglycemia, have been developed on similar premises.
While there are a number of studies that have examined such matters in patients with pDPN, PHN, or fibromyalgia, nearly all focus on younger, commercially insured individuals or are conducted among patients who have received a specific treatment [9–19]. Other published studies have reported only limited cost and utilization data, do not have information on comorbidities, or are based on relatively less contemporary data [20–24]. This has left a gap in published information on older patients with pDPN, PHN, or fibromyalgia.
Thus, the objective of this study was to describe the characteristics, treatment, and health care expenditures of a sample of Medicare supplemental-insured patients with pDPN, PHN, or fibromyalgia.
This was a retrospective cohort study.
Setting and Data
The setting of this study was U.S. clinical practice, as reflected within a database comprising administrative claims from 2.3 million older adults participating in Medicare supplemental insurance programs (Truven Health MarketScan (Ann Arbor, MI, USA) Medicare Supplemental and Coordination of Benefits [Medicare Supplemental] database). This database represents a convenience (non-probability) sample formed from administrative claims data that have been contributed from over 300 self-insured employers and health plans. This database is generalizable to U.S. population with Medicare supplemental insurance (15.3 million retirees in the U.S.), which represents approximately 31% of 49.4 million patients in the U.S. Medicare population .
The Medicare Supplemental database comprises inpatient medical, outpatient medical, and outpatient prescription drug information for retirees with employer-sponsored Medicare supplemental insurance. Both the Medicare-covered portion of payments (represented as Coordination of Benefits Amount, or COB) and the employer-paid portion of payments are included within the Medicare Supplemental database.
The MarketScan databases have formed the basis of numerous health economic and epidemiologic studies, including many in the area of chronic pain . All MarketScan database records are de-identified and fully compliant with U.S. patient confidentiality requirements, including the Health Insurance Portability and Accountability Act (HIPAA) of 1996. The databases are certified to satisfy the conditions set forth in Sections 164.514 (a)-(b)1ii of the HIPAA privacy rule regarding the determination and documentation of statistically de-identified data. Because this study used only de-identified patient records and did not involve the collection, use, or transmittal of individually identifiable data, Institutional Review Board approval to conduct this study was not necessary.
In order to be considered for the study, subjects (patients) were required to have at least one inpatient or outpatient non-diagnostic claim (i.e., excluding claims associated with procedures that may be associated with an attempt to rule out a condition, such as radiology or laboratory claims) with an International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis code, in any position, for DPN (250.6×, 357.2×), PHN (053.1×), or fibromyalgia (729.1) at any point during the period from January 1, 2008 to December 31, 2009. To ensure complete capture of administrative claims data for the study period, patients were also required to be continuously enrolled in medical and prescription benefits throughout years 2008 and 2009. Patients were required to be aged ≥65 years at time of first DPN, PHN, or fibromyalgia diagnosis during the period from January 1, 2008 to December 31, 2009. Since the ICD-9-CM diagnosis codes for DPN and fibromyalgia are nonspecific to painful DPN and fibromyalgia (e.g., fibromyalgia is coded under 729.1, which corresponds to myalgia and myositis, unspecified), patients were required to have at least one claim for a medication or pain intervention procedure used in treating pain associated with pDPN, PHN, or fibromyalgia within 60 days after any diagnosis of DPN, PHN, or fibromyalgia during the period from January 1, 2008 to December 31, 2009.
In order to exclude patients with conditions or in settings that could potentially complicate the nature of the analysis, those meeting any of the following criteria from January 1, 2008 to December 31, 2009 were excluded from the study sample: any claims with an ICD-9-CM diagnosis code, in any position, for epilepsy (which shares some common treatments with pDPN and fibromyalgia; 345.xx, 780.39); any claims with an ICD-9-CM diagnosis code and/or procedure code, in any position, indicative of transplant surgery; long-term care facility residence for ≥90 total days; at least one inpatient or two outpatient non-diagnostic claims on different days, at least 21 days apart, with an ICD-9-CM diagnosis code, in any position, for cancer except for basal cell and squamous cell skin cancers and benign neoplasms (140.xx-172.xx, 174.xx-208.xx, 235.xx-239.xx). Lists of procedures and codes used in the selection of subjects are available from the authors upon request.
Study measures included utilization of, and expenditures on, pain-related and all-cause pharmacotherapy and medical interventions in 2009 as outcomes and patient characteristics for the purpose of describing the cohorts. See the Appendix for a listing of all oral and topical medications included in the analysis.
Measures were based on enrollment data and inpatient medical, outpatient medical, and outpatient pharmaceutical claims data using ICD-9-CM diagnosis and procedure codes, Current Procedural Terminology (CPT) codes, Healthcare Common Procedure Coding System (HCPCS) codes, and National Drug Codes (NDCs), as appropriate. Nosologists (medical coders) were consulted when compiling the lists of codes used for the measures. Pain-related utilization categories listed below were drawn from prior studies [27,28].
Pain-related pharmacotherapy included: opioid analgesics, non-opioid analgesics, antidepressants used in the treatment of pain, anticonvulsants, other oral and topical pain-related medications, and muscle relaxants (fibromyalgia cohort only).
Pain-related medical interventions included: physician office visits with a diagnosis of DPN, PHN, or fibromyalgia; physical therapy; acupuncture; chiropractic procedures; introduction of anesthetic agent (nerve block); neurostimulators; nerve destruction by neurolytic agent; nerve decompression; other outpatient services with a diagnosis of DPN, PHN, or fibromyalgia; emergency department visits with a diagnosis of DPN, PHN, or fibromyalgia in the primary diagnosis position, and inpatient visits with a diagnosis of DPN, PHN, or fibromyalgia in the primary diagnosis position.
Health care utilization and total expenditures from all causes were classified according to the following places of service and settings: inpatient, emergency department, physician office, other outpatient, and outpatient prescriptions. Health care expenditures were measured as the gross covered payments for all health care services or products and were adjusted to 2010 constant dollars using the Medical Care component of the Bureau of Labor Statistics Consumer Price Index .
Patient demographic characteristics, measured as of January 1, 2009, included: age in years; sex; population density; U.S. Census Bureau Geographic region of residence; insurance plan type and presence of capitated payment arrangements (a payment mechanism in which health care is paid for on a per-person, rather than a per-service, basis) in the health plan.
Patient clinical characteristics, measured throughout 2008, included: “incident” (vs “prevalent”) disease in 2009, defined as prevalent, if in 2008, the patient had a diagnosis for DPN, PHN, or fibromyalgia or evidence of a pain intervention; comorbid pDPN, PHN, and/or fibromyalgia; indices of health status (Deyo–Charlson Comorbidity Index, which is a measure ranging from 0 to 37 that assigns a score to patients based on the presence of claims coded with selected comorbidities ; count of unique NDCs, which is a measure ranging from 0 to no particular maximum [i.e., it is bounded by the number of NDC codes corresponding to the universe of medications in the United States, and therefore the true maximum would not be reached] that assigns a score of 1 to patients for each unique NDC code recorded on their prescription claims and therefore correlates directly with the number of medication types that they have received ; count of unique three-digit ICD-9-CM diagnosis codes, which is a measure ranging from 0 to no particular maximum [i.e., it is bounded by the number of three-digit ICD-9-CM diagnosis codes, and therefore the true maximum would not be reached] that assigns a score of 1 to patients for each unique three-digit ICD-9-CM diagnosis code recorded on their medical claims and therefore correlates directly with the number of comorbidities for which they have received a diagnosis ); specified comorbidities and concomitant medications.
Because this study was descriptive in nature, no statistical inference or hypothesis testing was conducted. Bivariate descriptive statistics were used to characterize all study measures within each of the three cohorts. Categorical variables are presented as the count and percentage of patients in each category. Continuous variables are summarized by providing the mean and standard deviation (SD) of the variable distribution. A post-hoc bivariate analysis was also conducted in which selected study measures were examined by age groups (65–74 years, 75–84 years, and 85+ years). Analyses were conducted using SAS System Version 9 for Microsoft Windows, Copyright 2002 (SAS Institute Inc., Cary, NC, USA).
Table ⇓ displays the summary of the study sample attrition associated with the application of each study inclusion and exclusion criterion. The final sample included 25,716 patients with pDPN, 4,712 patients with PHN, and 25,246 patients with fibromyalgia.
Table ⇓ displays the patient demographic characteristics of each study cohort. The mean (SD) age and percentage of females were: 75.2 (6.3) years and 51.2% in the pDPN cohort, 77.2 (6.9) years and 63.9% in the PHN cohort, and 74.4 (6.4) years and 73.0% in the fibromyalgia cohort, respectively.
Table ⇓ displays the patient clinical characteristics of each study cohort. A total of 33.5% of the pDPN cohort, 44.7% of the PHN cohort, and 43.8% of the fibromyalgia cohort had no evidence of treatment or diagnosis for their condition in 2008. In general, all three cohorts were characterized by numerous comorbidities and polypharmacy, as indicated by the mean (SD) number of unique three-digit ICD-9-CM codes and unique NDC codes: 15.5 (9.0) and 20.6 (9.6) in the pDPN cohort, 13.4 (8.5) and 16.0 (9.3) in the PHN cohort, and 15.6 (8.8) and 17.5 (9.8) in the fibromyalgia cohort, respectively.
Antidepressants not used to treat specified condition
Irritable bowel syndrome medications
Deyo–Charlson Comorbidity Index range 0–37; Unique NDCs range 0 to no maximum; Unique three-digit ICD-9-CM diagnosis codes 0 to no maximum.
ICD-9-CM = International Classification of Diseases, Ninth Revision, Clinical Modification; NDC = National Drug Code; pDPN = painful diabetic peripheral neuropathy; PHN = post-herpetic neuralgia; SD = standard deviation.
Table ⇓ displays pain-related health care utilization and expenditures for pharmacotherapy. In all three cohorts, the most commonly used pain-related pharmacotherapy was opioid analgesics, used by approximately two thirds of patients in each cohort. Mean (SD) pain-related expenditures for pharmacotherapy in 2009 were: $720 ($1,489) in the pDPN cohort, $794 ($1,577) in the PHN cohort, and $932 ($2,187) in the fibromyalgia cohort.
Table ⇓ displays pain-related health care utilization and expenditures for medical interventions and by place of service. In all three cohorts, treatment provided on an outpatient basis accounted for the largest share of pain-related health care expenditures, followed by pharmacotherapy. Mean (SD) pain-related total health care expenditures in 2009 were: $1,632 ($3,371) in the pDPN cohort, $1,403 ($2,725) in the PHN cohort, and $1,635 ($3,116) in the fibromyalgia cohort.
Table ⇓ displays all-cause health care utilization and expenditures by place of service. Considerably higher than pain-related health care expenditures, mean (SD) all-cause total health care expenditures in 2009 were $24,740 ($37,542) in the pDPN cohort, $16,579 ($24,417) in the PHN cohort, and $18,320 ($24,870) in the fibromyalgia cohort. The proportions of all-cause health care expenditures not classified as being pain-related were: 93.4% in the pDPN cohort (i.e., [$24,740 all cause—$1,632 pain related] ÷ $24,740); 91.5% in the PHN cohort, and 91.1% in the fibromyalgia cohort (data not shown in table). In contrast to pain-related health care expenditures, for which pain-related inpatient admissions contributed only 6–8% of the total, all-cause inpatient admissions comprised a substantial proportion (25–30%) of all-cause total health care expenditures.
Tables ⇓, ⇓, and ⇓ presents selected study measures, stratified by age group, for the pDPN, PHN and fibromyalgia cohorts, respectively. In all the three cohorts, the proportions of patients with the use of each pain medication category generally decreased as age increased, with the exception of slight non-monotonic changes in the use of opioid analgesics in the fibromyalgia cohort. The proportions of patients with pain-related physician office visits also decreased as age increased in all the three cohorts; monotonically among fibromyalgia and pDPN patients and non-monotonically among the PHN patients. The proportions of patients with all-cause inpatient admissions and emergency room visits increased substantially as age increased in all the three cohorts. Though the proportions of patients with pain-related inpatient admissions and emergency room visits differed by only slight magnitudes across the age groups, the relative changes were sometimes large (e.g., a five-fold difference was observed in the proportion of PHN patients with pain-related inpatient admissions between ages 65–74 [0.4%] and ages 85 + [2.0%]) and varied in direction across the cohorts. In all the three cohorts, mean pain-related expenditures decreased steadily as age increased. In the fibromyalgia cohort, all-cause expenditures were highest among patients ages 75–84 and lowest among patients ages 65–74. All-cause expenditures fell moderately with age in the pDPN cohort and rose moderately with age in the PHN cohort.
Patients with any tricyclic antidepressant use (N, %)
Patients with other antidepressant use (N, %)
Patients with any gabapentin use (N, %)
Patients with any pregabalin use (N, %)
Patients with other anticonvulsant use (N, %)
Patients with any other oral medication use (N, %)
Patients with other topical medication use (N, %)
Patients with muscle relaxant use (N, %)
Patients with any pain-related physician office visits (N, %)
Patients with any all-cause physician office visits (N, %)
Patients with any pain-related inpatient admissions (N, %)
Patients with any all-cause inpatient admissions (N, %)
Patients with any pain-related emergency department visits (N, %)
Patients with any all-cause emergency department visits (N, %)
Total pain-related expenditures (Mean, SD)
Total all-cause expenditures (Mean, SD)
SD = standard deviation.
The objective of this study was to describe the characteristics, treatment, and health care expenditures of Medicare supplemental-insured patients with pDPN, PHN, or fibromyalgia, finding that patients with these chronic pain conditions in Medicare populations are very complex, having numerous comorbidities and often being treated with polypharmacy. As one of the first studies to examine such aspects of these chronic pain conditions in the Medicare population, this study adds uniquely to a growing literature on common conditions in the Medicare population and advances the overall understanding of the “real-world” circumstances of these patients.
Several findings from this study demonstrate the high complexity of Medicare-supplemental enrolled patients with these chronic pain conditions. Since chronic pain often presents concomitantly with other conditions, it was not surprising that in each of the three cohorts patients typically had numerous comorbidities, and many were treated with polypharmacy. This was particularly evident from the observation that high numbers of unique three-digit ICD-9-CM codes (medians ranging from 12 to 14) and unique NDC codes (medians ranging from 15 to 19) were recorded within patients' administrative claims data histories. While high prevalence rates of comorbidities and concomitant medications in these chronic pain conditions are not necessarily unique to the older, Medicare-enrolled population, studies of these chronic pain conditions in which results were reported separately for commercially insured and Medicare-eligible patients have shown that the Medicare-enrolled patients typically had more comorbidities (generally demonstrated through the Charlson Comorbidity Index) and/or concomitant medications than commercially insured patients, differences which are likely to be related to the more advanced age of Medicare-enrolled patients [18,19,22,23].
Greater health care costs can also be indicative of greater patient complexity. Average all-cause health care expenditures were substantial in the present study, ranging from $16,579 in the PHN cohort to $24,740 in the pDPN cohort. Because of methodological differences between the present study and other related studies of commercially insured patients (e.g., in the databases, patients, and time periods examined), cross-study comparisons of cost outcomes data are difficult. For example, a large proportion of prior studies were conducted among patients who have specifically received either pregabalin or duloxetine, a group of patients that may be inherently more costly than the average patient from a sample such as the one used in the present study, which comprises both potentially “new” or “incident” cases as well as “existing” or “prevalent” cases. In studies that used a design that is relatively comparable to the present study's design, average all-cause health care expenditures have been estimated at $10,199 (in 2005 dollars, estimated among patients aged 18–64 years, based on 2005 data) and $10,911 (in 2004 dollars, estimated among patients aged 18–64 years, based on 2001–2004 data) for fibromyalgia patients, at $14,062 (in 2003 dollars, estimated among patients aged 18 years or older, based on 2002–2003 data) for pDPN patients, and at $10,054 (in 2004 dollars, estimated among patients aged 18–64 years, based on 2001–2003 data) for PHN/subacute herpetic neuralgia patients [10,11,22,23]. Thus, even when accounting for the approximate 26% growth in the cost of medical care between 2003 and 2009 (according to the Medical Care component of the Bureau of Labor Statistics Consumer Price Index) our Medicare supplemental-based health care expenditures estimates are greater than the aforementioned estimates that were largely based on populations of patients who were aged 18–64 years .
Additional research examining health care cost and utilization outcomes on samples of commercially insured patients that are more contemporary and formed in a way that is similar to that of the present study may help to further advance the understanding of key differences in the characteristics, treatment, and health care expenditures between older and younger patients.
The high proportion of total expenditures not classified as being pain related within this study also underscores the role that non-pain-related conditions play in the complexity of these patients. Similar findings have also been observed in prior studies of patients with pDPN, PHN, or fibromyalgia [11,12,15,17,22]. These findings may be driven partially by misclassification error, in which some pain-related health care utilization is not coded according to the algorithms used in the studies.
A notable finding was that in all the three cohorts, the most commonly used pain-related pharmacotherapy was opioid analgesics. Neuropathic pain guidelines have generally recommended opioid analgesics for second-line treatment after non-response to first-line medications such as anticonvulsants and serotonin-norepinephrine reuptake inhibitors [1,2]. Given that this study focused on a sample of individuals that was reflective of individuals who have had their conditions for various durations of time, the high use rates of second-line agents is not surprising. While the use of opioids can be safe in older individuals, a recent study identified opioid analgesics as one of the most commonly implicated medications in emergency hospitalizations for adverse drug events in older U.S. adults, both in absolute terms and when adjusting for person-year exposure .
In our post-hoc age-stratified analyses, we found that in all the three cohorts, the proportions of patients with the use of pain medications typically decreased as age increased, as did expenditures on pain-related health care overall. Interestingly, all-cause expenditures did not follow a consistent pattern across the three cohorts, although the proportions of patients with all-cause inpatient admissions and emergency room visits increased as age increased. Though we are unable to fully elucidate the reasons for these patterns in the present analysis, one plausible explanation for the decrease in pain-related expenditures is competing health demands (as evidenced by increases in the use of acute all-cause services). Prior research and expert commentary have implicated competing health demands, such as the need to treat a primary condition or alternative comorbidity, as a reason why pain management may sometimes be delivered suboptimally from a health care provider faced with the often complex needs of a patient [34–36]. Because this study did not quantify whether care was delivered in an optimal fashion, however, future analyses examining whether competing health demands affect the quality of pain management for older individuals may be warranted. Other equally plausible reasons for the aforementioned age-related patterns are better control over pain (as potentially evidenced by a decrease in the prevalence of pain-related medication use, although such decreases may also be due to competing health demands) or age-related differences in the underlying severity of pain—a factor which was unavailable in the study database.
This study has limitations. Administrative claims data are not collected for research purposes and are subject to coding error, which can lead to suboptimal sensitivity and specificity when measuring conditions. Physicians may fail to record ICD-9-CM diagnosis codes for conditions that do not increase reimbursement, which can also lead to suboptimal sensitivity when measuring conditions. Because acupuncture and chiropractic services may not be covered by all insurance companies, it is possible that patients may have paid for such services out of pocket, resulting in an underestimation of the use rates for such services. The Medicare Supplemental database is a convenience (non-probability) sample of individuals with employer-sponsored Medicare supplemental insurance. Because these individuals have insurance that supplements coverage gaps of traditional Medicare insurance, they may have better general access to pharmacological and medical treatments. Consequently, these data may not be generalizable to the entire U.S. Medicare population, which may have lower reimbursed costs and rates of covered medication use. Additionally, because the diagnosis codes for pDPN and fibromyalgia are nonspecific to these conditions, we required evidence of pain treatment. Patients who did not receive pain treatment would not have been included in this study, and therefore the studied sample may not be fully representative of all patients with pDPN and fibromyalgia. Additionally, because we required only one non-diagnostic medical claim with an ICD-9-CM code of interest for cohort selection, it is possible that some of these patients may have been identified as having the pain condition when in fact they did not. Administrative claims data also do not have information to distinguish with certainty between pain that is chronic vs acute for fibromyalgia, pDPN, and PHN. We determined that 67.1% of the pDPN patients, 49.6% of the PHN patients, and 42.8% of the fibromyalgia patients had two or more non-diagnostic claims (on different days) with the ICD-9-CM codes of interest for their cohort selection. Ultimately, the magnitude of these matters being potential analytical limitations is dependent on how many patients who receive both a diagnosis and subsequent treatment for pain within 60 days thereafter do not actually have one of the conditions of interest and/or experience pain that can be thought of as chronic, an issue that warrants investigation in future analyses. It is important to note that despite this particular limitation, these data were drawn from over 2.3 million individuals residing in all regions of the United States, and therefore, they possess greater generalizability than data that are drawn from a single practice or an individual payer.
The clinical significance of this study is reflected in the finding that patients with these chronic pain conditions in Medicare populations are very complex, having numerous comorbidities and often being treated with polypharmacy. The use of pain-related medications and overall expenditures on pain-related health care tends to decrease as age increases. The explanation for such decreases remains unknown and warrants further investigation. The information that this study provides may be helpful in the design of interventions to address the multi-faceted nature of older individuals' complex health care needs.
The numerous comorbidities, polypharmacy, and magnitude of expenditures in this sample of Medicare Supplemental-insured patients with pDPN, PHN, or fibromyalgia underscore the complexity and importance of appropriate management of these chronic pain patients.
Appendix Appendix 1 Oral and topical medications included in analysis
Disclosure: This study was sponsored by Pfizer, Inc. Stephen S. Johnston, Donna McMorrow, and Robert Fowler are full-time employees of Truven Health Analytics who were paid consultants to Pfizer Inc for the conduct of this study and for the development of this manuscript. Daniel Mullins was a paid consultant to Pfizer for the conduct of this study and for the development of this manuscript. Margarita Udall and Jose Alvir are full-time employees of Pfizer Inc.
This study describes the characteristics, treatment, and health care expenditures of Medicare Supplemental-insured patients with painful diabetic peripheral neuropathy, post-herpetic neuralgia, or fibromyalgia. Patients were characterized by numerous comorbidities, polypharmacy, and a high magnitude of expenditures. The study findings underscore the complexity and importance of appropriate management of these chronic pain patients.
Evaluation of healthcare resource utilization and costs in employees with pain associated with diabetic peripheral neuropathy treated with pregabalin or duloxetine. J Med Econ. 2010;13(4):738–747. [Epub 2010 Nov 24].
. Comparison of medication adherence and healthcare costs between duloxetine and pregabalin initiators among patients with fibromyalgia. Pain Pract 2011;11(3):204–216.doi: 10.1111/j.152500.2010.00412.x. [Epub 2010 Aug 27].
. (Podium presentation) Comparative performance of risk adjustment measures in a sample of commercially-insured patients under age 65—two simple measures outperform current standards. Value Health 2010;13:A4.