The Role of Managed Care in Improving Outcomes in Epilepsy

Supplements and Featured Publications, Strategies for Improving Outcomes in the Management of Epilepsy: Clinician and Payer Perspectives - , Volume 17, Issue 10 Suppl

Abstract

The burden of epilepsy for patients and payers is large and onerous. The heterogeneous nature of the condition and the lack of diagnostic or treatment biomarkers present considerable clinical challenge. Despite expansion of the treatment armamentarium, selection of maximally appropriate therapy for individual patients remains a challenge, especially in those with treatment-refractory epilepsy. While numerous factors play a role in resolving these uncertainties, 3 key factors stand out. An overwhelming quantity of clinical data support the notion that the most effective therapies—measured by the ability to control seizures while minimizing adverse effects—are typically the most cost-effective. A second consideration is maximizing adherence to prescribed therapies, which has been an ongoing dilemma in the treatment of epilepsy because the condition occurs episodically. Poor adherence has a profound effect on treatment and costs, and several strategies for improving adherence have been identified. If properly observed, these can improve outcomes and lower expenditures. The third consideration is the monitoring of therapeutic response. This is essentially through ongoing care by the prescriber with careful assessment of seizures, patient satisfaction with treatment, dose-related adverse effects, and the subjective aspects of idiosyncratic toxicity; however, laboratory testing can also have a role. The resources needed to provide this monitoring vary across treatments. From a systemic point of view, engagement of all stakeholders—patients, payers, and physicians—in overseeing the effective and efficient use of healthcare resources will benefit all parties involved.

(Am J Manag Care. 2011;17:S263-S270)

Appropriate Treatment of Epilepsy

Epilepsy is a complex neurological disorder that can be difficult to accurately diagnose due to its diverse and heterogeneous presentation, and the absence of a diagnostically specific biomarker. The diagnostic challenges common to all forms of epilepsy have a detrimental effect on appropriate application of available therapies and, therefore, patient outcomes. The goal of achieving diagnostic accuracy, in addition to employing appropriate treatments, must ultimately involve improved understanding of available therapies and strategies among all involved parties, including managed care organizations, primary care physicians, and specialists.1

The annual cost burden of epilepsy in the United States was estimated at $12.5 billion in 1995; more recent data suggest this figure is likely higher.2 A study conducted by Ivanova et al, which examined the direct and indirect costs of epilepsy in 2005 among privately insured US patients, determined that only 20% to 29% of costs associated with epilepsy arise directly from the disorder itself and variations in cost were dependent upon the clinical manifestation of the epilepsy.3,4 Nevertheless, the greatest cost burdens by far in patients with epilepsy are indirect costs, that is, related lost opportunities and non-medical expenses due to seizure occurrences.3-5 Variability in the use of healthcare resources, and consequent increased costs, are also dependent on the severity of the disease and frequency of seizures.6 Thus, medication-resistant seizures are the primary driver of epilepsy costs, and the bulk of the costs are not medical. Effective control of epilepsy through early diagnosis and appropriate treatment has a significant effect on the ultimate cost burden. This was demonstrated in a Spanish pharmacoeconomic analysis which used expenditure data from pediatric neurologists to determine the costs associated with childhood epilepsy. The investigators observed that the mean annual cost per child with epilepsy in 1998 was $1853 for controlled epilepsy compared with $4950 for uncontrolled epilepsy.7

The burden of uncontrolled seizures, in terms of human suffering and economic costs, is vast. Patients whose disease remains uncontrolled experience much higher rates of bodily injuries and psychiatric illnesses, particularly mood disorders, in addition to a variety of other, often debilitating, comorbidities.3,8,9 These patients also experience psychosocial isolation, limitations in daily activities, and premature death. This group has the highest risk for sudden unexpected death in epilepsy. At the same time, restrictions these patients face due to uncontrolled seizures often result in unemployment or underemployment. Inability to lead productive work lives forces many of these patients to fall outside the managed care system, often into public aid, or outside of the healthcare system altogether.3 An early epilepsy diagnosis and access to treatment, however, can eliminate much of this suffering and reduce total costs.

The Heterogeneous Nature of Epilepsy

The traditional classification of epilepsy divides the disorder into 2 categories: focal and generalized.10 Focal epilepsy (sometimes referred to as localization related) is manifested by partial-onset seizures, which may remain confined to 1 region of 1 cerebral hemisphere or become generalized. The extent of the seizure’s spread within the brain relates to the seizure’s manifestation; the terms simple partial and complex partial have been used to describe the degree of cognitive impairment during the seizure. Focal epilepsy may emerge in childhood or adulthood. Generalized epilepsy is manifested by generalized-onset seizures, which differ from focal seizures by initially appearing in both cerebral hemispheres; however, the seizures are similar in that they may be manifested as episodes of only cognitive impairment (similar to complex partial seizures in appearance) or as episodes of full body convulsion (similar to partial seizures that spread to become generalized).

Both focal epilepsy and generalized epilepsy can occur in types that are typically successfully treated with medications and in types that are rarely treatable with full seizure control. Both types can also be related to genetic abnormalities, structural abnormalities of the brain, or both.10 Particular epilepsy syndromes have been described within the larger, dichotomous classification system, and these syndromes provide for more accurate prognostication and sometimes better treatment selection than the broader class. Overall, the approach to classification continues to evolve as better scientific understanding of seizures and epilepsies emerges.

Traditional epilepsy classifications have been challenged in recent years, and the International League Against Epilepsy (ILAE) Commission on Classification and Terminology released a document in 2010 suggesting revisions of the epilepsy classifications and terminology.11 The proposed changes include revised definitions for focal and generalized seizures, as well as idiopathic and symptomatic/secondary dichotomies, although syndromic classifications have not been revised in this document (except with regard to nomenclature). The ILAE revisions should be regarded as a positive development since the proposed changes more accurately describe epilepsy as it is now understood. However, these changes are still provisional and the traditional terminology remains operative at present.

The task of differential diagnosis in epilepsy includes several distinct aspects, the first of which is to accurately identify the seizure type, or types, that the patient exhibits, and to rule out other potential confounding conditions that may resemble epileptic seizures, such as syncope or psychogenic nonepileptic seizures. In addition, a key aspect of differential diagnosis is identification of the epilepsy syndrome, although this may not always be immediately possible.12 Classification of a specific epilepsy syndrome requires consideration of several different features, in addition to the particular seizure type. The timing and clinical context of seizures play a role in classification, as do age at onset, electroencephalographic expression, and concurrent clinical features (including cognitive function).13

Accurately determining the epilepsy syndrome is enormously useful for several reasons. Prognostically, identifying the epilepsy syndrome can significantly assist the clinician in determining a likely pattern of disease progression. Moreover, determination of a specific syndrome is invaluable in helping the clinician select the most appropriate treatment, as certain syndromes are more likely to respond to particular epilepsy therapies.14 An awareness of epilepsy syndrome can also allow for greater insight into the influence of genotype on disease manifestation, and for a better understanding of the particular etiology, in cases of familial epilepsy.15 Because epilepsy is such a heterogeneous disorder, as exemplified by the variety of its syndromes and the potential genetic component, it is important for patients and clinicians to have maximum access to different antiepileptic therapies with numerous mechanisms of action. This will increase the likelihood of matching a patient to the most effective treatment.

Nonclinical Challenges to Appropriate Treatment

Aside from the challenges of appropriately diagnosing and selecting therapy for patients with epilepsy, it is important to understand some of the nonclinical hurdles that must be addressed to achieve successful outcomes. Among these challenges are the issues of adherence to a prescribed therapeutic regimen, generic substitution, and therapeutic drug monitoring.

Adherence

The issue of patient adherence to medication regimens is a common problem across a spectrum of medical conditions. In the treatment of epilepsy, nonadherence is a well-recognized problem in the clinical setting and has long been an area of primary concern.16 The association of nonadherence with breakthrough seizures has been well documented. In a retrospective study, insurance claims for over 18,000 patients with epilepsy were analyzed.17 Patients were aged 21 to 64 years, and seizure was defined as admission to a hospital or emergency department for epilepsy or nonfebrile convulsions. Nonadherence was defined as a medication possession ratio (MPR) less than 0.8; mean follow-up was 133 days for patients who had a seizure and 305 days for those who did not. A total of 2467 (14%) patients experienced a seizure, and the risk of seizure was 21% greater among patients nonadherent to their antiepileptic drug (AED) regimen compared with those who remained adherent (hazard ratio = 1.205, P = .0002).

The risks associated with medication nonadherence are not, however, limited to the recurrence of seizures. A long-term, retrospective study using Medicaid claims from 1997 to 2006 originating from patients in Florida, Iowa, and New Jersey—the RANSOM study (Research on Antiepileptic Nonadherence and Selected Outcomes in Medicaid)—included more than 33,000 patients with a diagnosis of epilepsy or nonfebrile convulsions.18 The definition of nonadherence was an MPR less than 0.8. During the study, 5405 deaths occurred, with nonadherent patients almost 5 times more likely to die than adherent patients (hazard ratio = 4.96; 95% CI, 4.66-5.27), a risk that remained significant and more than 3-fold higher in nonadherent patients after multivariate adjustments (hazard ratio = 3.32; 95% CI, 3.11-3.54). In addition, patients nonadherent to their AED regimen were more likely to experience motor vehicle injuries, fractures, hospitalizations, and emergency department visits than those who were adherent (Table 1).18

As these data suggest, nonadherence to therapy exerts a significant cost burden in epilepsy. Indeed, nonadherence may be the single largest factor related to increased costs in this patient population.19 The influence of nonadherence on treatment-specific costs was accessed using data from the PharMetrics Integrated Outcomes Database, a data source that included (at the time) 75 managed care health plans, 40 million patients, and 2 billion healthcare transactions.20 The analysis used data from 2000 to 2005, and included information for almost 11,000 patients 21 years and older (mean age, 44 years) with an epilepsy or nonfebrile convulsion diagnosis. Nonadherence was defined as an MPR less than 0.8. A total of 39% of patients were nonadherent to treatment. After regression analysis, the investigators found that hospitalization was 11% more likely in nonadherent patients (P = .013), and that inpatient costs were $1799 higher per nonadherent patient per year (P = .001). Emergency department admission was 48% more likely in nonadherent patients (P <.0001) and was associated with an increase in costs of $260 per patient per year (P <.001). The net effect on healthcare costs was an increase of $1466 in annual costs after the savings in AED prescriptions ($701) and other prescription drugs ($358) were deducted.

The pursuit of increased medication adherence involves multiple considerations, including drug formulations, compliance monitoring, and selection of therapies associated with greater tolerability and more favorable pharmacokinetics. With regard to drug formulations, increased adherence has been observed when patients are prescribed medications with once-daily or twice-daily dosing as opposed to agents that require 3 or 4 daily doses.21,22 The adherence gap between AED regimens with 2 and 3 daily doses was significant, with an even larger difference in adherence among regimens with 3 and 4 daily doses. Dosing intervals may also play an important role in adherence; agents that require dosing just before sleep and immediately upon waking are associated with lower adherence rates.

Therapeutic drug monitoring can be a useful tool to assess AED treatment adherence since poor adherence is characterized by a high degree of fluctuations in AED blood concentrations.23 In older patients, however, drug monitoring must be undertaken cautiously, as inherent absorption variability in these patients can appear to indicate nonadherence when patients are taking their medication faithfully.

Ultimately, selecting an AED possessing characteristics that optimize the likelihood of adherence may be the most important means of keeping patients on their treatment regimen and avoiding the serious events that undertreatment, or inappropriate treatment, can cause. Some newer-generation AEDs offer benefits supportive of adherence, such as dosing and titration schedules, while possessing demonstrated efficacy similar to earlier-generation AEDs.24,25 A literature review to evaluate the relative efficacy and adverse event profile of second-generation AEDs found no significant differences with regard to efficacy, but notably fewer severe side effects and drug interactions compared with earlier standard AED therapies.24,26 Although more studies are needed, the cognitive effects of some of the newer epilepsy agents appear to be less than that seen with “traditional” AEDs, which could have a positive effect on treatment adherence.25

In general, pharmacokinetic considerations appear to favor newer AEDs, and they may be associated with improved treatment adherence. As previously noted, less frequent dosing is associated with greater adherence, and some of the newer agents for epilepsy possess longer half-lives, allowing for once- or twice-daily dosing.26 Several newer AEDs also have a lower potential for drug interactions because they are minimally bound, or do not bind, to plasma proteins. The propensity of most of the earlier generation AEDs for hepatic enzyme induction—creating an elevated risk for drug interactions—is also absent with several of the newer AEDs.

Unfortunately, there are limited data from recent studies to directly compare adherence rates among patients with epilepsy receiving different antiepileptic agents. There are, however, recently reported data, from a study of 6373 patients (the majority were male) 66 years and older in the Veteran’s Health Administration healthcare system who received 1 or more of 9 different AEDs representing new- and older-generation drugs.27 The authors determined that AEDs associated with fewer adverse events, including cognitive dysfunction, as well as those that offered more convenient dosing schedules, were more likely to elicit treatment adherence within the study population. Such AEDs, in most cases, were newer- rather than older-generation agents.

Patients respond differently, and to some extent, unpredictably, to AEDs. Some AEDs are more or less effective, and more or less tolerable, in different types of epilepsy and specific patients. The diversity of agents in the current AED armamentarium—with regard to mechanisms of action and particular tolerability and safety profiles—is a direct response to patient variability and the heterogeneous nature of epilepsy. Thus, it is of the highest importance that patients have maximal access to epilepsy treatments so they have the best chance at epilepsy control and avoidance of the clinical and cost burdens associated with inadequate treatment.

Generic Substitution

The question of substitution with generic AEDs is one of great interest to patients, clinicians, and third-party payers. At present, data on the comparative efficacy and tolerability of generic products versus branded formulations are inadequate to reach any definite conclusions. Nevertheless, a variety of studies have examined the effects of switching from branded to generic AEDs on efficacy, tolerability, and costs. The economic burden associated with generic and branded AEDs in the United States was assessed in a study that used data from almost 34,000 patients from the PharMetrics database.28 This retrospective, open-cohort design study included 3 patient populations: those who received only branded AEDs, those who received only generic AEDs, and those who received both. Cost analyses were conducted by dividing patients into exclusive periods; the mean period of observation for each patient was 4.2 years. Unadjusted total costs were significantly higher for patients receiving generic AEDs, with an annualized cost difference (ACD) of $3398 (P <.05). Cost differences were associated with medical services: all-cause medical costs (ACD = $3282, P <.05) and epilepsy-specific costs (ACD = $827, P <.05). Moreover, costs for users of generic AEDs were also higher for non-AED medications (ACD = $520, P <.05); greater expenditures more than offset the savings gained by purchasing generic rather than branded drugs. The ACD remained significant after adjustment for potentially confounding factors, with total annualized costs 25.8% higher in those using generics (ACD = $3254, P <.05).

The risk of increased treatment costs with generic agents has been confirmed by a number of other studies that also indicate that generic substitution of AEDs may be associated with a greater incidence of breakthrough seizures, increased injuries and hospitalizations, and overall poor acceptance rates among patients and physicians.29-33

Therapeutic Drug Monitoring

Therapeutic drug monitoring by measuring serum concentrations of the medication was discussed above in the context of its utility for measuring adherence to medication regimens, but it sometimes has several additional benefits. Drug monitoring can, for example, help identify instances of therapeutic failure due to underdosage or overdosage (which may precipitate seizures or increase risk of adverse events or neurotoxicity), and pharmacokinetic factors that may contribute to inadequate therapy.23 This is most appropriate for AEDs that have narrow therapeutic windows or unpredictable pharmacokinetics. A systematic review of the utility of therapeutic drug monitoring in patients receiving older-generation AEDs demonstrated that monitoring was associated with better control of seizures, fewer side effects, and improved cost-effectiveness.34 The study authors also observed that therapeutic drug monitoring was useful in titrating AEDs for patients with difficult-to-control epilepsy and in situations where drug-drug interactions are suspected.

Improving Epilepsy Care

In recent years, there has been a movement to shift the structure of healthcare insurance reimbursement toward a pay-for-performance model—a need that has particularly urgency in the chronic care setting.35 This payment model is based on the notion of creating incentives for practitioners to encourage proactive improvements in patient care and the achievement of specific therapeutic goals. The treatment of epilepsy is an ideal therapeutic area for this kind of model. In order to determine appropriate treatment goals, the Centers for Disease Control and Prevention (CDC) funded a study to identify and define quality indicators that could be used to evaluate the care of adults with epilepsy. Table 2 includes a partial list of these quality indicators.35 The authors employed an array of resources to formulate the indicators, including national guidelines, expert panelists, and patient perceptions. A similar project, conducted in cooperation with the CDC, focused on quality indicators for the treatment of pediatric epilepsy in the primary care setting.36 In this case, the authors identified 30 quality indicators based on expert consensus, noting that the next step would be to validate these indicators by examining their relationship to clinical outcomes and healthcare utilization.

Implementation of these quality indicators would play an important role in moving toward a system of incentivizing healthcare practitioners to improve patient care. With healthcare reform at our doorsteps, and the number of insured expected to rise, getting patients more involved in their care is also necessary. Value-based insurance design (VBID) is one method to increase patient involvement. VBID links cost-sharing to the therapeutic value of a service: the more beneficial the service, the lower the cost share to the patient. This may be a particularly effective model to follow for the management of epilepsy since it would enable patients to take a more active role in their care, while incentivizing them to do so.37,38 Approaches to VBID include eliminating copayments for: (1) certain fundamental drugs or services; (2) patients with specific diagnoses; (3) high-risk patients with certain diagnoses; and (4) patients with specific diagnoses who participate in disease management programs.38 For a chronic disease that imposes a high burden on the healthcare system, creative methods are needed to improve care. Persuading patients to get involved in their care and providing them with incentives to do so is an important means of improving outcomes.

Conclusions

The complexity of diagnosing and treating epilepsy, and the significant burden that epilepsy exacts upon the healthcare system, necessitates a frank assessment of the true expenditures and the most effective means of improving epilepsy management. It is clear that several key clinical and nonclinical factors, if appropriately addressed, can achieve the dual goal of reducing the cost burden while improving outcomes. With regard to clinical factors, timely diagnosis and treatment with the most effective available therapies will ultimately produce the best outcomes for patients while reducing overall costs to payers. Thus, it is necessary for patients to have access to a variety of AEDs so they have the best chance at epilepsy control. Nonclinical factors, such as adherence to prescribed therapies and sometimes integration of therapeutic drug monitoring into disease management, will provide additional benefits to patients and payers. Finally, increasing the involvement of physicians, payers, and patients in the management of epilepsy, and clarifying their roles and giving each party a stake in success, is fundamental to achieving their respective goals.

Author affiliation: Department of Neurology, University of California, Los Angeles.

Funding source: This supplement was supported by UCB, Inc.

Author disclosure: Dr Stern reports serving as a consultant/advisory board member for Sunovion, Supernus Pharmaceuticals, and UCB, Inc. He reports lectureship for and receipt of honoraria from GlaxoSmithKline and UCB, Inc.

Authorship information: Concept and design; acquisition of data; analysis and interpretation of data; drafting of the manuscript; and critical revision of the manuscript for important intellectual content.

Address correspondence to: John M. Stern, MD, Department of Neurology, University of California, Los Angeles, 710 Westwood Plaza, Los Angeles, CA 90095. E-mail: jstern@ucla.edu.

1. Montouris G. How will primary care physicians, specialists, and managed care treat epilepsy in the new millennium? Neurology. 2000;55(11 suppl 3):S42-S44.

2. Begley CE, Famulari M, Annegers JF, et al. The cost of epilepsy in the United States: an estimate from population-based clinical and survey data. Epilepsia. 2000;41(3):342-351.

3. Ivanova JI, Birnbaum HG, Kidolezi Y, Qiu Y, Mallett D, Caleo S. Economic burden of epilepsy among the privately insured in the US. Pharmacoeconomics. 2010;28(8):675-685.

4. Ivanova JI, Birnbaum HG, Kidolezi Y, Qiu Y, Mallett D, Caleo S. Direct and indirect costs associated with epileptic partial onset seizures among the privately insured in the United States. Epilepsia. 2010;51(5):838-844.

5. Pugliatti M, Beghi E, Forsgren L, Ekman M, Sobocki P. Estimating the cost of epilepsy in Europe: a review with economic modeling. Epilepsia. 2007;48(12):2224-2233.

6. Beghi E, Frigeni B, Beghi M, De Compadri P, Garattini L. A review of the costs of managing childhood epilepsy. Pharmacoeconomics. 2005;23(1):27-45.

7. Argumosa A, Herranz JL. Economic costs of childhood epilepsy in Spain. Revista de Neurologia. 2000;30(2):104-108.

8. Lacey CJ, Salzberg MR, Roberts H, Trauer T, D’Souza WJ. Psychiatric comorbidity and impact on health service utilization in a community sample of patients with epilepsy. Epilepsia. 2009; 50(8):1991-1994.

9. Kanner AM. Depression in epilepsy: prevalence, clinical semiology, pathogenic mechanisms and treatment. Biol Psychiatry. 2003;54:388-398.

10. French JA, Kenner AM, Bautista B, et al. Efficacy and tolerability of the new antiepileptic drugs II: treatment of refractory epilepsy: Report of the Therapeutics and Technology Assessment Subcommittee and Quality Standards Subcommittee of the American Academy of Neurology and the American Epilepsy Society. Neurology. 2004;62;1261-1273.

11. Berg AT, Berkovic SF, Brodie MJ, et al. Revised terminology and concepts for organization of seizures and epilepsies: report of the ILAE Commission on Classification and Terminology, 2005-2009. Epilepsia. 2010;51(4):676-685.

12. Holmes G. Classifying seizures. Epilepsy.com Professionals. http://professionals.epilepsy.com/page/seizures_classified.html. Accessed October 5, 2010.

13. Muthugovindan D, Hartman AL. Pediatric epilepsy syndromes. Neurologist. 2010;16(4):223-237.

14. Schachter SC. Treatment of seizures. In: Schachter SC, Schomer DL, eds. The comprehensive evaluation and treatment of epilepsy. San Diego, CA: Academic Press; 1997:61-74.

15. Duchowny M, Harvey AS. Pediatric epilepsy syndromes: an update and critical review. Epilepsia. 1996;37(suppl 1):S26-S40.

16. Hayes SM, Melin JD, Dupuis M, Murray S, Labiner DM. Assessing the true learning needs of health care professionals in epilepsy care. Epilepsy Behav. 2007;11(3):434-441.

17. Manjunath R, Davis KL, Candrilli SD, Ettinger AB. Association of antiepileptic drug nonadherence with risk of seizures in adults with epilepsy. Epilepsy Behav. 2009;14(2):372-378.

18. Faught E, Duh MS, Weiner JR, Guérin A, Cunnington MC. Nonadherence to antiepileptic drugs and increased mortality: findings from the RANSOM Study. Neurology. 2008;71(20):1572-1578.

19. Wahab A. Difficulties in Treatment and Management of Epilepsy and Challenges in New Drug Development. Pharmaceuticals. 2010; 3:2090-2110.

20. Davis KL, Candrilli SD, Edin HM. Prevalence and cost of nonadherence with antiepileptic drugs in an adult managed care population. Epilepsia. 2008;49(3):446-454.

21. Cramer J, Vachon L, Desforges C, Sussman NM. Dose frequency and dose interval compliance with multiple antiepileptic medications during a controlled clinical trial. Epilepsia. 1995;36:1111-1117.

22. Greenberg RN. Overview of patient compliance with medication dosing: a literature review. Clin Ther. 1984;6(5):592-599.

23. Patsalos, PN, Berry DJ, Bourgeois BF, et al. Antiepileptic drugs—best practice guidelines for therapeutic drug monitoring: a position paper by the subcommission on therapeutic drug monitoring, ILAE Commission on Therapeutic Strategies. Epilepsia. 2008;49:1239-1276.

24. LaRoche SM. A new look at the second-generation antiepileptic drugs: a decade of experience. Neurologist. 2007;13(3):133-139.

25. Meador KJ. Newer anticonvulsants: dosing strategies and cognition in treating patients with mood disorders and epilepsy. J Clin Psychiatry. 2003;64(suppl 8):30-34.

26. Sabers A, Gram L. Newer anticonvulsants: comparative review of drug interactions and adverse effects. Drugs. 2000;60:23-33.

27. Zeber JE, Copeland LA, Pugh MJ. Variation in antiepileptic drug adherence among older patients with new-onset epilepsy. Ann Pharmacother. 2010;44(12):1896-1904.

28. Helmers SL, Paradis PE, Manjunath R, et al. Economic burden associated with the use of generic antiepileptic drugs in the United States. Epilepsy Behav. 2010;18(4):437-444.

29. Duh MS, Paradis PE, Latrémouille-Viau D, et al. The risks and costs of multiple-generic substitution of topiramate. Neurology. 2009;72(24):2122-2129.

30. Berg MJ, Gross RA, Haskins LS, Zingaro WM, Tomaszewski KJ. Generic substitution in the treatment of epilepsy: patient and physician perceptions. Epilepsy Behav. 2008;13(4):693-699.

31. Berg MJ, Gross RA, Tomaszewski KJ, Zingaro WM, Haskins LS. Generic substitution in the treatment of epilepsy: case evidence of breakthrough seizures. Neurology. 2008;12;71(7):525-530.

32. LeLorier J, Duh MS, Paradis PE, et al. Clinical consequences of generic substitution of lamotrigine for patients with epilepsy. Neurology. 2008;70(22, pt 2):2179-2186.

33. Andermann F, Duh MS, Gosselin A, Paradis PE. Compulsory generic switching of antiepileptic drugs: high switchback rates to branded compounds compared with other drug classes. Epilepsia. 2007;48(3):464-469.

34. Touw DJ, Neef C, Thomson AH, et al; Cost-Effectiveness of Therapeutic Drug Monitoring Committee of the International Association for Therapeutic Drug Monitoring and Clinical Toxicology. Cost-effectiveness of therapeutic drug monitoring: a systematic review. Ther Drug Monit. 2005;27:10-17.

35. Pugh MJ, Berlowitz DR, Montouris G, et al. What constitutes high quality of care for adults with epilepsy? Neurology. 2007;69(21):2020-2027. Appendix: http://www.neurology.org/content/69/21/2020/suppl/DC1.

36. Caplin DA, Rao JK, Filloux F, Bale JF, Can Orman C. Development of performance indicators for the primary care management of pediatric epilepsy: expert consensus recommendations based on the available evidence. Epilepsia. 2006;47:2011-2019.

37. Spaulding A, Fendrick AM, Herman WH, et al. A controlled trial of value-based insurance design—the MHealthy: a focus on diabetes (FOD) trial. Implement Sci. 2009;4:19.

38. Fendrick AM, Eldin ML. Value-based insurance design landscape digest. National Pharmaceutical Council; 2009. http:/www.sph.umich.edu/vbidcenter/publications/pdfs/NPC_VBIDreport_7-22-09.pdf. Accessed October 11, 2010.