Expert Review of Pharmacoeconomics & Outcomes Research
ISSN: 1473-7167 (Print) 1744-8379 (Online) Journal homepage: http://www.tandfonline.com/loi/ierp20
Outcomes associated with switching from monotherapy to adjunctive therapy for patients with partial onset seizures Zhixiao Wang, Xuan Li, Annette Powers & Jose E Cavazos To cite this article: Zhixiao Wang, Xuan Li, Annette Powers & Jose E Cavazos (2015) Outcomes associated with switching from monotherapy to adjunctive therapy for patients with partial onset seizures, Expert Review of Pharmacoeconomics & Outcomes Research, 15:2, 349-355 To link to this article: http://dx.doi.org/10.1586/14737167.2015.989217
Published online: 04 Dec 2014.
Submit your article to this journal
Article views: 139
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ierp20 Download by: [University of Wisconsin Oshkosh]
Date: 17 November 2015, At: 02:52
Downloaded by [University of Wisconsin Oshkosh] at 02:52 17 November 2015
Research Report
Outcomes associated with switching from monotherapy to adjunctive therapy for patients with partial onset seizures Expert Rev. Pharmacoecon. Outcomes Res. 15(2), 349–355 (2015)
Zhixiao Wang*1, Xuan Li1, Annette Powers1 and Jose E Cavazos2,3 1 Eisai, Health Economics and Outcomes Research, 100 Tice Blvd, Woodcliff Lake, NJ 07677, USA 2 Department of Neurology, Pharmacology, and Physiology, University of Texas, San Antonio, TX, USA 3 San Antonio VA Epilepsy Center of Excellence, South Texas, San Antonio, TX, USA *Author for correspondence: [email protected]
Background: Some patients with partial onset seizures are drug-resistant and may benefit from adjunctive therapy. This study evaluated monotherapy/sequential monotherapy versus adjunctive therapy on use/costs. Methods: Retrospective analysis using commercial/Medicare database (1 January 2007 to 31 December 2009). Patients with ‡2 diagnoses for partial onset seizures who received ‡2 prescriptions of the same antiepileptic drug were included. Outcomes assessed in the 12-month follow-up period were hospitalizations, ER visits, outpatient visits and prescription costs for patients who received monotherapy but switched to adjunctive. Results: 1353 patients met criteria. After patients transitioned to adjunctive therapy, the average monthly percentage of patients with a hospitalization decreased from 5.3 to 3.0% (p < 0.0001). Similar results occurred with epilepsy-related hospitalizations (4.0 vs 1.7%, p < 0.0001). Adjusted costs decreased significantly (US$4205 vs 2944/month, p < 0.0001). Adjusted epilepsy-related costs decreased from US$1601 to 909/month (p < 0.0001). Conclusion: Adjunctive therapy in potentially drug-resistant patients with partial onset seizures can lead to reduced healthcare use and costs. KEYWORDS: adjunctive therapy • anti-epileptic drugs • epilepsy • partial onset seizures • sequential monotherapy
Background
Epilepsy is a disorder affecting 50 million people worldwide; approximately 2 million of them are Americans [1,2]. A specific class of seizures, partial onset seizures (POS), affects localized or focal regions of the brain accounting for approximately 60% of patients with epilepsy [3,4]. The primary goal in the treatment of epilepsy, regardless of the type of seizures, is freedom from seizures with as few side effects from medications as possible [5], with the typical standard of therapy consisting of the use of antiepileptic drugs (AEDs) [6]. Therapy is usually initiated with a single AED [7]; however, even when medications are taken properly, up to 30% of patients with POS still suffer from inadequate seizure control [8–10]. Oftentimes, these patients may not respond adequately to the AED and may be considered informahealthcare.com
10.1586/14737167.2015.989217
drug-resistant. The International League Against Epilepsy defines drug-resistant epilepsy as failure of adequate trials of two tolerated and appropriately chosen and used AED schedules, either in monotherapy or in combination [11]. For patients who experience inadequate seizure control, physicians are faced with two treatment options: sequential monotherapy, whereby patients are switched to another single AED, or adjunctive therapy, when physicians add additional AED medications. The use of adjunctive therapy has been debated over the years. With the use of firstgeneration AEDs, adjunctive therapy had the propensity to increase problematic side effects in patients, thus making sequential monotherapy the preferred option for patients not well controlled. However, for patients with potential drug-resistant epilepsy, the likelihood of benefiting from subsequent trials of
2014 Informa UK Ltd
ISSN 1473-7167
349
Research Report
Wang, Li, Powers & Cavazos
Table 1. Baseline patient characteristics for all cohorts. Monotherapy only (n = 10,400)
Monotherapy to adjunctive therapy (n = 1353)
Adjunctive therapy throughout (n = 1125)
Serial monotherapy (n = 2163)
Mean age (SD)
51 (19)
47 (17)
47 (17)
49 (18)
Male
46%
41%
48%
45%
Charlson Comorbidity Index (CCI) (SD)
1.05 (1.59)
1.05 (1.59)
1.03 (1.70)
1.10 (1.66)
Downloaded by [University of Wisconsin Oshkosh] at 02:52 17 November 2015
SD: Standard deviation.
monotherapy agents may be small, and initiating adjunctive therapy earlier in the treatment plan may improve outcomes [12]. The advent of newer AEDs has led to the development of medications with differing mechanisms of actions and improved side-effect profiles. In theory, combining agents may, therefore, present fewer difficulties than were associated with older treatment regimens and is clinically a viable means to improving seizure control in drug-resistant patients. Improvement in clinical outcomes can also lead to improved economic outcomes. The economic burden of POS is substantial, predominately because of the fact that this group represents the largest number of patients with epilepsy. It has been well established that patients with refractory POS (defined as ‡3 AEDs in other studies) incur higher resource use and/or costs [13–18]. In a study comparing patients with POS to patients with refractory POS, average annual costs for refractory patients were significantly higher than non-refractory patients (US$33,613 vs US$19,085) [16]. The clinical and economic burden of POS has been quantified, but the clinical and economic evidence to support the use of adjunctive therapy is still limited, furthering the need to evaluate treatment outcomes of adjunctive therapy for the treatment of potentially drug-resistant patients in a real-world clinical practice. Although it was not possible to determine clinically whether the patients in this study were drug-resistant or not, based on the definition set forth by International League Against Epilepsy for drug-resistant epilepsy, it was assumed that at least some portion of the patients switching from monotherapy to adjunctive therapy were likely drugresistant. For our purposes, we termed these patients ‘potentially drug-resistant’. The methodology and approach for patient selection for this study were similar to other current studies evaluating adjunctive therapy [19]. The purpose of this study was to determine the effect of monotherapy/ sequential monotherapy versus adjunctive therapy on healthcare resource use and costs for patients with potentially drugresistant POS. Methods
A retrospective claims analysis was conducted using the Truven Health MarketScan Commercial Claims and Encounters (Commercial) Database covering the time period from 1 January 2007 to 31 December 2009. This database contains 350
the healthcare experience of approximately 32 million privately insured individuals and their dependents covered under a variety of health plans. Detailed enrollment, use, costs and outcomes data are available for services performed in inpatient and outpatient settings, including medical and pharmacy claims. All database records are de-identified and fully compliant with US patient confidentiality requirements. Patients with at least two diagnosis codes for POS (ICD-9 CM codes 345.4 and 345.5 for POS; 345.9 for unspecific epilepsy; and 790.39 for other convulsions as it is commonly used for POS seizures) who received ‡2 prescriptions for at least one AED were included. The first prescription date was considered to be the index date. Patients were required to be continuously eligible in the 6-month pre-period and the 12-month post-period from the index date. As a result of these criteria, patients were likely newly diagnosed as they had not received an AED in the 6 months prior to the index date. Patients who were diagnosed with bipolar disorder, migraines or headache were excluded. Four possible treatment cohorts were established based on AED prescribing patterns during the study period: patients who received monotherapy only (one/same agent throughout the follow-up period); patients who received monotherapy (one agent or sequential monotherapy) but were switched to adjunctive therapy (received more than one AED, with AED therapy overlapping >60 days); patients who received adjunctive therapy from the beginning (received more than one AED and overlapped >60 days within the first 90 days of the index date); and patients who received sequential monotherapy (received more than one AED but different AEDs only overlapped once or for £60 days). For the study objective, the analysis focused on the patient cohort who changed from monotherapy to adjunctive therapy, comparing economic outcomes between the monotherapy phase and the adjunctive therapy phase. Baseline demographics included age, gender and Charlson Comorbidity Index score. The most commonly prescribed AEDs were captured and treatment duration was measured as the average number of days on therapy. To determine the effects of the different treatment regimens, outcomes were assessed in the form of all-cause hospitalizations, emergency room (ER) visits, outpatient visits and prescription use and costs. Epilepsy-related use and costs were also evaluated, defined as a medical service with a primary or secondary Expert Rev. Pharmacoecon. Outcomes Res. 15(2), (2015)
Downloaded by [University of Wisconsin Oshkosh] at 02:52 17 November 2015
Switching from monotherapy to adjunctive therapy in patients with partial onset seizures
diagnosis of epilepsy. Average monthly costs were assessed for each medical use category to compare monotherapy/sequential monotherapy to adjunctive therapy. For this analysis, costs are representative of the amount paid for the service or medication. A subgroup analysis was also conducted by the number of AEDs received as monotherapy/sequential monotherapy to evaluate whether treatment outcomes may be affected by the number of AEDs as monotherapy prior to adjunctive therapy. Descriptive statistics were calculated for continuous and categorical variables at baseline for the study patients. For continuous variables, means (standard deviations) and medians were generated. For categorical variables, percentages were reported. Within-subject differences were tested using paired t-tests or non-parametric Wilcoxon signed-rank tests for continuous data, and McNemar tests for categorical variables. Multivariate analysis was used for the cost component and included mixed regression models on the repeated measure analysis after adjusting for patient demographics (age, gender, CCI and number of previous ICD-9 diagnosis codes). To ensure the appropriateness of the 60-day overlap for cohort classification, a 90-day overlap window was also used as sensitivity analysis to address concerns that some patients may have required longer time to switch therapies. Results Study population
A total of 15,041 patients were identified as patients with POS meeting initial inclusion criteria. The study population consisted of 10,400 patients (69.1%) in the monotherapyonly cohort, 1353 patients (9%) in the monotherapy to adjunctive therapy cohort, 1125 patients (7.5%) in the adjunctive therapy from initiation cohort and 2163 (14.4%) in the serial monotherapy cohort. The mean age of the study population ranged from 47 to 51 years, depending on the treatment cohort, with less than half of patients being male (range 41–48% depending on treatment cohort). Baseline characteristics for all cohorts are provided in TABLE 1. All cohorts were included in this table for informational purposes. Treatment characteristics
In the monotherapy to adjunctive therapy cohort, 48.7% received one AED during the monotherapy phase, 44.4% received two AEDs and 6.8% received three or more AEDs before moving to adjunctive therapy. The overall average duration of the monotherapy phase was approximately 269 days (median 239 days). Levetiracetam was the most commonly prescribed agent during the monotherapy phase of treatment for this cohort with 29.4% of patients receiving it. For the adjunctive therapy phase, levetiracetam, regardless of dosage form, was most often used in combination with phenytoin or valproic acid. Additional information regarding AED therapies for the monotherapy to adjunctive therapy cohort is provided in TABLE 2. informahealthcare.com
Research Report
Table 2. Most commonly prescribed antiepileptic drugs in the monotherapy to adjunctive therapy cohort. Monotherapy phase
Adjunctive therapy phase
Levetiracetam
29.4%
Levetiracetam/phenytoin
13.6%
Phenytoin
18.6%
Levetiracetam/valproic acid
7.5%
Lamotrigine
9.3%
Phenytoin/valproic acid
6.1%
Valproic acid
8.1%
Levetiracetam/lamotrigine
5.7%
Topiramate
7.2%
Levetiracetam/clonazepam
4.1%
Carbamazepine
7.0%
Phenytoin/phenobarbital
3.6%
Clonazepam
5.3%
Clonazepam/lamotrigine
3.5%
Gabapentin
4.9%
Valproic acid/clonazepam
3.1%
Oxcarbazepine
4.0%
Levetiracetam/topiramate
2.9%
Pregabalin
2.9%
Levetiracetam/gabapentin
2.8%
Resource use & costs
Once patients in the monotherapy to adjunctive therapy cohort transitioned from the monotherapy phase to the adjunctive therapy phase, the average monthly percentage of patients having a hospitalization for any reason (all-cause) decreased significantly from 5.3 to 3.0% (p < 0.0001). Similarly, ER visits decreased from 8.2 to 4.8% (p < 0.0001). This trend also occurred with epilepsy-related hospitalizations (4.0 vs 1.7%, p < 0.0001) and ER visits (4.5 vs 2.2%, p < 0.0001). Results are shown in FIGURES 1 & 2. Congruent with the use results, overall healthcare costs decreased significantly for patients in the monotherapy to adjunctive treatment cohort after patients transitioned over to the adjunctive phase (average monthly cost US$4147 vs US$2886, p < 0.0001). When costs were adjusted for patient demographics, costs were US$4205 in the monotherapy phase and US$2994 in the adjunctive phase (p < 0.0001, FIGURE 3). Pharmacy epilepsyrelated costs increased when patients in the monotherapy transitioned to the adjunctive therapy phase from US$189 per month
10% 9% 8% 7% 8% 5% 4% 3% 2% 1% 0%
Monotherapy Adjunctive therapy
8.2%
5.3%
4.8% 3.0%
Inpatient
ER
Figure 1. Average monthly percentage of patients with an impatient hospitalization or emergency room visit for monotherapy versus adjunctive therapy phase (all-cause)*. *p < 0.0001 for all comparisons. ER: Emergency room.
351
Research Report 10% 9% 8% 7% 6% 5% 4% 3% 2% 1% 0%
Wang, Li, Powers & Cavazos
Monotherapy Adjunctive therapy 4.5%
4.0%
Inpatient
Downloaded by [University of Wisconsin Oshkosh] at 02:52 17 November 2015
2.2%
1.7%
ER
Figure 2. Average monthly percentage of patients with an impatient hospitalization or emergency room visit for monotherapy versus adjunctive therapy phase (seizure-related)*. *p < 0.0001 for all comparisons. ER: Emergency room.
to US$194 per month, but this increase was not significant (p = 0.2082). Despite the slight increase in pharmacy costs for this treatment cohort, total epilepsy-related monthly costs decreased from US$1577 to US$885 per month after transitioning to the adjunctive therapy phase. When adjusted for patient demographics, epilepsy-related costs were US$1601 for the monotherapy phase and US$909 for the adjunctive phase (p < 0.0001, FIGURE 3). All unadjusted costs are reported in TABLE 3. Healthcare costs by number of AEDs (1, 2 or 3 AEDs) as monotherapy prior to adjunctive therapy did not show significant difference (TABLE 4). Sensitivity analysis using a 90-day overlap versus a 60-day overlap resulted in an approximately 10% decrease in sample size for the monotherapy to adjunctive therapy group (n = 1353 for 60 days vs n = 1207 for 90 days). However, results were similar and maintained directionality and significance. Discussion
When examining the history behind the monotherapy versus adjunctive therapy debate, it is important to consider the history of AEDs and their pharmacokinetic and pharmacodynamics
US$5000 US$4500
US$4205
Monotherapy Adjunctive therapy
US$4000 US$3500
US$2994
US$3000 US$2500 US$2000
US$1601
US$1500
US$909
US$1000 US$500 US$0 All-cause
Seizure-related
Figure 3. Adjusted all-cause and seizure-related costs for monotherapy versus adjunctive therapy phase*. *p < 0.0001 for all comparisons.
352
properties. The concept of sequential monotherapy began in the 1970s [20,21]. Because of the agents available on the market at the time, when patients were switched to adjunctive therapy from monotherapy, only a relatively small percentage experienced a reduction in seizures (11–13%) and most had an increase in side effects [22]. These trends began to change with the advent of newer AEDs in the 1990s. Prior to this time, sodium channel blocking agents were the mainstay of therapy, resulting in an amplification of side effects when drugs with the same mechanism of action were combined. Newer agents provide novel mechanisms of action, which result in the prospect of additive efficacy. These agents also offer less pharmacokinetic interactions and exhibit fewer side effects when combined [23,24]. Consequently, recent literature suggested that the concept of adjunctive therapy should be revisited [7,23,24]. Faught et al. concluded that adjunctive therapy should not be overlooked. His argument is that second or third monotherapy treatment sequences should be reconsidered because early polytherapy is often effective and well tolerated with the newer AEDs available [7]. To date, limited clinical evidence exists in the comparison between monotherapy and adjunctive therapy. Deckers et al. conducted a double-blind randomized study that evaluated the carbamazepine alone versus carbamazepine and valproate in combination. There were 130 patients with tonic-clonic and partial seizures that participated in the study, with neurotoxicity serving as the primary outcome measure. Results from the study demonstrated no statistical differences in the reduction of seizure frequencies, overall neurotoxicity or in overall systemic toxicity. Adverse effects differed between the two groups of patients, and fewer patients receiving adjunctive therapy (polytherapy) withdrew because of adverse effects although the difference was not significant. Although the study demonstrated adjunctive therapy is potentially more tolerable, the trial was powered to detect differences in neurotoxicity, and not designed (powered) to evaluate differences in systemic toxicity or efficacy [25]. Beghi et al. conducted a multicenter, parallelgroup, open-label study in patients with cryptogenic or symptomatic partial epilepsy who were not controlled after single or sequential AED monotherapy. Patients were randomized to receive monotherapy with an alternative AED or adjunctive therapy with a second AED, with drug and dose determined by the physician’s best judgment. Outcomes measured were the proportion of patients continuing on the assigned treatment, proportion of patients seizure-free after achieving the targeted dose and adverse effect rates. There were 157 patients who participated in the study. More patients receiving adjunctive therapy continued the assigned treatment strategy, but the difference was not significant (65 vs 55%, p = 0.74). The probability of remaining seizure free was higher for the adjunctive group (16 vs 14%, p = 0.74), and similar adverse effects were noted between the groups. The authors did note, however, that power was also an issue for this study [26]. Results from our study using real-world data indicate that patients who initially received monotherapy and transitioned to adjunctive therapy experienced significant decreases in resource Expert Rev. Pharmacoecon. Outcomes Res. 15(2), (2015)
Switching from monotherapy to adjunctive therapy in patients with partial onset seizures
Research Report
Table 3. Unadjusted average monthly costs for monotherapy versus adjunctive therapy. Mean (SD)
All-cause
Seizure-related
Monotherapy (US$)
Adjunctive (US$)
p-value
Monotherapy (US$)
Adjunctive (US$)
p-value
Inpatient
1862 (7370)
1044 (5705)
ISSN: 1473-7167 (Print) 1744-8379 (Online) Journal homepage: http://www.tandfonline.com/loi/ierp20
Outcomes associated with switching from monotherapy to adjunctive therapy for patients with partial onset seizures Zhixiao Wang, Xuan Li, Annette Powers & Jose E Cavazos To cite this article: Zhixiao Wang, Xuan Li, Annette Powers & Jose E Cavazos (2015) Outcomes associated with switching from monotherapy to adjunctive therapy for patients with partial onset seizures, Expert Review of Pharmacoeconomics & Outcomes Research, 15:2, 349-355 To link to this article: http://dx.doi.org/10.1586/14737167.2015.989217
Published online: 04 Dec 2014.
Submit your article to this journal
Article views: 139
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ierp20 Download by: [University of Wisconsin Oshkosh]
Date: 17 November 2015, At: 02:52
Downloaded by [University of Wisconsin Oshkosh] at 02:52 17 November 2015
Research Report
Outcomes associated with switching from monotherapy to adjunctive therapy for patients with partial onset seizures Expert Rev. Pharmacoecon. Outcomes Res. 15(2), 349–355 (2015)
Zhixiao Wang*1, Xuan Li1, Annette Powers1 and Jose E Cavazos2,3 1 Eisai, Health Economics and Outcomes Research, 100 Tice Blvd, Woodcliff Lake, NJ 07677, USA 2 Department of Neurology, Pharmacology, and Physiology, University of Texas, San Antonio, TX, USA 3 San Antonio VA Epilepsy Center of Excellence, South Texas, San Antonio, TX, USA *Author for correspondence: [email protected]
Background: Some patients with partial onset seizures are drug-resistant and may benefit from adjunctive therapy. This study evaluated monotherapy/sequential monotherapy versus adjunctive therapy on use/costs. Methods: Retrospective analysis using commercial/Medicare database (1 January 2007 to 31 December 2009). Patients with ‡2 diagnoses for partial onset seizures who received ‡2 prescriptions of the same antiepileptic drug were included. Outcomes assessed in the 12-month follow-up period were hospitalizations, ER visits, outpatient visits and prescription costs for patients who received monotherapy but switched to adjunctive. Results: 1353 patients met criteria. After patients transitioned to adjunctive therapy, the average monthly percentage of patients with a hospitalization decreased from 5.3 to 3.0% (p < 0.0001). Similar results occurred with epilepsy-related hospitalizations (4.0 vs 1.7%, p < 0.0001). Adjusted costs decreased significantly (US$4205 vs 2944/month, p < 0.0001). Adjusted epilepsy-related costs decreased from US$1601 to 909/month (p < 0.0001). Conclusion: Adjunctive therapy in potentially drug-resistant patients with partial onset seizures can lead to reduced healthcare use and costs. KEYWORDS: adjunctive therapy • anti-epileptic drugs • epilepsy • partial onset seizures • sequential monotherapy
Background
Epilepsy is a disorder affecting 50 million people worldwide; approximately 2 million of them are Americans [1,2]. A specific class of seizures, partial onset seizures (POS), affects localized or focal regions of the brain accounting for approximately 60% of patients with epilepsy [3,4]. The primary goal in the treatment of epilepsy, regardless of the type of seizures, is freedom from seizures with as few side effects from medications as possible [5], with the typical standard of therapy consisting of the use of antiepileptic drugs (AEDs) [6]. Therapy is usually initiated with a single AED [7]; however, even when medications are taken properly, up to 30% of patients with POS still suffer from inadequate seizure control [8–10]. Oftentimes, these patients may not respond adequately to the AED and may be considered informahealthcare.com
10.1586/14737167.2015.989217
drug-resistant. The International League Against Epilepsy defines drug-resistant epilepsy as failure of adequate trials of two tolerated and appropriately chosen and used AED schedules, either in monotherapy or in combination [11]. For patients who experience inadequate seizure control, physicians are faced with two treatment options: sequential monotherapy, whereby patients are switched to another single AED, or adjunctive therapy, when physicians add additional AED medications. The use of adjunctive therapy has been debated over the years. With the use of firstgeneration AEDs, adjunctive therapy had the propensity to increase problematic side effects in patients, thus making sequential monotherapy the preferred option for patients not well controlled. However, for patients with potential drug-resistant epilepsy, the likelihood of benefiting from subsequent trials of
2014 Informa UK Ltd
ISSN 1473-7167
349
Research Report
Wang, Li, Powers & Cavazos
Table 1. Baseline patient characteristics for all cohorts. Monotherapy only (n = 10,400)
Monotherapy to adjunctive therapy (n = 1353)
Adjunctive therapy throughout (n = 1125)
Serial monotherapy (n = 2163)
Mean age (SD)
51 (19)
47 (17)
47 (17)
49 (18)
Male
46%
41%
48%
45%
Charlson Comorbidity Index (CCI) (SD)
1.05 (1.59)
1.05 (1.59)
1.03 (1.70)
1.10 (1.66)
Downloaded by [University of Wisconsin Oshkosh] at 02:52 17 November 2015
SD: Standard deviation.
monotherapy agents may be small, and initiating adjunctive therapy earlier in the treatment plan may improve outcomes [12]. The advent of newer AEDs has led to the development of medications with differing mechanisms of actions and improved side-effect profiles. In theory, combining agents may, therefore, present fewer difficulties than were associated with older treatment regimens and is clinically a viable means to improving seizure control in drug-resistant patients. Improvement in clinical outcomes can also lead to improved economic outcomes. The economic burden of POS is substantial, predominately because of the fact that this group represents the largest number of patients with epilepsy. It has been well established that patients with refractory POS (defined as ‡3 AEDs in other studies) incur higher resource use and/or costs [13–18]. In a study comparing patients with POS to patients with refractory POS, average annual costs for refractory patients were significantly higher than non-refractory patients (US$33,613 vs US$19,085) [16]. The clinical and economic burden of POS has been quantified, but the clinical and economic evidence to support the use of adjunctive therapy is still limited, furthering the need to evaluate treatment outcomes of adjunctive therapy for the treatment of potentially drug-resistant patients in a real-world clinical practice. Although it was not possible to determine clinically whether the patients in this study were drug-resistant or not, based on the definition set forth by International League Against Epilepsy for drug-resistant epilepsy, it was assumed that at least some portion of the patients switching from monotherapy to adjunctive therapy were likely drugresistant. For our purposes, we termed these patients ‘potentially drug-resistant’. The methodology and approach for patient selection for this study were similar to other current studies evaluating adjunctive therapy [19]. The purpose of this study was to determine the effect of monotherapy/ sequential monotherapy versus adjunctive therapy on healthcare resource use and costs for patients with potentially drugresistant POS. Methods
A retrospective claims analysis was conducted using the Truven Health MarketScan Commercial Claims and Encounters (Commercial) Database covering the time period from 1 January 2007 to 31 December 2009. This database contains 350
the healthcare experience of approximately 32 million privately insured individuals and their dependents covered under a variety of health plans. Detailed enrollment, use, costs and outcomes data are available for services performed in inpatient and outpatient settings, including medical and pharmacy claims. All database records are de-identified and fully compliant with US patient confidentiality requirements. Patients with at least two diagnosis codes for POS (ICD-9 CM codes 345.4 and 345.5 for POS; 345.9 for unspecific epilepsy; and 790.39 for other convulsions as it is commonly used for POS seizures) who received ‡2 prescriptions for at least one AED were included. The first prescription date was considered to be the index date. Patients were required to be continuously eligible in the 6-month pre-period and the 12-month post-period from the index date. As a result of these criteria, patients were likely newly diagnosed as they had not received an AED in the 6 months prior to the index date. Patients who were diagnosed with bipolar disorder, migraines or headache were excluded. Four possible treatment cohorts were established based on AED prescribing patterns during the study period: patients who received monotherapy only (one/same agent throughout the follow-up period); patients who received monotherapy (one agent or sequential monotherapy) but were switched to adjunctive therapy (received more than one AED, with AED therapy overlapping >60 days); patients who received adjunctive therapy from the beginning (received more than one AED and overlapped >60 days within the first 90 days of the index date); and patients who received sequential monotherapy (received more than one AED but different AEDs only overlapped once or for £60 days). For the study objective, the analysis focused on the patient cohort who changed from monotherapy to adjunctive therapy, comparing economic outcomes between the monotherapy phase and the adjunctive therapy phase. Baseline demographics included age, gender and Charlson Comorbidity Index score. The most commonly prescribed AEDs were captured and treatment duration was measured as the average number of days on therapy. To determine the effects of the different treatment regimens, outcomes were assessed in the form of all-cause hospitalizations, emergency room (ER) visits, outpatient visits and prescription use and costs. Epilepsy-related use and costs were also evaluated, defined as a medical service with a primary or secondary Expert Rev. Pharmacoecon. Outcomes Res. 15(2), (2015)
Downloaded by [University of Wisconsin Oshkosh] at 02:52 17 November 2015
Switching from monotherapy to adjunctive therapy in patients with partial onset seizures
diagnosis of epilepsy. Average monthly costs were assessed for each medical use category to compare monotherapy/sequential monotherapy to adjunctive therapy. For this analysis, costs are representative of the amount paid for the service or medication. A subgroup analysis was also conducted by the number of AEDs received as monotherapy/sequential monotherapy to evaluate whether treatment outcomes may be affected by the number of AEDs as monotherapy prior to adjunctive therapy. Descriptive statistics were calculated for continuous and categorical variables at baseline for the study patients. For continuous variables, means (standard deviations) and medians were generated. For categorical variables, percentages were reported. Within-subject differences were tested using paired t-tests or non-parametric Wilcoxon signed-rank tests for continuous data, and McNemar tests for categorical variables. Multivariate analysis was used for the cost component and included mixed regression models on the repeated measure analysis after adjusting for patient demographics (age, gender, CCI and number of previous ICD-9 diagnosis codes). To ensure the appropriateness of the 60-day overlap for cohort classification, a 90-day overlap window was also used as sensitivity analysis to address concerns that some patients may have required longer time to switch therapies. Results Study population
A total of 15,041 patients were identified as patients with POS meeting initial inclusion criteria. The study population consisted of 10,400 patients (69.1%) in the monotherapyonly cohort, 1353 patients (9%) in the monotherapy to adjunctive therapy cohort, 1125 patients (7.5%) in the adjunctive therapy from initiation cohort and 2163 (14.4%) in the serial monotherapy cohort. The mean age of the study population ranged from 47 to 51 years, depending on the treatment cohort, with less than half of patients being male (range 41–48% depending on treatment cohort). Baseline characteristics for all cohorts are provided in TABLE 1. All cohorts were included in this table for informational purposes. Treatment characteristics
In the monotherapy to adjunctive therapy cohort, 48.7% received one AED during the monotherapy phase, 44.4% received two AEDs and 6.8% received three or more AEDs before moving to adjunctive therapy. The overall average duration of the monotherapy phase was approximately 269 days (median 239 days). Levetiracetam was the most commonly prescribed agent during the monotherapy phase of treatment for this cohort with 29.4% of patients receiving it. For the adjunctive therapy phase, levetiracetam, regardless of dosage form, was most often used in combination with phenytoin or valproic acid. Additional information regarding AED therapies for the monotherapy to adjunctive therapy cohort is provided in TABLE 2. informahealthcare.com
Research Report
Table 2. Most commonly prescribed antiepileptic drugs in the monotherapy to adjunctive therapy cohort. Monotherapy phase
Adjunctive therapy phase
Levetiracetam
29.4%
Levetiracetam/phenytoin
13.6%
Phenytoin
18.6%
Levetiracetam/valproic acid
7.5%
Lamotrigine
9.3%
Phenytoin/valproic acid
6.1%
Valproic acid
8.1%
Levetiracetam/lamotrigine
5.7%
Topiramate
7.2%
Levetiracetam/clonazepam
4.1%
Carbamazepine
7.0%
Phenytoin/phenobarbital
3.6%
Clonazepam
5.3%
Clonazepam/lamotrigine
3.5%
Gabapentin
4.9%
Valproic acid/clonazepam
3.1%
Oxcarbazepine
4.0%
Levetiracetam/topiramate
2.9%
Pregabalin
2.9%
Levetiracetam/gabapentin
2.8%
Resource use & costs
Once patients in the monotherapy to adjunctive therapy cohort transitioned from the monotherapy phase to the adjunctive therapy phase, the average monthly percentage of patients having a hospitalization for any reason (all-cause) decreased significantly from 5.3 to 3.0% (p < 0.0001). Similarly, ER visits decreased from 8.2 to 4.8% (p < 0.0001). This trend also occurred with epilepsy-related hospitalizations (4.0 vs 1.7%, p < 0.0001) and ER visits (4.5 vs 2.2%, p < 0.0001). Results are shown in FIGURES 1 & 2. Congruent with the use results, overall healthcare costs decreased significantly for patients in the monotherapy to adjunctive treatment cohort after patients transitioned over to the adjunctive phase (average monthly cost US$4147 vs US$2886, p < 0.0001). When costs were adjusted for patient demographics, costs were US$4205 in the monotherapy phase and US$2994 in the adjunctive phase (p < 0.0001, FIGURE 3). Pharmacy epilepsyrelated costs increased when patients in the monotherapy transitioned to the adjunctive therapy phase from US$189 per month
10% 9% 8% 7% 8% 5% 4% 3% 2% 1% 0%
Monotherapy Adjunctive therapy
8.2%
5.3%
4.8% 3.0%
Inpatient
ER
Figure 1. Average monthly percentage of patients with an impatient hospitalization or emergency room visit for monotherapy versus adjunctive therapy phase (all-cause)*. *p < 0.0001 for all comparisons. ER: Emergency room.
351
Research Report 10% 9% 8% 7% 6% 5% 4% 3% 2% 1% 0%
Wang, Li, Powers & Cavazos
Monotherapy Adjunctive therapy 4.5%
4.0%
Inpatient
Downloaded by [University of Wisconsin Oshkosh] at 02:52 17 November 2015
2.2%
1.7%
ER
Figure 2. Average monthly percentage of patients with an impatient hospitalization or emergency room visit for monotherapy versus adjunctive therapy phase (seizure-related)*. *p < 0.0001 for all comparisons. ER: Emergency room.
to US$194 per month, but this increase was not significant (p = 0.2082). Despite the slight increase in pharmacy costs for this treatment cohort, total epilepsy-related monthly costs decreased from US$1577 to US$885 per month after transitioning to the adjunctive therapy phase. When adjusted for patient demographics, epilepsy-related costs were US$1601 for the monotherapy phase and US$909 for the adjunctive phase (p < 0.0001, FIGURE 3). All unadjusted costs are reported in TABLE 3. Healthcare costs by number of AEDs (1, 2 or 3 AEDs) as monotherapy prior to adjunctive therapy did not show significant difference (TABLE 4). Sensitivity analysis using a 90-day overlap versus a 60-day overlap resulted in an approximately 10% decrease in sample size for the monotherapy to adjunctive therapy group (n = 1353 for 60 days vs n = 1207 for 90 days). However, results were similar and maintained directionality and significance. Discussion
When examining the history behind the monotherapy versus adjunctive therapy debate, it is important to consider the history of AEDs and their pharmacokinetic and pharmacodynamics
US$5000 US$4500
US$4205
Monotherapy Adjunctive therapy
US$4000 US$3500
US$2994
US$3000 US$2500 US$2000
US$1601
US$1500
US$909
US$1000 US$500 US$0 All-cause
Seizure-related
Figure 3. Adjusted all-cause and seizure-related costs for monotherapy versus adjunctive therapy phase*. *p < 0.0001 for all comparisons.
352
properties. The concept of sequential monotherapy began in the 1970s [20,21]. Because of the agents available on the market at the time, when patients were switched to adjunctive therapy from monotherapy, only a relatively small percentage experienced a reduction in seizures (11–13%) and most had an increase in side effects [22]. These trends began to change with the advent of newer AEDs in the 1990s. Prior to this time, sodium channel blocking agents were the mainstay of therapy, resulting in an amplification of side effects when drugs with the same mechanism of action were combined. Newer agents provide novel mechanisms of action, which result in the prospect of additive efficacy. These agents also offer less pharmacokinetic interactions and exhibit fewer side effects when combined [23,24]. Consequently, recent literature suggested that the concept of adjunctive therapy should be revisited [7,23,24]. Faught et al. concluded that adjunctive therapy should not be overlooked. His argument is that second or third monotherapy treatment sequences should be reconsidered because early polytherapy is often effective and well tolerated with the newer AEDs available [7]. To date, limited clinical evidence exists in the comparison between monotherapy and adjunctive therapy. Deckers et al. conducted a double-blind randomized study that evaluated the carbamazepine alone versus carbamazepine and valproate in combination. There were 130 patients with tonic-clonic and partial seizures that participated in the study, with neurotoxicity serving as the primary outcome measure. Results from the study demonstrated no statistical differences in the reduction of seizure frequencies, overall neurotoxicity or in overall systemic toxicity. Adverse effects differed between the two groups of patients, and fewer patients receiving adjunctive therapy (polytherapy) withdrew because of adverse effects although the difference was not significant. Although the study demonstrated adjunctive therapy is potentially more tolerable, the trial was powered to detect differences in neurotoxicity, and not designed (powered) to evaluate differences in systemic toxicity or efficacy [25]. Beghi et al. conducted a multicenter, parallelgroup, open-label study in patients with cryptogenic or symptomatic partial epilepsy who were not controlled after single or sequential AED monotherapy. Patients were randomized to receive monotherapy with an alternative AED or adjunctive therapy with a second AED, with drug and dose determined by the physician’s best judgment. Outcomes measured were the proportion of patients continuing on the assigned treatment, proportion of patients seizure-free after achieving the targeted dose and adverse effect rates. There were 157 patients who participated in the study. More patients receiving adjunctive therapy continued the assigned treatment strategy, but the difference was not significant (65 vs 55%, p = 0.74). The probability of remaining seizure free was higher for the adjunctive group (16 vs 14%, p = 0.74), and similar adverse effects were noted between the groups. The authors did note, however, that power was also an issue for this study [26]. Results from our study using real-world data indicate that patients who initially received monotherapy and transitioned to adjunctive therapy experienced significant decreases in resource Expert Rev. Pharmacoecon. Outcomes Res. 15(2), (2015)
Switching from monotherapy to adjunctive therapy in patients with partial onset seizures
Research Report
Table 3. Unadjusted average monthly costs for monotherapy versus adjunctive therapy. Mean (SD)
All-cause
Seizure-related
Monotherapy (US$)
Adjunctive (US$)
p-value
Monotherapy (US$)
Adjunctive (US$)
p-value
Inpatient
1862 (7370)
1044 (5705)
