Table 3.
Description of non-randomized controlled trials involving anti-epileptic drugs
| Source | Drugs studied | Population (setting) |
Study Design | Number of patients (mean age for cases) |
Newcastle- Ottawa Score* |
Results | Conclusion | Source of funding |
|---|---|---|---|---|---|---|---|---|
| Christopher et al., 1993 | Tegretol (Ciba-Geigy) vs. Apo-carbamazepine (Apotex) | Institutionalized young patients with epilepsy on Apo-carbamazepine for > 2 mos (Canada) | Nonrandomized prospective switching study (2 four-wk phases) | 10 in each arm (16.6) | 6 | No significant differences in mean number of seizures (4.7 +/− 8.4 versus 4.4 +/− 6.4) | No effect on seizure rates | Not listed |
| Andermann et al., 2007 | Lamictal (GSK), Frisium (Sanofi-Aventis), and Depakene (Abbott) vs. 5 versions of lamotrigine, 6 versions of clobazam, and 15 version of valproic acid | Patients with claims codes for epilepsy using brand-name AEDs who were switched to generic versions in a public payer database (Canada) | (1) Drug-by-drug determination of switchbacks and comparison to non-AEDs; (2) Retrospective cohort study of switchbacks to Lamictal vs. patients staying with lamotrigine (“non-switchbacks”) | (1) 5,062 (all AEDs) vs. 163,370 (non-AEDs); (2) 149 (switchbacks) vs. 993 (non-switchbacks) (38.5) | 6 | (1) Switchbacks for AEDs (12.9–s20.9%) higher than non-AEDs (1.5–2.9%); (2) Non-switchbacks had significant average dosage increase of 6.2% over baseline (p<0.0001) (vs. 0.9% increase over baseline for switchbacks (p=0.69)) and a significant average increase over baseline in codispensed AEDs (13.4%, p<0.0001) and non-AEDs (19.3%, p<0.0001) (vs. 4.6% (p=0.13) and 8.8% (p=0.04) decrease from baseline for switchbacks | More switchbacks among AED users despite pre-auth hurdles and generic AED use associated with higher average dosage and use of other drugs | GSK (maker of Lamictal) |
| LeLorier et al., 2008 | Lamictal, Neurontin (Pfizer), Tegretol CR (Novartis), and Frisium vs. lamotrigine, gabapentin, carbamazepine-CR, and clobazam | Patients with claims codes for epilepsy using brand-name AEDs who were switched to generic versions in a public payer database (Canada) | (1) Aggregate determination of switchbacks and comparison to non-AEDs; (2) Retrospective cohort study of people staying with lamotrigine and switching back to Lamictal | (1) 2,784 (all AEDs) vs. 15,762 (non-AEDs); (2) 449 (Lamictal users) vs. 222 (lamotrigine users) (39) | 9 | (1) Patients receiving AEDs more likely to switchback than non-AEDs (HR 2.46; CI 1.93-3.14; p<0.0001); (2) During generic use period, other AEDs more often dispensed (RR 1.17; CI 1.14- 1.20; p<0.0001); non-AEDs more often dispensed (1.30; 1.27–1.33; p<0.0001); higher mean number outpatient visits (1.13; 1.09–1.18; p<0.0001). No difference in mean number of inpatient visits (1.14; 0.96- 1.35; p=0.13). | Higher healthcare utilization when switching from brand-name to generic AED | GSK (maker of Lamictal) |
| Zachry et al., 2008 | Any brand-name AED with generic version available from July 2006 – Dec 2006 | Patients with claims codes for epilepsy and on AED treatment in database of managed Medicare, Medicaid, and commercially insured patients (US) | Case-control study with controls matched 1:3 by seizure diagnosis type and age (within 5 yrs) | 416 cases vs. 1248 controls (37.4) | 8 | OR 1.81 (CI 1.25–2.63) for association between switch in AED formulation and seizure | Switching associated with seizure | Abbott (maker of Zonegran, for which generic version first available in July 2006) |
| Duh et al., 2009 | Topamax (Ortho-McNeil) vs. topiramate | Patients with claims codes for epilepsy using Topamax who were switched to a generic version in a public payer database (Canada) | (1) Aggregate determination of AED switching metrics, with comparison to non-AEDs; (2) Retrospective cohort study of brand and generic use periods for patients on Topamax and/or topiramate, stratified by receipt of single or multiple generic versions | (1) 3,667 (users of AEDs with generic entry on or after 2000) vs. 16,781 (users of AEDs with generic entry before 2000) vs. 9,806 (non-AEDs); (2) 948 Topamax or topiramate users (33.7- 37.5) | 5 | (1) Patients receiving older AEDs more likely to switchback than those receiving newer AEDs (19.2% vs. 14.7%), and both have higher switchback rates than non-AEDs (7.8%); (2) Multiple-generic periods associated with higher incidence of hospitalization relative to brand-only use (IRR 1.65, CI 1.3–2.1), but difference between single-generic period and brand-only use not significant (IRR 0.95, CI 0.95–1.02). No difference in mean number of outpatient visits. | Higher healthcare utilization and costs when switching from brand-name to generic AED | Ortho-McNeil Janssen Scientific Affairs (maker of Topamax) |
| Rascati et al., 2009 | Any brand-name AED with generic version available from April 2005 – Dec 2006 | Patients with claims codes for epilepsy, aged 12–64, and on AED treatment in database of over 55 million patients (US) | Case-control study with controls matched 1:3 by seizure diagnosis type (generalized, partial or other, and intractable or not), sex, and age (within 5 yrs) | 991 cases vs. 2973 controls (35.6) | 9 | OR 1.84 (CI 1.44–2.36) for association between switch in AED formulation and seizure. Adjusting for confounding due to age, sex, region of residence, diagnosis, and use of multiple antiepileptic drugs gave OR of 1.51 (1.17–1.96). | Switching associated with seizure | Abbott (unrestricted educational grant) |
| Hansen et al., 2009 | Any brand-name AED with generic version available in 2006 | Patients with claims codes for epilepsy, aged 12–64, and on AED treatment in database of commercially insured patients (US) | Case-control study with controls matched 1:3 by seizure diagnosis type (generalized, partial or other, and intractable or not) and age (within 5 yrs) | 757 cases vs. 2271 controls (36.9) | 8 | Adjusted OR 1.57 (CI 1.17 – 2.10) for association between switch in AED formulation and emergently-treated epilepsy-related event | Switching associated with seizure | Abbott (unrestricted educational grant) and Lilly |
RR = risk ratio, CI = confidence interval, OR = odds ratio, AED = anti-epileptic drug, IRR = incidence rate ratio
*
Newcastle-Ottawa score ranges from 1–9 stars, with 9 stars indicating highest quality.30