Several epilepsy specialists1, 2 and societies, including the American Academy of Neurology,3 have expressed concern that generic substitution of branded drugs carries a greater potential risk of detrimental patient outcomes in epilepsy than in many other conditions. This increase in risk is attributed to the narrow therapeutic range of antiepileptic drugs (AEDs). Most states in the US, however, allow pharmacists to substitute any approved generic agent for a given brand. Indeed, the manufacturer used by an individual pharmacy might change frequently according to the availability and the cost of a drug. As many generics are available for most AEDs (Table 1), a patient returning monthly for their prescription could receive a generic AED from a different manufacturer each time. A study by Duh and colleagues has compared the overall health-care expenses of patients with epilepsy for whom generic substitution of AEDs was performed with the financial costs of treating individuals with branded agents.4 In contrast to previous studies, Duh et al. assessed the effects of using generic agents from both one and more than one manufacturer. The results from the study support the hypothesis that single and, particularly, multiple generic substitutions of AEDs increase the risk of injury and health-care costs.
Duh and colleagues compiled data about prescriptions and other medical claims from 948 patients in the province of Quebec, Canada.4 The study had a retrospective cohort design and examined a total of eight AEDs. Patients were fully reimbursed for the costs of all AEDs (both generic and brand), so that the prices did not influence the choice of generic versus brand. The investigators found that the rates of 'switchback'—changing back to the branded drug after a trial on a generic equivalent—were substantially higher for all AEDs than for other classes of drugs. Moreover, and in accordance with a study on lamotrigine,5 Duh et al. found that the overall cost of health-care utilization was higher in patients taking a generic equivalent of topiramate than in those individuals receiving the branded agent. The resulting increase in cost was attributed to more-frequent use of other prescription drugs, higher rates of hospitalization, and longer hospital stays. These authors specifically showed that, with topiramate, patients who received more than one generic equivalent were the most frequent users of health care and at the highest risk of injury. Indeed, the risk of head injury or fracture was determined to be nearly threefold higher in patients switched from one generic to another than in cases of uninterrupted brand use. This increase in the risk of injury was presumably related to an increase in the number of seizures from subtherapeutic AED levels and/or an increase in toxicity because of higher than expected anticonvulsant concentrations. The study found that the overall health-care costs in patients receiving multiple generic equivalents of topiramate were $1,716 per individual per year more than in individuals who remained on the brand-name drug.
The FDA sets strict standards for the approval of generic equivalents, as do other western countries. Generics must contain the same active compounds as the branded agent and be tested in healthy individuals against the branded drug. For both peak serum concentration and total absorption, the 95% CI of the generic must fall between 80% and 125% of that of the brand. For most conditions, this amount of variability should not be of concern to patients or physicians. In epilepsy, however, a relatively small decrease in the delivered dose could result in breakthrough seizures, with potentially disastrous consequences. By contrast, an increase in absorption could result in toxicity because of a higher than expected serum concentration. Changing from one generic with a relatively low resulting concentration of the active compound (compared with the brand) to another generic with a relatively high concentration could amplify potential fluctuations in serum concentration from one prescription to the next. In rare cases, patients can also develop adverse effects related to the inert matrix of the tablet or capsule, such as an allergic reaction to dye. If the development of such an effect requires stopping or changing medication, even temporarily, seizures in a previously well-controlled patient could occur and result in injury, driver's license suspension, or even death.
At least two other studies have examined the effect of substituting generics for branded agents.5, 6 The results from both investigations showed that the rates of switchback were substantially higher for all AEDs than for other classes of drugs. These results suggested a decreased satisfaction with generic AEDs compared with the branded agents, although the reasons could not be determined. Both studies, one could argue, were subject to bias, as the physicians and patients involved in the studies were fully aware of which agents were being prescribed. One of these studies also showed that overall health-care costs increased when generic AED equivalents, as opposed to brand-name lamotrigine, were used.5 Results from another investigation demonstrated an association between generic substitution of AEDs and an increase in the use of emergency care.7
The physicians who treated the patients assessed by Duh et al. would not have been expected to be aware of multiple generic substitutions. Thus, despite the open-label design, these results are more likely to be valid than data from previous studies where prescribers were well aware that the patient was taking the branded or generic agent. Patients in the Duh et al. study might have been aware that they had received drugs from more than one generic manufacturer, and, therefore, some degree of bias could still have existed. Overall, however, the findings support an increase in health-care costs when generic AEDs are used, particularly when multiple manufacturers are involved in supplying a drug for a given patient's prescription.
Objective data are lacking on the frequency of clinically relevant problems related to generic substitution of AEDs. At the very least, results from rigorous pharmacokinetic studies comparing generics from various manufacturers would be helpful. Even with such information, generic substitution might still be problematic for certain patients, such as those individuals with conditions that lead to increases or decreases in gastric motility. The truly devastating consequences of generic substitution, such as a breakthrough seizure with accidental injury or death, are likely to be unusual. Nevertheless, the increase in costs associated with the use of brand-name AEDs must be weighed against the high personal and health-care costs of such rare instances. Physicians and patients should undertake generic substitution with caution, ideally checking serum levels of the active compound whenever a change has been made (whether brand to generic, generic to brand, or generic to generic). Furthermore, a generic from a single manufacturer should be used for a given patient when possible. Many countries in the European Union regulate the free substitution of various generic manufacturers by pharmacists, and a few states in the US have done so as well. The results of the study by Duh et al. suggest that the practice of limiting substitution for conditions with narrow therapeutic windows, particularly epilepsy, should become universal.
