Expensive brand-name drugs are prescribed over older, less costly generics whose efficacy and risk profiles aren’t much different. Sometimes the financial issues involved are painfully obvious, such as when a drug company introduces a new, “improved” version of a medication that is merely a longer-acting version of the same chemical entity shortly before the patent expires on the original formulation. Patients do benefit from longer-acting formulations in some instances, and some of these formulations are even cost-effective. Often, however, no significant differences are detected in large-scale comparisons of short- and longer-acting versions of a drug. There are similarly many conditions for which several chemically distinct treatments are comparable to one another in terms of potential risks and benefits and differ primarily in price.
Medications used in the treatment of epilepsy represent one situation where available evidence clearly indicates that the best approach is to avoid using some of the old “stand-by” drugs when possible. In the case of epilepsy treatment, newer, more expensive drugs are worth the additional short-term cost, both in terms of patient outcomes and long-term health expenditures.
Certain older anti-seizure medications, including carbamazepine, phenytoin, and phenobarbital, are known as enzyme-inducing antiepileptic drugs, or EI-AEDs. They effectively prevent seizures but promote increased synthesis of certain enzymes that interfere in the metabolism of many other medications as a side-effect. The increased levels of enzymes in turn reduces levels of various other drugs in whose clearance the affected enzymes participate, including antiretrovirals, chemotherapeutics, anticoagulants, antipsychotics, and hormonal contraceptives. In addition to decreased efficacy of the drugs whose removal from the body is accelerated by EI-AEDs, toxicity and serious adverse drug reactions may result upon discontinuation or reduction in EI-AED dosage. Levels of some biological molecules, including nutrients and endogenous hormones, are also reduced by EI-AEDs.
Many clinicians continue to prescribe EI-AEDs to patients rather than using newer drugs like Keppra (levetiracetam) or Lamictal (lamotrigine), despite a growing body of evidence demonstrating that the former group of medications are both riskier for patients and no better, especially in older adults, at preventing seizures. Experts in the field have expressed an inability to provide a satisfactory explanation for the continued widespread use of EI-AEDs as a first-line therapy for new-onset epilepsy.
The consequences of EI-AEDs can be serious, not only decreasing the quality of patients’ lives but potentially even shortening them. The use of EI-AED in cancer and HIV/AIDS patients can be particularly dangerous, since they can decrease levels of treatment drugs, reducing therapeutic effect and promoting resistance. Pediatric leukemia patients taking EI-AEDs are reported to have approximately triple the relapse rates observed among patients not taking such drugs, and patients with certain brain tumors taking EI-AEDs have shorter average survival periods than patients taking non-EI-AEDs. Similarly, because of their effects on certain antiretroviral medications, guidelines for treating seizures in patients with HIV/AIDS specifically recommend avoiding EI-AED use in those on common HIV treatment regimens.
EI-AEDs also affect the metabolism of hormonal contraceptives. A recent study of more than 1,100 epileptic women reported that failure rates for hormonal contraceptives were more than four times higher among women taking EI-AEDs than among those taking non-EI-AEDs. Unintended pregnancies are especially problematic for epileptic women, who are at greater risk for complications, such as congenital abnormalities and decreased IQ scores among offspring prenatally exposed to anti-seizure drugs.
In addition to their effects on the metabolism of other medications, EI-AEDs affect levels of some biological molecules, including sex hormones and vitamin D. Use of EI-AEDs is associated with sexual problems, including decreased sexual satisfaction among women and increased sexual dysfunction in men compared to patients treated with non-EI-AEDs. Patients using EI-AEDs are also at increased risk for fractures due to drug-related bone loss.
Long-term use of EI-AEDs may also increase vascular risks. Shortly after switching from an EI- to a non-EI-AED, persistent reductions in total serum cholesterol and C-reactive protein – major risk factors for cardiovascular-related as well as all-cause mortality – are observed. Long-term use of EI-AEDs has been linked to progression of atherosclerosis, as well as associated with a 46% increased risk of myocardial infarction relative to the risk observed among patients taking non-EI-AEDs for the same duration.
The downsides to using EI-AEDs are so pronounced that epilepsy treatment guidelines specifically recommend against their initial use in new patients, especially older adults. The Quality Indicators for Epilepsy Treatment (QUIET) guidelines expressly state: “IF a newly diagnosed person with epilepsy is over the age of 60 and is not currently on any AED therapy, THEN use of enzyme inducing AEDs (phenobarbital, phenytoin, carbamazepine) should not be started unless at least two other AEDs have been unsuccessful in stopping seizures or have intolerable adverse effects.” Original iteration publicly available here; updated QUIET indicators available in supplemental file here.
A recent examination of Medicare D claims to investigate prescribing practices related to the QUIET-9 guideline found that over 40% of patients identified as new probable epileptics initially received an EI-AED during 2009, despite the existing recommendations that such drugs be avoided in new patients over 60 until they had tried and failed at least 2 non-EI-AED alternatives. Additional analyses revealed no significant disparities in receipt of QUIET-9 concordant treatment based on race or ethnicity after controlling for other factors. However, equality in substandard care is nothing to celebrate. The findings are consistent with those of other studies and demonstrate a need for improvement in this area.
Why are EI-AEDs so commonly prescribed as first-line treatment to new patients despite the explicit recommendations? No one really knows, but the reported pattern for epileptic treatment is consistent with a myriad of findings in various other areas of health care. Satisfactory explanations for such “quality problems” – ongoing patterns of medical care that fail to incorporate well-established evidence and clinical guidelines into actual practice – are lacking.
How can we improve the care received by older epileptics and prevent EI-AEDs from being used as first-line therapy? An overview of systematic reviews of interventions to promote various types of quality-related behavior change among medical providers published between 1966 and 1998 concluded that educational outreach and prescribing reminders as well as multifaceted interventions were most effective, and that passive approaches were generally unsuccessful. A more recent overview of reviews published between 2005 and 2015 drew similar conclusions.
It might be useful to include prescribing reminders in electronic health systems to notify providers inputting EI-AED prescriptions for patients over 60 that the drugs should not be used in the case of a new-onset seizure until at least 2 non-EI-AEDs had first been tried. This approach is consistent with the conclusions drawn by the authors of the more recent overview of reviews mentioned, who reported that “clinical decision support systems were found to be beneficial in improving knowledge, optimizing screening rate and prescriptions, enhancing patient outcomes, and reducing adverse events.”
It is important to note that the authors of the Medicare D claims analysis referenced above found that older patients were 83% more likely to receive QUIET-9 concordant care if seen by a neurologist close to the occurrence of their first seizure than those who did not see a specialist. Moreover, patients not seen by a neurologist close to the time of their first seizure were more likely to be African American, eligible for the low-income Part D subsidy, and reside in a high-poverty zip code. It might therefore be helpful to examine interventions to ensure and encourage appropriate specialty referrals for new epilepsy cases among Medicare beneficiaries, both to improve outcomes among older Americans in general and to reduce racial and socioeconomic disparities in epilepsy care
Barbie Zabielski