BACKGROUND/RATIONALE:
The past decade has witnessed a change in recommendations for epilepsy treatment due to findings from randomized clinical trials and the development of new antiepileptic drugs (AEDs). Randomized clinical trials have demonstrated that that phenobarbital is unequivocally a suboptimal drug, and that phenytoin has pharmacokinetic and pharmacodynamic properties that make it potentially problematic for the elderly. Instead, carbamazepine, valproate, and newly developed AEDs (e.g., lamotrigine and gabapentin,) are considered more appropriate first-line alternatives due to their more favorable side-effect profiles in elderly patients. Consistent recommendations to avoid phenytoin in the elderly suggest that it is important to determine the state of treatment for these patients in the VA.
OBJECTIVE(S):
Objective 1: Identify patient, provider, and system factors predicting adoption of recommended AEDs for treatment of newly diagnosed older veterans (FY00-04).
Objective 2: Identify barriers to use of recommended AED in initial therapy for newly diagnosed older veterans with epilepsy.
METHODS:
Using existing national VA outpatient, inpatient, pharmacy and Medicare databases, we identified the extent to which treatment for older veterans newly diagnosed with epilepsy changed between FY00-FY04, and identify predictors of change at the patient (age, sex, race, disease burden), provider (specialty vs. primary care), and system levels. We used semi-structured interviews of primary care, geriatric and neurology providers to begin to identify barriers to use of recommended AED.
FINDINGS/RESULTS:
Of the 9,682 older veterans who met criteria for new-onset epilepsy between FY00-04, 2.5% received phenobarbital and 67% received phenytoin. Rates of phenobarbital decreased 42% (3.2% to 1.9%), while rates of phenytoin decreased 7% from 70.6% to 66.1%. Use of new AEDs increased significantly from 12.9% to 19.8%, due primarily to use of lamotrigine, levetiracetam and topiramate. The strongest predictors of receiving recommended AEDs were receiving the initial diagnosis of epilepsy in a neurology setting, and having a pre-existing diagnosis of multiple psychiatric comorbidities, neuropathic pain or migraine. Contrary to expectations, we found that patients treated in VA facilities that either housed epilepsy monitoring beds or that participated in VA Cooperative Study 428 were no more likely than patients in other facilities to receive recommended AEDs. System level analysis using General Estimating Equations (facility as random effect) found no significant difference for neurology "saturation" (proportion of cohort receiving neurology care) when only VA neurology clinic visits were included. However, when using both VA and Medicare data for neurology care, we found that patients in locales with low neurology saturation (less than median) were more likely to receive recommended AEDs than patients who received care in locales with high neurology saturation. Interestingly, there was no significant difference in AED type for individuals who were initially diagnosed in VA vs. Medicare settings.
Qualitative findings suggest that geriatric providers were more likely to focus on age related changes in assessing patients with new onset epilepsy or complications of chronic epilepsy treatment than Neurology or Primary Care Providers. Moreover, geriatricians and general internists were more likely than neurologists to say they would refer the patients in each scenario to an expert in geriatric epilepsy. Barriers in care focused on lack of availability of epilepsy specialists or clinical pharmacists who are knowledgeable of epilepsy medications. Providers in facilities with high rates of recommended AED use frequently acknowledged academic resources as facilitators to providing care for older veterans with epilepsy; academic resources were not identified at sites with moderate- or low-use of recommended AEDs.
IMPACT:
This study found that there has been little change in epilepsy treatment for older veterans with new onset epilepsy, despite wider availability of newer AEDs. Quantitative data suggest that geographic norms are the strongest predictor of recommended AED use, while qualitative data suggest that academic affiliates are viewed as a resource for providers at facilities with high rates of recommended AED use. Because primary care providers and geriatricians felt underprepared to care for older veterans with epilepsy, and it is likely that this population will increase, it is important to begin developing interventions to provide clinicians with the support they need to provide high quality care. While increasing the availability of neurology clinicians is ideal, clinicians identified resources such as increased home care, availability of neurologically focused clinical pharmacists, interdisciplinary clinics, and additional training as possible avenues to improve support to generalists caring for these patients.
PUBLICATIONS:
Journal Articles
- Berlowitz DR, Pugh MJ. Pharmacoepidemiology in community-dwelling elderly taking antiepileptic drugs. International Review of Neurobiology. 2007; 81: 153-63.
- Pugh MJ, Berlowitz DR, Kazis L. The impact of epilepsy on older veterans. International Review of Neurobiology. 2007; 81: 221-33.
- Pugh MJ, Foreman PJ, Berlowitz DR. Prescribing antiepileptics for the elderly: differences between guideline recommendations and clinical practice. Drugs & Aging. 2006; 23(11): 861-75.
