Changing the approach to treatment choice in epilepsy using big data.

PURPOSE: A UCB-IBM collaboration explored the application of machine learning to large claims databases to construct an algorithm for antiepileptic drug (AED) choice for individual patients. METHODS: Claims data were collected between January 2006 and September 2011 for patients with epilepsy > 16 years of age. A subset of patient claims with a valid index date of AED treatment change (new, add, or switch) were used to train the AED prediction model by retrospectively evaluating an index date treatment for subsequent treatment change. Based on the trained model, a model-predicted AED regimen with the lowest likelihood of treatment change was assigned to each patient in the group of test claims, and outcomes were evaluated to test model validity. RESULTS: The model had 72% area under receiver operator characteristic curve, indicating good predictive power. Patients who were given the model-predicted AED regimen had significantly longer survival rates (time until a treatment change event) and lower expected health resource utilization on average than those who received another treatment. The actual prescribed AED regimen at the index date matched the model-predicted AED regimen in only 13% of cases; there were large discrepancies in the frequency of use of certain AEDs/combinations between model-predicted AED regimens and those actually prescribed. CONCLUSIONS: Chances of treatment success were improved if patients received the model-predicted treatment. Using the model's prediction system may enable personalized, evidence-based epilepsy care, accelerating the match between patients and their ideal therapy, thereby delivering significantly better health outcomes for patients and providing health-care savings by applying resources more efficiently. Our goal will be to strengthen the predictive power of the model by integrating diverse data sets and potentially moving to prospective data collection.

Newer antiepileptic drug use and other factors decreasing hospital encounters.

A retrospective analysis was conducted in one claims database and was confirmed in a second independent database (covering both commercial and government insurance plans between 11/2009 and 9/2011) for the understanding of factors influencing antiepileptic drug (AED) use and the role of AEDs and other health-care factors in hospital encounters. In both datasets, epilepsy cases were identified by AED use and epilepsy diagnosis coding. Variables analyzed for effect on hospitalization rates were as follows: (1) use of first-generation AEDs or second-generation AEDs, (2) treatment changes, and (3) factors that may affect AED choice. Lower rates of epilepsy-related hospital encounters (encounters with an epilepsy diagnosis code) were associated with use of second-generation AEDs, deliberate treatment changes, and treatment by a neurologist. Epilepsy-related hospital encounters were more frequent for patients not receiving an AED and for those with greater comorbidities. On average, patients taking ≥1 first-generation AED experienced epilepsy-related hospitalizations every 684days, while those taking ≥1second-generation AED were hospitalized every 1001days (relative risk reduction of 31%, p<0.01). Prescriptions for second-generation AEDs were more common among neurologists and among physicians near an epilepsy center. Use of second-generation AEDs, access to specialty care, and deliberate efforts to change medications following epilepsy-related hospital encounters improved outcomes of epilepsy treatment based on average time between epilepsy-related hospital encounters. These factors may be enhanced by public health policies, private insurance reimbursement policies, and education of patients and physicians.