Model identifies opportunities to streamline pre-surgical evaluation in epilepsy

Key takeaways:

• Researchers identified 13 factors that may prolong pre-surgical evaluations in epilepsy, nine of which significantly impacted consultations.
• Future research will use the data to inform best practices.

LOS ANGELES — Researchers used data-driven modeling techniques to identify variables that significantly impact pre-surgical consultations for patients with epilepsy, which could lead to more streamlined service by health care providers.

“We were trying to identify contributors to epilepsy pre-surgical evaluations that make the duration of the evaluation long,” M. Scott Perry, MD, medical director of the genetic epilepsy clinic at Cook Children’s Hospital Fort Worth, told Healio at the American Epilepsy Society annual meeting. “We found that certain things like getting MEG scans or SPECT scans or having a nonlesional MRI all made evaluation longer.” We take that data and look at all the institutions in the database and find the ones that do those things most efficiently.”

Although prior research has established that longer epilepsy duration is linked to worse outcomes, the impact of factors that delay pre-surgical evaluation in epilepsy are not well documented.

Perry and colleagues sought to examine the factors that lead to these delays as well as to compare average duration among 20 similar institutions and share best practices toward identifying and eliminating the variables that contribute to delays.

Their study collected data from the Pediatric Epilepsy Research Consortium surgery database. From an initial pool of 2,318 individuals, the final analysis included 1,682 patients with complete records of pre-surgical evaluation from 23 sites over a median duration of 8 weeks.

The researchers compared patient and evaluation characteristics between those with a longer (75th percentile and above) and those with shorter (25th percentile and below) evaluative intervals.

According to the results, 458 patients underwent short-term evaluations and 425 endured workups of longer duration.

Using a data-driven model, the researchers were able to predict which variables would lead to longer evaluative duration for 15 of 20 locations. Their analysis revealed 13 variables significantly associated with longer duration of pre-surgical evaluations, with nine of 13 centering on sociodemographics, geographic distance, type and length of surgery required, number of ancillary tests required, presence of brain lesions and method of non-invasive imaging.

Based on these factors, Perry and colleagues believe that the model can accurately root out the cause of pre-surgical delays, with future research requiring process mapping to provide a better framework for best practices among the member institutions.

“We know that outcomes are worse the longer the duration your drug-resistant epilepsy [lasts],” Perry said. “There are lots of ways to shorten the duration; you can identify people sooner, you can do the work up faster and you can get them to the operating room sooner. This helps you get the work up faster.”