Subscribe to eTOC

Neurology Ups the Ante With Levels of Evidence Classification


All papers describing therapeutic clinical trials submitted to Neurology after Jan. 15 will be required to include level of evidence classification.

If you've been thoroughly reading your Neurology lately, you will have likely noticed that all papers describing therapeutic clinical trials submitted after Jan. 15 will be required to include level of evidence (LOE) classification. LOE is a standardized system that organizes the evidence in studies into meaningful levels of recommendations for physicians. This helps them make better assessments about whether a particular study should influence their clinical practice — supporting the AAN goal to make clinical decision-making increasingly evidence-based.

“We agree that this new service to the readership of neurology will allow them to more clearly and concisely understand the strength of the evidence being provided so they can consider how they will use the data to inform their clinical decisions,” Karen C. Johnston, MD, the Harrison Distinguished Professor and chair of the department of neurology at the University of Virginia, told Neurology Today.

Dr. Johnston co-authored an editorial with Neurology Deputy Editor Robert A. Gross, MD, PhD, in the Jan. 6 issue, announcing the change and explaining the AAN classifications for LOE.

Currently only papers on therapeutic trials — representing one-fifth of the 10 papers published each week — now include LOE, but the journal's goal is to ultimately include LOE with all articles.


Using criteria developed by the AAN Quality Standards Subcommittee (QSS), the evidence is classified into one of four classes, from Class 1 (randomized, blinded, controlled trials in a representative population that signify the strongest evidence and a high level of reliability for practice) to Class IV (observational studies with no controls representing the least reliability for practice). A lower level of evidence does not indicate a flawed scientific study but instead refers to weaker evidence as applied to making therapeutic decisions.


No caption available.

“Basically, the rating tells the reader how much judgment they need to use when applying the results of the study to their patients,” said Gary S. Gronseth, MD, vice-chair and associate professor of neurology at the University of Kansas Medical Center. Dr. Gronseth, former co-chair of the AAN QSS, added: “A study providing Class IV evidence needs to be applied with considerable judgment, a study providing Class I evidence can be applied with much less judgment. The other element to the formalized rating process is to clearly define the patient population treated, the intervention used, and the outcomes measured. Clearly defining these features will also help the neurologist decide how applicable the study is to his or her patients.”

“I do not think this will be a huge change for the trial investigators,” said Dr. Gronseth. “Some may be a bit disappointed if their study is rated Class III or IV. It is important for them to understand that the strength-of-evidence rating does not take into account the impact of their findings. Many significant observations have come from case reports or case series. Even though such studies would be rated Class IV, their findings are potentially extremely important. After seeing the rating of such a paper, a neurologist would know to use considerable caution before applying the findings to their patients.”

Journal authors will assign an LOE for each clinical question the study was designed to address. QSS members then make the final determination of the LOE designation. The LOE system does not replace practice parameters — which assess all the available studies on a single topic in order to provide even stronger recommendations than the LOE — but rather helps in developing those guidelines.


No caption available.


Could the new system have an effect on the amount or quality of submissions? Patricia Baskin, managing editor of Neurology, said the journal will review any changes in submission rates regularly. The journal currently publishes fewer than 20 percent of original submissions (with solicited articles and Letters to the Editor, the percentage increases to 22- to 25-percent), so a drop or increase in submissions may not have much impact.

“However, because we still want the best papers to be submitted to us,” she pointed out, “it is important that the authors understand that these assessments indicate strength of the scientific evidence for making therapeutic decisions rather than implying they indicate good or bad science; the journal only accepts superb papers.” The journal's review process will be increased by two days for manuscripts assigned this new classification, she said.

“As far as we know, Neurology is the first journal in the field to require LOE for original submissions,” said Baskin. She noted that a number of other medical specialty journals require it, including Obstetrics & Gynecology, Journal of Hand Therapy, Anesthesia & Analgesia, Otolaryngology-Head and Neck Surgery, Cancer Control, and many surgical and orthopedic journals. Numerous journals have some type of rating system for quality of evidence and strength of recommendations and there are many published articles describing them.

Said Dr. Johnston: “We really feel like this is the way of the future and that before long, most if not all journals will be offering such a service to their readerships to help facilitate the dissemination of information that will improve patient care.”