New Website "Unravel Medicine" Aims to Do Just That

By Megan Brooks

May 14, 2019

NEW YORK (Reuters Health) - An international team of researchers headed by The University of British Columbia has made it their mission to critically appraise "high-impact" observational studies published in the medical literature and inform healthcare providers, policymakers, researchers and the general public about the strengths and limitations of these studies. And they'll do it through a first-of-a-kind website called UnravelMedicine.com.

"Observational studies are an important source of evidence in guiding clinical decision making and guiding and informing public health and policy," co-project leader Dr. Mahyar Etminan of the departments of ophthalmology and neurology at The University of British Columbia told Reuters Health by phone.

"However, observational studies are also prone to systematic sources of error and methodological limitation that may limit the validity of the study. This is reflected through a large number of published studies with contradictory results, which can have negative repercussions for patients, physicians and policy makers," Dr. Etminan explained.

Dr. Etminan has 15 years of experience in conducting large epidemiologic studies using big data that have been published in leading journals. His co-project leader is Dr. Mohammad Ali Mansournia of the Tehran University of Medical Sciences in Iran, a leader in the area of causal inference and associate editor of the American Journal of Epidemiology.

Unravel Medicine attempts to dissect the main methodological components of studies pertaining to effectiveness or harm of a medical intervention (drug, diet, device) and appraise these studies based on established epidemiological and statistical methodologies with a focus on bias. The developers created a bias score that can identify 18 different types of biases at play in observational studies that use big data.

Studies are carefully identified for review and are appraised by a group of epidemiologists and updated on a weekly basis. Priority is given to studies that have the highest public-health impact.

"The articles are selected in all specialties of medicine including general medicine, oncology, and cardiology," Dr. Etminan said. "The assessment for each study is adjudicated as a committee. We are open to reviewing more studies as our team grows in size."

The project received no funding. The website was developed and funded by the founders of Unravelmedicine.com.

Continuous and ongoing appraisal of the medical literature is important, Dr. Harlan Krumholz of Yale School of Medicine, in New Haven, Connecticut, who is not involved in the project, noted in an email to Reuters Health.

"It is good to have experts - and groups of experts - appraise the literature. There is such heterogeneity in the quality of medical studies, even among papers published in top venues. More voices, more discussion, more active critique may help reveal some of these differences and point a path to improvement. And this heterogeneity is not limited to observational studies, though it may be that more heterogeneity exists there," said Dr. Krumholz.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....