Informatics consults connect providers to clinical evidence, thorough analysis

Healthcare providers often need a lot of information in very little time, making it hard to provide value-based care to their patients. One potential solution to such scenarios, according to a new analysis published in the Journal of the American College of Radiology, is an informatics consult.

Informatics consults help connect physicians with challenging questions to clinical researchers and data scientists who can help provide answers. The idea for these consults came when a team researchers feared patients weren’t being provided with the very best care possible.

“Our healthcare system is plagued by missed opportunities, waste and harm,” wrote lead author Alejandro Schuler, MS, with the center for biomedical informatics research at Stanford University in Stanford, California, and colleagues. “Data generated in the course of care are often underutilized, scientific insight goes untranslated, and evidence is overlooked. To address these problems, we envisioned a system where aggregate patient data can be used at the bedside to provide practice-based evidence.”

Knowing the question at hand

Schuler et al. explained an informatic consult begins with a question. One example they provided was a hospitalist considering the use of imaging after spinal fusion surgery.

“The hospitalist requested a consult to determine how many patients who underwent spinal fusion surgery also had a spinal x-ray performed during the inpatient stay when the surgery was performed and in the two weeks after surgery,” the authors wrote. “We found that the majority of spinal fusion surgery patients had an x-ray taken during their inpatient stay, and fewer than 5 percent also had a second x-ray taken in the two weeks postsurgery.”

Collecting the data

The authors wrote that before the analysis even begins, an appropriate data set is extracted from the electronic health record (EHR). National claims data may also be used if necessary. Consults aren’t all the same, of course, as different questions may require different kinds of data.

To sort through their large data sets, the authors added, they use the Stanford Advanced Temporal Language Aided Search (ATLAS) engine using phenotyping, sometimes with the help of supervised machine learning. But there is still work to be done before the data can be analyzed.

“Because phenotype definitions are difficult to evaluate without expert-labeled data, stability analyses are a good way to detect potential biases,” the authors wrote. “If there are multiple alternative phenotype rules or models, the same analysis should be performed using each of them and the final results compared.”

Performing the analysis

Once the data is finalized, the researchers must to determine the type of analysis they need to perform. Options include descriptive or exploratory analysis, inferential and causal analysis, and predictive analysis.

From that point forward, it’s a matter of digging deep into the data and providing the best answer possible to the provider. According to Schuler and colleagues, 48 hours or less is the ideal amount of time it should take for the consult to take place.

“Our informatics consult service attempts to create a synergy between a thorough study using sound methods and clinical judgment, enabling the rapid generation of applicable clinical evidence where there was none before,” the authors concluded. “The informatics consult is a step towards a fully integrated learning healthcare system.”