New study demonstrates excellent prospective performance of HDAI's Risk Stratification Models across multiple data sources

BOSTON, Jan. 24, 2023 /PRNewswire/ -- New research conducted by Health Data Analytics Institute (HDAI), in collaboration with researchers at the Cleveland Clinic, has been published in Anesthesiology, the American Society of Anesthesiologists journal. The results¹ show that HDAI's models, using administrative claims history, accurately predict the likelihood of an adverse event for adult patients at hospital admission. The report is titled "Extended-age Out-of-Sample Validation of Risk Stratification Index 3.0 Models Using Commercial All-payer Claims" and is available online here.

"The objective of this research was to determine whether our suite of risk prediction models, which were built using health records from a large, Medicare population, could be applied similarly to a younger population of patients with commercial health coverage. Dataset shift is a recognized limitation of many previous predictive algorithms, so it was critical to show that our technology could be applied successfully using independent validation datasets. We observed similar or better performance when our models were applied to patients who were younger and healthier than the typical Medicare beneficiary," said Nassib Chamoun, Founder, President, and CEO of HDAI.

Methodology: Predicted outcomes included unplanned hospital admissions, in-hospital mortality, acute kidney injury, sepsis, pneumonia, respiratory failure, and a composite of major cardiac complications. Patient demographic and coding history in the year before admission provided features used to predict adverse events through 90-days post-admission. Risk Stratification Index 3.0 models, described in another recent feature article,² were prospectively applied to two large independent All Payor Claims Datasets from Utah and Oregon. Similar acceptance criteria for performance as were used previously were also utilized in this validation study.

Results: In the Utah dataset, there were 55,109 All Payer Claims admissions across 40,710 patients. In the Oregon dataset, there were 21,213 admissions from 16,951 patients. Model performance on the two state datasets exceeded the pre-specified acceptance criteria and were similar or better than in Medicare patients, with an average area under the curve of 0.83 (0.71 to 0.91). Model calibration was reasonable with an R2 of 0.93 (0.84 to 0.97) for Utah and 0.85 (0.71 to 0.91) for Oregon. The mean sensitivity for the highest 5% risk population was 28% (17 to 44) for Utah and 37% (20 to 56) for Oregon.

"This impressive extended validation of HDAI's models increases our confidence that we can successfully deploy these tools for preoperative optimization for all of our patients in an effort to reduce complications," said Kamal Maheshwari, MD, MPH, Departments of General Anesthesiology and Outcomes Research at the Cleveland Clinic and one of the study authors.

About HDAI

Health systems and Accountable Care Organizations use HDAI's predictive analytic product, named Health Vision, to support their care teams at the population and patient level, generating over 20M predictions every week. These innovative organizations are deploying the system as a stand-alone system fed by claims data in as little as 48 hours. Some are also enhancing the immediacy of the data through real-time EHR integration.

The Health Vision platform leverages the underlying models validated in this study, and enables care teams to rapidly identify patients at high risk for adverse events, along with the underlying factors that contribute to each risk. Subsequently, clinicians can leverage the patient's curated health history, including all encounters, medications, providers, tests, diagnoses, and interventions, to generate a personalized care plan, pre- and post-hospitalization.

The complete findings of the study, funded by HDAI, can be viewed online Extended-age Out-of-sample Validation of Risk Stratification Index 3.0 Models Using Commercial All-payor Claims | Anesthesiology | American Society of Anesthesiologists (asahq.org)

References

¹ Extended-age Out-of- sample Validation of Risk Stratification Index 3.0 Models Using Commercial All-payer Claims Scott Greenwald, Ph.D., George F. Chamoun, B.S., Nassib G. Chamoun, M.S., David Clain, B.S., Zhenyu Hong, M.S., Richard Jordan, Ph.D., Paul J. Manberg, Ph.D., Kamal Maheshwari, M.D., Daniel I. Sessler, M.D. Anesthesiology 2022; https://doi.org/10.1097/ALN.0000000000004477

² Greenwald S, Chamoun GF, Chamoun NG, Clain D, Hong Z, Jordan R, Manberg PJ, Maheshwari K, Sessler DI: Risk Stratification Index 3.0, a broad set of models for predicting adverse events during and after hospital admission. Anesthesiology 2022; 137:673–86

Company contact:
Carola Endicott
[email protected]
617-314-6135 

Press contact:
Erik Milster
[email protected]

SOURCE Health Data Analytics Institute