RSNA 2014 

Abstract Archives of the RSNA, 2014


INE100

A Predictive Diagnostic Imaging Calculator as a Clinical Decision Support Tool

Education Exhibits

Presented on November 30, 2014
Presented as part of INS-SUB: Informatics Sunday Poster Discussions

Participants

Jose Morey MD, Presenter: Nothing to Disclose
Nora Marie Haney BS, Abstract Co-Author: Nothing to Disclose
Penny B. Cooper, Abstract Co-Author: Nothing to Disclose

BACKGROUND

In preparation for a community hospital's transition into an Accountable Care Organization, a quality improvement group of radiologists was asked to monitor the hospital’s imaging utilization. The group developed a predictive imaging calculator based on patient, physician, and department averages.

EVALUATION

An analysis was conducted of all patients of radiologic modalities from 2009 through 2013. Success of linear regressions were determined to be statistically significant by P<0.05. Diagnostic, Nuclear Medicine, CT, MRI, Ultrasound, Women's Imaging, and Special Procedures modalities were analyzed for Inpatient, Outpatient, Referred, Emergency and Surgical Day Care departments. Patient age, sex, Charlson Comorbidity Index, and primary diagnosis were analyzed. The data from 2009-2013 was used as the baseline for the predictive imaging calculator using the aforementioned patient, physician, and department specifications.

DISCUSSION

With this data, the group developed an imaging calculator with multiple advantages. One goal is to aid physicians with image utilization and diagnosis based on age, sex, comorbidity, and indices regarding primary diagnosis. If a physician goes beyond the hospital standard of what is typically ordered by a standard deviation, then an indication comes up with cheaper modality alternatives or advise that an alternative diagnosis should be considered to assist in patient management.

CONCLUSION

Predictive analytics based on patient demographics can help streamline patients for more efficient care, as well as create bundling payment models. Imaging information could be monitored within an individual institution or shared across institutions for more complex patient cases in need of imaging. Monitoring trends in imaging utilization is pivotal to enhance efficiency, decrease unnecessary imaging, reduce radiation and improve the quality of care. As the healthcare community transforms from a volume based to value based system, implementing a predictive calculator is a method radiology departments can use to improve care and demonstrate value to their ACO's.

FIGURE (OPTIONAL)

http://abstract.rsna.org/uploads/2014/14006520/14006520_v6o9.jpg

Cite This Abstract

Morey, J, Haney, N, Cooper, P, A Predictive Diagnostic Imaging Calculator as a Clinical Decision Support Tool.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14006520.html