ARRS: Decision support guides the way to more positive chest CT findings
Checkmark - 112.54 Kb
Using a decision support system for ordering chest CT exams results in positive examinations more than three-quarters of the time, according to a study presented May 1 at the annual meeting of the American Roentgen Ray Society in Vancouver.

The study, conducted at Massachusetts General Hospital in Boston by Supriya Gupta, MD, and colleagues, was designed to assess the correlation between the appropriateness score and findings in chest CT exams. “The implementation of decision support systems for ordering examinations has greatly enhanced imaging utilization. Our hypothesis was that examinations with a high utility score in the decision support system have more positive findings in radiology reports,” wrote the authors in the study abstract.

The decision support system provides a utility score for the test being ordered, according to Gupta. When given a low utility score, referring physicians are given the option to modify their imaging request, cancel the exam or proceed with the order as is.

Nearly 94,000 chest CT exams completed with the assistance of decision support from 2005 to 2010 were included in the study.

Results confirmed the authors’ hypothesis, as 76 percent of the chest CT reports in the study contained positive findings. Furthermore, 12 percent of the exams with a low utility score resulted in a positive finding, compared with 87 percent for those with a high utility score.

In addition to looking at the probability of a positive finding, the study also reviewed utilization and found it was highest when abnormal chest radiography, shortness of breath or persistent cough was listed as the indication for performing the chest CT.

"Rising concerns about radiation risks, imaging growth and reimbursement costs has highlighted inappropriate imaging use," said Gupta in a statement. The authors concluded that the results of the study could serve as a reference standard to provide feedback to ordering physicians.