Physicians who had access to a health information exchange (HIE) ordered fewer lab tests for patients with prior test results after the HIE was formed than they did previously, according to a new study published in the Archives of Internal Medicine.
This finding appears to contradict the results of a controversial paper recently published in Health Affairs in early March. In that study, researchers determined that physicians with access to the results of a patient's previous imaging and blood tests in an electronic health record (EHR) ordered more tests than those who did not have an EHR.
The Health Affairs paper set off a fierce debate over the value of EHRs. Farzad Mostashari, the national coordinator of health IT, publicly criticized the study, prompting a retort from its authors on the Health Affairs blog.
The new Archives study is based on the medical records of more than 117,000 patients treated in the outpatient departments of Boston-based Brigham & Women's and Massachusetts General Hospitals between 1999 and 2004. In 2000, the two hospitals--both part of Partners Healthcare--created a health information exchange (HIE).
The researchers found that 346 of the patients treated at one of the hospitals had recently received tests at the other hospitals, according to Reuters. Forty-four of those patients were tested prior to the HIE's establishment. Nearly 22,000 patients who had not recently received tests were treated at one of the hospitals before the HIE existed.
Overall, the number of tests ordered for each patient in the study fell from an average of seven before the HIE's advent to four afterward. While the number of tests ordered for people who had not undergone previous tests rose slightly, the number ordered for those with prior tests fell by 49 percent, indicating that physicians reduced the amount of redundant test ordering.
The contrast between results of the Archives study and the paper in Health Affairs may be related to the different settings in which they were conducted, Danny McCormick, a coauthor of the study, told Reuters. "We studied a different population," he noted.
Whereas the patients in the Archives study were treated in hospital outpatient departments in Boston, the other study was based on a national cohort of office visits. In addition, noted Alexander Turchin, one of the Archives paper's co-authors, physicians may be more likely to order redundant imaging tests than redundant lab tests.