In an observational study, only 7% of incidental findings were of major clinical significance. The rest were considered medically minor or moderate.
Patient hospital stays that are longer than necessary waste money and resources, not to mention increase the chance that a patient will suffer a hospital-acquired condition, such as an infection or a fall.
Inappropriate imaging studies may contribute to longer-than-needed stays, researchers say.
Reporting their findings in the Journal of Hospital Medicine, researchers analyzed the medical records of 376 chest pain patients admitted to Johns Hopkins Bayview Medical Center over the course of 24 months.
They found that for 197—more than half—of those patients, diagnostic images yielded unexpected, incidental findings that were not related to their chest pain complaints. Only 7% of these findings were of major clinical significance.
Half were considered medically minor and 42% moderate.
Although most of the incidental findings were not of major significance, the follow-up testing they generated may be significant in terms of resource use. According to the researchers, the incidental findings are associated with a 26% increase in length of hospital stay.
"We could only say that there is a positive association; we cannot say that there is a causality," first author Venkat Gundareddy, MD, MPH, a director of operations for Collaborative Inpatient Medicine Service at Johns Hopkins Bayview Medical Center, says.
But the findings do add to evidence that incidental findings are a factor in increased length-of-stay.
He says that as imaging gets better, it's gotten more sensitive, so it will often pick up things that the care team wasn't looking for and that doesn't apply to the initial diagnosis. Such incidental findings raise two main questions.
Is it Necessary?
The first is whether such imaging was necessary in the first place. He says clinicians should remember that often, "the test might not add much to the overall diagnosis of the patient" and in some cases, can actually hurt more than they help.
For instance, for "patients coming in with chest pains, does it really make sense to subject this patient to a CT scan with contrast?" he asks.
Gundareddy says clinicians should keep in mind whether the tests they're ordering are appropriate, citing initiatives such as Choosing Wisely that aim to decrease the use of unneeded testing.
For instance, a keyword search of "chest pain" on the Choosing Wisely website's clinician recommendation list yields seven results. Among them:
- "Don't perform cardiac imaging for patients who are at low risk," from the American Society of Nuclear Cardiology
- "Don't perform stress cardiovascular magnetic resonance (CMR) in the initial evaluation of chest pain patients with low pretest probability of coronary artery disease," from the Society for Cardiovascular Magnetic Resonance
- "Don't perform stress CMR in patients with acute chest pain and high probability of coronary artery disease," from the Society for Cardiovascular Magnetic Resonance
"If it's inappropriate, it's inappropriate, whether you're in or out of the hospital," Gundareddy says. He says healthcare leaders can also help reduce unneeded testing by issuing recommendations at a system level that are to be followed by all providers, regardless of setting.
Related Link: Lown Offers Alternatives to Choosing Wisely Campaign
Gundareddy points to Providers for Responsible Ordering, which was founded at Johns Hopkins Bayview Medical Center.
Determining Appropriate Followup
The second question is whether the incidental findings should be investigated while the patient is in the hospital or whether patients should be seen after discharge by their PCPs.
"These findings might or might not be clinically significant, and although they may need attention, they don't necessarily need inpatient hospital attention," Gundareddy said in a statement about the study.
Following up with incidental findings in the outpatient setting also raises concerns about handoffs, discharge, and coordination of care.
Key to this is communication between providers, which is often lacking when they don't adequately share patient information. Gundareddy notes that using health information exchanges could be helpful.
Alexandra Wilson Pecci is an editor for HealthLeaders.
