The nation’s emergency rooms have not been consistently “choosing wisely” when it comes to CT scans of the head, according to our latest analysis of Crimson data.
For several common conditions that lead patients to visit the emergency room—headache, syncope (fainting), or minor head trauma—more than four out of five head CT scans performed in 2013 didn’t meet specialty society recommendations for appropriate use. In other words, most of these scans—which are costly—were administered even though patients didn’t need them, the results didn’t inform the course of care, and each scan exposed the patient to radiation unnecessarily.
To be fair, this is a worst-case estimate; some of these scans could have been clinically appropriate, but physicians did not record risk factors accurately in the patients’ medical records, for example. Still, based on our analysis, hospitals and physicians should be much more selective about which emergency department patients with head-related conditions really need to be sent the CT scanner.
Testing the “Choosing Wisely” guidelines
In recent years, specialty societies have developed recommendations about when it is appropriate for physicians to order diagnostic procedures, especially ones, like CT scans, that result in risks such as radiation exposure. Many of these recommendations are being promoted through the “Choosing Wisely” initiative, a campaign developed by the ABIM Foundation that has collected and communicated guidelines from across the medical community.
To date, however, there has not been much aggregate information available about how health care practitioners have been adopting these recommendations.
Because the “Choosing Wisely” guidelines on appropriate use of head CT scans in emergency patients are relatively uncontroversial and widely accepted, we decided to assess actual compliance with these guidelines using Crimson’s national clinical data set. Our Crimson data set includes charge-level information for more than a quarter of all U.S. inpatient and outpatient hospital encounters, so it provides a unique and timely way to assess the real state of health care practice in America today.
For this analysis, we looked at ED visits and imaging procedures across 2013, drawing on data feeds from over 700 hospitals in 45 states.
Real-world practice not living up to guidelines
We compared real-world behavior to three specific “Choosing Wisely” guidelines.
The first, from the American College of Emergency Physicians (ACEP), states that doctors should “avoid computed tomography (CT) scans of the head in emergency department patients with minor head injury who are at low risk based on validated decision rules.”
Our examination of 605,065 minor head injury cases found that nearly two-thirds of them resulted in head CT scans—ten times more than recommended by the "Choosing Wisely" criteria.
Guideline compliance for patients with syncope was only slightly better: 43% of syncope patients received head CT scans, but only about one of five met the Choosing Wisely criteria. (For syncope, the ACEP Choosing Wisely guidelines state: “Avoid CT of the head in asymptomatic adult patients in the emergency department with syncope, insignificant trauma and a normal neurological evaluation.”)
The last guideline we examined was on headache: patients who come to the emergency room with a headache but no other complications or risk factors should not get CT scans, according to the American College of Radiology’s “Choosing Wisely” guideline.
But again, more than four in five CT scans done on uncomplicated headache patients were not done on patients who had complications or other risk factors.
Corroboration for our findings
Lest our conclusion that over 80% of emergency head CT scans are unnecessary seem extreme, other researchers have been publishing similar findings recently.
Just last month, a study in the American Journal of Roentgenology concluded that most CT scans in the ED for syncope were unnecessary. The researchers examined the actual results of the head CT scans as well as patients’ clinical outcomes.
The study suggested that certain groups who could benefit from CT scanning include patients who are "older, have a focal neurologic deficit, or have a history of recent head trauma."
Indeed, our data shows that patients coming to the emergency room complaining of syncope are scanned much more often when they are over the age of 65. However, when we strictly applied the Choosing Wisely criteria for syncope, only 11% of older patients qualified to get a head CT scan, even though more than half actually did receive a scan.
Considering the opportunity
If emergency departments were to hew more closely to the "Choosing Wisely" guidelines, the impact on the volume of CT scans could be substantial. In total, headache, syncope, and minor head trauma accounted for 6% of all emergency department visits in our data set.
Therefore, there is almost certainly a large opportunity for hospitals to reduce the use of head CT scans in ED patients, even if some of our observations can be explained by imperfect documentation and coding. When we looked hospital by hospital at how many head-related emergency room visits met the clinical criteria for getting a CT scan, very few hospitals had over 30% of all patients meeting the clinical criteria—and that was far below the proportion of patients who actually got scans.
Financial ramifications mixed
Unfortunately, the current health care payment system does not necessarily reward providers for doing the right thing. In a fee-for-service payment environment, better compliance with these treatment guidelines is not necessarily good news for the hospital’s bottom line.
Guidelines are not gospel
In most cases, hospitals bill separately for CT scans performed on patients who come to the emergency department; if emergency department physicians order fewer CT scans of the head, the hospital gets less outpatient imaging revenue. Imaging tends to be an especially profitable service when it is billed separately in the hospital outpatient setting.
On the other hand, for health systems that bear financial risk for the cost and quality of care, fewer CT scans of the head could be beneficial, because fewer scans translates to lower overall costs.
Regardless of the financial incentives, though, this Crimson analysis should motivate emergency department physicians and administrators to review their data and revisit their protocols for headache, syncope, and minor head trauma to improve the quality of care for emergency department patients.