A new study by investigators from Rutgers New Jersey Medical School’s department of emergency medicine suggests that concussions are “grossly” underdiagnosed in pediatric patients who present to the ED, but it is not clear why or what the ramifications are for patients who do not receive an explicit diagnosis of concussion. The single-center study involved chart reviews of patients who presented to the pediatric ED at University Hospital in Newark, NJ, during an 18-month study period.
- Out of 627 children who were evaluated for a head injury during the study period, 233 demonstrated criteria compatible with a diagnosis of concussion, but investigators found that just 87 of these patients were diagnosed with concussion, for an overall underdiagnosis rate of 62.7%.
- Investigators found that children were more likely to be diagnosed with concussions if they arrived after a motor vehicle collision, and that they were less likely to be diagnosed if they came in after a sports-related injury or they received a CT scan.
- Despite the high rate of underdiagnoses, investigators found that providers exhibited a 95% compliance rate with Pediatric Emergency Care Applied Research Network guidelines regarding when children presenting with head injuries require a CT scan.
A provocative study conducted by investigators from Rutgers New Jersey Medical School’s department of emergency medicine suggests that pediatric concussion is “grossly” underdiagnosed in the ED, a curious finding given all the attention focused on head injuries in recent years. Investigators found that more than 60% of the children who presented with signs of concussion during an 18-month period in the dedicated pediatric ED at University Hospital in Newark, NJ, did not receive a diagnosis of concussion.
Although the results stem from just one single-center, retrospective chart review, Katie Myers, MD, the lead investigator and a third-year resident, notes that University Hospital is a Level I Trauma Center.
“We essentially see all the major head traumas that occur in northern New Jersey, and that includes pediatrics,” she observes. “The annual volume in our pediatric ED is about 17,000 patients per year.” Further, given the low bar for diagnosing concussion, spelled out in a consensus statement on concussion published in 2013, and then later updated in 2017,1 the study findings are cause for concern, Myers observes. “A head injury plus a persistent headache equals concussion, according to the consensus statement,” she explains. “If a child has had a head injury that is bad enough that [he or she] comes to the ED, we really need to be picking up these diagnoses.”
Examine Criteria, Diagnoses
A different kind of investigation prompted the study, according to Myers.
“I started off doing a retrospective chart review looking to see if we were compliant with guidelines regarding head imaging of children in the ED,” she explains. “As I started looking back through charts, I was reading these stories about these kids coming in with head injuries, and I realized there were a lot of kids that I felt met the criteria for concussion where I wasn’t seeing concussion anywhere in the chart.”
Myers decided to assess whether providers were adequately diagnosing concussion in children. She obtained institutional review board approval to conduct the study and began collecting data with her team of investigators. Specifically, they conducted retrospective chart reviews on two cohorts of patients: one group included children presenting to the ED between July and December 2013, and a second group was comprised of children presenting between January and December 2015.
Investigators used the consensus statement published in 2013 to determine whether patients met the criteria for a diagnosis of concussion, and then determined whether the diagnosis was made. Roughly two-thirds of the study population consisted of boys, and the average age of the participants was 11 years.
Among the 228 patients evaluated for a head injury between July and December 2013, 95 patients exhibited criteria indicative of concussion, but only 23 patients were diagnosed with concussion, resulting in an underdiagnosis rate of 65.2%. Among the 399 patients evaluated for a head injury between January and December 2015, 138 patients demonstrated criteria indicative of concussion, but only 54 received a diagnosis of concussion, for an underdiagnosis rate of 60.9%.
Putting both groups together, out of 627 children who were evaluated for a head injury during the study periods, 233 exhibited criteria compatible with a diagnosis of concussion, but investigators found that just 87 were diagnosed with concussion, for an overall underdiagnosis rate of 62.7%.
“We did an analysis to see if there was any difference in the diagnostic rate between the two years, and there was no statistically significant difference, so we combined the data to see what factors predicted under-diagnosis,” Myers notes. “We found that kids were more likely to be diagnosed with concussions if they came in after a motor vehicle collision, and they were less likely to be diagnosed if they came in after a sports-related injury or they had received a CT scan.”
Consider Varied Terminology
It is unclear why the sports-related injuries and CT scans were predictive of underdiagnoses of concussion, but Myers stresses that it would be wrong to conclude that emergency physicians take these types of patients less seriously.
“My initial study was actually to see whether or not we were following the PECARN [Pediatric Emergency Care Applied Research Network] guidelines,” Myers says, referring to the well-regarded rules for determining when imaging is recommended for young patients who present with head trauma.
“Our overall compliance rate with PECARN was 95%, so we appropriately scan and specifically rule out clinically important traumatic brain injury in children the way that is recommended, and that is regardless of whether children came in after a sports-related injury or a motor vehicle collision,” Myers adds.
Myers stresses that the high compliance rate with PECARN indicates that emergency providers demonstrate the appropriate level of concern for every patient with head injury who walks in the door. But why is the underdiagnosis rate for concussion so high? Myers speculates that it could be related to a misunderstanding about precisely when a specific diagnosis of concussion is appropriate, and that the patient does not need to exhibit persistent symptoms for 24 to 48 hours to meet the criteria. Then again, it could all boil down to semantics.
“There are so many different words these days for head injury, whether it be minor traumatic brain injury or head injury or concussion,” she says. “Different providers may not use the same terminology.”
Furthermore, if a patient is not diagnosed with a concussion specifically, he or she still may receive the appropriate information and guidance for a patient with concussion. Myers notes that, generally, such recommendations would include:
- a delayed return to sports or any other activities that could cause head injuries;
- brain rest, such as a delayed return to school or other cognitive activities;
- symptom management for headache, nausea, or other effects;
- neurocognitive testing to evaluate the cognitive impact of the head injury.
Indeed, the next steps for Myers are to work with colleagues to make sure that providers are aware of what the diagnostic criteria are for concussion and create new ways to facilitate the necessary follow-up care that patients who present with concussion require.
“We are working with both the department of physical medicine and rehabilitation at our institution at Rutgers as well as the department of neurology to establish a concussion follow-up clinic where we will be able to refer our patients so that they get hooked into treatment down the road, whether it be for neuro-cognitive testing or physical therapy — whatever they need in terms of outpatient care in a dedicated concussion clinic,” Myers explains.
Another focus is to make sure that both young patients and their parents receive appropriate education about what care is needed following a concussion.
“That tends to be a big problem in our patient population,” Myers observes. “We have a very urban and low health literacy patient population, so we are working on trying to provide better education.”
Myers also would like to see her study repeated at other institutions.
“If we are underdiagnosing concussion here at a Level I Trauma Center, then it is most likely being underdiagnosed at other places as well,” she says. “But maybe not. Maybe community hospitals do a better job of recognizing this.”
- McCrory P, Meeuwisse W, Dvožák J, et al. Consensus statement on concussion in sport – the 5th international conference on concussion in sport held in Berlin, October 2016. Br J Sports Med 2017;51:838-847.
- Katie Myers, MD, Emergency Physician, Department of Emergency Medicine, Rutgers New Jersey Medical School, Newark, NJ. Email: firstname.lastname@example.org.