CLINICIAN'S CAPSULE
What is known about the topic?
In Canada, there is no standardized discharge or follow-up protocol for patients diagnosed with a concussion in the emergency department (ED).
What did this study ask?
What proportion of patients return to the ED within 14 days of a concussion diagnosis, and what are their characteristics?
What did this study find?
Of those patients with a concussion, 10% returned to the ED within 2 weeks, most commonly for a persistent headache.
Why does this study matter to clinicians?
Evidence-based discharge instructions and specialized follow-up for patients diagnosed with a concussion in the ED may decrease return rates.
INTRODUCTION
Greater public awareness has led to an increase in the number of patients seeking medical care with possible concussions.Reference McCrory, Meeuwisse and Aubry1 In the last 5 years, emergency department (ED) visits for sport-related brain injuries in Ontario and Alberta have increased by 46%, with 94% being concussion-related.2 Individuals discharged with a concussion diagnosis often experience ongoing symptoms, which can prompt recovery concerns.Reference McCrory, Meeuwisse and Aubry1 Many Canadians do not have a family physician,3 and, thus, a return to the ED may be their only available option for additional care. Ganti et al.Reference Ganti, Conroy and Bodhit4 observed a 5% rate of return to the ED within 72 hours of the index diagnosis of a mild traumatic brain injury at one hospital in the United States. Post-concussive syndrome was the most common complaint, including headache, vomiting, or altered mental status, followed by pain.Reference Ganti, Conroy and Bodhit4 Little is known about the ED return rate among patients diagnosed with a concussion in the Canadian context.
This study sought to identify what proportion of patients discharged from the ED with a diagnosis of a concussion return within 14 days of the index visit in a Canadian setting. A subsequent focus was to describe the population of ED returners and compare them with non-returners.
METHODS
Study design
A health records review was conducted on a consecutive sample of adult patients who were diagnosed with a concussion at either the Hamilton Health Sciences (HHS) ED or Urgent Care Centre (UCC) between January and December 2016. Charts were identified using International Classification of Disease (ICD) 10 codes S060 (concussion, brain) and F07.2 (post-concussional syndrome) and reviewed by two independent researchers (LM and RT, Appendix 1). Demographics and clinical encounter details were recorded in a password-protected Microsoft Excel sheet designed a priori. Inclusion criteria were ages ≥ 18, clinical presentation adhering to the Zurich Consensus definition of concussion,Reference McCrory, Meeuwisse and Aubry1 and injury within 14 days of the index visit (Appendix 2). The diagnosis of a concussion included one or more of subjective symptoms (e.g., headache, nausea, irritability) and physical signs (e.g., loss of consciousness, amnesia, vomiting). The Zurich definition states that a concussion “typically results in the rapid onset of short-lived neurological function that resolves spontaneously.”Reference McCrory, Meeuwisse and Aubry1 Patients were included even if only subjective symptoms were noted and no neurological impairment was documented. Additional information was collected on patients who returned to an HHS site within 14 days of the index visit. Patients were excluded if they had an intracranial hemorrhage identified on a computed tomography (CT) scan or were admitted to the hospital at the index visit. Each patient's electronic health record for 2016 was manually reviewed to identify any missed return visits, previous visits, or duplicate entries. All patients had equal access to imaging, and patients seen at the UCC who required a CT scan were transported to an appropriate site. The Hamilton Integrated Research Ethics Board approved this study.
Statistical analysis
The proportion of patients returning to the ED within 14 days is calculated from the total sample of adult patients discharged from the ED with a concussion diagnosis. Dichotomous variables are presented as frequencies and percentages and were compared between groups (i.e., ED returners v. non-returners) using a Fisher exact test. Continuous variables are presented as medians (range) or means (standard deviation [SD]) and were compared using a Kruskal–Wallis or t-test, respectively. All statistical tests were performed using STATA/SE version 13.1 (StataCorp LP, College Station, TX, USA).
RESULTS
Of the 389 identified adult patients discharged from the ED with a diagnosis of a concussion, 38 (9.8%) returned to the ED within 14 days. The demographics and clinical characteristics of returners and non-returners are described in Table 1. Patients who suffered an assault-related concussion were more likely to return (21% v. 6%, p = 0.01), whereas those who suffered a sport-related concussion or were referred to a concussion clinic were less likely to return (5% v. 23%, p = 0.03; and 5% v. 25%, p = 0.04, respectively). Additionally, the distribution of the Canadian Triage and Acuity Scale (CTAS) scores, and the time between injury and index visit were statistically significantly different between returners and non-returners.
Patient demographics and visit characteristics on the index visit
a GCS data were recorded for 219 patients only.
b Red flags to return were defined as persistent nausea/vomiting, worsening headache, or new neurologic deficit.Reference McCrory, Meeuwisse and Aubry1
c A copy of the HIR can be found at http://hamiltonhealthsciences.ca/documents/Patient%20Education/HeadInjuryAndConcussion-trh.pdf.
Patients returned to the ED on average within a few days of the index visit (mean = 3.7; SD = 1.8) (Table 2, Appendix 3). The most common reason for returning to the ED was the presence of a headache (66%). Almost half of the returning patients underwent a CT scan (42%, compared with 33% of all patients on the index visit. Note: No intracranial abnormality was found on any of the CT scans in this study population). Approximately 42% of returning patients received new discharge instructions (i.e., different than what they received at their index visit) and/or a Head Injury Routine (HIR) informational handout (Appendix 4).
DISCUSSION
Interpretation
We observed that 10% of patients diagnosed with a concussion returned to the ED within 14 days of their index visit. Headache, followed by dizziness and nausea or vomiting, respectively, was the most common complaint on a return to the ED. Of the return visitors, many underwent a CT scan and/or received different discharge instructions than from their index visits. Given that there were no intracranial abnormalities identified among the return visit CT scans and patients were within the expected window of symptom occurrence, these visits theoretically could have been prevented with specific discharge education and arrangements for follow-up. The Canadian CT Head Rule applies only to patients presenting within 24 hours of injury; however, future studies may look to identify which patients returning after more than 24 hours with a symptomatic minor head injury can safely be managed without neuroimaging.Reference Stiell, Wells and Vandemheen5
Previous studies
Multiple studies highlight that patients’ understanding of discharge instructions and their recovery outcomes are improved when both verbal and written instructions are given.Reference Bergman and Louis6,Reference Rowe, Eliyahu and Lowes7,Reference Seeley, Maimarisc and Allanson8 Snell et al. (2011) have shown that patient understanding on concussions, their recovery expectations, strategies for symptom management, and instructions for a gradual return-to-activity improve recovery outcomes.Reference Snell, Siegert, Hay-Smith and Surgenor9 The HIR handout, including information on the previous topics, was available at the study centres. However, despite a relatively high compliance (83% of visits) among physicians in our study in documenting some form of discharge instruction, only 43% of patients received the HIR. Although we observed no association between receiving the handout and the ED return rate, details regarding the discharge encounter and HIR adherence were not captured in this study.
Patients who were referred to a concussion clinic were less likely to return to the ED. The clinic provides patients a forum to ask questions and access to specialty care (with a neurologist) if symptoms persist. Although studies investigating specialized follow-up after a mild head injury show no long-term benefit in symptom duration or final outcome, our study found that those referred were less likely to return to the ED.Reference Andersson, Emanuelson, Björklund and Stålhammar10–Reference Matuseviciene, Borg and Stålnacke12 The relative low rate-of-return among athletes could reflect either a proclivity to avoid healthcare professionals in order to prematurely return-to-sport or the opposite phenomenon whereby athletes are more likely to follow up with their team physician if available to be cleared for sport.Reference Yard and Comstock13 However, research points towards the former because one study reported that 40% of athletes will return to sports on the day of injury.Reference Yard and Comstock13 A higher rate-of-return among assault victims could represent a perceived necessity to document ongoing symptoms for those wishing to pursue legal action, comparable with the trends of work-place-related injuries requiring compensation.Reference Brown, McDonough, Mustard and Shannon14
Limitations
The term concussion lacks a robust definition making it difficult to succinctly identify patients. We identified patients using ICD codes and the Zurich Consensus definition. Adherence to ICD coding may result in an underrepresentation of presenting cases, and, although the Zurich criteria are internationally recognized, they were designed for use in sport-related concussions. We also did not rigorously follow the Zurich Consensus definition of a concussion in that we included patients without a history of neurological impairment. Details not charted by the attending physician cannot be captured in the data collection and thus may not be an accurate depiction of the clinical conversation. Furthermore, we could not capture return visits outside of the study centres, and so data may underestimate the proportion of returners. Finally, although this study involved multiple centres, physicians are from a single group, which may affect generalizability.
Clinical implications
Our study suggests the potential to reduce the burden to both the healthcare system and the patient through careful discharge instructions outlining anticipated symptoms following a concussion (specifically, headache) or a referral to a concussion clinic.
CONCLUSION
Approximately 10% of patients diagnosed with a concussion in a Canadian hospital setting returned to the ED within 14 days, most commonly for a headache. There were no missed intracranial abnormalities discovered at any of the return visits. Factors associated with a lower rate of return included sports injuries and a referral to a specialized clinic.
SUPPLEMENTARY MATERIAL
The supplementary material for this article can be found at https://doi.org/10.1017/cem.2019.22
