Abstract
Concussion often results in symptoms, including difficulty concentrating, focusing, and remembering, that are typically managed with cognitive and physical rest. Often, the school environment is not conducive to cognitive rest and may lead to worsening or prolonged symptoms that can contribute to impaired academic performance. The objective of the review was to identify and summarize literature concerning the effects of concussion or mild traumatic brain injury (mTBI) on academic outcomes. MEDLINE, Embase, Scopus, and CINAHL were searched until June 1, 2016. Studies must have been primary literature examining students enrolled in primary, secondary, or post-secondary education, have received a physician diagnosis of concussion or mTBI, and have post-injury academic outcomes assessed in numeric or alphabet grade/grade point average (GPA), school attendance records, or national examination scores. Data were extracted and checked by a second reviewer for accuracy and completeness. Nine studies were included. Among four studies that examined grades, one found a significant difference in pre- and post-grades only in the subject Afrikaans. Three examined national test scores and no significant differences were found between cases and controls. Four examined school absenteeism and found that students who developed post-concussion syndrome missed significantly more school days and took longer to return to school than students with extremity injuries. Although mTBI or concussion is associated with missed school, the results demonstrate minimal impact on school grades and national examination scores at a group level. Further research is needed to identify risk factors for impaired school functioning following mTBI and concussion in individual patients.
Introduction
I
Studies measuring academic achievement or academic performance using objective methods of assessment (e.g., standardized testing results or cognitive testing) have generally found no adverse long-term effects of a concussion. 11,12 The majority of these studies utilize cognitive, neurobehavioral, and neuropsychological tests, including the Wechsler Intelligence Scale for Children, Trail Making Test, California Verbal Learning Test, and Wide Range Achievement Test (WRAT) to assess intelligence, adaptive problem solving, memory, and academic skills, such as reading, spelling, and arithmetic. 13
Given the growing attention on short- and long-term outcomes of pediatric concussion, there is an increasing need to address the extent to which objective academic outcomes are affected following a concussion in order for education professionals to adequately develop and implement a plan for effective reintegration of concussed youth back into the classroom. Therefore, the objective of this study was to systematically identify and summarize literature concerning the effects of a concussion or mild traumatic brain injury (mTBI) on selected academic outcomes among students.
Methods
Sources and selection
Search terms were used to systematically search four electronic databases for eligible studies (Table 1). MEDLINE, Embase, Scopus, and CINAHL were searched using earliest date of coverage through June 1, 2016, using key terms brain concussion, concussion, achievement, educational measurement, educational status, craniocerebral trauma, head injury, brain injury, academic performance, academic outcomes, education performance, and education outcomes. Titles and abstracts were then independently reviewed by two reviewers and identified as potentially relevant or excluded. Next, the full text of all potentially relevant studies was obtained and the two reviewers independently applied a priori inclusion and exclusion criteria. Differences were assessed and resolved through discussion between the two reviewers.
Inclusion criteria
Studies were included in the systematic review providing the article met all four inclusion criteria: 1) primary research published in English, including randomized control trials, cohort, case-control, cross-sectional, longitudinal, or case-series; 2) students enrolled in post-secondary (including university, college, and trade/apprenticeship programs), secondary, or primary education (all participants had to be enrolled in school or the study had to include a subgroup analysis of only students); 3) students received a physician diagnosis of a concussion or mTBI; and 4) post-injury academic outcomes were assessed in the form of numeric or alphabet grade/grade point average (GPA), school attendance records, or national examination scores. Case reports were excluded. The aim was to examine the “real world” experience of returning to the classroom, and thus studies assessing academic achievement or academic performance only using cognitive, neurobehavioral, neuropsychological, or standardized intelligent tests (various IQ tests) tests or self-report questionnaires and telephone surveys also were excluded.
Assessment of methodological quality
The methodological quality of the included studies was evaluated using the Risk of Bias Tool from the Cochrane Handbook. 14 The tool was applied independently by two reviewers. Indicators of quality included selection bias (non-random and consecutive recruitment), detection bias (data collection methods), performance bias (blinding), attrition bias (withdrawals and dropouts), reporting bias (outcomes measured, but not reported), and other sources of bias (possible confounders). 14 Scores of high, medium, or low risk were given for each indicator and an overall risk of high, medium, or low was assigned. Discrepancies were resolved through consensus.
Data extraction and analysis
Data extracted included study design, method of recruitment, study population (age of participants, sample size, sex), setting (inpatient, outpatient), diagnosis and diagnostic criteria of concussion or mTBI, cause of injury, inclusion and exclusion criteria, duration of study, academic outcome measures, and numeric results with confidence intervals (CIs) or p values, and statistical methods used for each reported outcome. Data extraction was performed by one reviewer and checked for completeness and accuracy by the second. Due to substantial heterogeneity in study design, definition of concussion or mTBI, age of participants, and academic outcomes measured, a meta-analysis was not performed. A summary of the study design, mTBI definition, sample size, age range, academic outcomes measured, and results of all included studies are presented in Table 1.
Results
Literature search
Search results are outlined in Figure 1. A total of nine studies met the inclusion criteria for this review. Six full text articles were included from the initial literature search (one was from when the search was updated). Two additional studies were identified from a systematic review by Lloyd and colleagues 12 and the final study was found through a literature search for another systematic review in the SPORTDiscus database.
Preferred Reporting Items for Systematic Reviews and Meta-Analyses
Study characteristics
The nine studies were published from 1998 to 2016: five in the U.S., one in Canada, one in Wales, one in New Zealand, and one in South Africa. The number of students ranged from 30 to 8240 youth, age 0 to 18 years old at time of mTBI or concussion, and age 5–18 years old at time of academic assessment. Two studies only included children younger than 10 years old at time of injury. 17,18 The duration of study periods varied from end of concussed school year up to 12 years. Two studies were prospective cohorts, four were retrospective cohorts, and three were case-series. The characteristics and results from the included studies are summarized in Table 2. The risk of bias in the included studies ranged from low to medium. Six studies were scored as medium risk. 15,19 –23 Three studies were scored as low risk. 16 –18
mTBI, mild traumatic brain injury; ED, emergency department; GCS, Glasgow Coma Scale; SD, standard deviation; vmTBI, very mild traumatic brain injury; LOC, loss of consciousness; ICD-10, International Statistical Classification of Diseases and Related Health Problems, Tenth Revision; KS1, Key Stage 1; ICD-9, International Statistical Classification of Diseases and Related Health Problems, Ninth Revision; OR, odds ratio; CI, confidence interval; GPA, grade point average; DSM-IV, Diagnostic and Statistical Manual of Mental Disorders, 4th Edition; PCS, post-concussion syndrome; RR, relative risk; PPCS, persistent post-concussion syndrome; IQR, interquartile range.
Grade point average
Three cohort studies assessed academic performance based on grades. One prospective study analyzed grades in mTBI, other injury, and non-injured controls from the quarter or semester immediately before injury and again at closest to 1 year later; grades were equated to a 5-point rating scale and averaged over all academic classes. 15 There were no significant main effects for school grades and the magnitude of differences were small (range of effect sizes: 0.13 to 0.38). 15 There were no significant differences in the mean pre-injury and post-injury school grades for head injury, other injury, and non-injury group. 15
One retrospective study collected grades over 2 years from students with very mild TBI (vmTBI), mTBI, and non-injured controls in the subjects Afrikaans, English, math, and science. There were no statistically significant differences in pre- and post-test results for English, math and science. 16 There was a statistically significant decrease in academic performance in Afrikaans in both the vmTBI (61.5% to 58.6%; p = 0.017) and mTBI (65.4% to 59.2%; p = 0.016) students and no significant decrease among controls (69.5% to 67.2%; p = 0.567). 16 A retrospective population-based cohort study included high school students with and without a concussion and found no change in overall grade-point average in the year prior to and year of concussion (concussed students: −0.90%; 95% CI: −1.88, 0.08; non-concussed students: −0.57%; 95% CI: −1.32, 0.19). 17
National examination scores
Three studies measured academic outcomes defined as national examination scores. The first study acquired achievement test scores from standardized test records, which were averaged over national percentile scores for all academic subtests available, providing a mean achievement test score (in percentile) for concussed students, other injury, and non-injured controls at pre-injury and 1 year post-injury. 15 The head- and other-injury group were not significantly different, nor were the non-injury and head-injury group. The authors concluded that mild head injuries do not affect academic outcomes when measured using national examination scores. 15 The second study used the Wales Electronic Cohort for Children (WECC), a population-anonymized but linkable e-cohort study of all children born between 1990 and 2008. Academic performance on the Key Stage 1 (KS1) National Curriculum assessment in children 5–7 years old from 2003 to 2008 was used to assess changes in academic performance. 18 Students were classified as either level achieved or level not achieved. KS1 results were available for 90,661 students and 290 (0.3%) of these children sustained a head injury: skull fracture (n = 153), intracranial injury (n = 107), or concussion (n = 30). There was no statistically significant association between head injury and satisfactory performance on KS1 assessment (adjusted odds ratio: 0.87; 95% confidence interval [CI]: 0.31, 2.49). 18 The final study used performance on national examinations to assess 15–16 year olds who had sustained a mild head injury between the ages of 0 to 10. 21 The 132 students were divided into an inpatient group who stayed overnight in hospital (n = 36) and an outpatient group (n = 96); in 20 cases, multiple head injuries were sustained and the most severe injury was used to assign group (seven assigned to inpatient, 13 to outpatient). 19 Effects of mild head injury before age 10 were examined. McKinlay and colleagues found that the inpatient group tended to score more poorly on the cognitive and academic measures, but no significant group or pair wise effects were identified. 19
Attendance
Three case-series and one prospective cohort study inquired about the number of school day absences through follow-ups via telephone interviews ranging from 3 days to 3 months post-injury. Two studies measured attendance by comparing students who did and did not develop PCS. 22,23 Babcock and colleagues found a mean of 3.8 missed school days but the 29.3% of students who developed PCS (based on Diagnostic and Statistical Manual of Mental Disorders, 4th Edition criteria) missed a mean of 7.4 days, compared with 2.2 days for those who did not develop PCS (statistical significance not reported). 20 At the 3-month follow-up, 44.3% with PCS missed >2 days of school but only 22.5% without PCS missed >2 days of school (statistical significance not reported). Further, students who missed >2 days of school were at an increased risk of developing PCS, compared with those who missed fewer days (relative risk adjusted: 1.89; 95% CI: 1.39, 2.58). 20 Grubenhoff and colleagues found that the median number of school days missed was 2 (interquartile range [IQR] 1–3), 21% of students developed PCS (based on the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision criteria) and missed a median of 3 days (IQR 2–5) compared with those who did not develop PCS and missed a median of 1.5 days (IQR 1–2, p < 0.001). 21 At the 30 day follow-up, 69% of students without PCS missed at least 1 day of school and 82% with PCS missed at least 1 day of school (p = 0.27). 21 In a case-series, Thomas and colleagues assessed post-concussive symptoms, outcomes, and complications at 3 days, 2 weeks, 6 weeks, and 3 months follow up if symptoms persisted. The median number of missed school days was 2 (IQR 1–4). 22 Wasserman and colleagues compared students 1 week and 1 month post injury who sustained a concussion or musculoskeletal extremity injury. Concussed students took an average of 5.4 days to return to school compared to 2.8 days for a musculoskeletal extremity injury (statistical significance was not reported). 22
Discussion
Among the studies identified in this systematic review, only three measured grades and the authors did not find statistically or clinically significant group differences in grades before and after sustaining a concussion or mTBI when compared with controls. Overall, these studies concluded that any effect of mTBI or concussion on academic performance was negligible. Laubscher and colleagues only found a statistically significant decrease in the subject Afrikaans, but not in English, math, or science. 16 Russell and colleagues found a significant decrease in social studies grades and a significant increase in foreign languages and no changes in math, English, or science. 17 Three studies measured academic outcomes based on results of national examination scores. 15,17 –19 No study measuring national examination scores found significant group difference in scores before and after sustaining an mTBI or mild head injury when compared with controls.
Four studies that measured attendance found that students with a concussion or mTBI did miss school after their injury. Two studies compared students who did and did not developed PCS and found that students with PCS missed significantly more school days. 20,21 Babcock and colleagues 20 reported higher rates of PCS and a greater amount of absenteeism in both students with and without PCS compared with Grubenhoff and colleagues, 21 but this could be attributed to a longer follow-up period (3 months vs. 30 days) and characteristics of the data (reported mean vs. median). Wasserman and colleagues found that students with a concussion took almost twice as long as students with an extremity injury to return to school and 24% of concussed students had not returned to school after 1 week, but there was no mention of the severity of concussion symptoms, recommended cognitive rest, or a combination of both. 23 A study that did not meet inclusion criteria but is worth mentioning determined the time until a concussed patient returned to school full-time without academic accommodations (i.e., no homebound education, half-days, full days with breaks, elimination of examination, examinations with extra time, and/or note cards). They found that patients with vestibular deficits, defined as those with symptomatic vestibulo-ocular reflex testing and impaired tandem gait at initial examination, took a significantly longer time to return to school (median 59 days vs. 6 days; p = 0.001). 24
Many studies that examined mTBI or concussion on academic performance or achievement opted for student and parental assessment of academic difficulties, or cognitive, neurobehavioral, and neuropsychological measures associated with successful learning instead of directly looking at GPA or national examination scores. 12,25 These tests evaluate attributes, including intelligence, memory, language, and executive function, which play a role in academic success in the classroom. These assessments included standardized tests, such as the WRAT. The WRAT measures reading recognition, spelling, and arithmetic, which are commonly used as indicators of academic proficiency in many studies looking at educational outcomes following an mTBI or concussion. 13 Other studies used subjective, self-report surveys or questionnaires, such as the Child Behaviour Checklist, that are completed by students, parents, or teachers. 11 Although only a few studies directly measured GPA or national examination scores, the findings of these studies support the conclusion that mTBI or concussion has no significant effect on academic outcomes at a group level. This is consistent with findings in other studies using objective tests to measure academic outcomes, where students with an mTBI or concussion show minor to negligible long-term deficits with cognitive and neuropsychological testing. 11 –13,26 –29
Limitations
Among the included studies, there was notable heterogeneity with regard to different data sources, methods to compute results, and included age groups. One study categorized examination scores as level achieved or level not achieved. 18 Three studies took place outside of North America (Wales, New Zealand, South Africa). While all studies reached similar conclusions, it is challenging to compare results across a limited amount of studies that use different methods to assess academic outcomes in children at different levels of education.
There also was heterogeneity in the definition of mTBI or concussion. Several studies imposed time limits on loss of consciousness (LOC) to be classified as an mTBI (i.e., LOC less than 1 h with neurological deficits; less than 6 h without such deficits; LOC less than 20 min). 15,19 –21 Three studies classified mTBI or concussion as a closed-head injury. 15,18,19 Some studies incorporated Glasgow Coma Scale score into the diagnosis. 15,20 –22 Therefore, a concussion or mTBI in one study may not be included in another study or may be categorized as a more severe TBI. Differing definitions between studies can result in an overall mTBI or concussion group with a wide range of severities, which may differ in post-injury academic outcomes.
Our goal was to describe the in-class experience of returning to school after sustaining a concussion or mTBI. Although beyond the scope of this systematic review, other outcomes may be important, such as parents and teachers perception of academic performance, receiving special accommodations or special education, or IQ testing. Finally, one author noted difficulties in obtaining complete data. Light and colleagues described logistical difficulties and problems with compliance from some schools resulting in missing or unusable school data and obtaining only approximately half of school grades and test scores. 15
Future directions
Although studies show that mTBI or concussion have minimal effects on academic performance at a group level, some students returning to school or increasing mental activity may experience exacerbation of their concussion symptoms that may have a transient negative impact on school functioning and academic outcomes. 30,31 Customized Return-to-Learn programs have emerged as a useful tool to help monitor symptoms and work with students, teachers, and parents to implement guidelines and individualized plans to optimize recovery and minimize absenteeism while minimizing symptom exacerbation. Some jurisdictions in the United States have enacted legislation that address Return-to-Learn programs; however, it is unclear what impact these laws have had on academic outcomes or whether such steps are indeed necessary. 32 In addition, the optimal timing of returning patients to the classroom after concussion warrants further study but is likely to remain highly individualized.
Future studies should determine if there are risk factors for short-term poor academic performance and if students are receiving appropriate school-related accommodations, such as excused absences, reduced work load, delayed tests, or extensions on assignments. A previous study found that only 46% of school districts and 53% of schools had policies, guidelines, or protocols in place to assist students returning to school following a concussion. 33 Additionally, future studies also should examine the differences in academic performance among those who develop PCS and those who do not.
Footnotes
Acknowledgments
We would like to thank Janet Rothney, MLIS, Dentistry Librarian, and Tania Gottschalk, MLIS, Medicine Librarian, Neil John Maclean Health Sciences Library, University of Manitoba, for conducting the literature searches.
Author Disclosure Statement
No competing financial interests exist.