[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
Purchase Options:
[Skip to Content Landing]
Figure 1.
Flow Diagram of Participants With and Without Early Physical Activity Following Acute Concussion
Flow Diagram of Participants With and Without Early Physical Activity Following Acute Concussion

ED indicates emergency department; RA, research assistant.

aIncludes those for whom reason was not specified or was missing data due to the fact that 1 of the 9 research ethics boards did not permit the collection of reasons for meeting exclusion criteria due to provincial regulations.

Figure 2.
Distribution of Propensity Scores in the Physical Activity Group and the Rest Group
Distribution of Propensity Scores in the Physical Activity Group and the Rest Group

Inverse probability of treatment-weighting analysis includes only those patients with complete data on physical activity and all 43 covariates included in the propensity analysis. For intervals along the x-axis, the area under the probability density curve represents the probability of those propensity scores. Smoothing was via the kernel density estimate.28

Table 1.  
Baseline Characteristics Total Sample, Unweighted Sample, Propensity Score-Matched Sample, and Inverse Probability of Treatment-Weighted Sample
Baseline Characteristics Total Sample, Unweighted Sample, Propensity Score-Matched Sample, and Inverse Probability of Treatment-Weighted Sample
Table 2.  
Summary of Results of the Primary Analysis
Summary of Results of the Primary Analysis
Table 3.  
Summary of Sensitivity Analysis 1 and 2
Summary of Sensitivity Analysis 1 and 2
1.
McCrory  P, Meeuwisse  WH, Aubry  M,  et al.  Consensus statement on concussion in sport.  Br J Sports Med. 2013;47(5):250-258.PubMedGoogle ScholarCrossref
2.
Zemek  R, Duval  S, Dematteo  C.  Guidelines for Diagnosing and Managing Pediatric Concussion. Toronto, Canada: Ontario Neurotrauma Foundation; 2014.
3.
DiFazio  M, Silverberg  ND, Kirkwood  MW, Bernier  R, Iverson  GL.  Prolonged activity restriction after concussion.  Clin Pediatr (Phila). 2016;55(5):443-451.PubMedGoogle ScholarCrossref
4.
Cancelliere  C, Hincapié  CA, Keightley  M,  et al.  Systematic review of prognosis and return to play after sport concussion.  Arch Phys Med Rehabil. 2014;95(3)(suppl):S210-S229.PubMedGoogle ScholarCrossref
5.
Schneider  KJ, Iverson  GL, Emery  CA, McCrory  P, Herring  SA, Meeuwisse  WH.  The effects of rest and treatment following sport-related concussion.  Br J Sports Med. 2013;47(5):304-307.PubMedGoogle ScholarCrossref
6.
Majerske  CW, Mihalik  JP, Ren  D,  et al.  Concussion in sports.  J Athl Train. 2008;43(3):265-274.PubMedGoogle ScholarCrossref
7.
Thomas  DG, Apps  JN, Hoffmann  RG, McCrea  M, Hammeke  T.  Benefits of strict rest after acute concussion.  Pediatrics. 2015;135(2):213-223.PubMedGoogle ScholarCrossref
8.
Willer  B, Leddy  JJ.  Management of concussion and post-concussion syndrome.  Curr Treat Options Neurol. 2006;8(5):415-426.PubMedGoogle ScholarCrossref
9.
Gagnon  I, Grilli  L, Friedman  D, Iverson  GL.  A pilot study of active rehabilitation for adolescents who are slow to recover from sport-related concussion.  Scand J Med Sci Sports. 2016;26(3):299-306.PubMedGoogle ScholarCrossref
10.
Leddy  JJ, Kozlowski  K, Donnelly  JP, Pendergast  DR, Epstein  LH, Willer  B.  A preliminary study of subsymptom threshold exercise training for refractory post-concussion syndrome.  Clin J Sport Med. 2010;20(1):21-27.PubMedGoogle ScholarCrossref
11.
Silverberg  ND, Iverson  GL.  Is rest after concussion “the best medicine?”.  J Head Trauma Rehabil. 2013;28(4):250-259.PubMedGoogle ScholarCrossref
12.
Zemek  R, Osmond  MH, Barrowman  N,  et al.  Predicting and preventing postconcussive problems in paediatrics (5P) study.  BMJ Open. 2013;3(8):e003550.PubMedGoogle ScholarCrossref
13.
Zemek  R, Barrowman  N, Freedman  SB,  et al; Pediatric Emergency Research Canada (PERC) Concussion Team.  Clinical risk score for persistent postconcussion symptoms among children with acute concussion in the ED.  JAMA. 2016;315(10):1014-1025.PubMedGoogle ScholarCrossref
14.
Harris  PA, Taylor  R, Thielke  R, Payne  J, Gonzalez  N, Conde  JG.  Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support.  J Biomed Inform. 2009;42(2):377-381.PubMedGoogle ScholarCrossref
15.
Gioia  GA, Collins  M, Isquith  PK.  Improving identification and diagnosis of mild traumatic brain injury with evidence.  J Head Trauma Rehabil. 2008;23(4):230-242.PubMedGoogle ScholarCrossref
16.
Sady  MD, Vaughan  CG, Gioia  GA.  Psychometric characteristics of the postconcussion symptom inventory in children and adolescents.  Arch Clin Neuropsychol. 2014;29(4):348-363.PubMedGoogle ScholarCrossref
17.
McCrory  P, Meeuwisse  WH, Aubry  M,  et al.  Child SCAT3.  Br J Sports Med. 2013;47(5):263-266.PubMedGoogle Scholar
18.
Steindel  SJ.  International classification of diseases, 10th edition, clinical modification and procedure coding system: descriptive overview of the next generation HIPAA code sets.  J Am Med Inform Assoc. 2010;17(3):274-282.PubMedGoogle Scholar
19.
Newcombe  RG.  Two-sided confidence intervals for the single proportion: comparison of seven methods.  Stat Med. 1998;17(8):857-872.PubMedGoogle ScholarCrossref
20.
Haukoos  JS, Lewis  RJ.  The propensity score.  JAMA. 2015;314(15):1637-1638.PubMedGoogle ScholarCrossref
21.
Austin  PC.  An introduction to propensity score methods for reducing the effects of confounding in observational studies.  Multivariate Behav Res. 2011;46(3):399-424.PubMedGoogle ScholarCrossref
22.
Kurth  T, Walker  AM, Glynn  RJ,  et al.  Results of multivariable logistic regression, propensity matching, propensity adjustment, and propensity-based weighting under conditions of nonuniform effect.  Am J Epidemiol. 2006;163(3):262-270.PubMedGoogle ScholarCrossref
23.
Brookhart  MA, Schneeweiss  S, Rothman  KJ, Glynn  RJ, Avorn  J, Stürmer  T.  Variable selection for propensity score models.  Am J Epidemiol. 2006;163(12):1149-1156.PubMedGoogle ScholarCrossref
24.
Rosenbaum  P, Rubin  D.  The central role of the propensity score in observational studies for causal effects.  Biometrika. 1983;70(1):41-55. doi:10.1093/biomet/70.1.41Google ScholarCrossref
25.
Williamson  EJ, Forbes  A.  Introduction to propensity scores.  Respirology. 2014;19(5):625-635.PubMedGoogle ScholarCrossref
26.
Youden  WJ.  Index for rating diagnostic tests.  Cancer. 1950;3(1):32-35.PubMedGoogle ScholarCrossref
27.
Rosenbaum  P, Rubin  D.  Constructing a control group using multivariate matched sampling methods that incorporate the propensity score.  Am Stat. 1985;39:33-38. doi:10.1017/CBO9780511810725.019Google Scholar
28.
Scott  D.  Multivariate Density Estimation: Theory, Practice and Visualization. Wiley Series in Probability and Mathematical Statistics. New York, NY: John Wiley & Sons; 1992.
29.
Khan  NA, Hillman  CH.  The relation of childhood physical activity and aerobic fitness to brain function and cognition.  Pediatr Exerc Sci. 2014;26(2):138-146.PubMedGoogle ScholarCrossref
30.
Hillman  CH, Erickson  KI, Kramer  AF.  Be smart, exercise your heart.  Nat Rev Neurosci. 2008;9(1):58-65.PubMedGoogle ScholarCrossref
31.
Veerbeek  JM, van Wegen  E, van Peppen  R,  et al.  What is the evidence for physical therapy poststroke?  PLoS One. 2014;9(2):e87987.PubMedGoogle ScholarCrossref
32.
Longmuir  PE, Colley  RC, Wherley  VA, Tremblay  MS.  Canadian Society for Exercise Physiology position stand: benefit and risk for promoting childhood physical activity.  Appl Physiol Nutr Metab. 2014;39(11):1271-1279.PubMedGoogle ScholarCrossref
33.
Rush  E, Simmons  D.  Physical activity in children.  Med Sport Sci. 2014;60:113-121.PubMedGoogle Scholar
34.
McKay  HA, Petit  MA, Schutz  RW, Prior  JC, Barr  SI, Khan  KM.  Augmented trochanteric bone mineral density after modified physical education classes.  J Pediatr. 2000;136(2):156-162.PubMedGoogle ScholarCrossref
35.
Annesi  JJ.  Correlations of depression and total mood disturbance with physical activity and self-concept in preadolescents enrolled in an after-school exercise program.  Psychol Rep. 2005;96(3 Pt 2):891-898.PubMedGoogle ScholarCrossref
36.
Howie  EK, Pate  RR.  Physical activity and academic achievement in children.  J Sport Health Sci. 2012;1(3):160-169. doi:10.1016/j.jshs.2012.09.003Google ScholarCrossref
37.
Chaddock-Heyman  L, Hillman  CH, Cohen  NJ, Kramer  AF  III.  The importance of physical activity and aerobic fitness for cognitive control and memory in children.  Monogr Soc Res Child Dev. 2014;79(4):25-50.PubMedGoogle ScholarCrossref
38.
Moor  HM, Eisenhauer  RC, Killian  KD,  et al.  The relationship between adherence behaviors and recovery time in adolescents after a sports-related concussion.  Int J Sports Phys Ther. 2015;10(2):225-233.PubMedGoogle Scholar
39.
Perrey  S.  Promoting motor function by exercising the brain.  Brain Sci. 2013;3(1):101-122.PubMedGoogle ScholarCrossref
40.
Farquhar  WB, Greaney  JL.  Autonomic exercise physiology in health and disease.  Auton Neurosci. 2015;188:1-2.PubMedGoogle ScholarCrossref
41.
Leddy  JJ, Kozlowski  K, Fung  M, Pendergast  DR, Willer  B.  Regulatory and autoregulatory physiological dysfunction as a primary characteristic of post concussion syndrome.  NeuroRehabilitation. 2007;22(3):199-205.PubMedGoogle Scholar
42.
Adamo  KB, Prince  SA, Tricco  AC, Connor-Gorber  S, Tremblay  M.  A comparison of indirect versus direct measures for assessing physical activity in the pediatric population.  Int J Pediatr Obes. 2009;4(1):2-27.PubMedGoogle ScholarCrossref
Original Investigation
December 20, 2016

Association Between Early Participation in Physical Activity Following Acute Concussion and Persistent Postconcussive Symptoms in Children and Adolescents

Author Affiliations
  • 1Children’s Hospital of Eastern Ontario Research Institute, Ottawa, Ontario, Canada
  • 2Sports Concussion Clinic, Boston Children’s Hospital, Boston, Massachusetts
  • 3Department of Pediatrics, Alberta Children’s Hospital, Alberta Children’s Hospital Research Institute, University of Calgary, Alberta, Canada
  • 4Department of Psychology, Alberta Children’s Hospital Research Institute, and Hotchkiss Brain Institute, University of Calgary, Alberta, Canada
  • 5Department of Pediatrics, Hospital Ste Justine, University of Montreal, Montreal, Quebec, Canada
  • 6Department of Pediatrics, Montreal Children’s Hospital, McGill University Health Center, Montreal, Quebec, Canada
  • 7Department of Pediatrics, Hospital for Sick Children, Toronto, Ontario, Canada
  • 8Sport Injury Prevention Research Centre, Faculty of Kinesiology, University of Calgary, Calgary, Alberta, Canada
  • 9Department of Pediatrics, Children’s Hospital of Eastern Ontario, University of Ottawa, Ottawa, Ontario, Canada
JAMA. 2016;316(23):2504-2514. doi:10.1001/jama.2016.17396
Key Points

Question  Is participation in physical activity within 7 days following acute concussion associated with lower rates of persistent postconcussive symptoms in children and adolescents compared with conservative rest?

Findings  In this prospective, multicenter cohort study of 3063 children and adolescents aged 5.00 to 17.99 years after propensity matching, the proportion with postconcussive symptoms at 28 days was 28.7% with participation in early physical activity vs 40.1% with conservative rest, a significant difference.

Meaning  Participation in physical activity within 1 week after injury may benefit symptom recovery following acute concussion in children and adolescents.

Abstract

Importance  Although concussion treatment guidelines advocate rest in the immediate postinjury period until symptoms resolve, no clear evidence has determined that avoiding physical activity expedites recovery.

Objective  To investigate the association between participation in physical activity within 7 days postinjury and incidence of persistent postconcussive symptoms (PPCS).

Design, Setting, and Participants  Prospective, multicenter cohort study (August 2013-June 2015) of 3063 children and adolescents aged 5.00-17.99 years with acute concussion from 9 Pediatric Emergency Research Canada network emergency departments (EDs).

Exposures  Early physical activity participation within 7 days postinjury.

Main Outcomes and Measures  Physical activity participation and postconcussive symptom severity were rated using standardized questionnaires in the ED and at days 7 and 28 postinjury. PPCS (≥3 new or worsening symptoms on the Post-Concussion Symptom Inventory) was assessed at 28 days postenrollment. Early physical activity and PPCS relationships were examined by unadjusted analysis, 1:1 propensity score matching, and inverse probability of treatment weighting (IPTW). Sensitivity analyses examined patients (≥3 symptoms) at day 7.

Results  Among 2413 participants who completed the primary outcome and exposure, (mean [SD] age, 11.77 [3.35] years; 1205 [39.3%] females), PPCS at 28 days occurred in 733 (30.4%); 1677 (69.5%) participated in early physical activity including light aerobic exercise (n = 795 [32.9%]), sport-specific exercise (n = 214 [8.9%]), noncontact drills (n = 143 [5.9%]), full-contact practice (n = 106 [4.4%]), or full competition (n = 419 [17.4%]), whereas 736 (30.5%) had no physical activity. On unadjusted analysis, early physical activity participants had lower risk of PPCS than those with no physical activity (24.6% vs 43.5%; Absolute risk difference [ARD], 18.9% [95% CI,14.7%-23.0%]). Early physical activity was associated with lower PPCS risk on propensity score matching (n = 1108 [28.7% for early physical activity vs 40.1% for no physical activity]; ARD, 11.4% [95% CI, 5.8%-16.9%]) and on inverse probability of treatment weighting analysis (n = 2099; relative risk [RR], 0.74 [95% CI, 0.65-0.84]; ARD, 9.7% [95% CI, 5.7%-13.7%]). Among only patients symptomatic at day 7 (n = 803) compared with those who reported no physical activity (n = 584; PPCS, 52.9%), PPCS rates were lower for participants of light aerobic activity (n = 494 [46.4%]; ARD, 6.5% [95% CI, 5.7%-12.5%]), moderate activity (n = 176 [38.6%]; ARD, 14.3% [95% CI, 5.9%-22.2%]), and full-contact activity (n = 133 [36.1%]; ARD, 16.8% [95% CI, 7.5%-25.5%]). No significant group difference was observed on propensity-matched analysis of this subgroup (n = 776 [47.2% vs 51.5%]; ARD, 4.4% [95% CI, −2.6% to 11.3%]).

Conclusions and Relevance  Among participants aged 5 to 18 years with acute concussion, physical activity within 7 days of acute injury compared with no physical activity was associated with reduced risk of PPCS at 28 days. A well-designed randomized clinical trial is needed to determine the benefits of early physical activity following concussion.

Introduction

Quiz Ref IDRest has long been considered the cornerstone of concussion management,1 and pediatric guidelines universally recommend an initial period of cognitive and physical rest following a concussion.1,2 Cognitive rest recommendations include modification of school attendance and mental activities.1 Physical rest recommendations advocate avoidance of physical activity until postconcussive symptoms have resolved, endorsing gradual resumption of activities only if symptoms are not exacerbated.1,2

Due to limited high-quality evidence, existing physical rest guidelines are based on consensus and precautionary principles.2,3 There is limited evidence that following these guidelines results in a positive effect on prognosis.4,5 Although strenuous exercise in patients recovering from concussion may be deleterious and increase re-injury risk,6 recent literature suggests that protracted rest may hamper concussion recovery,7 leading to secondary symptoms of fatigue, depression, anxiety, and physiological deconditioning.3,8 Increasing evidence suggests the introduction of controlled, light aerobic physical activity following pediatric concussion may be safe9 while promoting recovery9 by enhancing physical, psychological, and academic outcomes.6,9,10 These preliminary findings indicate that gradual resumption of pre-injury activities could begin as soon as tolerated provided there is no increased risk of re-injury.11

The objective of this study was to examine the association between participation in physical activity within 7 days postinjury and the occurrence of persistent postconcussive symptoms (PPCS) following concussion in children and adolescents. It was hypothesized that early participation in physical activity would be associated with lower PPCS rates compared with no physical activity.

Methods

This research comprises a planned secondary analysis of the Predicting Persistent Postconcussive Problems in Pediatrics (5P) study,12,13 a prospective, multicenter cohort study that recruited participants from August 2013 until June 2015 at 9 Pediatric Emergency Research Canada (PERC) network tertiary pediatric emergency departments (EDs).

Participants

This study enrolled 3063 participants aged 5.00 to 17.99 years with ED presentation for acute head injury12,13 occurring within the preceding 48 hours, who met concussion diagnosis criteria according to the 2012 Zurich consensus statement.1 Exclusion criteria were a Glasgow Coma Scale score of 13 or less; any abnormality on brain computed tomography or magnetic resonance imaging; neurosurgical intervention, intubation, or intensive care unit admission; multisystem injury requiring hospitalization; severe preexisting neurological developmental delay resulting in communication difficulties; intoxication; absence of trauma as the primary event; previously enrolled in this same study; insurmountable language barrier; or inability to follow-up by phone or electronic-mail. The 5P study was approved by ethics committees of each participating institution, and a written informed consent and assent was obtained from all participants or parents as appropriate.

Study Protocol

Trained research assistants completed standardized assessments of all patients in the ED.12 Data were collected and managed using Research Electronic Data Capture (REDCap) tools hosted at the Children’s Hospital of Eastern Ontario Research Institute.14 Patients and parents provided information on demographics and past history (ie, prior concussion, headache, developmental or psychiatric conditions), as well as injury characteristics using the Acute Concussion Evaluation inventory,15 a validated scale used to identify concussion in children and adolescents aged 3 to 18 years. Patients and parents quantified pre-injury and current symptoms (ie, physical, emotional, cognitive, and sleep) using the Post-Concussion Symptom Inventory (PCSI).16 Cognitive status, physical examination, and balance assessments were completed using the Child-Sport Concussion Assessment Tool-3rd Edition (Child-SCAT3) evaluation.1,17

Follow-up Procedures

Enrolled patients were offered web-based survey or telephone follow-up at 7 and 28 days postenrollment.12 Patients received email reminders 24 hours after each survey deadline; research assistants telephoned nonresponders as many as 5 times and offered verbal interviews. Surveys were parent reported for children aged 5.00 to 7.99 years and patient reported for all other participants. Current level of physical activity was self-categorized as no activity (eg, physical rest), light aerobic exercise (eg, walking, swimming, or stationary cycling), sport-specific exercise (eg, running drills in soccer or skating drills in ice hockey), noncontact training drills (eg, complex passing drills), full-contact practice (eg, normal training activities), and return to competition (eg, normal game play).12 Early physical activity participation was defined as any level of physical activity other than no activity at 7 days postenrollment. Early physical activity subcategories were defined as no activity, light aerobic exercise, moderate exercise (sport-specific exercise or noncontact training drills), or full exercise (full-contact practice or return to competition). Questions regarding physical activity were based on Zurich Consensus Statement on Concussion in Sport return-to-play1 steps; these questions have not been validated.

Primary Outcome Measure

Primary outcome was the presence of PPCS, defined as at least 3 new or worsening individual symptoms compared with the preconcussion status measured at day 28 according to the validated PCSI.12,13,18 An individual symptom was defined as a positive difference between the current minus the perceived pre-injury symptom rating as completed 28 days postenrollment.12,13

Statistical Analysis

Frequencies and descriptive statistics were used to summarize patient baseline characteristics for the overall sample and by early physical activity. Missing data were managed via list-wise deletion (ie, participants were excluded from the analysis if any single values were missing).

The proportion of PPCS in each group was computed, along with a Wilson score 95% CI, a method for obtaining a CI for a proportion.19 The unadjusted association between early physical activity and PPCS was estimated using the sample relative risk (RR) and the sample Absolute risk difference (ARD).

Propensity scores20 were developed to account for potential confounding by observed baseline characteristics.21,22 A propensity score was derived to reflect the probability of a participant having engaged in early physical activity given an observed set of baseline characteristics. Propensity score methods replace an entire set of baseline characteristics with a single composite score, and this can be accomplished with numbers of potential confounders in excess of what is possible with conventional regression methods.2325 Clinically relevant variables (defined a priori) and those that may be associated with early physical activity were included in the models. Continuous variables were categorized based on the Youden index26 or through visualization using locally weighted polynomial regression (LOESS) curves. The following variables were included as predictors of early activity using multivariate logistic regression to calculate the propensity score: age group, sex, duration of prior concussion (no prior concussion or concussion with symptoms lasting <1 week vs prior concussion with symptoms lasting ≥1 week), personal history of migraines, family history of migraines, learning disability, attention-deficit/hyperactivity disorder, developmental disorder, anxiety, depression, sleep disorder, other psychiatric disorder, loss of consciousness duration (did not lose consciousness or loss of consciousness <3 minutes vs loss of consciousness ≥3 minutes), time between head injury and triage, seizure, early symptoms on the Acute Concussion Evaluation (appears dazed and confused, confused about events, answers questions slowly, repeats questions, forgetful), balance tandem stance (0-3 errors vs ≥4 errors or physically unable), sports injury, all 20 parent reported indicators of the Postconcussion Symptom Inventory,16 and site.

To examine the outcome associated with early activity,21 participants who did and did not engage in early physical activity were matched 1:1 in random order on the logit of the propensity scores using a greedy algorithm and nearest-neighbor approach (maximum caliper distance, 0.1) using the MatchIt package in R (R Project for Statistical Computing).27 Equivalence between matched participants (activity vs nonactivity groups) was assessed by testing for differences in covariates using χ2 analyses and Mann-Whitney U tests where appropriate. Standardized mean differences were calculated using the R package Tableone. After obtaining a matched data set, the association between early participation in physical activity and PPCS was estimated using the sample RR and the sample ARD.

Inverse probability of treatment weighting (IPTW) was used to investigate the association of early participation in physical activity among the entire population of youth recovering from acute concussion when this population is hypothetically moved from no early activity to participation in early activity. Participants were weighted by the inverse of the probability of engaging in physical activity at day 7. The association between early participation in physical activity and PPCS was estimated using the RR obtained from log-binomial regression and the ARD obtained from identity link binomial regression. In both cases, the IPTW weights were used with a quasibinomial model to obtain robust variance estimates. To avoid convergence issues, the R package glm2 was used. Group differences were assessed by calculating IPTW proportions, weighted medians, and standardized mean differences.

Because the self-report questionnaire at day 7 does not differentiate between the timing of activity and symptoms within the first week postinjury, 2 sensitivity analyses were performed. First, the original analyses were repeated by replacing the total ED symptom load with the total score at day 7. Second, a subanalysis of only patients remaining symptomatic with at least 3 symptoms at day 7 was performed, thus excluding recovered patients and those with minimal symptomatology. A sensitivity analysis was also conducted to investigate a possible interaction between age and physical activity in the model for PPCS. A quasibinomial model with a log link included an effect for early exercise, age (as a continuous variable), and the product of these 2 variables. Two sided P values of less than 0.05 were considered statistically significant. All analyses were performed using IBM SPSS Statistics version 23 (IBM Corp) and R version 3.0.2.

Results

In total, 2584 of 3063 (84.4%) patients completed the primary outcome assessment (Figure 1). Of these, 171 were excluded because of missing data on participation in physical activity at day 7, resulting in a cohort of 2413 patients. Baseline characteristics for the total cohort, and for participant groups with and without early physical activity are summarized in Table 1. Overall, 733/2413 (30.4%) patients met criteria for PPCS.

Early Participation in Physical Activity

At 7 days postenrollment, 1677 (69.5%) patients reported participating in physical activity including light aerobic exercise (795 [32.9%]), sport-specific exercise (214 [8.9%]), noncontact training drills (143 [5.9%]), full contact practice (106 [4.4%]), or return to competition (419 [17.4%]), and 736 patients (30.5%) reported no physical activity. Of the 1677 patients who engaged in early physical activity, 523 (31.3%) were symptom free and 803 (48.0%) had at least 3 persistent or worsening postconcussive symptoms at day 7. Of those reporting engaging in no physical activity at day 7, 584 (79.5%) had at least 3 persistent or worsening postconcussive symptoms at day 7.

Bivariable Analysis

In bivariable analysis (unweighted sample), early participation in any type of physical activity compared with no physical activity was associated with lower risk of PPCS (413 [24.6%] patients vs 320 [43.5%] patients; RR, 0.75 [95% CI, 0.70-0.80]; ARD, 18.9% [95% CI, 14.7%-23.0%]). When early physical activity subcategories were distinguished, participation in light aerobic exercise (250 patients [31.4%]; RR, 0.82 [95% CI, 0.76-0.89]; ARD, 12.0% [95% CI, 7.2%-16.8%]), moderate exercise (87 patients [24.4%]; RR, 0.75 [95% CI, 0.69-0.81]; ARD, 19.1% [95% CI, 13.2%-24.6%]), and full exercise (76 patients [14.5%]; RR, 0.66 [95% CI, 0.61-0.71]; ARD, 29.0% [95% CI, 24.2%-33.5%]) were all associated with significantly lower risk of PPCS as compared with the no activity group (320 patients [43.5%]; Table 2).

Propensity Score-Matched Analysis

Prior to matching, median propensity to engage in physical activity in the activity group was 0.74 (interquartile range [IQR], 0.65-0.80) vs 0.66 (IQR, 0.55-0.75) in the nonactivity group. Matching resulted in 554 children and adolescents participating in early physical activity matched to 554 children not participating in activity. Because more participants reported engaging in physical activity than not, 900 who participanted in physical activity were unmatched in contrast to 91 nonactivity participants. The distribution of propensity scores in the early activity and nonactivity groups are shown in before matching (Figure 2A) and after matching (Figure 2B). Following propensity score matching, mean (SD) propensity for early physical activity was similar for those reporting activity (0.641 [0.176]) vs no physical activity (0.627 [0.171]) and also resulted in between-group balance on baseline characteristics (Table 1). In propensity score–matched bivariable analysis (n = 1108), early participation in physical activity remained significantly associated with lower PPCS risk (n = 159 [28.7%]) vs n = 222 [40.1%]; RR, 0.84 [95% CI, 0.77-0.92]; ARD, 11.4% [95% CI, 5.8%-16.9%]).

Inverse Probability Treatment Weighting Analysis

IPTW, formed by those with complete data on exercise and all 43 covariates included in the propensity analysis, also resulted in between-group balance on baseline characteristics (Table 1; n = 2099). Figure 2C shows the weighted distribution of propensity scores in the early activity and nonactivity groups. Mean (SD) weight was 2.00 (1.52) with a range of 1.03 to 12.9. The highest values for standardized mean differences in the weighted data were 0.113 for wearing a mouth guard and 0.101 for wearing a helmet; all other baseline variables had standardized mean difference values of less than 0.1. PPCS remained significantly less likely in the early physical activity group in IPTW log-binomial regression analysis compared with the no physical activity group (RR, 0.74 [95% CI, 0.65-0.84]; ARD, 9.7% [95% CI, 5.7%-13.7%]).

Sensitivity Analyses

When the total ED symptom score was replaced with the total score at day 7, only the association in the unweighted sample remained significant with similar magnitude and directionality as in the primary analyses (bivariable analyses RR, 0.75 [95% CI, 0.70-0.80]; Table 3).

In the second sensitivity analysis, the analytical sample was limited to children and adolescents with at least 3 symptoms at day 7 (n = 1387). Despite current guidelines strongly advocating physical rest until the patient is asymptomatic, 584 of 1387 (57.9%) participants engaged in some form of physical activity (ie, were nonadherent with current recommendations). Although the directionality of the association remained similar, the propensity score–matched analysis and IPTW analysis no longer reached statistical significance (bivariable analysis RR, 0.83 [95% CI, 0.74-0.92]); propensity score–matched RR, 0.92 [95% CI, 0.80-1.05]; IPTW RR, 0.92 [95% CI, 0.82-1.04]; Table 3). When early physical activity subcategories were distinguished within this symptomatic cohort, children and adolescents who participated in physical activity had lower risk of PPCS (light aerobic exercise absolute risk, 46.4% [RR, 0.88 {95% CI, 0.78-0.99}; ARD, 6.6% {95% CI, 0.6%-12.5%}]; moderate exercise absolute risk, 38.6% [RR, 0.77 {95% CI, 0.66-0.89}; ARD, 14.3% {95% CI, 5.9%-22.2%}]; and full exercise absolute risk, 36.1% [RR, 0.74 {95% CI, 0.63-0.86}; ARD, 16.8% {95% CI, 7.5%-25.5%}]) compared with the no activity group (absolute risk, 52.9%). Finally, there was no statistically significant interaction between age and physical activity in an unadjusted model for PPCS; for each additional year of age, RR increased by a factor of 1.01 (95% CI, 0.97-1.05; P = .52).

Discussion

Quiz Ref IDIn this prospective cohort study, 69.5% of children and adolescents participated in physical activity within 7 days following an acute concussion—primarily with light aerobic exercise. The resumption of physical activity within 7 days postconcussion was associated with a lower risk of PPCS as compared with no physical activity. This finding was consistent across analytic approaches and intensity of exercise.

Quiz Ref IDEvidence about the importance of physical activity in childhood for maintaining physical and cognitive health is unequivocal.29 Physical activity is considered an effective method for improving cognitive function and brain health.30 Compared with other conditions in which the latest insights regarding the benefits of early physical rehabilitation have been adopted, including stroke,31 the field of pediatric concussion lags behind.2 Overwhelming evidence supports the overall benefits of physical activity in youth 32 including better body composition,33 skeletal health,34 and cardiorespiratory fitness, as well as improvement of depression, anxiety, self-concept,35 cognitive performance, and academic achievement.36,37 Preliminary studies in concussed adolescents found that participants engaging in moderate levels of activity reported lower symptom levels and superior neurocognitive performance compared with those with physical rest, although the optimal timing for re-introducing physical activity remains undetermined.6,11 Available evidence suggests that gradual resumption of physical activity should begin as soon as tolerated following an acute concussion,3,11 with the exception of activities likely to increase the risk of re-injury.6,9,11 Rest exceeding 3 days postinjury was similarly or less effective than treatment regimens allowing for earlier participation in physical activity following a concussion11,38; if prolonged, rest may predispose to secondary symptoms of fatigue, reactive depression, physiological deconditioning, and delayed recovery.7,8Also in symptomatic adolescents, pilot evidence suggests that gradual resumption of aerobic physical activities results in superior symptom recovery from concussion compared with complete rest.9,10

Quiz Ref IDA proposed mechanism by which exercise may improve recovery is through the promotion of neuroplasticity mechanisms and from possible effects on cardioregulatory mechanisms, possibly leading to improved cerebral blood flow.39 This is of particular importance in pediatric concussion, since autoregulatory dysfunction and abnormal cerebral blood flow regulation have been associated with PPCS in school-aged children.40,41 Controlled aerobic exercise may improve recovery by restoring normal cerebral blood flow regulation10 with the rate of symptom improvement relating directly to the exercise intensity achieved.10 Conversely, physical inactivity may predispose patients to PPCS through an activity restriction cascade model; it has been theorized that the psychological consequences of removal from life-validating activities, combined with physical deconditioning, may contribute to the development of PPCS after mild traumatic brain injury in youth.3

The results of this study should be considered in the context of study limitations. Because of the observational design, the authors cannot account for unmeasured confounding due to factors that may have been associated with physical activity shortly after concussion, nor can causation be determined. Although potential confounding by observed baseline characteristics was accounted for by conducting a propensity analysis,20 unmeasured confounders and intermediaries may have influenced the results. Because the lowest odds of PPCS were observed in children participating in full exercise at day 7, children who simply felt better may have started physical activity earlier and subsequently resumed full competition despite still having symptoms. This possibility was examined through sensitivity analyses in which 1-week symptoms replaced ED symptoms and the inclusion of only those children with 3 or more symptoms at day 7. Given the limitation of possible confounding variables, a well-designed and adequately powered randomized clinical trial is needed to confirm the benefits of early return to physical activity.

Second, physical activity was rated via self-report questionnaires. Although direct measures of physical activity have greater precision, no single criterion standard exists and self-rated measures remain the most common and feasible method of measuring physical activity in large settings due to their practicality, low cost, low participant burden, and general acceptance.42

Third, because objective data on physical activity (eg, actigraphy) was not collected, information regarding duration and frequency of physical activity is limited; uncertainty remains as to whether exercise intensity exacerbates symptoms and how total activity load may be associated with PPCS risk.

Fourth, it is possible that patients who did not resume physical activity may have participated in more cognitive activity and therefore may have been potentially more symptomatic. Measuring cognitive rest is challenging because it is poorly defined across the literature and difficult to objectively measure. Since objective cognitive activity data were not collected, conclusions regarding the benefit or detriment of cognitive rest were unattainable.

Fifth, because all participants received care as usual by treating physicians, rest and activity recommendations likely varied across sites and clinicians both in the ED and subsequent follow-up.

Sixth, the influence of interim activity (eg, participation in physical activities between days 7 and 28 postenrollment) was not considered.

Divergence from conservative rest recommendations following pediatric concussion toward early active physical rehabilitation would be a new approach in concussion management, potentially affecting the well-being of millions of children and families worldwide. Early physical activity could mitigate the undesired effects of physical and mental deconditioning associated with prolonged rest. Quiz Ref IDRegardless of potential benefit, caution in the immediate postinjury period is prudent; participation in activities that might introduce risk for collision (eg, resumption of contact sports) or falls (eg, skiing, skating, bicycling) should remain prohibited until clearance by a health professional to reduce the risk for a potentially more serious second concussion during a period of increased vulnerability. To be noted, results of this study do not infer any evidence of benefit or harm in association with return to practice or play.

Conclusions

Among children and adolescents aged 5 to 18 years with acute concussion, participation in physical activity within 7 days of acute injury compared with no physical activity was associated with lower risk of PPCS at 28 days. A well-designed randomized clinical trial is needed to determine the benefits of early physical activity following concussion.

Back to top
Article Information

Corresponding Author: Roger Zemek, MD, Children’s Hospital of Eastern Ontario Research Institute, 401 Smyth Rd, Ottawa, ON K1H 8L1, Canada (rzemek@cheo.on.ca).

Author Contributions: Dr Zemek had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Grool, Aglipay, Meehan III, Freedman, Gravel, Gagnon, Boutis, Meeuwisse, Barrowman, Osmond, Zemek.

Acquisition, analysis, or interpretation of data: Grool, Aglipay, Momoli, Meehan III, Freedman, Yeates, Gravel, Gagnon, Boutis, Barrowman, Ledoux, Osmond, Zemek.

Drafting of the manuscript: Grool, Aglipay, Barrowman, Zemek.

Critical revision of the manuscript for important intellectual content: Grool, Aglipay, Momoli, Meehan III, Freedman, Yeates, Gravel, Gagnon, Boutis, Meeuwisse, Ledoux, Osmond, Zemek.

Statistical analysis: Aglipay, Momoli, Barrowman, Zemek.

Obtained funding: Meehan, Freedman, Gravel, Gagnon, Boutis, Meeuwisse, Osmond, Zemek.

Administrative, technical, or material support: Ledoux.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Meehan reports receipt of royalties from ABC-Clio and from Wolders Kluwer for sales of authored publications; contracts with ABC-Clio and Springer International for a future book; and research funding, in part, by a grant from the National Football League Players Association and by philanthropic support from the National Hockey League Alumni Association through the Corey C. Griffin Pro-Am Tournament. Dr Freedman reports receipt of support from the Alberta Children’s Hospital Foundation Professorship in Child Health and Wellness. Dr Zemek reports receipt of support by the University of Ottawa Brain and Mind Research Institute as a clinical research chair in pediatric concussion. No other disclosures were reported.

Funding/Support: This study was supported by a Canadian Institutes of Health Research (CIHR) operating grant (MOP 126197); a CIHR–Ontario Neurotrauma Foundation Mild Traumatic Brain Injury team grant (TM1 127047); and CIHR planning grant (MRP 119829).

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Group Information: Additional PERC Predicting Persistent Post-Concussive Problems in Pediatrics (5P) Concussion Team members: Candice McGahern, BA (Children’s Hospital of Eastern Ontario, Ontario); Gurinder Sangha, MD (Department of Pediatrics, London Children’s Hospital, London, Ontario); Darcy Beer, MD (Department of Pediatrics, Manitoba Children’s Hospital; Winnipeg, Manitoba); William Craig, MDCM (Department of Pediatrics, Stollery Children’s Hospital, Edmonton, Alberta); Emma Burns, MD (Department of Pediatrics, IWK Health Centre, Halifax, Nova Scotia); Ken J. Farion, MD (Department of Pediatrics; Children’s Hospital of Eastern Ontario); Angelo Mikrogianakis, MD (Department of Pediatrics, Alberta Children’s Hospital); Karen Barlow, MD (Department of Pediatrics and Clinical Neurosciences, Alberta’s Children’s Hospital, Calgary, Alberta); Alexander S. Dubrovsky, MDCM, MSc (Department of Pediatrics, Montreal Children’s Hospital, Montreal, Québec;); Gerard Gioia, PhD (Department of Neuropsychology, Children’s National Health System, George Washington University School of Medicine, Rockville, Maryland); Miriam H. Beauchamp, PhD (Ste Justine Research Center, University of Montreal, Montreal, Quebec); Yael Kamil, BSc (Children’s Hospital of Eastern Ontario); Blaine Hoshizaki, PhD (Department of Kinesiology, University of Ottawa); Peter Anderson, PhD (Department of Psychology, Children’s Hospital of Eastern Ontario); Brian L. Brooks, PhD (Alberta Children’s Hospital Research Institute, Calgary, Alberta); Michael Vassilyadi, MDCM, MSc (Department of Neurosurgery, Children’s Hospital of Eastern Ontario); Terry Klassen, MD (Department of Pediatrics, Manitoba Children’s Hospital, Winnipeg, Manitoba); Michelle Keightley, PhD (Bloorview Research Institute, Holland Bloorview Kids Rehabilitation Hospital, Toronto, Ontario); Lawrence Richer, MD (Department of Neurology, Stollery Children’s Hospital, Edmonton, Alberta); Carol DeMatteo, MSc (School of Rehabilitation Science, McMaster University, Hamilton, Ontario). Mss McGahern and Kamil received compensation in association with their contributions to this article. None of the other aforementioned individuals received compensation.

Previous Presentation: Preliminary results of this study were presented as an abstract at the 11th Congress on Brain Injury, The Hague, the Netherlands, March 2016; the Pediatric Academic Societies Meeting, Baltimore, Maryland, May 2016; and the Fifth International Consensus Conference on Concussion in Sport, Berlin, Germany, October 2016.

Disclaimer: This is a substudy of the Predicting and Preventing Postconcussive Problems in Pediatrics (5P) study (recently published in JAMA), and while it includes necessary overlap in baseline patient characteristic data, the primary and secondary outcomes are unique.

Additional Contributions: We thank the parents and children who enrolled in this study and acknowledge the research coordinators and research assistants across the 9 sites responsible for patient recruitment, enrollment, and follow-up. Student volunteers at Children’s Hospital of Eastern Ontario, Alberta Children’s Hospital, Le Centre hospitalier universitaire Sainte-Justine (CHUSJ), and the Hospital for Sick Children provided assistance in patient screening at the emergency department. We appreciate the collaboration and assistance of all the treating physicians of the emergency departments across the sites. We are grateful to Pediatric Emergency Research Canada for making this study possible. None of those mentioned were compensated for their contribution.

References
1.
McCrory  P, Meeuwisse  WH, Aubry  M,  et al.  Consensus statement on concussion in sport.  Br J Sports Med. 2013;47(5):250-258.PubMedGoogle ScholarCrossref
2.
Zemek  R, Duval  S, Dematteo  C.  Guidelines for Diagnosing and Managing Pediatric Concussion. Toronto, Canada: Ontario Neurotrauma Foundation; 2014.
3.
DiFazio  M, Silverberg  ND, Kirkwood  MW, Bernier  R, Iverson  GL.  Prolonged activity restriction after concussion.  Clin Pediatr (Phila). 2016;55(5):443-451.PubMedGoogle ScholarCrossref
4.
Cancelliere  C, Hincapié  CA, Keightley  M,  et al.  Systematic review of prognosis and return to play after sport concussion.  Arch Phys Med Rehabil. 2014;95(3)(suppl):S210-S229.PubMedGoogle ScholarCrossref
5.
Schneider  KJ, Iverson  GL, Emery  CA, McCrory  P, Herring  SA, Meeuwisse  WH.  The effects of rest and treatment following sport-related concussion.  Br J Sports Med. 2013;47(5):304-307.PubMedGoogle ScholarCrossref
6.
Majerske  CW, Mihalik  JP, Ren  D,  et al.  Concussion in sports.  J Athl Train. 2008;43(3):265-274.PubMedGoogle ScholarCrossref
7.
Thomas  DG, Apps  JN, Hoffmann  RG, McCrea  M, Hammeke  T.  Benefits of strict rest after acute concussion.  Pediatrics. 2015;135(2):213-223.PubMedGoogle ScholarCrossref
8.
Willer  B, Leddy  JJ.  Management of concussion and post-concussion syndrome.  Curr Treat Options Neurol. 2006;8(5):415-426.PubMedGoogle ScholarCrossref
9.
Gagnon  I, Grilli  L, Friedman  D, Iverson  GL.  A pilot study of active rehabilitation for adolescents who are slow to recover from sport-related concussion.  Scand J Med Sci Sports. 2016;26(3):299-306.PubMedGoogle ScholarCrossref
10.
Leddy  JJ, Kozlowski  K, Donnelly  JP, Pendergast  DR, Epstein  LH, Willer  B.  A preliminary study of subsymptom threshold exercise training for refractory post-concussion syndrome.  Clin J Sport Med. 2010;20(1):21-27.PubMedGoogle ScholarCrossref
11.
Silverberg  ND, Iverson  GL.  Is rest after concussion “the best medicine?”.  J Head Trauma Rehabil. 2013;28(4):250-259.PubMedGoogle ScholarCrossref
12.
Zemek  R, Osmond  MH, Barrowman  N,  et al.  Predicting and preventing postconcussive problems in paediatrics (5P) study.  BMJ Open. 2013;3(8):e003550.PubMedGoogle ScholarCrossref
13.
Zemek  R, Barrowman  N, Freedman  SB,  et al; Pediatric Emergency Research Canada (PERC) Concussion Team.  Clinical risk score for persistent postconcussion symptoms among children with acute concussion in the ED.  JAMA. 2016;315(10):1014-1025.PubMedGoogle ScholarCrossref
14.
Harris  PA, Taylor  R, Thielke  R, Payne  J, Gonzalez  N, Conde  JG.  Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support.  J Biomed Inform. 2009;42(2):377-381.PubMedGoogle ScholarCrossref
15.
Gioia  GA, Collins  M, Isquith  PK.  Improving identification and diagnosis of mild traumatic brain injury with evidence.  J Head Trauma Rehabil. 2008;23(4):230-242.PubMedGoogle ScholarCrossref
16.
Sady  MD, Vaughan  CG, Gioia  GA.  Psychometric characteristics of the postconcussion symptom inventory in children and adolescents.  Arch Clin Neuropsychol. 2014;29(4):348-363.PubMedGoogle ScholarCrossref
17.
McCrory  P, Meeuwisse  WH, Aubry  M,  et al.  Child SCAT3.  Br J Sports Med. 2013;47(5):263-266.PubMedGoogle Scholar
18.
Steindel  SJ.  International classification of diseases, 10th edition, clinical modification and procedure coding system: descriptive overview of the next generation HIPAA code sets.  J Am Med Inform Assoc. 2010;17(3):274-282.PubMedGoogle Scholar
19.
Newcombe  RG.  Two-sided confidence intervals for the single proportion: comparison of seven methods.  Stat Med. 1998;17(8):857-872.PubMedGoogle ScholarCrossref
20.
Haukoos  JS, Lewis  RJ.  The propensity score.  JAMA. 2015;314(15):1637-1638.PubMedGoogle ScholarCrossref
21.
Austin  PC.  An introduction to propensity score methods for reducing the effects of confounding in observational studies.  Multivariate Behav Res. 2011;46(3):399-424.PubMedGoogle ScholarCrossref
22.
Kurth  T, Walker  AM, Glynn  RJ,  et al.  Results of multivariable logistic regression, propensity matching, propensity adjustment, and propensity-based weighting under conditions of nonuniform effect.  Am J Epidemiol. 2006;163(3):262-270.PubMedGoogle ScholarCrossref
23.
Brookhart  MA, Schneeweiss  S, Rothman  KJ, Glynn  RJ, Avorn  J, Stürmer  T.  Variable selection for propensity score models.  Am J Epidemiol. 2006;163(12):1149-1156.PubMedGoogle ScholarCrossref
24.
Rosenbaum  P, Rubin  D.  The central role of the propensity score in observational studies for causal effects.  Biometrika. 1983;70(1):41-55. doi:10.1093/biomet/70.1.41Google ScholarCrossref
25.
Williamson  EJ, Forbes  A.  Introduction to propensity scores.  Respirology. 2014;19(5):625-635.PubMedGoogle ScholarCrossref
26.
Youden  WJ.  Index for rating diagnostic tests.  Cancer. 1950;3(1):32-35.PubMedGoogle ScholarCrossref
27.
Rosenbaum  P, Rubin  D.  Constructing a control group using multivariate matched sampling methods that incorporate the propensity score.  Am Stat. 1985;39:33-38. doi:10.1017/CBO9780511810725.019Google Scholar
28.
Scott  D.  Multivariate Density Estimation: Theory, Practice and Visualization. Wiley Series in Probability and Mathematical Statistics. New York, NY: John Wiley & Sons; 1992.
29.
Khan  NA, Hillman  CH.  The relation of childhood physical activity and aerobic fitness to brain function and cognition.  Pediatr Exerc Sci. 2014;26(2):138-146.PubMedGoogle ScholarCrossref
30.
Hillman  CH, Erickson  KI, Kramer  AF.  Be smart, exercise your heart.  Nat Rev Neurosci. 2008;9(1):58-65.PubMedGoogle ScholarCrossref
31.
Veerbeek  JM, van Wegen  E, van Peppen  R,  et al.  What is the evidence for physical therapy poststroke?  PLoS One. 2014;9(2):e87987.PubMedGoogle ScholarCrossref
32.
Longmuir  PE, Colley  RC, Wherley  VA, Tremblay  MS.  Canadian Society for Exercise Physiology position stand: benefit and risk for promoting childhood physical activity.  Appl Physiol Nutr Metab. 2014;39(11):1271-1279.PubMedGoogle ScholarCrossref
33.
Rush  E, Simmons  D.  Physical activity in children.  Med Sport Sci. 2014;60:113-121.PubMedGoogle Scholar
34.
McKay  HA, Petit  MA, Schutz  RW, Prior  JC, Barr  SI, Khan  KM.  Augmented trochanteric bone mineral density after modified physical education classes.  J Pediatr. 2000;136(2):156-162.PubMedGoogle ScholarCrossref
35.
Annesi  JJ.  Correlations of depression and total mood disturbance with physical activity and self-concept in preadolescents enrolled in an after-school exercise program.  Psychol Rep. 2005;96(3 Pt 2):891-898.PubMedGoogle ScholarCrossref
36.
Howie  EK, Pate  RR.  Physical activity and academic achievement in children.  J Sport Health Sci. 2012;1(3):160-169. doi:10.1016/j.jshs.2012.09.003Google ScholarCrossref
37.
Chaddock-Heyman  L, Hillman  CH, Cohen  NJ, Kramer  AF  III.  The importance of physical activity and aerobic fitness for cognitive control and memory in children.  Monogr Soc Res Child Dev. 2014;79(4):25-50.PubMedGoogle ScholarCrossref
38.
Moor  HM, Eisenhauer  RC, Killian  KD,  et al.  The relationship between adherence behaviors and recovery time in adolescents after a sports-related concussion.  Int J Sports Phys Ther. 2015;10(2):225-233.PubMedGoogle Scholar
39.
Perrey  S.  Promoting motor function by exercising the brain.  Brain Sci. 2013;3(1):101-122.PubMedGoogle ScholarCrossref
40.
Farquhar  WB, Greaney  JL.  Autonomic exercise physiology in health and disease.  Auton Neurosci. 2015;188:1-2.PubMedGoogle ScholarCrossref
41.
Leddy  JJ, Kozlowski  K, Fung  M, Pendergast  DR, Willer  B.  Regulatory and autoregulatory physiological dysfunction as a primary characteristic of post concussion syndrome.  NeuroRehabilitation. 2007;22(3):199-205.PubMedGoogle Scholar
42.
Adamo  KB, Prince  SA, Tricco  AC, Connor-Gorber  S, Tremblay  M.  A comparison of indirect versus direct measures for assessing physical activity in the pediatric population.  Int J Pediatr Obes. 2009;4(1):2-27.PubMedGoogle ScholarCrossref
×