Article Text

Fatigue recovery and connected factors following paediatric concussion
  1. Fabian Fabiano1,
  2. Michael Takagi1,
  3. Nicholas Anderson1,
  4. Franz E Babl1,2,
  5. Silvia Bressan3,
  6. Cathriona Clarke1,
  7. Katie Davies1,
  8. Gavin A Davis1,4,
  9. Kevin Dunne1,5,
  10. Stephen Hearps1,
  11. Vera Ignjatovic6,7,
  12. Vanessa C Rausa1,
  13. Marc Seal1,
  14. Vicki Anderson1,8
  1. 1 Clinical Sciences Research, Murdoch Children's Research Institute, Melbourne, Victoria, Australia
  2. 2 Emergency Department, Royal Children's Hospital, Melbourne, Victoria, Australia
  3. 3 Department of Women's and Children's Health, Università degli Studi di Padova, Padova, Italy
  4. 4 Cabrini Health, Melbourne, Victoria, Australia
  5. 5 Department of Rehabilitation, Royal Children's Hospital, Melbourne, Victoria, Australia
  6. 6 Institute for Clinical and Translational Research, Johns Hopkins All Children's Hospital, St Petersburg, Florida, USA
  7. 7 Departments of Pediatrics, The Johns Hopkins University School of Medicine, Baltimore, Maryland, USA
  8. 8 Melbourne School of Psychological Sciences, University of Melbourne, Melbourne, Victoria, Australia
  1. Correspondence to Fabian Fabiano, Murdoch Childrens Research Institute, Melbourne, Victoria, Australia; fabian.fabiano{at}


Objective Using a biopsychosocial framework and the three-factor fatigue model, we aimed to (1) plot recovery of fatigue over the 3 months following paediatric concussion and (2) explore factors associated with persisting fatigue during the first 3 months postconcussion.

Methods 240 children and adolescents aged 5–18 years (M=11.64, SD=3.16) completed assessments from time of injury to 3 months postinjury. Separate linear mixed effects models were conducted for child and parent ratings on the PedsQL-Multidimensional Fatigue Scale to plot recovery across domains (General, Cognitive, Sleep/Rest) and Total fatigue, from 1 week to 3 months postinjury. Two-block hierarchical regression analyses were then conducted for parent and child ratings of fatigue at each time point, with age, sex and acute symptoms in block 1 and child and parent mental health variables added to block 2.

Results There was a significant reduction in both child and parent ratings across the 3 months postinjury for all fatigue domains (all p<0.001). For both child and parent fatigue ratings, child mental health was the most significant factor associated with fatigue at all time points. Adding child and parent mental health variables in the second block of the regression substantially increased the variance explained for both child and parent ratings of fatigue.

Conclusion Our findings confirm that fatigue improves during the first 3 months postconcussion and highlights the importance of considering child and parent mental health screening when assessing patients with persisting postconcussive symptoms.

  • Fatigue
  • Pediatrics
  • Brain Concussion
  • Child Health

Data availability statement

Data are available on reasonable request.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • Fatigue is among the most commonly reported persisting symptoms following paediatric concussion yet remains an under-researched area. Comprehensive, developmentally appropriate assessment of fatigue is rarely conducted by clinicians during treatment of paediatric concussion.


  • We plot the recovery of fatigue over the first 3 months following paediatric concussion and explore associated factors of persisting fatigue. We found that fatigue was associated with emotional, more so than physical, symptoms.


  • This study highlights the potential importance of including mental health screening in the management of paediatric concussion.

Concussion, a subset of mild traumatic brain injury (TBI), is the result of a complex pathophysiological process induced by biomechanical forces acting on the head, and accounts for up to 90% of all paediatric TBI.1 2 Immediately following injury, concussion manifests as a constellation of transient physical, cognitive, emotional and postconcussive symptoms (PCS).3 4 Examples of PCS include headache, nausea, fatigue, sleep disturbance, irritability, memory or concentration difficulties, mental slowness, visual disturbance, and sensitivity to light or noise.5 Research indicates that most children recover spontaneously within 2 weeks of their concussion.4 6 7 However, up to one-third remain symptomatic at 1 month, 10% at 3 months and less than 5% at 1 year.3 6 Persistent PCS may interfere with return to normal life, including participation in social, school, and sporting activities.8

Fatigue is among the most commonly reported persisting symptoms following all forms of paediatric acquired brain injury,9 and in concussion specifically.7 It is defined as ‘an overwhelming sense of tiredness, lack of energy and a feeling of exhaustion, associated with impaired physical and/or cognitive functioning…’.10(Shen J, 10 p70) There is no consensus on the specific number and nature of fatigue dimensions, although, in the context of TBI, most authors embrace a multidimensional framework, including physical, cognitive, emotional, and motivational dimensions.11 Varni et al 12 propose a three factor model: (1) general fatigue (ie, lack of physical energy); (2) sleep/rest fatigue (ie, sleep-wake cycle disturbance and daytime sleepiness) and (3) cognitive fatigue (ie, reduced mental endurance and attention). All aspects of fatigue are likely to have wide-ranging functional and clinical impacts on the child in daily life, and yet paediatric postconcussion fatigue remains under-researched, and comprehensive and reliable assessment of fatigue is rarely included in clinical treatment or management.13

In other paediatric conditions, fatigue has been associated with multiple functional consequences: reduced school, psychosocial, and physical participation, lower cognitive function, mood and quality of life, and increased parent burden.9 14–19 The handful of studies that assessed fatigue following paediatric concussion provide inconsistent findings. Barlow et al 3 reported that, of all PCS, parents rated fatigue the most common symptom to have increased at 1-month postinjury. Eisenberg et al 6 found that 64.7% of 11–22 years endorsed fatigue symptoms within 72 hours of a concussion, with an additional 15.4% of the adolescent patients developing symptoms more than 7 days postinjury, while Babcock et al 20 reported that 11% of children aged 5–18 years endorsed symptoms of fatigue 3 months following mild TBI. These inconsistent results likely reflect methodological issues including differing study designs, timing of measurement and quality and sensitivity of measurement tools.

Using a biopsychosocial framework and the three-factor fatigue model, we aimed to: (1) plot recovery of fatigue and (2) explore the factors associated with persisting fatigue at 3 months postconcussion. We hypothesised that all fatigue domains would decrease with time and that persisting fatigue would be associated with emotional, more so than physical, symptoms, with parent mental health also playing a role.



Data were collected as part of the ‘Take CARe’ (Concussion Assessment and Recovery Research) study, a single site, prospective, longitudinal study of children and adolescents (referred to as children) presenting to the emergency department (ED) of a tertiary paediatric hospital with a diagnosis of concussion. See Bressan et al 21 and Takagi et al 22 for full details.


Participants aged 5–18 years who presented to the ED between 2014 and 2017 were recruited. Consistent with the Berlin Consensus statement,23 concussion was defined as a blunt injury to the head associated with an altered mental state or any of the following symptoms: headache, dizziness, fatigue, drowsiness, nausea/vomiting, poor balance, memory or concentration problems.

Exclusion criteria were: intubated, neurosurgical operative intervention or general anaesthesia for injury management, structural/haemorrhagic intracranial injury on CT scan, clinical evidence of cerebrospinal fluid leak, intellectual disability with inability to complete the testing measures, injury from child abuse or assault, alcohol or drug intoxication at presentation, insufficient English to complete study requirements, multiple trauma, presence of fever at time of initial assessment (T>37.5°), no clear history of trauma as primary event (eg, seizure, syncope or migraine as primary event) and already enrolled in the study.

We screened 2031 children who presented to the ED with a concussion sustained in the previous 48 hours. As seen in online supplemental figure S1, 1236 met study inclusion criteria and 394 agreed to participate, 24 of which were excluded post enrolment, resulting in a study sample of 370. Of the 370 in the study sample, 130 withdrew from the study, mostly due to difficulties in arranging follow-up sessions, resulting in a final sample of 240 participants for analyses.

Supplemental material


Participants completed assessments over five time points: T0: during the initial ED admission; T1: 2–4 days postinjury; T2: 2 weeks; T3: 1 month and T4: 3 months postinjury. At each time point, families completed several questionnaires as part of the larger study, which are not all included in this investigation.


Clinical and demographic data: Data were collected at T0 using a standard study clinical report form which included items regarding injury mechanism, symptoms and medical history.

Primary outcome

Child Fatigue (primary outcome): The Pediatric Quality of Life Multidimensional Fatigue Scale (PedsQL MFS) is an 18-item scale developed to measure child fatigue.24 General, sleep/rest and cognitive fatigue symptoms are rated on a 5‐point Likert scale (0=never a problem; to 4=almost always a problem). Higher scores indicate higher quality of life (ie, fewer fatigue symptoms). This measure was completed by parents and children at T1, T3 and T4. An additional data collection time point was introduced at T2 in 2017 but these data were omitted from the analysis in this study. Independent samples t-tests found no distinction between T2 and other time points on all demographic variables.


Acute postconcussion function (T0)

We assessed acute postconcussion function in the ED within 48 hours of injury using the Sport Concussion Assessment Tool (SCAT3) for children 13 years and older and the Child-SCAT3 for children 5–12 years.2 The SCAT3/Child-SCAT3 comprises three core measurement components: postconcussion symptoms, cognitive function and postural stability, detailed below. SCAT3 and ChildSCAT3 scales were standardised, and older and younger versions were collapsed into one variable for analyses to form the following constructs:

Acute PCS (T0): Acute PCS were derived from the SCAT3 Post-Concussion Symptom Scale (PCSS) for children aged 13–18 years, and the Child-SCAT3 Health and Behaviour Inventory (HBI) for children aged 5–12 years. The SCAT3 PCSS is a 22-item scale that measures common symptoms following concussion. Presence and severity of child symptoms are rated on a six-point Likert scale (0=none to 6=severe). The Child-SCAT3 incorporates the child report and parent report 20-item HBI scale that measures common symptoms following concussion. Parents and children rated presence and severity of child symptoms on a four-point Likert scale (0=never a problem to 4=often a problem).

Acute physical function (T0): The modified balance error-scoring system (BESS)25 26 assesses postural stability (Child-SCAT3 double and tandem stance, and SCAT3 includes the additional single leg stance), with scores reflecting total errors (maximum error score for SCAT3 is 30, and for Child-SCAT3 is 20).

Acute cognitive function (T0): The Standardised Assessment of Concussion (SAC/SAC-C)27 28 assesses orientation, immediate memory, concentration and delayed recall. The maximum possible score (equal to the number of correct answers for the different items) is 30, with higher scores representing better performance.

Child and parent mental health (T4)

Parents were asked to complete self-reported measures of their child’s and their own mental health status 3 months following the injury using the following validated tools:

Child Behaviour Checklist

The Child Behaviour Checklist (CBCL) provides a parent rating of the child’s behaviour problems with three summary scores: Internalising Behaviour, Externalising Behaviour and Total Behaviour Score, with higher scores reflecting more behaviour problems.29

Child PTSD Symptom Scale

The Child PTSD Symptom Scale (CPSS) is a 24-item self-report measure assessing the severity of child post-traumatic stress across three clusters: Re-experiencing, Avoidance and Arousal. Responses are on a 4-point Likert scale: 0 (not at all), 1 (once a week or less), 2 (2–4 times a week), to 3 (5 or more times a week). Higher scores reflect greater post-traumatic symptom severity.30

Kessler 10 Scale

The Kessler 10 Scale (K10) measures non-specific parent psychological distress using 10 symptoms and level of functional impairment in the previous month. A total score is calculated, with higher scores indicating higher levels of distress.31 32

Patient and public involvement

No patients or members of the public were involved in the design or interpretation of this study. All participation was voluntary. Patients helped to motivate community involvement during and beyond the study. We carefully assessed the burden of study procedures on patients and updated the study protocol as appropriate. We intend to disseminate the main results to trial participants and will seek patient and public involvement in the development of an appropriate method of dissemination.

Equity, diversity and inclusion

Our study included all identified cases of concussion in children and adolescents inclusive of all genders, race/ethnicities and socioeconomic levels. The author team included early, middle and late career researchers with balance from people who identify as male and female. The authors’ disciplines include neuropsychology, physiotherapy, biomedical sciences and social sciences.

Data analysis

Rates of missing data among demographic variables ranged from 0 (age and sex) to 16.67% (CPSS; see table 1 for demographic information and counts and proportions of missing data). As such, multiple imputations were executed using the mice package in R.33 The overall percentage of missingness for all variables was 3.95%, and, as such, 20 imputed datasets were created, meeting recommended guidelines.34

Table 1

Participant preinjury and injury characteristics taken at enrolment (N=240)

Separate linear mixed effects models (with time as a categorical predictor) were conducted for PedsQL MFS child and parent fatigue ratings to plot recovery across the three fatigue domains (General, Cognitive, Sleep/Rest) as well as Total Fatigue scores at T1, T3 and T4. Two-block hierarchical regression analyses were conducted for each fatigue time point on Total Fatigue symptom scores. The first block contained age at enrolment, sex and acute PCS (PCSS/HBI), acute physical function (BESS), and acute cognitive function (SAC). The second block added child behaviour problems (CBCL Total), child post-traumatic stress (CPSS) and parent psychological distress (K10). Separate hierarchical regression analyses were performed for parent and child ratings of fatigue.

All analyses were conducted by using R V.


Sample characteristics

Total sample size was 240, with the mean age at enrolment being 11.64 years (SD=3.16, range=5–17 years) and 72.5% were male participants, as presented in table 1. Postconcussion characteristics and cognitive and mental health function are illustrated in table 2.

Table 2

Descriptive statistics

Fatigue recovery

Separate linear mixed effects models were conducted for child and parent ratings of fatigue over time and across fatigue domains and total fatigue scores (figures 1 and 2). There was a significant change in fatigue ratings in each domain for both parent and child ratings over time (all p values<0.001). Examination of the mean fatigue scores demonstrated a general reduction in fatigue across the three time points with a reduction across all three fatigue domains and total fatigue in both child and parent assessment (reflected by an increase in ratings).

Figure 1

Mean Child Ratings on PedsQL MFS for each Fatigue Domain across Time (N=240). Y-axis has been reversed to reflect increased scores indicating lower fatigue. All changes across the three time points were significant (all p values<0.001). PedsQL MFS, Pediatric Quality of Life Multidimensional Fatigue Scale.

Figure 2

Mean parent ratings on PedsQL MFS for each Fatigue Domain across Time (N=240). Y-axis has been reversed to reflect increased scores indicating lower fatigue. All changes across the three time points were significant (all p values<0.001). PedsQL MFS, Pediatric Quality of Life Multidimensional Fatigue Scale.

Factors associated with fatigue

Hierarchical regression analyses for child and parent fatigue ratings are provided in tables 3 and 4 along with standardised coefficients and CIs for factors associated with fatigue. For child ratings at T1, the first block (block 1), which included age, sex and acute PCS, accounted for approximately 23% of the variance in fatigue ratings, and adding the child and parent mental health variables (block 2) explained an additional 12% of the variance in fatigue scores at 1–4 days. An even larger increase in variance explained was seen at T3 (11% in block 1 to 40% in block 2) and T4 (12% in block 1 to 59% in block 2). Regarding individual predictors, acute PCS severity was consistently the most significant predictor in block 1, reaching significance at all three time points, and was rendered non-significant in block 2 at all but T1. Child post-traumatic stress was the strongest predictor in block 2.

Table 3

Hierarchical regression analysis of factors associated with child-rated fatigue on the Pediatric Quality of Life Multidimensional Fatigue Scale across time points (N=240)

Table 4

Hierarchical regression analysis of factors associated with parent-rated fatigue on the Pediatric Quality of Life Multidimensional Fatigue Scale across time points (N=240)

For parent ratings at T1, block 1 accounted for approximately 9% of the variance, and adding child and parent mental health variables increased the variance explained to 28% in block 2. An even larger increase from block 1 to block 2 in the magnitude of variance explained was seen at T3 (12%–32%) and T4 (8%–50%). Child behaviour problems were the most significant factor related to fatigue when incorporated in a regression with other covariates.


Our study found that fatigue decreased between the acute postinjury stage to 3 months in children following a concussion. Both child and parent ratings displayed the predicted significant decreases over time in all fatigue domains and total fatigue, but child ratings indicated a decrease mostly from 1 to 4 days to 1 month, whereas parent ratings indicated a decrease mostly from 1 to 3 months. This could be due to a lag in parent ratings, as decreases in the subjective experience of fatigue might occur in the child before it is noticed by the parent. This is potentially a novel finding and warrants further investigation. Moreover, child ratings of fatigue indicated a larger reduction in fatigue than parent ratings, which might also indicate a lag effect, and ratings might have eventually equalised.

In contrast to Bogdanov et al,36 who found that parent ratings led to more robust linear models, child ratings were more robust in our study. This is an encouraging finding, demonstrating that child ratings can potentially be just as efficacious as parent ratings. The decreases were relatively consistent and synchronous across the general, cognitive and sleep/rest fatigue domains, and further research is required to unpack whether different fatigue domains change differently over time, particularly in the context of targeted interventions.

While paediatric research in postconcussion fatigue is scarce, adult studies have explored factors contributing to persisting fatigue postconcussion. Age has often failed to predict fatigue outcomes postconcussion (eg, Norup et al 37) and our study findings support an inconsistent link between age and fatigue. Sex has also been associated with fatigue in some studies,38 but not in others.39 In this study, sex was the least substantial correlate of fatigue, with no relationship found at any time point postinjury.

The association between acute PCS severity and ongoing fatigue is also yet to be established,40 and while this study found a relationship between acute PCS severity and fatigue, there was little to no relationship between fatigue and acute physical and cognitive measures. Furthermore, the relationship between acute PCS severity and fatigue dissolved when mental health was included in our models.

As others have demonstrated,7 with time since injury, child mental health becomes increasingly important in driving outcomes. Depression, anxiety and post-traumatic stress have all been associated with fatigue recovery.41 42 Consistent with this previous work, our study identified child behaviour problems and traumatic stress as highly associated with postconcussion fatigue. Whether this effect is injury related or caused by difficulties adjusting to injury-related limitations (eg, exclusion for school and sports) requires further investigation.

Parent distress is established as contributing to poorer recovery outcomes in children after TBI of any severity40 and was moderately associated with fatigue in our study. When combined with child post-traumatic stress and child behaviour problems, however, its importance is reduced. This is an interesting finding, possibly suggesting that child mental health status may mediate the relationship between parent mental health and fatigue.

Clinical implications

This study demonstrates that fatigue should reduce over the first 3 months following concussion, perhaps to preinjury levels. Our findings also highlight the importance of considering both child and parent measures when assessing fatigue, as they might display differences in the rate of recovery. Additionally, links between fatigue and child mental health symptoms indicate the critical role of mental health screening in clinical follow-up and management of child concussion. The complex interplay of factors contributing to postconcussion fatigue highlights the importance of multidisciplinary approaches to concussion recovery (eg, psychopharmaceutic,43 psychotherapeutic44 and physiotherapeutic45) and close collaboration between healthcare professionals, such as paediatricians, sports medicine specialists, psychologists and physiotherapists, to ensure comprehensive care that addresses the complex biopsychosocial nature of concussion recovery. Recognising the potential impact of psychosocial constructs on fatigue outcomes also underscores the need for increased patient and family education regarding the role of mental health in recovery, and the importance of seeking appropriate evidence-based support throughout recovery.


This study was not without limitations. Mental health was assessed 3 months postinjury for a retrospective report of the prior months, which introduces the possibility of bias, as does retrospective reporting of preinjury symptoms compared with current symptoms. Additionally, with research showing that only 12% of children with concussion present to the ED,46 recruitment from this setting does not necessarily reflect the general community of children with concussion. Attrition may introduce the possibility of bias due to the potential for non-random dropouts, particularly if those who are either recovered or experiencing severe symptoms are less likely to return for follow-up. As such, a thorough comparison (see online supplemental table S2) was conducted between participants who completed the study and those who did not across all analysed variables and found very minor differences between those who completed all time points and those lost to follow-up. Finally, sample descriptive data were compared with normative samples in the absence of a healthy control group.

Supplemental material


This study used a prospective longitudinal design in a paediatric concussion sample and found that psychosocial and mental health constructs such as child behaviour problems and post-traumatic stress were strongly associated with fatigue during the first 3 months following injury, as was parent mental health. Future investigation into this matter should continue to expand on often-cited variables into more intricate social factors, such as parent mental health, child behaviour problems and post-traumatic responses to the injury event, and thereby explore the looping relationship between these constructs and fatigue.

Data availability statement

Data are available on reasonable request.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants and was approved by the Royal Children’s Hospital (RCH) Human Research Ethics Committee (33122). Participants gave informed consent to participate in the study before taking part.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Twitter @VickiAn28323584

  • Contributors FF drafted the initial manuscript and analysed the data, and VA drafted the initial manuscript, and conceptualised, designed, and is guarantor of the study. FEB, SB, KD, GAD, SH, VI, MS and MT conceptualised and designed the study and critically revised the manuscript for important intellectual content. NA, CC, KD and VCR made substantial contributions to the conception of the study, the acquisition of data, and critically revised the manuscript for important intellectual content. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

  • Funding This work was supported by The Royal Children’s Hospital Research Foundation (2014-370). VA and FEB are supported by Australian National Health and Medical Research Council practitioner fellowships.

  • Disclaimer The funding organisations did not have a role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or the decision to submit the manuscript for publication.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.