Abstract
Objective
To determine sex differences in the recovery and prognosis after mild traumatic brain injury (MTBI) in adults and children.
Data Sources
We analyzed all scientifically admissible primary studies in the World Health Organization (WHO) (n=120) and International Collaboration on Mild Traumatic Brain Injury Prognosis (ICoMP) (n=101) systematic reviews regarding prognosis of MTBI for sex-stratified findings. They searched MEDLINE and other databases from 1980 through 2000 (WHO) and 2001 through 2012 (ICoMP) for published, peer-reviewed reports in English and other languages.
Study Selection
We selected controlled trials and cohort and case-control studies that assessed the effect of sex on outcomes after MTBI.
Data Extraction
Data from the eligible studies from both systematic reviews combined (n=14, 7%) were extracted into evidence tables.
Data Synthesis
Prognostic information relating to sex was prioritized according to design as exploratory or confirmatory, and a best-evidence synthesis was conducted. After MTBI, females may have a higher risk of epilepsy (children, young adults) and suicide, and use more health care services; males may be at higher risk for schizophrenia. Most studies did not find a sex difference for postconcussion symptoms in children and adults. No sex difference was found for risk of dementia and primary brain tumor, return to work, or posttraumatic stress syndrome.
Conclusions
Sex is not a well-studied prognostic indicator for recovery after MTBI, but small sex differences were found for some outcomes. More well-designed studies are needed that report outcomes according to sex and control for potential confounders.
Keywords
List of abbreviations:
CI (confidence interval), ED (emergency department), ICoMP (International Collaboration on Mild Traumatic Brain Injury Prognosis), MTBI (mild traumatic brain injury), OR (odds ratio), PCS (postconcussion symptoms), PTSD (posttraumatic stress disorder), RPQ (Rivermead Postconcussion Symptoms Questionnaire), RR (relative risk), RTW (return to work), TBI (traumatic brain injury), WHO (World Health Organization)Traumatic brain injury (TBI) is a major cause of disability worldwide.
1
Concussion or mild TBI (MTBI) is the most common form of TBI and has an incidence in excess of 600 cases per 100,000 people per year.2
The economic effect of MTBI is substantial, accounting for approximately 44% of the $60 billion annual cost of TBI in the United States.3
As a result, MTBI has become a major public health problem. In order to improve patient management and outcomes, it is important to identify all potential prognostic indicators, such as sex, that may impact recovery after MTBI.Sex plays an important role in health because individuals may experience various processes differently based on their biology.
4
For example, males and females respond differently to drugs and therapeutics because of differences in body composition, metabolism, and hormones.4
Sex differences in outcome after MTBI have been addressed in a number of studies, of which some found females to have a poorer outcome than males.5
, 6
, 7
However, the evidence is limited and conflicting.8
, 9
Therefore, the objective of this report is to determine the prognostic significance, if any, of sex on recovery after MTBI in adults and children. This has important implications for rehabilitation and prognosis in the field of MTBI.Methods
We analyzed all the scientifically admissible primary studies in 2 systematic reviews
10
, 11
regarding the prognosis of MTBI for sex-stratified findings. One review10
was conducted by the World Health Organization (WHO) Collaborating Centre for Neurotrauma, Prevention, Management and Rehabilitation Task Force (WHO Task Force, 2004). The second review11
was conducted by the International Collaboration on Mild Traumatic Brain Injury Prognosis (ICoMP, 2014). The methodologies for these reviews are outlined in detail elsewhere.12
, 13
WHO Task Force review
Briefly, the WHO Task Force systematically searched the published English literature using MEDLINE and PsycINFO (1980–2000), CINAHL (1982–2000), and Embase (1988–2000). Study types included controlled trials; and cohort, case-control, cross-sectional, and case-series designs. Cohort studies identifying prognostic factors were classified as phase I, II, or III studies.
14
Phase I studies are hypothesis-generating investigations that explore the associations between potential prognostic factors and disease outcomes in a descriptive or univariate way. Phase II studies are extensive exploratory analyses that focus on particular sets of prognostic factors or attempt to discover which factors have the highest prognostic value. Phase III studies are large confirmatory studies of explicit prestated hypotheses that allow for a focused examination of the strength, direction, and independence of the proposed relationship between a prognostic factor and the outcome of interest. Accepted studies (ie, those with a low risk of bias) were summarized in evidence tables and included in the best-evidence synthesis. The best-evidence synthesis links summary statements and conclusions to the evidence tables so that the strength of the evidence on which these statements are based is obvious. Strength of the evidence considers both the design of the study and methodological quality. Information from high-quality phase III studies is confirmatory and considered the strongest evidence, followed by evidence from methodologically sound phase II studies. Phase I studies do not consider confounding and are considered more limited evidence, but still potentially more informative about prognosis than cross-sectional and case-series designs.ICoMP review
The main objective of the ICoMP review was to update the WHO Task Force findings.
11
Briefly, MEDLINE, PsycINFO, Embase, CINAHL, and SPORTDiscus were systematically searched from 2001 to 2012 for published literature in English, French, Swedish, Norwegian, Danish, and Spanish. MTBI was defined as an acute brain injury resulting from mechanical energy to the head from external physical forces and (1) 1 or more of the following: confusion or disorientation, loss of consciousness for ≤30 minutes, posttraumatic amnesia for <24 hours, and/or other transient neurologic abnormalities such as focal signs, seizures, and intracranial lesions not requiring surgery; and (2) Glasgow Coma Scale score of 13 to 15 after 30 minutes postinjury or later on presentation for health care.10
The ICoMP also categorized the evidence on prognostic factors as exploratory or confirmatory using the phases of study framework described above, and performed a best-evidence synthesis.Study selection
Studies had to assess the effect of sex on outcomes after MTBI. For this report, we only analyzed scientifically admissible controlled trials and cohort and case-control studies from both systematic reviews because these designs can infer causality. We excluded admissible primary studies that had cross-sectional and case-series designs (n=1), and studies that included assault cases (n=3). The rationale for this is that recovery from assault may be complicated by feelings of victimization, and to understand their prognosis would require stratifying results by injury intent.
Data extraction
Data from 14 eligible studies (16 articles; 4 phase III, 11 phase II, 1 phase I) were extracted into an evidence table.
Results
The WHO Task Force critically reviewed 427 articles and accepted 120 (28%) as scientifically admissible. Of these, 3 studies (5 articles) are relevant to our report. The ICoMP critically reviewed 299 articles and accepted 101 (34%) as scientifically admissible. Of these, 11 are relevant for our report. Thus, 14 studies (16 articles) spanning from 1980 through 2012 form the basis of our synthesis regarding sex as a prognostic indicator after MTBI (table 1).
Table 1Studies (n=14) examining sex differences in the recovery and prognosis after MTBI
| Author, Year, Country, Study Phase | Source Population, Study Size, Participation, Follow-Up | Inclusion/Exclusion Criteria | MTBI Case Definition | Prognostic Factors/Self-Reported Outcomes | Findings |
|---|---|---|---|---|---|
| Outcomes in children after MTBI | |||||
| Christensen et al, 15 2009; DenmarkPhase III cohort | Danish population born in Denmark between Jan 1, 1977, and Dec 31, 2002 Population-based cohort: n=1,605,216 | Danish children born in Denmark between Jan 1, 1977, and Dec 31, 2002 | ICD-8 850.99; ICD-10 S06.0. Diagnosis based on ACRM definition: Direct head trauma that results in LOC, amnesia, confusion, disorientation, or focal temporary neurologic deficit. LOC no longer than 30min, GCS 14 or 15 after 30min, PTA ≤24h | Prognostic factors: TBI severity; age at injury; sex; length of hospital stay; family history of epilepsy Outcome: Time to diagnosis of epilepsy (ICD-8 345 from 1977–1993 and ICD-10 G40, G41 from 1994–2002) | RR of epilepsy after MTBI was higher among females (2.49; 95% CI, 2.25–2.76, P=.003) than among males (2.01; 95% CI, 1.83–2.22). RR of epilepsy after MTBI (compared with no TBI) was 2.22 (95% CI, 2.07–2.38). Risk was highest during the 1st year after injury (RR for first 6mo, 5.46; 95% CI, 4.67–6.37) and decreased over time. |
| Outcomes in adults after MTBI | |||||
| Mehta et al, 17 1999; The NetherlandsPhase III cohort | Community-based: Rotterdam, The Netherlands Total cohort: N=6645 59.1% women Average F/U: 2y | Inclusion: Dementia-free status at baseline | Patient interview: Head trauma, duration of LOC, and PTA. | Prognostic factors: MTBI with LOC <15min, adjusting for sex, number of TBIs, and time since trauma Outcome: Dementia | No increased dementia (RR=1.0; 95% CI, 0.5–2.0) or AD (RR=0.8; 95% CI, 0.4–1.9) risk with MTBI There were no major differences in the RRs of AD and dementia between men and women. Patients who developed incident dementia or AD were older, more often women, and were less educated. |
| Nygren et al, 18 2001; SwedenPhase III cohort | All patients from the Swedish Inpatient Register from the period 1965–1994 discharged from hospital with a diagnosis of skull trauma Total cohort: N=311,006 patients contributing 3,222,317 person-years Male: 192,090 Female: 118,916 Mean F/U: 10.4y | Inclusion: Discharge diagnosis of skull trauma Exclusion: Death during first hospitalization, prevalent cancer at the time of entry to cohort, inconsistencies during record linkage | ICD-7, 8, and 9 injury codes | Prognostic factors: TBI adjusted for sex and TBI severity Outcome: Primary brain tumor | No increased risk of brain tumor after a head injury of any severity. The overall SIR was 1.0 (0.9–1.2) with little variation by sex. |
| Nielsen et al, 19 2002; DenmarkPhase III case control | Danish population. 8288 cases and 82,880 controls. Database study so complete participation and F/U. Time between MTBI and schizophrenia was split into <1y, 1–5y, and >5y | Inclusion criteria: Cases were all admissions to Danish psychiatric hospitals or psychiatric wards with diagnosis of schizophrenia (ICD-8 295) from Jan 1, 1978, to Dec 31, 1993. Excluded cases were those ever diagnosed with bipolar affective disorder. Controls were randomly selected from the Central Person Register (total Danish population), matched to sex, year of birth, and being alive on Jan 1, 1978. Excluded controls were those who had ever been admitted to a psychiatric hospital on or before the date of the case's first psychiatric admission. | Any admission to hospital with MTBI (ICD-8 code 850) | Prognostic factors: MTBI or severe head injury vs no TBI, adjusted for “accident proneness” (non–skull fractures, leading to hospitalization). Outcome: First psychiatric hospitalization of an individual who was then or subsequently diagnosed with schizophrenia. Time lag from MTBI to psychiatric hospitalization was <1y, 1–5y, and >5y. | MTBI may increase the risk of schizophrenia in males. Where they exist, effect sizes are modest. In unadjusted analyses, overall OR of association between MTBI and schizophrenia was .936 (NS). Association between MTBI and schizophrenia was affected by both sex and time lag. In adjusted analyses, males with schizophrenia were more exposed to MTBI (OR=1.501, P<.001). Effect was seen at <1y (OR=2.034, P=.004) and >5y (OR=1.499, P=.012). No effect in adjusted analyses for sex for any time frame. |
| Dischinger et al, 26 2009; United StatesPhase II cohort | 180 MTBI patients admitted to a level I trauma center 64% male Baseline measures were made within 3–10d postinjury and F/U at 3mo postinjury | Inclusion: 18–64y of age, acceptable score on Mini-Mental State Examination, and English speaking Exclusion: Brain lesion requiring intervention, moderate/severe multiple injuries, focal neurologic findings, skull fracture requiring clinical intervention, cerebrospinal fluid leak requiring clinical intervention, prior moderate or severe brain injury, new or prior seizures, history of psychiatric disorder requiring hospitalization or hallucinations, recent history of substance abuse, current probation/parole, and active duty in military | ACRM: admission GCS, 13–15; presence of ≥1 of the following: LOC <30min, loss of memory of events immediately before or after injury, or alteration of mental state after injury (eg, confusion, disorientation, feeling dazed). PTA >24h. | Prognostic factors: PCS: 6 physical (headache, dizziness, blurry/double vision, fatigue, sensitivity to light, sensitivity to noise), 3 cognitive (difficulty concentrating, memory problems, trouble thinking), and 3 emotional (anxiety, depression, irritability). Also sociodemographic factors: Age, sex, education, history of substance abuse and depression Outcomes: Postconcussion syndrome defined by ≥4 PCS | At baseline, 84.2% had ≥4 PCS, and 41.4% reported postconcussion syndrome at 3mo. Postconcussion syndrome was associated with female sex (OR=2.4; 95% CI, 1.10–5.32). Anxiety was associated with postconcussion syndrome in women only (OR=48.66; 95% CI, 7.50–315.8). |
| Gil et al, 27 2005; IsraelPhase II cohort | 120 MTBI patients recruited from 2 surgical wards of a hospital 58% male F/U: 1wk, 1mo, and 6mo | Inclusion criteria: Age 18–50y and fluent in Hebrew Exclusion criteria: Actively receiving psychiatric care, prior history of head trauma, cognitive deficit, substance abuse, and major untreated medical condition | GCS 13–15 at the time of admission. None had LOC at the time of admission. | Prognostic factors: Memory of the injury event within the first 24h of injury. Also measures of depression and anxiety at 1wk using the Beck Depression and Anxiety Inventories, acute posttraumatic symptoms using PTSS and CA-PTSDS, history of psychiatric disorder, age, sex, education, ISS, marital status, country of origin, history of physical illness. Outcome: PTSD diagnosis based on CA-PTSDS and PTSS | By 6mo, 14% had developed PTSD. PTSD was more prevalent in those with a memory of the injury event (23%) than those without memory (6%). PTSD was associated with memory of event (OR= 2.2; 95% CI, 1.0–10.1), acute posttraumatic symptoms (OR=5.3; 95% CI, 1.1–9.3 for CA-PTSDS; and OR=5.2; 95% CI, 1.0–9.4 for PTSS), anxiety (OR=4.9; 95% CI, 1.0–9.1), depression (OR=5.1; 95% CI, 1.0–9.2), and history of psychiatric disorder (OR=3.7; 95% CI, 1.1–8.9). Sex was not associated with PTSD at 6mo. |
| Hou et al, 28 2012; United KingdomPhase II cohort | 126 MTBI patients seen at an ED of a general hospital 63% male F/U: 3 and 6mo | Inclusion: 18–60y of age Exclusion: Those with multitrauma requiring hospitalization and those with major neurologic or psychiatric disorders | Traumatically induced physiological disruption of brain function with at least one of the following: LOC ≤15min, PTA ≤60min, any alteration in mental state at the time of the injury, lack of focal neurologic deficit, and GCS 13–15 | Prognostic factors: BIPQ measures patients' perceptions of their injury, BRIQ measures behavior after onset of illness/injury and the “all or nothing” subscale measures patterns of activity and rest, HADS, IES measures distress after the injury, SSQ measures perceived social support including availability and satisfaction, sex, age, GCS 14 or 15, LOC, PTA, education, marital status, occupation, and litigation status Outcome: PCS defined by ICD-10 criteria of ≥3 of headache; dizziness; fatigue; irritability; insomnia; concentration problems; memory difficulty; or intolerance of stress, emotion or alcohol measured using RPQ | Postconcussion syndrome was present in 22% at 3mo and 21% at 6mo. Fatigue, forgetfulness and sleep disturbance were most commonly reported at 3mo. Headache, fatigue and sleep disturbance were most commonly reported at 6mo. No significant differences between postconcussion syndrome cases and noncases with respect to sex at 3 and 6mo postinjury |
| Stulemeijer et al, 29 2008; The NetherlandsPhase II cohort | 201 MTBI patients admitted to a level I trauma center 62% male F/U: 6mo postinjury | Inclusion: 18–60y of age, able to speak and write in Dutch, no premorbid mental retardation or dementia Exclusion: Questionnaires that were completed >6wk postinjury | European Federation of Neurological Societies' definition of MTBI: History of impact to head with or without LOC ≤30min, with or without PTA, and admission GCS of 13–15 | Prognostic factors: 1. Preinjury: Age, sex, education, emotional problems, physical comorbidities, or prior head injury 2. Peri-injury: GCS, LOC, PTA duration, brain CT abnormality, early symptoms (ie, dizziness, nausea/vomiting, headache), additional extracranial injuries (ie, score ≥2 on AISS) 3. Early postinjury: PCS (RPQ), posttraumatic stress (IES with scores >26 classified as severe), severe fatigue (AFQ with a cutoff value of 20), pain severity score in 5 body regions, self-efficacy (ie, GSES median split) Outcome: RPQ—recovered defined as a score <3 on at least 13 of 16 PCS and RTW at 6mo | Sex did not predict low PCS or full RTW. Absence of comorbid physical problems (OR=3.5; 95% CI, 1.6–7.8), low levels of early PCS (OR=5.5; 95% CI, 2.3–13.2), and low levels of early posttraumatic stress (OR=10.0; 95% CI, 2.3–42.9) predicted low PCS (ie, 90% chance of remaining free of PCS). Discriminative ability was good with AUC=.73. Prediction of full RTW: High level of education (≥11y): OR=6.4 (95% CI, 2.3–18.3); middle level of education (11–14y): OR=4.6 (95% CI, 1.7–12.6); absence of nausea/vomiting on admission: OR=5.1 (95% CI, 1.8–14.3); absence of additional extracranial injuries: OR=3.4 (95% CI, 1.6–7.3); no severe pain early after injury: OR=2.3 (95% CI, 0.9–5.9). |
| Cassidy et al, 24 2014; CanadaPhase II cohort | 1716 MTBI patients treated after a traffic injury 53% female F/U: 6wk, 3, 6, 9, and 12mo | Inclusion: All Saskatchewan residents aged ≥18y who were treated for, or made an insurance claim for, a traffic injury over a 2-year period Exclusion: Those who made a claim, or were treated for a traffic injury more than 42d after the collision, or sustained a serious injury (ie, died, could not answer questionnaire because of injury), or could not understand English. Also excluded were workers' compensation claims and those with LOC >30min. | Answered yes to “Did you hit your head in the collision” and answered “yes” or “don't know” to 1 of the following: LOC, PTA, disorientation or confusion. Also had to answer “yes” to having at least 1 of the following symptoms: dizziness or unsteadiness, memory problems or forgetfulness, and concentration or attention problems. | Prognostic factors: Demographics (age, sex, income, marital status, education), position in vehicle, days in hospital, symptoms (checklist of PCS), fractured bones, LOC, PTA, pain intensity (NRS-11: neck, headache, face, low back, midback, arms, hands, leg, foot, abdomen, chest or groin), prior health, current health, expectations for recovery, depression (CES-D cut point, 16), and number of comorbid health conditions Outcome: Self-reported recovery (ie, “all better or cured” or “very much improved”) | Median time to recovery was 99d (95% CI, 97–102). Sex was not associated with time to recovery of PCS. Factors associated with prolonged recovery were age >50y vs 18–23y (HRR=.76; 95% CI, .63–.91); > high school education vs some high school (HRR=1.24; 95% CI, 1.07–1.44); expectations for recovery—never get better (HRR=.26; 95% CI, .14–.50), —don't know when will get better (HRR=.52; 95% CI, .43–.63), and —will get better slowly (HRR=.79; 95% CI, .67–.94) compared with —get better soon; depression (HRR=.99; 95% CI, .99–1); arm numbness (HRR=.83; 95% CI, .73–.94); hearing problems (HRR=.75; 95% CI, .59–.96); confusion after collision—don't know (HRR=.78; 95% CI, .62–.96) vs none; low back pain intensity (HRR=.97; 95% CI, .95 .98); headache intensity (HRR=.98; 95% CI, .96–1); and midback pain intensity (HRR=.97; 95% CI, .95–.99). |
| Kristman et al, 20 2014; CanadaPhase II cohort | 728 injured workers with MTBI who made a workers' compensation claim 65.8% males F/U: up to 2y | Inclusion: Lost-time workers' compensation claim made over 1-y period in Ontario. Age ≥18y. Exclusion: Skull fracture and more severe brain injuries | Workers' compensation claim codes for part of body injured (brain–1100) and nature of injury (concussion–6200). Positive predictive value for MTBI is 92%. | Prognostic factors: Age, sex, preclaim health utilization, claim duration, specialty of clinician, and diagnostic treatment codes Outcome: Weekly moving averages of health care services per 1000 claimants per day before and after the injury | Health care utilization was constant before the injury at 67.6 services per 1000 claimants per day. During the first 4wk postinjury, it peaked at 319/1000. From month 2 to 1y, it averaged 99.2/1000. Women had higher rates of health care utilization except for the 1-mo postclaim. |
| Covassin et al, 25 2007; United States Phase II cohort | 79 concussed varsity athletes from 5 northeastern universities (full range of varsity sports) AAN concussion grading: Grade 1 (62%): 24 men, 25 women Grade 2 (34%): 14 men, 13 women Grade 3 (4%): 3 men F/U: Preseason (baseline), up to 3d (time 2), 7–10d postconcussion (time 3) | Inclusion: Voluntary participation; varsity athletes who sustained concussions that required serial testing across all 3 periods (baseline/preseason, up to 3d, 7–10d) | AAN criteria | Prognostic factors: Sex, time Outcomes: PCS (Postconcussion Symptom Checklist), neurocognitive testing (ImPACT version 2.0) | No greater likelihood of sustaining a grade 2 or 3 concussion as a function of sex (P=.50) No between-subject multivariate effect of sex (P=.69) and no significant sex-by-time interaction (P=.59) were identified. |
| Preiss-Farzanegan et al, 16 2009; United StatesPhase II cohort | Nested cohort from NIH-funded TBI registry designed to evaluate epidemiology and 3-mo outcomes of concussion. Individuals who presented to a regional trauma center ED with concussion Feb 3, 2003, to Sept 20, 2003, were recruited (n=1438). 260 were admitted with a primary sports-related concussion; 215 eligible for analysis. Adult (>17y): n=78 47 male 31 female Minor (≤17y): n=137 97 male 40 female F/U: 3mo after initial assessment | Inclusion: Subset of NIH registry patients who reported their mechanism of injury involved a sport; did not report intentions to file a lawsuit as a result of their injury | ACRM: A blow to head or acceleration/deceleration movement of head resulting in ≥1 of the following: LOC <30min or PTA <24h or altered mental status at the time of injury; GCS ≥13 measured ≥30min after injury | Prognostic factors: Sex Adjustment factors: Age, source of postconcussion symptom reporting (self, proxy, or interviewer), previous head injury or LOC, sport Outcome: PCS (RPQ) | Compared with males, adult females (≥18y) are at greater risk for elevated RPQ scores (OR=2.57; 95% CI, 1.09–6.08; P=.013) but not female minors (≤17y) (OR=1.07; 95% CI, .52–2.19). Adult females, compared with males, appear to have elevated risk for specific symptoms of headache (OR=4.5; 95% CI, 1.6–12.4), dizziness (OR=2.8; 95% CI, 1.0–7.9), fatigue (OR=2.8; 95% CI, 1.0–7.4), irritability (OR=2.8; 95% CI, 1.0–7.7), and concentration problems (OR=3.0; 95% CI, 1.1–8.4) at 3mo after sports-related concussion. |
| Paniak et al, 22 2002; CanadaPhase II cohort | Presentations to ED. MTBI: n=118 52% female Controls: n=118 Age-, sex-, and education-matched controls (staff and students) 60% female F/U: Within 1mo | Inclusion: See MTBI definition. Exclusion: A history of inpatient treatment for any psychiatric disorder; diagnosis of mental retardation; inability to read fluently in English; history of TBI more severe than MTBI at any time in their life; an MTBI within 1y before involvement in this study; any ongoing central nervous system disorder or concurrent pregnancy | ACRM: GCS 13–15, LOC <30min, PTA <24h. | Prognostic factors: MTBI, sex, age, education, and SES Outcome: Symptom presence and severity (Problem Checklist) | MTBI averaged 18.7 symptoms (severity, 1.7) vs controls (12.6 symptoms; severity, 0.8). No relationship between symptoms and sex, age, or education |
| Paniak et al, 22 2002; CanadaPhase II cohort | Consecutive admissions to 2 EDs MTBI: n=68 Controls: n=118 Age-, sex-, and education-matched controls (staff and students). F/U: 3 and 12mo | Inclusion: See MTBI definition. Exclusion: A history of inpatient treatment for any psychiatric disorder; diagnosis of mental retardation; inability to read fluently in English; history of TBI more severe than MTBI at any time in their life; a second MTBI within 1y before involvement in this study; any ongoing central nervous system disorder or concurrent pregnancy | ACRM: GCS 13–15, LOC <30min, PTA <24h | Prognostic factors: Compensation seeking, SES, ISS (non–brain injuries), treatment satisfaction, age, sex, ethnicity, head injury severity, medication use, and prior psychological treatment Outcomes: Symptom presence and severity (Problem Checklist) | Demographic and injury-related variables did not predict compensation seeking. Compensation seeking/receiving associated with higher symptom report. None of age, years of education, premorbid SES, sex, or ethnic affiliation significantly affected the decision to seek compensation. |
| Paniak et al, 21 2000; CanadaPhase II cohort | Consecutive admissions to 2 ERs. Participants part of randomized controlled trial. MTBI: n=118 62 women 56 men F/U: 3mo | Inclusion: See MTBI definition. | ACRM: GCS 13–15, LOC <30min, PTA <24h | Prognostic factors: Age, sex, injury severity, alcohol use, prior health and life events, medication use, SES, and financial compensation Outcome: Preinjury vocational status | Overall, 74% had returned to full-time preinjury employment by 3mo. Best predictor of RTW was seeking or receiving financial compensation at F/U. No effect of sex as predictor of RTW 3–4mo postinjury |
| Teasdale and Engberg, 30 2001; DenmarkPhase I cohort | Population of Denmark. Data sources were National Bureau of Health registry of all hospitalization discharges in Denmark from 1979 to 1993, and national register of deaths. Total: n=145,440 MTBI: n=126,114 60.2% men | Inclusion: Hospital discharge between 1979 and 1993 with a main or secondary diagnosis of TBI: ICD-8 850 (MTBI), 800, 801, 803 (cranial fracture); 851–854 (cerebral lesion) Exclusion: Children aged <15y by end of study, died in hospital or did not survive at least 1mo past discharge | ICD-8 code 850 | Prognostic factors: Type of TBI, age at injury, sex, days in hospital, codiagnosis of substance misuse Outcome: Death by suicide, as recorded in a national register of deaths up to and including 1993 | .59% of those with MTBI committed suicide. SMRs stratified by age and sex showed that the incidence of suicide in MTBI was 3.02 (95% CI, 2.82–3.25) compared with the general population. Suicide rates were higher for females (SMR=3.89; 95% CI, 3.43–4.42) than males (SMR=2.74; 95% CI, 2.51–2.99). |
Abbreviations: AAN, American Academy of Neurology; ACRM, American Congress of Rehabilitation Medicine; AD, Alzheimer disease; AFQ, Abbreviated Fatigue Questionnaire; AISS, Abbreviated Injury Severity Score; AUC, area under the curve; BIPQ, Brief Illness Perception Questionnaire; BRIQ, Behavioral Responses to Illness Questionnaire; CA-PTSDS, Clinician-Administered Posttraumatic Stress Disorder Scale; CES-D, Center for Epidemiological Studies Depression Scale; CT, computed tomography; F/U, follow-up; GCS, Glasgow Coma Scale; GSES, General Self-Efficacy Scale; HADS, Hospital Anxiety and Depression Scale; HRR, hazard rate ratio; ICD, International Classification of Diseases; IES, Impact of Event Scale; ImPACT, Immediate Postconcussion Assessment and Cognitive Testing; ISS, Injury Severity Score; LOC, loss of consciousness; NIH, National Institutes of Health; NS, not significant; NRS, numeric rating scale; PCS, postconcussion symptoms; PTA, posttraumatic amnesia; PTSS, Posttraumatic Stress Scale; SES, socioeconomic status; SIR, standardized incidence ratio; SMR, standardized mortality ratio; SSQ, Social Support Questionnaire.
Overall, most of the evidence suggests that sex is not a strong prognostic indicator for recovery after MTBI. However, small sex differences were found for some outcomes. After MTBI, females may have a higher risk of epilepsy (children, young adults) and suicide, and use more health care services during the 2 years postinjury, and males may be at increased risk of schizophrenia. After MTBI, sex did not predict recovery from postconcussion symptoms (PCS) (children, adults), risk of dementia, primary brain tumor development, return to work (RTW), or posttraumatic stress disorder (PTSD).
Children
We accepted 2 studies: 1 phase III study from Denmark assessing the risk of epilepsy after MTBI, and 1 phase II study from the United States assessing PCS after MTBI.
Risk of epilepsy after MTBI
In a large phase III Danish population-based cohort study, Christensen et al
15
followed up 1,605,216 people born in Denmark ≥10 years after TBI for a total of 19,527,337 person-years. Compared with no brain injury, the risk of epilepsy was 2 times higher after MTBI (RR=2.22; 95% confidence interval [CI], 2.07–2.38), and remained high ≥10 years postinjury (RR=1.51; 95% CI, 1.24–1.85). The risk was higher among females (RR=2.49; 95% CI, 2.25–2.76; small-medium effect size) than among males (RR=2.01; 95% CI, 1.83–2.22). Interestingly, there was no interaction with sex for patients with skull fractures or severe brain injury.Risk of self-reported PCS after MTBI
Preiss-Farzanegan et al
16
(phase II, United States) assessed 137 minors (aged ≤17y) who presented to a regional trauma center emergency department (ED) and who were not involved in litigation. Three months after a sports-related MTBI, females were not at greater risk for elevated scores on the Rivermead Postconcussion Symptoms Questionnaire (RPQ).Adults
Of the 13 studies, 3 are phase III (The Netherlands,
17
Sweden,18
Denmark19
), 9 are phase II (Canada, 320
, 21
, 22
, 23
, 24
; United States, 316
, 25
, 26
; Israel, 127
; United Kingdom, 128
; The Netherlands, 129
), and 1 is phase I (Denmark30
). The findings are presented according to outcome: (1) PCS,16
, 22
, 23
, 24
, 25
, 26
, 28
, 29
(2) RTW,21
, 29
(3) risk of dementia,17
(4) risk of primary brain tumor development,18
(5) risk of schizophrenia,19
(6) PTSD,27
(7) health care utilization,20
and (8) suicide.30
Self-reported PCS after MTBI
We accepted 7 phase II studies.
16
, 22
, 23
, 24
, 25
, 26
, 28
, 29
Most of the participants were ED admissions; 1 study comprised American collegiate athletes,25
and another was a Canadian population-based study.24
Mean ages ranged from 30 to 38 years. Follow-up periods ranged from 1 week to 12 months, the most common being 3 months. PCS were predominantly assessed with standard instruments: the RPQ,16
, 28
, 29
Problem Checklist,22
, 23
Immediate Postconcussion Assessment and Cognitive Testing,25
and Concussion Symptom Checklist.26
The findings were mixed, but most studies (n=5)
22
, 23
, 24
, 25
, 28
, 29
found that sex did not predict PCS after MTBI. Paniak et al22
, 23
studied 118 consecutive admissions to 2 Canadian EDs. Symptom complaints, assessed with the Problem Checklist at 1 month postinjury, were compared with those of 118 matched control participants without MTBI.22
The authors did not find a relationship between symptoms and sex, age, or education. A subset analysis (n=68) revealed that sex was also not associated with the decision to seek compensation at 3 and 12 months postinjury. In the study by Hou et al,28
107 ED patients with MTBI in the United Kingdom completed the RPQ at 3 and 6 months postinjury. No significant differences were found between postconcussion syndrome cases and noncases with respect to sex at both periods. Stulemeijer et al29
developed a prediction rule for recovery at 6 months postinjury. They assessed 201 MTBI patients admitted to a level 1 trauma center in The Netherlands, and also found that sex did not predict low PCS as assessed by the RPQ. Covassin et al25
similarly concluded that sex does not predict PCS in collegiate athletes, as assessed by the Immediate Postconcussion Assessment and Cognitive Testing within 1 week postinjury. Lastly, in a large Canadian population-based study,24
a final model based on 1476 participants found that sex, along with loss of consciousness, posttraumatic amnesia, and comorbid health conditions, was not associated with time to recovery of PCS after traffic collisions.Two U.S. studies
16
, 26
found that female sex predicted PCS 3 months after MTBI. Dischinger et al26
assessed 180 level I trauma center patients with MTBI and found that females had approximately 2.5 times the risk of postconcussion syndrome than males (odds ratio [OR]=2.40; 95% CI, 1.10–5.32) using the Concussion Symptom Checklist. Further, anxiety was associated with postconcussion syndrome in females only (OR=48.66; 95% CI, 7.50–315.8). Preiss-Farzanegan16
assessed 78 adults who presented to a regional trauma center ED with a sports-related concussion and who were not involved in litigation. Compared with males, adult females (≥18y) were at greater risk of PCS as assessed with the RPQ (OR=2.57; 95% CI, 1.09–6.08; large effect size), but not female minors (≤17y). Additionally, adult females appeared to have an elevated risk for headache (OR=4.5; 95% CI, 1.6–12.4), dizziness (OR=2.8; 95% CI, 1.0–7.9), fatigue (OR=2.8; 95% CI, 1.0–7.4), irritability (OR=2.8; 95% CI, 1.0–7.7), and concentration problems (OR=3.0; 95% CI, 1.1–8.4).16
Return to work after MTBI
Two phase II studies from Canada
21
and The Netherlands29
found that sex did not predict RTW. Paniak et al21
studied 118 adults who had received specialized treatment soon after MTBI and were interviewed 3 to 4 months later to evaluate whether they had returned to full-time preinjury vocational activity. Stulemeijer et al29
did not find sex to predict RTW at 6 months after MTBI among 201 ED patients. Patients were classified as having full RTW when they were not on sick leave at the time of follow-up, or reported no change in working status to partial or lower-level employment because of the MTBI.Risk of dementia after MTBI
One phase III study
17
from The Netherlands investigated the relation between head trauma and the incidence of dementia in 6645 participants of the prospective, population-based Rotterdam Study, aged ≥55 years, who were free of dementia at baseline. Follow-up assessments occurred after an average ± SD of 2.1±0.8 years. The results suggested that MTBI is not a major risk factor for incident dementia, and there were no major differences in the relative risks between men and women.Risk of primary brain tumor development after MTBI
One phase III cohort study
18
investigated patients hospitalized for TBI from 1965 to 1994 by using the Swedish Inpatient Register (n=311,006; 3,225,317 person-years). A total of 281 cases of brain tumors were diagnosed during follow-up at approximately 10.5 years. The results indicated no excess risk of any kind of brain tumor after a head trauma of any severity, and no variation by sex; however, the follow-up period may have been relatively short for this outcome.Risk of schizophrenia after MTBI
A large Danish population-based study
19
of residents without a history of psychiatric illness (phase III case-control) found a modest association between concussion and onset of schizophrenia for males only. After adjusting for premorbid “accident proneness” (defined as history of non–skull fractures), there was an MTBI-related increased risk of schizophrenia for 5 years subsequent to the injury, although the risk was greatest during the first year postinjury (males, OR=2.034; females, OR=1.663; small-medium effect size). Nielsen et al19
concluded that MTBI may contribute to the risk for schizophrenia in males, and while this relation may also exist in females, it is paralleled by an increased liability to traumas in general.Risk of PTSD after MTBI
One phase II study (Israel)
27
investigated 120 patients with MTBI recruited from 2 surgical wards. All participants underwent psychiatric evaluation and self-assessment of their memory of the traumatic event. Symptoms of PTSD were assessed with the Clinician-Administered PTSD Scale and the Posttraumatic Stress Scale. The results indicated that sex was not associated with an increased risk of PTSD at 6 months.Health care utilization after MTBI
One phase II study (Canada)
20
compared the health care use of injured workers during the year before and 2 years after making a workers' compensation claim for MTBI. Kristman et al20
found that filing a claim for MTBI is associated with a long-term increase in the utilization of health care services, especially for females. For example, during the 13- to 24-week period after MTBI, the cumulative rate of health care utilization per 1000 workers per day was 133.7 (95% CI, 120.5–146.8) for females but only 102.4 (95% CI, 95.4–109.3) for males.Risk of suicide after MTBI
One phase I Danish population-based study
30
reported that .59% of those with a documented history of concussion committed suicide. The follow-up periods varied from 1 month to 14 years. The incidence of suicide after concussion was 3 times that of the general population (3.02; 95% CI, 2.82–3.25), with a higher risk for females (3.89; 95% CI, 3.43–4.42) than for males (2.74; 95% CI, 2.51–2.99).Discussion
Only 14 (7%) of 221 accepted primary studies of the WHO and ICoMP systematic reviews were eligible for our analysis. Overall, sex is not a well-studied prognostic indicator for recovery after MTBI, and only small sex differences were found for some outcomes.
For most serious disease-related outcomes, including brain tumor development and dementia, MTBI did not increase the risk for disease, and as a result, sex cannot be prognostic for the outcome. Regarding the objective outcomes where sex may be a prognostic factor (ie, schizophrenia, health care usage, epilepsy, suicide), these must be understood in the context of possible reverse causality, known population trends, absolute risk of disease, and premorbid health status. For example, Neilsen
19
found that males appear to be at an increased risk for schizophrenia after MTBI. However, the highest odds of receiving a schizophrenia diagnosis occurred less than 1 year after MTBI, suggesting an alternative explanation including possible reverse causality or protopathic bias, whereby schizophrenia precedes the MTBI. This same trend of increased odds with diminished time lag occurred for sustaining a fracture and having a subsequent schizophrenia diagnosis. A more recent population-based Danish study31
found no sex interaction between schizophrenia and head injury, irrespective of injury severity.Compared with males, females were at increased risk for epilepsy and suicide after MTBI, and used more health care services after MTBI. However, it is well known that men generally use fewer medical services than women.
32
With respect to an increased risk of epilepsy in female children and young adults, the absolute risk for any increase in epilepsy from MTBI, regardless of sex, remains very low at approximately 1 case per 1000 person-years.15
Furthermore, the higher detection in females may be because they are more likely to be registered in the National Hospital Register because of sex-specific factors such as pregnancy. Finally, the association between MTBI and suicide risk in females may have resulted from concomitant and premorbid characteristics, such as psychosocial disadvantage, rather than the MTBI itself.30
Sex was not shown to be a strong prognostic indicator for self-reported PCS after MTBI in our study; however, a recent Swedish population-based cohort study
7
found sex differences in PCS. Three years after MTBI, postconcussion syndrome was found in 50% of the women and 30% of the men. Moreover, women with MTBI more frequently reported back pain than men. Interestingly, pain (eg, headache, neck and back pain) is the most common PCS and is associated with overall recovery after MTBI.7
, 24
, 33
A wide body of literature suggests that compared with men, women have increased pain sensitivity and an increased risk for clinical pain, and are more likely to report pain.34
Thus, sex differences in PCS might be partially explained by pain. Of note, with the exception of 1 study,24
all studies we reviewed on self-reported outcomes after MTBI did not account for pain as a potential factor affecting prognosis; however, Cassidy et al24
found that having back pain or headache was associated with delayed recovery in those with MTBI, while sex was not. Women may experience pain more than men because of multiple biological (eg, sex hormones) and psychosocial factors (eg, coping).34
Study strengths
This report is based on the best-evidence syntheses conducted by 2 international groups of MTBI experts and methodologists. Both systematic reviews included comprehensive search strategies and in-depth methodological quality assessment of individual studies.
Study limitations
Our report also has limitations. We were only able to analyze a small number of studies (n=14, 7%) because outcome data were rarely classified by sex. Other limitations include those inherent to most systematic reviews such as publication bias and the potential for missing relevant studies. However, given the rigorous methodology used in both the WHO Task Force and ICoMP reviews, we feel that their potential for missing relevant studies was low.
Conclusions
We analyzed the best available evidence of the WHO Task Force and ICoMP systematic reviews, and based our results on 14 studies (7%) that reported sex-stratified outcomes after MTBI. We found that sex is not a consistent or strong prognostic indicator for recovery after MTBI, irrespective of outcome or patient population. However, this should be interpreted with caution given that so few studies reported sex differences.
Implications for future research
Future research should explore possible sex differences in self-reported PCS, especially pain, as this is an area where controversy may still remain. Studies should incorporate more pain-related questions when assessing PCS and include sufficient follow-up (ie, >3mo) to better determine the effect of pain on possible sex-related differences in MTBI recovery. Outcomes for males and females should be analyzed separately given increasing evidence for a sex-specific approach to care.
35
Studies should control for other potential confounders such as premorbid and concomitant conditions, coping mechanisms, external causes of injury, treatment variables, and differences in reporting and health care use between males and females. Finally, replicating initial findings is necessary to demonstrate consistency in other patient populations. A better understanding of any sex-related differences regarding MTBI prognosis has implications for treatment and rehabilitation. Males and females might need to be managed differently given the differences in biological and psychosocial factors.References
- The impact of traumatic brain injuries: a global perspective.NeuroRehabilitation. 2007; 22: 341-353
- Incidence, risk factors and prevention of mild traumatic brain injury: results of the WHO Collaborating Centre Task Force on Mild Traumatic Brain Injury.J Rehabil Med. 2004; : 28-60
- Incidence and lifetime costs of injuries in the United States.Inj Prev. 2006; 12: 212-218
- Better science with sex and gender: a primer for health research.Women's Health Research Network, Vancouver2007
- Sex differences in outcome after mild traumatic brain injury.J Neurotrauma. 2010; 27: 527-539
- Sex differences in depressive symptoms and their correlates after mild-to-moderate traumatic brain injury.J Neurosci Nurs. 2009; 41 (quiz 310-11): 298-309
- Sex differences in symptoms, disability and life satisfaction three years after mild traumatic brain injury: a population-based cohort study.J Rehabil Med. 2013; 45: 749-757
- Are there gender differences in cognitive function, chronic stress, and neurobehavioral symptoms after mild-to-moderate traumatic brain injury?.J Neurosci Nurs. 2012; 44: 124-133
- Sex differences in the long-term neuropsychological outcome of mild traumatic brain injury.Brain Inj. 2009; 23: 809-814
- Prognosis for mild traumatic brain injury: results of the WHO Collaborating Centre Task Force on Mild Traumatic Brain Injury.J Rehabil Med. 2004; : 84-105
- Protocol for a systematic review of prognosis after mild traumatic brain injury: an update of the WHO Collaborating Centre Task Force findings.Syst Rev. 2012; 1: 17
- WHO Collaborating Centre Task Force on Mild Traumatic Brain Injury. Systematic search and review procedures: results of the WHO Collaborating Centre Task Force on Mild Traumatic Brain Injury.J Rehabil Med. 2004; : 11-14
- Systematic search and review procedures: results of the International Collaboration on Mild Traumatic Brain Injury Prognosis.Arch Phys Med Rehabil. 2014; 95: S101-S131
- A systematic review of the prognosis of acute whiplash and a new conceptual framework to synthesize the literature.Spine. 2001; 26: E445-E458
- Long-term risk of epilepsy after traumatic brain injury in children and young adults: a population-based cohort study.Lancet. 2009; 373: 1105-1110
- The relationship between gender and postconcussion symptoms after sport-related mild traumatic brain injury.PM R. 2009; 1: 245-253
- Head trauma and risk of dementia and Alzheimer's disease: the Rotterdam Study.Neurology. 1999; 53: 1959-1962
- Primary brain tumors following traumatic brain injury—a population-based cohort study in Sweden.Cancer Causes Control. 2001; 12: 733-737
- Is head injury a risk factor for schizophrenia?.Schizophr Res. 2002; 55: 93-98
- Health care utilization of Workers' Compensation claimants associated with mild traumatic brain injury: a historical population-based cohort study of workers injured in 1997-1998.Arch Phys Med Rehabil. 2014; 95: S295-302
- Prediction of vocational status three to four months after treated mild traumatic brain injury.J Musculoskelet Pain. 2000; 8: 193-200
- Patient complaints within 1 month of mild traumatic brain injury: a controlled study.Arch Clin Neuropsychol. 2002; 17: 319-334
- A longitudinal study of the relationship between financial compensation and symptoms after treated mild traumatic brain injury.J Clin Exp Neuropsychol. 2002; 24: 187-193
- Population-based, inception cohort study of the incidence, course, and prognosis of mild traumatic brain injury after motor vehicle collisions.Arch Phys Med Rehabil. 2014; 95: S278-S285
- Sex differences in neuropsychological function and post-concussion symptoms of concussed collegiate athletes.Neurosurgery. 2007; 61 (discussion 350-1): 345-350
- Early predictors of postconcussive syndrome in a population of trauma patients with mild traumatic brain injury.J Trauma. 2009; 66 (discussion 296-7): 289-296
- Does memory of a traumatic event increase the risk for posttraumatic stress disorder in patients with traumatic brain injury? A prospective study.Am J Psychiatry. 2005; 162: 963-969
- When a minor head injury results in enduring symptoms: a prospective investigation of risk factors for postconcussional syndrome after mild traumatic brain injury.J Neurol Neurosurg Psychiatry. 2012; 83: 217-223
- Early prediction of favourable recovery 6 months after mild traumatic brain injury.J Neurol Neurosurg Psychiatry. 2008; 79: 936-942
- Suicide after traumatic brain injury: a population study.J Neurol Neurosurg Psychiatry. 2001; 71: 436-440
- Head injury as risk factor for psychiatric disorders: a nationwide register-based follow-up study of 113,906 persons with head injury.Am J Psychiatry. 2014; 171: 463-469
- Are men shortchanged on health? Perspective on health care utilization and health risk behavior in men and women in the United States.Int J Clin Pract. 2010; 64: 475-487
- Mild traumatic brain injury after motor vehicle collisions: what are the symptoms and who treats them? A population-based 1-year inception cohort study.Arch Phys Med Rehabil. 2014; 93: S286-S294
- Sex differences in pain: a brief review of clinical and experimental findings.Br J Anaesth. 2013; 111: 52-58
- Sex, gender, and health: the need for a new approach.BMJ. 2001; 323: 1061-1063
Article Info
Publication History
Published online: February 06, 2015
Footnotes
Publication of this article was supported by the American Congress of Rehabilitation Medicine.
Disclosures: none.
Identification
Copyright
© 2016 American Congress of Rehabilitation Medicine. Published by Elsevier Inc. All rights reserved.
