Outcomes after Concussion Recovery Education: Effects of Litigation and Disability Status on Maintenance of Symptoms

Abstract

This study examined the hypothesis that people who receive concussion recovery education would have better outcomes than those who received usual discharge paperwork from the emergency department (ED) and tested whether participants who were in litigation or seeking disability compensation had more symptoms than individuals not engaged in these activities. Two hundred and fifty-five persons with a diagnosis of concussion were assigned randomly to a brief education group (one-page double-sided document), a longer education group (10-page document), and usual care (standard ED discharge instructions), and were these documents in the ED. A (non-concussion) trauma comparison group was enrolled to determine the symptom rate unrelated to brain injury. The Concussion Symptom Checklist (CSC) and litigation and disability status questions were completed by telephone at one week, three months, and six months. Neither long nor brief information handouts had a significant impact on symptoms over time; the standard form had an average decrease of 1.20 symptoms compared with the brief instructional intervention group (p = 0.031). Litigation status and disability seeking status were significant predictors of symptoms on CSC over time: disability seeking (p = 0.017) and litigation status (p = 0.05). Persons seeking Social Security disability or legal compensation endorsed more symptoms over time than those who were not. Number of symptoms on the CSC for the trauma control group was the same as those who sustained concussion. Type of recovery material was not as important as noting that concussion symptoms resolve over time, and that remaining symptoms are not specific to brain injury. Litigation and disability seeking behavior accounted for maintained symptoms, rather than the concussion itself.

Introduction

Symptoms associated with mild traumatic brain injury (TBI) or concussion include headache, fatigue, malaise, concentration and memory difficulties, sleep disturbance, and irritability. Cognitive deficits after TBI follow a dose-response relationship in which increased severity of TBI is associated with increased cognitive impairment and recovery time. Unlike moderate and severe TBI, which by definition show significant neurologic, cognitive, and emotional impairments for an extended period, after concussion, most symptoms resolve spontaneously one week to three months after injury.^1,2 Education about concussion promotes recovery; in contrast, negative expectancy and misattribution of symptoms to the concussion are associated with persistent problems and poor outcome.^3

–6

Psychosocial factors, psychiatric history, comorbid physical conditions, and litigation or compensation seeking likely account for most lingering post-concussion symptoms after the usual recovery timeline.^2,5

–8 Symptoms that may have been neurologically based in the acute stage because of temporary biochemical and metabolic dysfunction⁹ or structural changes^10,11 may be maintained by psychological factors, such as a maladaptive coping style and cognitive appraisal of being a “brain injury victim.”^4,12 Symptom rates are higher for patients with concussion who are not given an explanation for their symptoms.^5,13

Treatment for persisting symptoms after concussion typically involves components of education, reassurance, and reattribution of symptoms to benign causes. Rates of symptoms among patients who receive treatment involving these components are lower than controls who receive standard discharge information or who received no treatment at all.^3
–5,14,15 A notable drawback to these interventions is that they have been time intensive in terms of contact hours between patients and clinician.

A concussion educational document developed by Mittenberg and associates¹⁶ has been used with some success in emergency departments (EDs); however, the original educational materials (10 pages) and time to review them with patients are lengthy. A brief, one-page version of the document was developed and piloted in an urban Level I Trauma Center.¹⁷ The present study reports a large-scale trial of this intervention. The central hypothesis was that persons who experience the educational intervention would have better outcomes post-concussion at one week, three months, and six months post-injury than participants who received usual discharge paperwork from the ED. It was also predicted that participants who were engaged in litigation or seeking disability compensation would have higher rates of symptom reporting than individuals who were not engaged in these activities.

Methods

Study design and setting

This was a single-center, randomized controlled design conducted in the ED of an urban Level I trauma center. The study was approved by the university's Institutional Review Board. Authors were blinded to randomization throughout the study.

Study protocol

Patients who consecutively presented to the ED with concussion were assigned to one of three groups: (1) Education Intervention–Brief (Brief Concussion Recovery Guide); (2) Education Intervention–Long (Original Concussion Recovery Guide); and (3) Usual Care (standard ED discharge instructions). The minimization method¹⁸ was used for random assignment to limit imbalance between the number of subjects in each treatment group over a number of factors. A (non-concussion) trauma comparison group from the ED (e.g., orthopedic, lacerations, strain/sprain) was also enrolled to determine the rate of symptoms in patients with similar medical experience who did not experience brain injury.

After informed consent, an ED research team member reviewed the resource guide (orally and with the written document)^19,20 with the participants who were randomized to one of the intervention groups. Baseline data for demographics and employment status were gathered before the concussion intervention. Participants who received the recovery guides were also administered a Patient Review Questionnaire, a 10-item scale developed to determine understanding and retention of the concussion educational material. Any wrong answers were corrected and discussed before the patients left the ED.

Follow-up evaluations were completed by telephone at one week, three months, and six months after discharge. The primary outcome was the Concussion Symptom Checklist (12 items),¹³ which quantifies severity, frequency, and number of symptoms reported. Two additional questions were asked of the concussion participants at each follow-up assessment: (1) Are you involved in a lawsuit as related to your injury at this time? and (2) Are you seeking disability insurance as related to your injury at this time? At the one-week follow-up, all four groups completed the protocol; the trauma comparison group was not contacted after the first follow-up. Participants were paid $25 for each assessment.

Selection of participants

Individuals who met inclusion criteria for the concussion groups were adults (≥18 years of age) diagnosed with concussion as evidenced by Glasgow Coma Scale 13–15, history of head trauma with alteration or loss of consciousness of up to 30 min, and/or post-traumatic confusion of up to 24 h.²¹ In addition, all eligible participants were discharged from the ED to home. The trauma comparison group comprised adults admitted to the ED for non-concussion related issues also discharged to home the same day. Exclusion criteria for all participants included non-English-speaking, comorbid injuries that required hospitalization, and pre-morbid cognitive deficits that would preclude participation (e.g., dementia, psychosis).

Analyses

Data were analyzed with Stata 15 using a linear mixed modeling approach to examine change in the total number of reported concussion symptoms on the Concussion Symptom Checklist over time, at one week, three months, and six months post-injury, as well as predictors (covariates) of this change. This trajectory is commonly expressed as a linear function of time containing two unknown individual latent growth factors: an intercept and slope. The individual intercept parameter represented the mean number of concussion symptoms reported at one week post-injury. The individual slope parameter represented the rate of change in the number of concussion symptoms reported over time, up to six months post-injury.

After this unconditional Level 1 model was specified to represent the individual change over time, a Level 2 (conditional) model was specified, in which predictor variables hypothesized to affect the individual growth parameters were entered into the model. The eight Level 2 predictors included: type of information on concussion received in the ED (long, brief, or standard information handouts), age at time of injury, race (Black or White), sex, years of education, employment status (employed vs. not), litigation status, and disability compensation status. This random coefficient model was developed with the prediction that the slope representing the rate of linear change, π _1i, also varied across participants, in addition to the intercept term.

Results

Characteristics of participants

Two hundred and fifty-five adults with concussion were enrolled: 85 (33.3%) Brief Education Intervention, 84 (32.9%) Long Education Intervention, and 86 (33.7%) Usual Care. Table 1 presents descriptive statistics for demographic and predictor characteristics at baseline for the concussion groups. The trauma comparison group included 63 adults. The three concussion groups were equivalent in age (F _2,252 = 1.54, p = 0.217) and years of education (F _2,247 = 0.12, p = 0.884) at injury, as well as proportions for sex, χ ² ₍₂₎ = 0.53, p = 0.767, and race χ ² ₍₂₎ = 2.60, p = 0.272.

Table 1.

Sample Demographic and Predictor Variable Characteristics

	Min	Max	\documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document}$$\bar x$$ \end{document}	SD
Age at injury	18	87	38.80	16.09
Years of education	6	18	12.50	1.76

	n	%
ED information received
Brief	85	33.3
Long	84	32.9
Standard	86	33.7
Sex
Female	119	46.7
Male	136	53.3
Race
Black	222	87.1
White	33	12.9
Employed at baseline
Yes	136	53.3
No	119	46.7

ED, emergency department.

Main analyses

Unconditional analyses

Similar rates of missing data on the Concussion Symptom Checklist were found at each follow-up assessment: one week = 38.4%; three months = 43.1%; and six months = 42.4% post-injury. An advantage of the linear mixed model is the ability to handle missing data on the response variable without dropping the cases from analysis. The simplest model—the unconditional means model—describes and partitions the variation in the Concussion Symptom Checklist. This model contains a random intercept, but no explanatory covariates. Estimation of this model indicated that γ₀₀, the mean intercept (i.e., grand mean), was 5.55 (i.e., an average of about six concussion symptoms at one week post-injury), which is statistically significant from zero (p < 0.0001). In addition, the random intercept contained significant variability, as indicated by the likelihood ratio test (p < 0.001). The intraclass correlation coefficient was 0.644, representing the proportion of variance in the Concussion Symptom Checklist at one week post-injury between individuals. In other words, 64.4% of variation in symptoms was between individuals, whereas 34.8% was “within-person” variation.

Given this pattern of findings, there was support to fit a more complex model: the unconditional growth model, which contained both a random intercept and random slope. The likelihood ratio test provided evidence for preferring this more complex model over the simpler random intercept-only model, χ²(3) = 20.81, p = 0.0001; however, there was much less variation in Concussion Symptom Checklist slope (trajectory) compared with intercept between persons. In other words, there was greater variability between persons in their baseline Concussion Symptom Checklist (intercept) than in their change over time in the Concussion Symptom Checklist (slope or trajectory). Participants' symptom endorsement on the Checklist tended to decline slightly over time, with an average decrease of 0.005 symptoms per day, which represented a statistically significant decline over time (p = 0.006).

Conditional analyses

We then conducted conditional analyses to determine which of the eight covariates could explain the variation in symptom scores found in the unconditional linear mixed models. As shown in Table 2, litigation status and disability seeking status were significant predictors of symptom report on the Checklist over time: disability seeking (p = 0.017) and litigation status (p = 0.05). Participants seeking Social Security disability or legal compensation in relation to their injuries endorsed more symptoms over time than those who were not. On average, participants seeking disability endorsed an increase of 1.078 symptoms, whereas those seeking legal compensation endorsed an average increase of 0.803 symptoms. This is in contrast to the overall trend among participants as a whole, which indicated an average decrease of 0.004 symptoms per time point since injury (p = 0.018).

Table 2.

Random Intercept and Random Slope Model with the Covariates

Fixed effects	Coefficient	SE	z	p
Days	-0.004	0.002	-2.37	0.018
Group
Brief	base
Long	−0.494	0.561	−0.88	0.378
Standard	−1.202	0.558	−2.16	0.031
Age	0.002	0.015	0.10	0.919
Ethnicity	0.791	0.709	1.12	0.264
Sex	−0.243	0.472	−0.51	0.607
Education	−0.021	0.136	−0.15	0.880
Employment	−0.322	0.377	−0.85	0.393
Litigation	0.803	0.418	1.92	0.055
Disability-seeking	1.078	0.453	2.38	0.017
Constant	5.423	2.652	2.04	0.041

Random effects	Estimate	SE	[95% CI]
Variance (slope)	0.0001	0.00006	0.00006, 0.0003
Variance (intercept)	7.434	1.324	5.244, 10.54
Covariance	-0.002	0.007	-0.017, 0.011
Variance (residual)	3.482	0.479	2.659, 4.561

SE, standard error; CI, confidence interval.

Neither the long nor the brief information handouts had a significant impact on symptom endorsement over time, whereas the standard form (e.g., usual discharge paperwork) was associated with an average decrease of 1.20 symptoms compared with the brief instructional intervention group (p = 0.031). Age, race, sex, education, and employment status were not statistically significant predictors of Concussion Symptom Checklist endorsement in this model. Taken together, this pattern of findings suggests that there are additional predictors needed to explain adequately the variability in symptom endorsement after concussion.

Results for the non-concussion trauma group were compared with the three concussion groups at one week after discharge. Interestingly, the number of endorsed symptoms on the Concussion Symptom Checklist was essentially the same as that of the participants who sustained concussion. For the comparison group, the average number of symptoms was 5.18 (standard deviation [SD] = 3.54), and for participants with concussion diagnoses, the average number of symptoms endorsed on the Concussion Symptom Checklist was 6.03 (SD = 3.38). Thus, even persons with concussions showed a rapid rate of symptom recovery within one week. Overall, both the trauma comparison group and concussion groups showed commensurate levels of total number of symptoms at approximately one week post-injury. Over time, the concussion participants continued to report a similar number of symptoms, because the mean score at three months was 5.44 (SD = 3.76) and at six months was 5.18 (SD = 3.85) for this group.

Discussion

This study supports previous research demonstrating that over time, nearly all individuals who experience a concussion have good outcome, and the type of education about symptom recovery in the ED does not seem to make a significant difference in symptom reporting. Empirical studies have consistently shown that after concussion, most symptoms resolve spontaneously one week to three months post-injury;^1,2 consistent with this finding, in this study, symptom reporting rates mirrored ED control participants who had not sustained concussion. Among participants with a concussion, at one week post-injury, individuals endorsed approximately six symptoms. The control group reported a similar number (five symptoms on average). This similarity and the continued endorsement of symptoms by participants in the treatment condition (average of about five symptoms at three months) may be related to the generally broad and not necessarily specific nature of post-concussive symptoms.

These symptoms have been found in multiple populations, including outpatients seen for psychological treatment,²² college students and their relatives,^23,24 and chronic pain patients.^25,26 Rates of concussion symptoms have been found to be similar to physical trauma controls²⁷(as in this study), veterans with exposure to trauma-inducing situations,²⁸ and healthy community volunteers.²⁹

Individuals who were involved in litigation or were seeking disability related to their concussion reported a higher number of symptoms than the individuals who were not engaged in these activities. This finding is consistent with previous literature demonstrating that litigation has been found to be a factor in prolonged symptom recovery.³⁰ Research has also shown that individuals who have had a concussion and are involved in litigation report a higher number of symptoms compared with those who have had a concussion and are not involved in litigation, even when severity of injury was considered.³¹ Similarly, among groups of individuals with concussion and with moderate-severe TBI, litigation has been found to be predictive of a greater number of symptoms.³²

Although the specific type of educational material did not differentially affect outcome among these participants, this finding does not undermine the longstanding and robust literature about the benefits of psychoeducation in the healthcare setting. Instead, given the very low rate of lasting symptoms in any of the groups, these findings may suggest that educating patients with concussion about the natural recovery process can occur successfully regardless of the format in which the education is presented. As per the literature about negative expectancies and health outcomes, psychoeducation can help patients with concussion develop the expectation that they should not have longstanding difficulties as a result of a mild brain injury as well as manage anxieties about the recovery process.

There are several potential reasons that the three educational formats did not differ in effectiveness. In retrospect, it is possible that standard instructions delivered by our ED and ED staff are richer and more detailed than most, given that we host a Level I trauma center and model system of care that specialize in TBI. Hence, the expert healthcare staff members are well versed in the different sequelae and trajectories of concussion versus moderate or worse injuries, and they are keenly aware of medical and public issues in these regards. Between the original Mittenberg Concussion Recovery Guide (10-page document), the brief one-page version, and the usual four-page discharge paperwork from the ED, the length of the education document and amount of detail covered did not appear to make a difference in symptom recovery.

The present findings are not inconsistent with good outcomes observed in previous studies that used long educational guides.^3,5 This study also included a lengthy and detailed guide in one of the three conditions; however, all three conditions did extremely well, regardless of the length of the educational guide. In this regard, a null finding for length of the educational guide may tacitly support using the shortest version, given the value of time in the medical environment.

Although timing of psychoeducation has been shown to affect effectiveness of outcomes, it is unlikely that the timing of the interventions explains the pattern of findings. The timing of the present design was similar to those of previous studies, because typically these education guides are given immediately after the injury—in the ED or within a week of the injury.

Ultimately, brain injury providers must be aware that factors other than the concussion itself may account for the maintenance of symptoms after the acute recovery period, and information about recovery sets up the expectation that symptoms will resolve on their own rather quickly. If these symptoms persist, they may be related to the effects of litigation and disability seeking behavior, rather than the brain injury itself.

Footnotes

Acknowledgments

Additional thank you to Robert Kotasek, Daniela Ristova-Trendov, and Syed Ayaz for their help in data collection.

This project was funded by the National Institute of Disability, Independent Living, and Rehabilitation Research, through Health and Human Services, grant # 90IF0048.

Author Disclosure Statement

No competing financial interests exist.

References

McCrea

, Guskiewicz

K.M.

, Marshall

S.W.

, Barr

, Randolph

, Cantu

, Onate

, Yang

, and Kelly

(2003) Acute effects and recovery time following concussion in collegiate football players: the NCAA Concussion Study. JAMA, 290, 2556–2563.

Carroll

L.J.

, Cassidy

J.D.

, Peloso

P.M.

, Borg

, von Holst

, Holm

, Paniak

, and Pépin

(2004) Prognosis for mild traumatic brain injury: results of the WHO Collaborating Centre Task Force on Mild Traumatic Brain Injury. J. Rehabil. Med.1, 43, Suppl, 84–105.

Mittenberg

, Canyock

E.M.

, Condit

, and Patton

(2001) Treatment of post-concussion syndrome following mild head injury. J. Clin. Exp. Neuropsychol., 23, 829–836.

Mittenberg

, Tremont

, Zielinski

R.E.

, Fichera

, and Rayls

K.R.

(1996) Cognitive-behavioral prevention of postconcussion syndrome. Arch. Clin. Neuropsychol.. 11, 139–145.

Ponsford

, Willmott

, Rothwell

, Cameron

, Kelly

, Nelms

, and Curran

(2002) Impact of early intervention on outcome following mild head injury in adults. J. Neurol. Neurosurg. Psychiatry, 73, 330–332.

Iverson

G.L.

, Lange

R.T.

, Brooks

B..L.

, and Rennison

V.L.

(2010) “Good old days” bias following mild traumatic brain injury. Clin. Neuropsychol., 24, 17–37.

Binder

L.M.

, and Rohling

M.L.

(1996). Money matters: a meta-analytic review of the effects of financial incentives on recovery after closed-head injury. Am. J. Psychiatry, 153, 7–10.

Rohling

M.L.

, Binder

L.M.

, and Langhinrichsen-Rohling

(1995) Money matters: a meta-analytic review of the association between financial compensation and the experience and treatment of chronic pain. Health Psychol., 14, 537–547.

Iverson

G.L.

, Lange

R.T.

, Gaetz

, and Zasler

N.D.

(2006) Mild TBI, in: N.D. Zasler, D., Katz, D., and R. Zafonte R (eds). Brain Injury Medicine: Principles and Practice. Demos Medical Publishing: New York, 333–371.

10.

Brolinson

P.G.

, Manoogian

, McNeely

, Goforth

, Greenwald

, and Duma

(2006) Analysis of linear head accelerations from collegiate football impacts. Curr. Sports Med. Rep., 5, 23–28.

11.

Guskiewicz

K.M.

, Mihalik

J.P.

, Shankar

, Marshall

S.W.

, Crowell

D.H.

, Oliaro

S.M.

, Ciocca

M.F.

, and Hooker

D.N.

(2007) Measurement of head impacts in collegiate football players: Relationship between head impact biomechanics and acute clinical outcome after concussion. Neurosurgery, 61, 1244–1252.

12.

Binder

L.M.

(1997) A review of mild head trauma: Part II: Clinical implications. J. Clin. Exp. Neuropsychol., 19, 432–457.

13.

Miller

L.J.

, and Mittenberg

(1998) Brief cognitive behavioral interventions in mild traumatic brain injury. Appl. Neuropsychol., 5, 172–183.

14.

Minderhoud

J.M.

, Huizenga

, and Saan

R.J.

(1980) Treatment of minor head injuries. Clin. Neurol. Neurosurg., 82, 127–140.

15.

Gronwall

(1986) Rehabilitation programs for patients with mild head injury: components, problems, and evaluation. J. Head Trauma Rehabil., 1, 53–63.

16.

Mittenberg

, Zielinski

R.E.

, and Fichera

(1993) Recovery from mild head injury: a treatment manual for patients. Psychotherapy in Private Practice., 12, 37–52.

17.

Scott

, Hanks

, Welch

, and O'Neil

(2010) Feasibility of a brief educational intervention regarding recovery from concussion. Arch. Phys. Med. Rehabil., 91, e33.

18.

Pocock

S.J.

, and Simon

(1975) Sequential treatment assignment with balancing for prognostic factors in the controlled clinical trial. Biometrics, 31, 103–115.

19.

Saunders

C.E.

, Cota

, and Barton

C.A.

(1986) Reliability of home observation for victims of mild closed-head injury. Ann. Emerg. Med., 15, 160–163.

20.

Levitt

M.A.

, Sutton

, Goldman

, Mikhail

, and Christopher

(1994) Cognitive dysfunction in patients suffering minor head trauma. Am. J. Emerg. Med., 12, 172–175.

21.

McCrea

, American Academy of Clinical

Neuropsychology.

(2008) Mild Traumatic Brain Injury and Postconcussion Syndrome: The New Evidence Base for Diagnosis and Treatment. Oxford University Press: New York.

22.

Fox

D.D.

, Lees-Haley

P.R.

, Earnest

, and Dolezal-Wood

(1995) Post-concussive symptoms: base rates and etiology in psychiatric patients. Clin. Neuropsychol., 9, 89–92.

23.

Machulda

M.M.

, Bergquist

T.F.

, Ito

, and Chew

(1998) Relationship between stress, coping, and postconcussion symptoms in a healthy adult population. Arch. Clin. Neuropsychol., 13, 415–424.

24.

Sawchyn

J.M.

, Brulot

M.M.

, and Strauss

(2000) Note on the use of the Postconcussion Syndrome Checklist. Arch. Clin. Neuropsychol., 15, 1–8.

25.

Gasquoine

P.G.

(2000) Postconcussional symptoms in chronic back pain. Appl. Neuropsychol., 7, 83–89.

26.

Iverson

G.L.

, and McCracken

L.M.

(1997) ‘Postconcussive’ symptoms in persons with chronic pain. Brain Inj., 11, 783–790.

27.

Meares

, Shores

E.A.

, Taylor

A.J.

, Lammel

, and Batchelor

(2011) Validation of the Abbreviated Westmead Post-traumatic Amnesia Scale: a brief measure to identify acute cognitive impairment in mild traumatic brain injury. Brain Inj., 25, 1198–1205.

28.

Fear

N.T.

, Jones

, Groom

, Greenberg

, Hull

, Hodgetts

T.J.

, and Wessely

(2009) Symptoms of post-concussional syndrome are non-specifically related to mild traumatic brain injury in UK Armed Forces personnel on return from deployment in Iraq: an analysis of self-reported data. Psychol. Med., 39, 1379–1387.

29.

Iverson

G.L.

, and Lange

,. (2003) Examination of “postconcussion-like” symptoms in a healthy sample. Appl. Neuropsychol., 10, 137–144.

30.

Fee

C.R.

, and Rutherford

W.H.

(1988) A study of the effect of legal settlement on post-concussion symptoms. Arch. Emerg. Med., 5, 12–17.

31.

Lange

R.T.

, Iverson

G.L.

, and Rose

(2010) Post-concussion symptom reporting and the “good-old-days” bias following mild traumatic brain injury. Arch. Clin. Neuropsychol., 25, 442–450.

32.

Tsanadis

, Montoya

, Hanks

R.A.

, Millis

S.R.

, Fichtenberg

N.L.

, and Axelrod

B.N.

(2008) Brain injury severity, litigation status, and self-report of postconcussive symptoms. Clin. Neuropsychol., 22, 1080–1092.