Original Contribution

Health status, not head injury, predicts concussion symptoms after minor injury^B

Samuel A. McLean MD, MPH^a,?, Ned L. Kirsch PhD^c, Cheribeth U. Tan-Schriner PhD^g, Ananda Sen PhD^e, Shirley Frederiksen RN, MS^b, Richard E. Harris PhD^d,

William Maixner DDS, PhD^h, Ronald F. Maio DO, MS^b,f

^aEmergency Medicine and the TRYUMPH Research Program, University of North Carolina, Chapel Hill, NC 27599, USA

^bEmergency Medicine, University of Michigan Medical Center, Ann Arbor, MI 48109, USA

^cPhysical Medicine and Rehabilitation, University of Michigan Medical Center, Ann Arbor, MI 48109, USA

^dInternal Medicine, University of Michigan Medical Center, Ann Arbor, MI 48109, USA

^eCenter for Statistical Consulting and Research, University of Michigan Medical Center, Ann Arbor, MI 48109, USA

^fInjury Research Center, University of Michigan Medical Center, Ann Arbor, MI 48109, USA

^gMichigan Public Health Institute, Okemos, MI 48864, USA

^hCenter for Neurosensory Disorders, University of North Carolina, Chapel Hill, NC 27599, USA

Received 18 January 2008; accepted 28 January 2008

Abstract

Objective: Postconcussion (PC) syndrome etiology remains poorly understood. We sought to examine predictors of persistent PC symptoms after minor injury.

Methods: Health status, symptom, and injury information were obtained on a sample of patients presenting to the emergency department after minor injury. Postconcussion and cognitive symptoms were assessed at 1, 3, and 12 months.

Results: Among 507 patients enrolled, 339 had head injury. Repeated-measures logistic regression modeling of PC and cognitive symptom presence across time indicated that baseline mental health status and physical health status were most predictive of Persistent symptoms. In contrast, head injury presence did not predict persistent PC syndrome.

Discussion: Baseline mental health status and physical health status were associated with persistent PC syndrome after minor injury, but head injury status was not. Further studies of PC syndrome pathogenesis are needed.

(C) 2009

^? This study was funded by the Centers for Disease Control (R49/ CCR523223-01). Doctor McLean is supported by NIH K23 RR017607-01.

* Corresponding author. Department of Emergency Medicine and

Center for Neurosensory Disorders, University of North Carolina, Chapel Hill, NC 27599-7455, USA. Tel.: +1 919 843 5931; fax: +1 919 966 3683.

E-mail address: [email protected] (S.A. McLean).

Introduction

Background

Patients commonly present to the emergency department (ED) for evaluation after Minor head injury. Persistent

0735-6757/$ – see front matter (C) 2009 doi:10.1016/j.ajem.2008.01.054

symptoms are commonly reported after such injuries, including headache, dizziness, nausea, and cognitive impair- ment [1,2]. Historically, the pathogenesis of such persistent symptoms has been attributed to mechanical brain trauma [3], giving rise to the term minor traumatic brain injury (MTBI). Current Centers for Disease Control (CDC) criteria

[4] define MTBI as head injury accompanied by either loss of consciousness (LOC) for 30 minutes or less, amnesia, or 2 or more “postconcussion” (PC) symptoms. Similarly, persistent symptoms have been termed the Postconcussion syndrome.

Despite the widespread use of these terms, increasing evidence is seemingly at odds with a mechanical injury hypothesis [5-9]. For example, a recent large prospective study from Lithuania found that PC symptoms were no more common after head injury with LOC than after other forms of minor injury occurring without head injury [6]. In addition, other studies have found presumed markers of trauma severity (such as amnesia and LOC) to be inferior predictors of symptom outcomes compared with other factors (eg, psychosocial factors [6,8,9]).

Interestingly, recent epidemiologic studies indicate that symptoms included in the definition of PC syndrome are common in nontrauma populations [10-12] and strongly associated with mental and physical health status [10,13,14]. Such symptoms exhibit marked interindividual variability [15-19] and seem to be modulated by physiologic systems involved in neurosensory processing [15-19]. These neuro- sensory processing systems are closely integrated with cardiovascular systems, as exemplified by the identified association between decreased arterial blood pressure and symptom vulnerability [20-25].

In this study, we examined injury characteristics, health status, MTBI status, and persistent PC symptoms among a cohort of patients presenting to the ED with minor injury. Consistent with the above data, we hypothesized that baseline health status and initial PC symptoms would be most predictive of persistent symptoms, whereas the presence or absence of head injury would have relatively less influence.

Methods

Study design and setting

This ED-based prospective cohort study was conducted at St. Joseph Mercy Hospital (Ann Arbor, MI), a level II trauma center with an annual adult ED census of approxi- mately 65 000.

Selection of participants

Patients 18 years or older who presented to the ED within 24 hours of a minor injury were recruited for the study.

Patients who were non-English speaking, those who had a Glasgow Coma Scale score of less than 13, prisoners, those transferred from another hospital, patients admitted to the hospital, patients who were clinically unstable, those with an LOC of 30 minutes or more, and those not competent to give consent were excluded.

Data collection and processing

• • 1. Study instruments and definition of MTBI

Participants completed the following self-report instru- ments within a structured clinical interview format adminis- tered by trained research assistants (RAs). Each instrument was chosen based on its psychometric properties and applicability to the minor head injury population.

• • • 1. Rivermead Post-Concussion Symptoms Question- naire [26]. The Rivermead Post-Concussion Symptoms Questionnaire (RPQ) was used to assess symptoms after injury. The RPQ is a 16-item questionnaire that assesses the presence and severity of the following PC symptoms: headaches, dizziness, nausea or vomiting, noise sensitivity, sleep disturbance, fatigue, irritability, depression, frustration, forgetfulness, poor concentration, slow thinking, Blurred vision, light sensitivity, Double Vision, and restlessness. Consistent with World Health Organization criteria [27], those with 3 or more PC symptoms rated at least “mild” on the RPQ were classified as having PC syndrome.
      2. Sickness Impact Profile Alertness Behavior subscale [28]. The sickness Impact Profile Alertness Behavior subscale (SIPab) was used to evaluate postinjury cognitive symptoms, including confusion, disorientation, forgetfulness, slow reactivity, difficulty completing tasks, reason and problem-solving difficulties, and problems with concentration and thinking. The SIPab has been shown to correlate with traumatic brain injury-specific measures, such as time to follow commands and GCS score [29].
      3. Short-Form 36 health survey [30]. The Short- Form 36 (SF-36) assesses 8 health concepts. Two distinct, higher-order summary scores may be calculated from these health concepts: a mental component summary (MCS) and a physical component summary (PCS). These summaries were used to assess baseline mental and physical health.
      4. MTBI definition. Presence of MTBI was defined according to criteria recommended by the CDC [4]. Patients with a GCS score of 13 or greater upon ED arrival who experienced head injury and had at least 1 of the following were classified as MTBI: (1) LOC for 30 minutes or more due to the trauma, (2) posttraumatic amnesia (PTA), or (3) 2 or more PC symptoms (symptoms rated at least “mild” on the RPQ). Because of the potential for brain trauma without cranial trauma (eg rotational injury), patients without head injury but with LOC or PTA due to trauma were also classified as having MTBI, if it was determined that there was no other demonstrable cause for the LOC (eg, alcohol intoxication).

Baseline data

Baseline information was collected from study partici- pants via personal interviews administered in the ED. Patients were recruited 8 hours a day, 7 days a week over a 15-month time period (April 2004 through June 2005). Seventy percent of the shifts were from 3 PM to 11 PM, 20% were from 7 AM to 3 PM, and 10% were from 11 PM to 7 AM. Potential study participants were identified via the ED electronic triage log and ED staff. Patients meeting inclusion/ exclusion criteria were approached for consent and provided with study information. A Mini-Mental Status Examination

[31] and Galveston Orientation and Amnesia Test [32] were administered to consenting patients. Only patients who were able to explain the essential elements of the study and had a Mini-Mental Status Examination score of greater than 18 and a Galveston Orientation and Amnesia Test score of 76 or greater (indicating no current PTA) were asked to participate. Because of the high prevalence of patients with minor injury in the ED (multiple potentially eligible patients often simultaneously in the ED) and the focus of the study on MTBI, patients with minor injury with triage history indicating head injury were given recruitment priority.

After consent, injury event history (occurrence of head trauma, LOC, PTA) was determined by the RA using both available medical records and participant interview. In cases where this information was discordant, the RA consulted with the patient’s medical care provider, who made the final determination. After data collection regarding the injury event, participants completed a structured interview that included the RPQ and SF-36, as well as information regarding participant educational status, Employment status, marital status, and race. Initial resting systolic Blood pressure and heart rate were determined from the ED triage note. Injury severity score (ISS) [33] was calculated from information in the medical record. Information on mechanism of injury was determined from the medical record and participant interview. After the initial interview, participants were given written and verbal information regarding the follow-up protocol and were instructed on how to contact the study team to inform them of any change in their contact information.

Outcome data

Patients were contacted by telephone 1, 3, and 12 months after their ED visit. At each follow-up time point, participants underwent a structured follow-up interview that included the RPQ and SIPab. Twelve-month follow-up interview also included questions regarding participant involvement in litigation related to the minor injury event. Before each telephone follow-up, a study reminder was mailed to the participant. A maximum of 10 attempts were made for each respondent for each follow-up interview. If the respondent dropped out of the study, he/she was not contacted for a follow-up interview.

The study was approved by institutional review boards at the Michigan Public Health Institute, St. Joseph Mercy Hospital, and the University of Michigan. Participants were initially compensated $25 for completing the initial ED interview and each telephone follow-up interview. Because the initial 3-month response rate was relatively low, compensation for completing the 12-month follow-up was subsequently increased to $75.

Data analyses

Data were evaluated for the presence of outliers via box plots. Injury group comparisons were performed using t tests for continuous variables and ?2 analysis for categorical variables. Associations between baseline characteristics and RPQ and SIPab results were compared using Spearman rank correlation. Because persistent symptoms (assessed via the RPQ and SIPab) were highly non-normal at follow-up time points, and a significant percentage of the sample had no symptoms at each follow-up time point, the contribution of baseline risk factors to symptom outcomes across time was assessed using repeated-measures logistic regression ana- lyses. In each model, baseline characteristics associated with Symptom burden in univariate analyses were included, and model fit and potential colinearity between independent variables were evaluated. Statistical analyses were performed using SPSS software (SPSS, Chicago, IL); repeated- measures logistic regression analysis was performed using SAS software (Cary, NC).

Results

Study sample characteristics

Study sample recruitment and retention data are shown in Fig. 1. Overall consent rate was 57%; the most common Reasons for refusal were no reason given or recorded (44%), physical symptoms preclude participation (22%), and concern would delay discharge (12%). Participant character- istics are shown in Table 1. Most participants were white; approximately 1 in 4 had experienced a motor vehicle collision . Mean ISS was low, reflecting the minor injury population evaluated in the study. Normalized SF-36 scores suggest that the health of the study sample was representative of that of the general population.

Follow-up rates

Telephone follow-up rates are shown in Fig. 1. At each time point, those in whom follow-up information was not obtained were significantly younger and less educated and more likely to be male and to have lower SF-36 mental health subscale scores than those in whom

Fig. 1 Study sample recruitment and retention data.

Participant characteristics and symptom outcomes according to initial injury and symptoms

Among individuals presenting to the ED with head injury, LOC was uncommon, occurring in only 13%. Most (70%) individuals with head injury who met CDC MTBI criteria [4] had no LOC or PTA and met MTBI criteria solely on the basis of having 2 or more PC symptoms in the ED (assessed via the RPQ). Participant characteristics according to initial injury features are shown in Table 2. Among patients without head injury, 46% reported 2 or more PC symptoms in the ED. Regardless of head injury status, when compared with patients without ED PC symptoms, individuals with 2 or more PC symptoms were significantly more likely to have worse self-reported mental health (t = 6.7, P b .0001), be younger (t = 4.1, P = .001), be female (?2 = 7.8, P =

.005), and have worse self-reported physical health (t = 2.3, P = .024).

follow-up information was obtained. At the 12-month time point, only 22 (6%) participants reported currently or previously being the plaintiff in a legal action related to their injury.

Table 1 Characteristics of study participants (N = 507)

Characteristic

Age, mean (SD), y Women, n (%) Education, n (%) bHigh school

High school or GED

Some college or trade school College graduate or beyond Unknown

Employment, n (%) Employed

Not working because of medical condition Not working because laid off/fired

Not working: student, retired, choice Race, n (%)

White

African American Hispanic

Asian Other/Unknown Marital status, n (%)

Married or in relationship Divorced, widowed, separated Never married

Unknown

Normalized SF-36 score, mean (SD) PCS

MCS

Head injury, n (%) Yes

No

ISS, mean (SD) Mechanism of injury, n (%) MVC

Non-MVC

39 (17)

271 (53.5)

53 (10.5)

127 (25.0)

170 (33.5)

151 (29.8)

6 (1.2)

378 (74.6)

29 (5.7)

18 (3.6)

82 (16.2)

394 (77.7)

89 (17.6)

12 (2.4)

3 (0.6)

9 (1.8)

215 (42.4)

92 (18.1)

195 (38.5)

5 (1.0)

51.6 (10.4)

49.4 (12.5)

339 (67)

168 (33)

1.88 (1.8)

131 (25.8)

376 (74.2)

Participant characteristics and symptom outcomes according to MTBI status

Of the 507 ED patients with minor injury in the initial cohort, 251 met criteria for MTBI. These patients were significantly more likely to be younger (37 vs 40 years; P =

.013) and have worse mental health (SF-36 MCS, 48 vs 51;

P = .003) and physical health (SF-36 PCS, 50 vs 53; P =

.019) than did those not meeting MTBI criteria. They also had more PC symptoms (RPQ score, 13.9 vs 3.7; P b .0001) and were more severely injured (ISS, 2.1 vs 1.7; P = .007) than those not meeting MTBI criteria. Individuals meeting CDC MTBI criteria [4] had a significantly higher incidence of PC syndrome and reported cognitive symptoms 1, 3, and 12 months after injury (Figs. 2A and 3A; P b.0001 for group differences at all time points). Among both those with and without initial MTBI, there seemed to be little change in symptom burden across time.

Fig. 2 Percentage of individuals with PC syndrome 1, 3, and 12 months after injury, according to (A) MTBI status and (B) injury and symptom characteristics. In panel B, dashed lines are non-MTBI groups; groups with head injury are indicated by shaded symbols.

Fig. 2B displays the percentage of participants meeting criteria for PC syndrome at 1, 3, and 12 months according to initial injury and symptom characteristics. At each follow-up time point, those with no head injury but had 2 or more PC symptoms in the ED were more likely to meet criteria for PC syndrome and report cognitive symptoms than were those who hit their head but did not have 2 or more PC symptoms in the ED. The percentage of individuals with PC syndrome was relatively stable across time within each subgroup, regardless of initial group characteristics (eg head injury vs no head injury). Fig. 3B displays the percentage of participants reporting 1 or more cognitive symptoms (assessed via SIPab) at 1, 3, and 12 months. Postconcussion and SIPab scores were highly correlated at all time points

(r = 0.742, 0.783, and 0.783 at 1, 3, and 12 months, respectively; P b .0001 at all time points).

Repeated-measures logistic regression models of symptom outcomes

Table 3 displays repeated-measures logistic regression models of PC syndrome presence and cognitive symptom presence across time. Baseline characteristics associated with PC symptoms in univariate analyses were included in the models. In both models, mental health status and physical health status before injury were most strongly associated with syndrome presence across time. In contrast, head injury presence or absence did not predict persistent symptoms in

Fig. 3 Percentage of individuals with 1 or more SIPab symptoms 1, 3, and 12 months after injury, according to (A) MTBI status and (B) injury and symptom characteristics. In panel B, dashed lines are non-MTBI groups; groups with head injury are indicated by shaded symbols.

either model. The interaction term between head injury and initial PC symptoms was also not significant in either model, indicating that the influence of initial PC Symptom severity in the ED on symptoms across time did not vary according to head injury status. Some variation in symptoms across time was observed, but, as seen in Fig. 3B, there was little evidence of symptom resolution over time.

Discussion

In the present study, initial PC symptoms assessed in the ED were a strong predictor of persistent PC symptoms, regardless of whether an individual hit his/her head or not, and PC syndrome presence or absence was relatively stable

across time. These data are consistent with the hypothesis that somatic/PC symptom vulnerability is, at least in part, an individual constitutional characteristic. When health status and ED symptoms were controlled for, head injury status did not predict the presence of persistent somatic or cognitive symptoms. This finding is consistent with the results from the recent longitudinal cohort study of Mickeviciene et al [6]. That study compared 217 individuals with head injury and LOC and 221 age- and sex-matched controls with minor nonhead injury and found few differences in PC symptoms over time between the groups [6].

In the present study, 7 in 10 individuals classified as MTBI had no LOC or PTA and met MTBI criteria solely on the basis of initial PC symptoms. Because PC symptoms are relatively stable across time, individuals with initial PC symptoms (who, per the CDC definition, have MTBI if they

Table 2 Characteristics of study participants according to injury and symptom characteristics

Characteristics No amnesia, LOC, or PC

symptoms

If head injury, met CDC MTBI criteria [4] (bold in table) F (P) PC symptoms (>=2) only Amnesia and LOC Amnesia only LOC only

Head injury (n = 339)
Subgroup, n (%) 93	(27)	173	(51)	40 (12)	30 (8)	3	(1)
Age, mean (SD), y 45	(20)	38	(17)	36 (15)	34 (17)	–		4.0	(.003)
Women, n (%) 38	(41)	103	(60)	17 (43)	16 (53)	–		13.2a	(.010)
SF-36 MCS 53	(9)	48	(13)	49 (13)	49 (17)	–		5.3	(b.0001)
SF-36 PCS 53	(9)	50	(12)	50 (12)	55 (7)	–		3.4	(.010)
ED RPQ score 0.9	(1.2)	14.6	(9.7)	14.0 (10.7)	11.1 (8.8)	–		42.8	(b.0001)
ISS 1.4	(1.1)	1.9	(2.1)	3.4 (2.3)	1.9 (1.8)			7.4	(b.0001)
No head injury (n = 168)
Subgroup, n (%) 87	(52)	78	(46)	2 (1)	0 (0)	1	(1)
Age, mean (SD), y 40	(16)	35	(13)	–	–	–		1.9	(.126)
Women, n (%) 46	(53)	49	(63)	–	–	–		2.5a	(.484)
SF-36 MCS 53	(8)	46	(15)	–	–	–		7.2	(b.0001)
SF-36 PCS 52	(8)	53	(10)	–	–	–		1.0	(.391)
ED RPQ score 1.1	(1.2)	10.1	(5.8)	–	–	–		74.7	(b.0001)
ISS 1.8	(1.5)	1.8	(1.4)	–	–	–		0.3	(.802)
a?2 statistic.

happened to hit their head) are much more likely to have continued symptoms. Although our data indicate that this is true regardless of head injury status, this classification system encourages attribution error (symptoms are because

Table 3 Repeated-measures logistic regression analyses assessing predictors of PC syndrome and sickness impact profile subscale score 1, 3, and 12 months after ED evaluation

Independent variable PC syndrome Sickness impact

profile

	?	P		?	P
Health status (SF-36) MCS	-0.0579	b.0001		-0.0714	b.0001
PCS	-0.0498	b.0001		-0.0554	b.0001
ED RPQ score	0.0738	.005		0.0158	.57
Sex Martial status	-0.4540	.03		-0.0747	.71
Divorced, widowed,	0.5918	.06	-0.0076		.98
separated Married or in	0.1358	.62	-0.2629		.30
relationship
Head injury	0.4118	.16	0.0681		.81
Heart rate divided by	1.1928	.17	0.2711		.16
systolic blood pressure Follow-up time point
1 month vs 12 month	-0.0716	.61	-0.1036		.45
3 month vs 12 month	-0.1730	.20	-0.4326		.001
Education	-0.0939	.41	-0.2077		.05
Age	0.0062	.46	0.0205		.01
ISS	-0.0397	.45	0.0627		.28
Head injury?Baseline PC symptom score	-0.0013	.97	0.0341		.27

of mechanical brain injury) and tends to falsely create an epidemic.

We believe that these findings have important clinical implications. First, because of increasing uncertainty regard- ing the pathogenesis of symptoms after injury, we suggest that ED care providers use neutral terms that do not imply a particular pathogenic model, such as minor head injury and posttraumatic symptoms, rather than the terms MTBI and postconcussion syndrome. This will reduce the risk of attribution error, which may be accompanied by patient distress and behavioral changes that may contribute to adverse outcomes. Second, we suggest that descriptive phrases be used rather than the CDC MTBI classification system [34]. The use of descriptive phrases may encourage researchers, ED care providers, and patients to think more broadly about the possible causes of persistent posttraumatic symptoms.

This study has several limitations that should be considered when interpreting the results. First, the RPQ

[26] combines symptom assessment and attribution (ie, current symptom level relative to symptom level before head injury), which likely increases attribution bias. However, this influence is biased toward finding increased PC syndrome prevalence in those with head injury, and yet in multivariate analyses, no differences were identified. The finding that ED somatic symptom severity, assessed via the RPQ, is strongly correlated with self-reported health status is also consistent with the results of large population and clinic-based studies assessing baseline somatic symptoms [10,13,14]. Second, consistent with World Health Organization criteria [27], PC syndrome presence was defined on the basis of 3 or more PC symptoms (symptoms rated at least “mild” on the RPQ). Experimental evidence indicates that individuals with

previous somatic symptoms may be most vulnerable to further increases in symptoms after stress or injury [35]. Our dichotomous outcome measure may have had a “ceiling effect,” which prevented us from detecting such postinjury increases. Future studies should consider the use of other measures, such as Visual analogue scales of current and past symptoms, to minimize attribution bias and ceiling effects. Finally, cohort follow-up was incomplete (81%, 67%, and 71% at 1, 3, and 12 months, respectively), and individuals lost to follow-up differed from those retained in the study. Although the degree of bias introduced into the study via loss to follow-up cannot be determined, logistic regression results regarding PC syndrome predictors were consistent across time when each time point was assessed separately.

In conclusion, the results of this study support the hypothesis that PC symptoms after minor trauma are not significantly influenced by the presence or absence of head trauma. Instead, the strongest predictors of symptoms after trauma are baseline mental and physical health status. Further studies that carefully assess preinjury symptoms and health and minimize attribution bias are needed. Such studies will be best able to further evaluate the role of head trauma in influencing minor injury outcomes and determine whether the link between symptoms and minor trauma is entirely related to attribution bias or whether minor trauma is capable of increasing symptoms in vulnerable individuals. The established biologic integration of stress and neurosen- sory systems [15-18] and evidence that minor injury events may result in disorders characterized by stress system dysregulation (eg, posttraumatic stress disorder [36-39]) suggest to us that this may indeed be the case.

References

1. Thurman D, Guerrero J. Trends in hospitalization associated with traumatic brain injury. Jama 1999;282(10):954-7.
2. Bazarian JJ, McClung J, Shah MN, Cheng YT, Flesher W, Kraus J. mild traumatic brain injury in the United States, 1998-2000. Brain Inj 2005;19(2):85-91.
3. Symonds C. Concussion and it’s sequelae. Lancet 1962;1:1-11.
4. Centers for Disease Control and Prevention AG. Mild traumatic injury tool kit. Unpublished meeting summary; 2001.
5. Mickeviciene D, Schrader H, Nestvold K, et al. A controlled historical cohort study on the post-concussion syndrome. Eur J Neurol 2002; 9(6):581-7.
6. Mickeviciene D, Schrader H, Obelieniene D, et al. A controlled prospective inception cohort study on the post-concussion syndrome outside the medicolegal context. Eur J Neurol 2004;11(6):411-9.
7. Rutherford WH. Sequelae of concussion caused by minor head injuries. Lancet 1977;1(8001):1-4.
8. Kay DW, Kerr TA, Lassman LP. Brain trauma and the postconcus- sional syndrome. Lancet 1971;2(7733):1052-5.
9. Ponsford J, Willmott C, Rothwell A, et al. Factors influencing outcome following mild traumatic brain injury in adults. J Int Neuropsychol Soc 2000;6(5):568-79.
10. Wolfe F. Pain extent and diagnosis: development and validation of the regional pain scale in 12,799 patients with rheumatic disease. J Rheumatol 2003;30(2):369-78.
11. Li C, Borgfeldt C, Samsioe G, Lidfeldt J, Nerbrand C. Background factors influencing somatic and psychological symptoms in middle- age women with different hormonal status. A population-based study of Swedish women. Maturitas 2005;52(3-4):306-18.
12. Kroenke K, Price RK. Symptoms in the community. Prevalence, classification, and psychiatric comorbidity. Arch Intern Med 1993; 153(21):2474-80.
13. Wolfe F, Rasker JJ. The Symptom Intensity Scale, fibromyalgia, and the meaning of fibromyalgia-like symptoms. J Rheumatol 2006; 33(11):2291-9.
14. Al-Windi A. Determinants of complaint symptoms in a Swedish health care practice–results of a questionnaire survey. J Psychosom Res 2004;57(3):307-16.
15. Diatchenko L, Slade GD, Nackley AG, et al. Genetic basis for individual variations in pain perception and the development of a chronic pain condition. Hum Mol Genet 2005;14(1):135-43.
16. Diatchenko L, Anderson AD, Slade GD, et al. Three major haplotypes of the b2 adrenergic receptor define psychological profile, blood pressure, and the risk for development of a common Musculoskeletal pain disorder. Am J Mol Genet 2006 In press.
17. Diatchenko L, Nackley AG, Slade GD, Fillingim RB, Maixner W. Idiopathic pain disorders–pathways of vulnerability. Pain 2006; 123(3):226-30.
18. Stein MB, Fallin MD, Schork NJ, Gelernter J. COMT polymorphisms and anxiety-related personality traits. Neuropsychopharmacology 2005;30(11):2092-102.
19. Diatchenko L, Nackley AG, Slade GD, et al. Catechol-O-methyl- transferase gene polymorphisms are associated with multiple pain- evoking stimuli. Pain 2006;125(3):216-24.
20. Girdler SS, Koo-Loeb J, Pedersen CA, Brown HJ, Maixner W. Blood pressure-related hypoalgesia in Bulimia nervosa. Psychosom Med 1998;60(6):736-43.
21. Fillingim RB, Maixner W, Bunting S, Silva S. Resting blood pressure and thermal pain responses among females: effects on pain unpleasant- ness but not pain intensity. Int J Psychophysiol 1998;30(3):313-8.
22. Pfleeger M, Straneva PA, Fillingim RB, Maixner W, Girdler SS. Menstrual cycle, blood pressure and Ischemic pain sensitivity in women: a preliminary investigation. Int J Psychophysiol 1997;27(2):161-6.
23. Maixner W, Fillingim R, Kincaid S, Sigurdsson A, Harris MB. Relationship between pain sensitivity and resting arterial blood pressure in patients with painful temporomandibular disorders. Psychosom Med 1997;59(5):503-11.
24. Bragdon EE, Light KC, Girdler SS, Maixner W. Blood pressure, gender, and parental hypertension are factors in baseline and poststress pain sensitivity in normotensive adults. Int J Behav Med 1997;4(1):17-38.
25. Bruehl S, Chung OY. Interactions between the cardiovascular and pain regulatory systems: an updated review of mechanisms and possible alterations in chronic pain. Neurosci Biobehav Rev 2004;28(4):395-414.
26. King NS, Crawford S, Wenden FJ, Moss NE, Wade DT. The Rivermead Post Concussion Symptoms Questionnaire: a measure of symptoms commonly experienced after head injury and its reliability. J Neurol 1995;242(9):587-92.
27. Organization WH. The ICD-9 classification of mental and behavioral disorders: clinical descriptions and diagnostic guidelines. Geneva: WHO; 1978.
28. Gilson BS, Gilson JS, Bergner M, et al. The Sickness Impact Profile. Development of an outcome measure of health care. Am J Public Health 1975;65(12):1304-10.
29. Temkin N, McLean Jr A, Dikmen S, Gale J, Bergner M, Almes MJ. Development and evaluation of modifications to the Sickness Impact Profile for head injury. J Clin Epidemiol 1988;41(1):47-57.
30. Ware JE, Sherbourne C. The MOS 36 Item Short Form Health Survey (SF 36). Conceptual framework and item selection. Med Care 1992(30):473-83.
31. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state.” A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12(3):189-98.
32. Levin HS, O’Donnell VM, Grossman RG. The Galveston Orientation and Amnesia Test. A practical scale to assess cognition after head injury. J Nerv Ment Dis 1979;167(11):675-84.
33. Baker SP, O’Neill B, Haddon W, Long WB. The injury severity score: a method for describing patients with multiple injuries and evaluating emergency care. J Trauma 1974;14:187-96.
34. CDC. Heads up: facts for physicians about mild traumatic brain injury . Unpublished meeting summary: National Center for Injury Prevention and Control; 2001.
35. McBeth J, Macfarlane GJ, Benjamin S, Silman AJ. Features of somatization predict the onset of chronic widespread pain: results of a large population-based study. Art Rheum 2001;44(4):940-6.
36. Kuch K, Cox BJ, Evans RJ. Posttraumatic stress disorder and motor vehicle accidents: a multidisciplinary overview. Can J Psychiatry 1996;41(7):429-34.
37. Mayou R, Bryant B, Duthie R. Psychiatric consequences of road traffic

accidents. BMJ 1993;307(6905):647-51.

1. McLean SA, Clauw DJ, Abelson JL, Liberzon I. The develop- ment of persistent pain and psychological morbidity after motor vehicle collision: integrating the potential role of stress response systems into a biopsychosocial model. Psychosom Med 2005; 67(5):783-90.
2. Yehuda R. Post-traumatic stress disorder. N Engl J Med 2002;346(2): 108-14.