PTSD After Severe Vehicular Crashes (2024)

As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsem*nt of, or agreement with, the contents by NLM or the National Institutes of Health.
Learn more: PMC Disclaimer | PMC Copyright Notice

Ann Adv Automot Med. 2009 Oct 5; 53: 177–193.

PMCID: PMC3256803

PMID: 20184843

Gabriel E. Ryb, MD MPH, Patricia C. Dischinger, PhD, Kathleen M. Read, MSW, and Joseph A. Kufera, MA

Author information Copyright and License information PMC Disclaimer

Abstract

Purpose:

To describe predictors of PTSD after motor vehicle crashes (MVC).

Methods:

MVC patients were interviewed during their hospitalization and at 6 and 12 months post-injury. Interviews included information about behavioral factors, circ*mstances around the crash, recovery and PTSD screening. PTSD was defined as the development of 3 or more of 7 PTSD symptoms. Association of risk factors with PTSD development at 6 and 12 months was analyzed using contingency tables. Multiple regression models were built for the prediction of PTSD.

Results:

367 and 317 patients completed the 6 and 12 month interviews respecively. PTSD developed in 27.5 % (n=101) and 24.3 % (n=77) of the population at 6 and 12 months repectively. PTSD occurred more frequently among females, those with a previous history of depression, violent injury, or other traumatic events, and those whose crashes involved a fatality. Those who were culpable for the crash, age<30, and sustained brain injuries were less likely to develop PTSD at 6 months. Occupant position, education, marital status, alcohol problems, injury severity, heart rate, and blood alcohol + status did not show any significant association with PTSD. In the multiple logistic regression, female gender, history of depression, culpabilty, prior violent injury, and a fatality in the crash were associated with PTSD at 6 months. Only prior violent injury, and a death in same crash were predictors at one year.

Conclusion:

PTSD occurs frequently after MVCs. Female gender, prior violent injury, death of another occupant and history of depression are associated with PTSD development.

INTRODUCTION

As trauma centers become increasingly proficient at saving lives, trauma specialists have been challenged to “pursue innovative strategies to further reduce the risk of injury related morbidity and mortality” (Gentilello, 2005). While injury prevention remains the best approach, the return of injured individuals to pre-injury functional and psychological health should remain a goal of those involved in trauma care. Recent studies have shown that psychological outcomes such as post-traumatic stress disorder (PTSD) can have a major impact on recovery from serious injury and greatly influence the long-term quality of life of those surviving such injuries (Zatzick et al, 2002; Blanchard et al, 2004; Donahue, 2006; Kupchik et al, 2007; Holbrook et al, 1999; Fitzharris et al, 2007; Read et al, 2004).

The term PTSD originated in the military, and is used to describe a psychiatric disorder characterized by recurrent and intense flashbacks and memories, hyperarousal, and avoidance. Studies of PTSD have greatly expanded recently to include various populations including Vietnam veterans, sexual assault victims, adult abuse survivors and survivors of disasters. Since the introduction of PTSD in the DSM-III, studies of psychiatric morbidity and predictors of the development of PTSD have taken on a broader approach often superseding other issues in explaining mental health symptoms. (McHugh and Treisman, 2007)

Among adults hospitalized following any type of injury, rates of PTSD have ranged from 14% to 46% (Feinstein and Dolan., 1991; Bryant and Harvey, 1995; Kuch, Cox and Evans., 1996; Blanchard et al., 1995; Shalev et al., 1996; Perry et al., 1992; Malt, 1988; Green et al., 1993; Landsman et al., 1990). These differences in prevalence estimates may be partly accounted for by different types of testing instruments used for diagnosis, different time periods for follow-up, and other factors such as multiple injury mechanisms, which make it difficult to understand the full magnitude of this often disabling problem.

Car crashes have been found to be the single leading cause of PTSD in the general population. (Blanchard, 1997; Kessler, 1995; Norris, 1992). Reports of the incidence of PTSD in victims of motor vehicle crashes have ranged from 8% to 46%. However, differences in populations studied, and the lack of systematic longitudinal follow-up studies have made it difficult to accurately characterize the importance of this condition among car crash victims.

Despite the high prevalence of exposure to traumatic stressors in modern society, a relatively small proportion of exposed people subsequently develop PTSD. According to Green, approximately 25% of people experiencing one or more traumas develop PTSD (Green, 1994). Thus, there is increasing interest in identifying both risk and protective factors for developing PTSD following a traumatic event. Better understanding of these factors will help guide the treatment of the disorder and the design of both prevention and early intervention strategies.

Several studies address risk factors and incidence rates of PTSD in survivors of motor vehicle crashes (Malt, 1988, 1993; Mayou, Bryand and Duthie 1993; Blanchard et al, 1995; Bryant and Harvey, 1996; Shalev et al, 1996). Many of these prior investigations, however, have small numbers of cases and have not systematically studied PTSD in relation to individual pre-injury history, crash reconstruction, and detailed injury factors. Thus, the purpose of this research is to expand the literature and identify additional individual, crash, and injury factors that may contribute to the development of PTSD among adult occupants of moderate and severe motor vehicle crashes who were admitted to a Level 1 trauma center.

METHODS

Study population

The study subjects were patients included in the Baltimore Crash Injury Research and Engineering Network (CIREN) database. CIREN is a multicenter research network with a rich resource of detailed crash data in combination with highly detailed injury data on crash-involved occupants. This injury information includes hospital discharge summaries, radiologic images and narratives, operative reports, autopsy reports, injury photographs, and long-term outcomes.

Psychosocial and PTSD assessment

Patients and their families were interviewed during their initial hospitalization in the trauma center and contacted again by telephone at 6 months and 12 months post-trauma to assess individual recovery and psychosocial adjustment. Each interview lasted 30 to 40 minutes and had several components including: pre-injury and trauma history, physical symptoms, behavioral and socioeconomic factors, as well as detailed information regarding circ*mstances of the crash. In addition, standardized instruments documenting pain, alcohol and drug use, and cognition were utilized as well as the Short Form-36 Health Survey (SF-36) (Ware, 1992).

Items meeting the diagnostic criteria for symptoms of PTSD were comprised of the following seven symptom clusters: (1) frequent, recurrent, re-experiencing of the event by intrusive, intense thoughts, flashbacks; physical reactions when thinking or reminded of the crash, (2) persistent avoidance and fear of stimuli associated with the trauma, (3) hyper-alertness, hyper-vigilance, being on guard, angry outbursts, (4) diminished, and serious loss of interest/emotional numbing, inability to have loving feelings because of crash, (5) significant sleep disturbances, (6) excessive cognitive difficulties in memory, concentration, avoiding thoughts of crash, and (7) survivor’s guilt.

If three or more of these seven symptom clusters resulted in functional impairment, patients were considered positive for PTSD. As PTSD descriptors can often be vague, since they do not take into account the degree of impairment or include general responses to any highly negative event, (Satel, 2004; McHugh and Treisman, 2007), only those symptoms that were specifically related to the exposure of the vehicular crash event were considered. Care was given to rule out those symptoms that may have been an exacerbation of a pre-existing mood, anxiety or personality disorder as well as to differentiate PTSD from other psychiatric distress, such as anxiety and depression or symptoms due to traumatic brain injury, since there is often overlap.

A six month time period was chosen for the first follow-up interview as it allowed time to look at the persistent symptoms that were impacting recovery. Blanchard et al (1996) and Ursano et al. (1999) reported that occupants of vehicular crashes met the DSM-III criteria for PTSD usually within 1 to 4 months following the crash and that nearly half had recovered by 6 months. Patients were very willing to respond at the 6 month follow-up point and demonstrated great insight into how their injuries and the crash aftermath affected them psychologically, socially, and financially, thus providing an overall picture of their own perception of recovery. The second follow-up interviews were conducted at one year.

Questions regarding cognitive changes consisted of four main components. Patients were asked: As compared to pre-injury status, have you experienced or has your family noticed any cognitive changes such as: 1) Diminished ability to think through/process information, 2) Short term memory deficits/ problems with recall, 3) Problems in concentration/focusing/overwhelmed easily, and 4) Attention deficits. A prior study underscored the need to supplement the SF-36 with a measure of cognitive function when evaluating outcome from trauma involving any head injury. (MacKenzie et al, 2002)

Two questions were asked to determine if depression was present pre-injury, based on a two-question screening instrument which was validated against six previously standardized instruments (Whooley et al, 1997).

In order to assess behavioral changes the following four items were measured. Patients were asked: “As compared to pre-injury status, have you experienced or has your family noticed any behavioral changes such as: 1) Increased irritability/outbursts, 2) Compulsive/impulsive behavior and/or actions, 3) Inappropriate behaviors; 4) Increased nervousness”.

A modification of the visual analogue pain rating scale and self rating scale was used as a simple way to record subjective estimates of pain. The scale is scored from 1 to 10, with 1 indicating no pain and 10 representing severe pain. This scale concentrates on the distress associated with pain versus intensity. (JCAHO, 2000)

All interviews were conducted by a clinical social worker (KR) with over 30 years experience and expertise in trauma and trauma recovery who was able to provide a comprehensive psychosocial assessment in addition to the standardized items.

Statistical analysis

Odds ratios and their 95% confidence intervals were calculated to determine the univariate association of various patient, injury and crash characteristics with the development of PTSD both at 6 months and 1 year. Multiple logistic regression models were then constructed to determine factors associated with the the development of PTSD at 6 months and at 1 year, using independent variables that were univariately associated with PTSD during follow-up. Regression results are displayed in terms of adjusted odds ratios and corresponding 95% confidence intervals. All analyses used p-value of 0.05 to indicate statistical significance.

RESULTS

Four hundred thirty eight cases admitted to the R Adams Cowley Shock Trauma Center were enrolled at the Baltimore CIREN center between April 1998 and February 2009. Six month follow up rate was 84% [12% were never interviewed (28 deaths, 12 dropped and 13 were missed), and 4% did not undergo the 6 month interview (4 deaths, 2 dropped, 1 refused and 11 could not be located)].

The resulting study sample consisted of 367 cases. Their mean age was 39.8 years. Fifty six percent were female, 52% were married and 84% had at least a high school diploma. Most (84%) subjects were drivers, involved in two-car vehicular collisions (74%), and tested negative for alcohol (85%). Their Injury Severity Score (ISS) was distributed as follows: ≤ 8: 10%, 9–15: 40%, 16–24: 27% and ≥25: 24%.

While 367 patients were seen at the 6 month follow-up interview, at the time of this report 317 (86%) had been interviewed and assessed for PTSD at 1 year (5 deaths, 9 could not be found, 3 did not answered the PTSD questions and 33 with pending interviews). The resulting 12 month interview follow up rate was 72% (317/438).

Three PTSD criteria were present in 101 cases (27.5%) and in 77 cases (24.3%) at the 6 and 12 month follow-up, respectively. For the remainder of the analysis we will refer to these patients as having screened positive for PTSD. Tables 1 through through33 show the prevalence of multiple individual characteristics, crash factors and injury severity among patients that screened positive and negative for PTSD.

Table 1:

Individual characteristics in relation to PTSD development at 6 months and one year

	Total (N = 367) (%)	6 month follow up (N=367)				One year follow up (N=317)¹
	Total (N = 367) (%)	PTSD + (N=101) (%)	PTSD− (N=266) (%)	P	Odd ratios (95% CI)	PTSD + (N=77) (%)	PTSD− (N=240) (%)	P	Odd ratios (95% CI)
Demographics
Age ≥30+	61	69	58	0.045	1.64 (1.01–2.68)	71	56	0.02	1.94 (1.11–3.39)
Female Gender	56	72	50	<0.001	2.57 (1.56–4.22)	69	56	0.04	1.78 (1.03–3.06)
≥ HS education	84	89	82	0.1	1.80 (0.90–3.59)	84	86	0.8	0.89 (0.44–1.81)
Married	45	48	43	0.5	1.18 (0.75–1.87)	48	44	0.49	1.20 (0.71–2.01)
Mental health history
Anxiety ²	22	36	16	0.01	2.88 (1.26–6.59)	33	18	0.1	2.25 (0.81–6.29)
Depression	28	43	23	<0.001	2.54 (1.56–4.15)	40	25	0.006	2.11 (1.23–3.64)
Depression or Apathy	38	51	34	0.003	1.99 (1.25–3.18)	49	35	0.03	1.78 (1.06–2.99)
CAGE ETOH ≥2	22	20	23	0.70	0.84 (0.48–1.49)	21	21	0.9	0.97 (0.52–1.83)
Prior trauma history
MVC injury	30	37	27	0.07	1.56 (0.96–2.54)	30	28	0.7	1.12 (0.64–1.97)
Assault ³	7	13	5	0.008	2.88 (1.28–6.44)	14	5	0.004	3.47 (1.44–8.36)
Other injury	27	32	25	0.2	1.41 (0.85–2.32)	33	25	0.2	1.41 (0.81–2.47)
Other trauma ²,⁴	23	33	18	0.05	2.25 (0.99–5.11)	46	13	<0.001	5.67 (2.00–16.08)

Open in a separate window

¹Fifty patients did not have one year follow up. PTSD grouping at one year does not overlap six month grouping (see text)

²Only 141 patients were asked this question (42 PTSD+) at 6 months and 101 (24 PTSD) at one year.

³Physical or Sexual assault

⁴House fires, disaster, etcetera

Table 3:

Injury factors in relation to PTSD development at 6 months and one year

	6 month follow up (N=367)				One year follow up (N=317)¹
	PTSD + (N=101) (%)	PTSD− (N=266) (%)	P	Odd ratios (95% CI)	PTSD + (N=77) (%)	PTSD− (N=240) (%)	P	Odd ratios (95% CI)
HR>100	35	32	0.8	1.14 (0.69–1.88)	39	31	0.5	1.38 (0.79–2.40)
ISS			0.5				0.7
≤8	9	10		Reference	11	10		Reference
9–15	46	37		0.75 (0.33–1.71)	39	41		1.08 (0.45–2.61)
16–24	23	28		1.13 (0.47–2.74)	25	28		1.19 (0.47–3.03)
≥25	22	24		1.01 (0.41–2.48)	25	21		0.88 (0.34–2.25)
Lower extremity injury ²	83	72	0.03	1.96 (1.07–3.59)	81	73	0.2	1.51 (0.80–2.85)
LOC/PTA ³	50	64	0.01	0.54 (0.34–0.87)	59	62	0.6	0.87 (0.52–1.47)
Initial GCS<13	5	12	0.04	0.37 (0.14–0.98)	9	13	0.7	0.94 (0.39–2.28)
Head injury ⁴	14	20	0.2	0.64 (0.33–1.24)	16	16	0.8	0.95 (0.47–1.90)
ICU days			0.5				0.2
0	70	76		Reference	75	73		Reference
1–2	7	4		0.48 (0.15–1.56)	11	4		0.32 (0.09–1.09)
3–7	10	5		0.52 (0.19–1.42)	7	8		1.27 (0.38–4.42)
>7	13	15		1.08 (0.47–2.43)	7	15		2.33 (0.69–7.72)
Ventilator days			0.9				0.1
0	77	80		Reference	84	78		Reference
1–2	8	6		0.69 (0.27–1.75)	8	7		0.93 (0.33–2.61)
3–7	5	4		0.83 (0.26–2.63)	3	5		1.66 (0.40–6.96)
>7	10	10		0.95 (0.42–2.11)	5	10		2.33 (0.71–7.61)

Open in a separate window

¹Fifty patients did not have one year follow up. PTSD grouping at one year does not overlap six month grouping (see text)

²AIS codes 8514xx.x -8536xx.x

³Loss of consciousness and peri-traumatic amnesia

Six month follow up

Demographic characteristics positively associated with PTSD were age > 30 and female gender. Prior history of depression, anxiety, and prior traumatic events were also associated with PTSD development, while there was no association with either educational achievement or problem drinking history (i.e. CAGE 2+) (Table 1).

Crash type (i.e. fixed object versus other vehicle), entrapment, change in velocity (delta V), occupant position and alcohol or other substance involvement were not linked with the development of PTSD. On the other hand, the occurrence of a fatality or the death or injury of a relative during the crash was positively associated with the disorder (Table 2). Being culpable for the crash, however, was negatively related to PTSD development.

Table 2:

Crash factors in relation to PTSD development at 6 months and one year

	6 month follow up (N=367)				One year follow up (N=317)¹
	PTSD + (N=101) (%)	PTSD− (N=266) (%)	P	Odd ratios (95% CI)	PTSD + (N=77) (%)	PTSD− (N=240) (%)	p	Odd ratios (95% CI)
Fixed object ²	20	28	0.2	0.65 (0.36–1.17)	18	27	0.2	0.66 (0.35–1.26)
ΔV ≥ 49kph ³	29	34	0.4	0.78 (0.43–1.40)	34	32	0.8	1.10 (0.59–2.05)
Entrapment	22	29	0.2	0.67 (0.38–1.18)	25	28	0.6	0.86 (0.48–1.56)
Driver position	82	85	0.5	0.79 (0.43–1.46)	88	85	0.5	1.29 (0.59–2.82)
BAC+	11	17	0.15	0.60 (0.30–1.21)	13	15	0.6	0.82 (0.39–1.74)
Any substance involvement	22	28	0.2	0.71 (0.41–1.23)	30	24	0.3	1.33 (0.74–2.38)
Culpable	49	64	0.006	0.52 (0.33–0.83)	51	63	0.07	0.62 (0.37–1.03)
Fatality in crash	21	6	<0.001	4.39 (2.16–8.92)	23	6	<0.001	4.92 (2.31–10.48)
Family member fatality	6	1	<0.001	8.34 (1.65–42.01)	8	1	0.003	10.06 (1.99–50.92)
Family member hurt	27	18	0.07	1.66 (0.96–2.84)	23	18	0.3	1.40 (0.75–2.60)
Fear of dying ⁴	62	24	<0.001	5.18 (2.31–11.64)	57	31	0.03	2.90 (1.10–7.61)

Open in a separate window

¹Fifty patients did not have one year follow up. PTSD grouping at one year does not overlap six month grouping (see text)

²Crash against a fixed object (i.e. pole, wall, etcetera)

³Only 260 cases had known ΔV at 6 months and 240 at one year.

⁴Only 128 cases assessed at 6 months and 96 at one year.

With respect to injury factors, patients with lower extremity injuries were significantly more likely to develop PTSD (Table 3). ISS, ICU and mechanical ventilator days, and admission heart rate were not related to the subsequent development of PTSD. However, loss of consciousness or peritraumatic amnesia, and admission GCS <13 were protective for PTSD development in the univariate analyses.

Multiple logistic regression models were constructed to combine individual, crash and injury factors found to be significant in the univariate analysis (hypothesized to be associated with PTSD development). The final model (Table 4) indicated that female gender, prior history of assault, and pre-injury history of depression were significantly associated with the development of PTSD, controlling for all other predictor variables. Among crash variables, the occurrence of a fatality remained significantly associated with subsequent PTSD; this association was even higher when the fatality was a family member. The protective effect of being culpable for the crash in relation to PTSD occurrence, seen in the univariate analysis, was also significant in the multiple logistic regression model. Several factors, such as history of anxiety, prior trauma (i.e. house fires, disasters etcetera) and fear of dying, could not be included in the models because they had been added to the interview later and therefore assessed in less than half of the population.

Table 4

Multiple logistic regression: PTSD at 6 months and one year

	6 month follow up (N=367)		One year follow up (N=317)¹
	Odds Ratio	95% CI	Odds Ratio	95% CI
Age ≥ 30	1.36	(0.80 – 2.31)	1.65	(0.90–3.00)
Female gender	2.05^*	(1.18 – 3.56)	1.50	(0.81–2.77)
Culpable	0.59	(0.35 – 0.99)	0.73	(0.41–1.32)
Prior assault injury	2.70^*	(1.12 – 6.52)	3.04^*	(1.18–7.84)
Family member fatality	9.32^*	(1.77 – 49.11)	11.33^*	(2.14–60.01)
Fatality in crash	3.05^*	(1.29 – 7.22)	3.40^*	(1.36–8.49)
Hx. of depression	2.04^*	(1.16 – 3.57)	1.71	(0.91–3.19)

Open in a separate window

1-Fifty patients did not have one year follow up. PTSD grouping at one year does not overlap six month grouping (see text)

*-statistically significant

Not surprisingly, PTSD at 6 months was associated with concomitant poor emotional, behavioral and cognitive outcomes (Table 5). There was also an association of PTSD with limited ambulation and persistence of pain at 6 months. While the unemployment rates of those employed prior to their injury were not significantly different for those with and without PTSD, the patients with PTSD were less likely to regain full time employment.

Table 5

Functional correlates of PTSD at 6 months and one year

	6 month follow up (N=367)				One year follow up (N=317)¹
	PTSD + (N=101) (%)	PTSD− (N=266) (%)	P	Odd ratios (95% CI)	PTSD + (N=77) (%)	PTSD− (N=240) (%)	P	Odd ratios (95% CI)
Depression	92	35	<0.001	21.63 (10.01–46.47)	95	28	< 0.001	47.12 (16.57–134.02)
Behavioral changes	56	23	<0.001	4.35 (2.68–7.08)	52	14	<0.001	6.78 (3.80–12.10)
Cognitive changes	61	34	<0.001	3.06 (1.90–4.91)	65	26	<0.001	5.32 (3.07—9.22)
Depression, behavioral or cognitive changes (any one)	96	54	<0.001	20.55 (7.34–57.47)	99	41	< 0.001	110 (15.06–805.23)
Depression, behavioral and cognitive changes (all 3)	39	11	<0.001	5.34 (3.05–9.36)	39	8	< 0.001	7.42 (3.86–14.30)
Limited ambulation	75	45	<0.001	3.73 (2.23–6.23)	70	34	< 0.001	4.61 (2.64–8.04)
Pain	83	42	<0.001	6.90 (3.88–12.27)	81	38	< 0.001	6.89 (3.70–12.82)
Pain rate ≥ 5/10 ²	60	32	<0.001	3.19 (1.77–5.74)	50	37	0.1	1.70 (0.88–3.28)
Unemployed ³	28	19	0.1	1.67 (0.85–3.29)	45	12	< 0.001	6.18 (2.92–13.10)
Full time employment ³	38	53	0.05	0.54 (0.30–0.99)	37	71	< 0.001	0.24 (0.12–0.48)

Open in a separate window

¹Fifty patients did not have one year follow up. PTSD grouping at one year does not overlap six month grouping (see text)

²Among those with pain (N=198 at 6 months and N=151 at one year).

³Among those employed prior to the injury (N=198 at 6 months and 195 at one year)

One year follow up

Of the 367 patients evaluated at 6 months, 86 % (n=317) were also evaluated for PTSD at one year. At the one year follow up interview, PTSD screen was positive in 24.3% (n=77) of patients

Within the cohort with one year follow up, 65% of the patients who screened positive for PTSD at 6 months were also positive at one year representing 77% (n=59) of those diagnosed to have PTSD at one year. Of the patients who screened negative for PTSD at 6 months (n=226), 8% (n=18) screened positive at one year, representing 23% of those diagnosed to have PTSD at one year.

The association of subject characteristics and crash factors with PTSD at one year was similar to that found at 6 months (Table 1 and and2).2). However, the negative association of being culpable for the crash, while of a similar magnitude, was not significant at one year.

The association of lower extremity injuries with PTSD also remained present, but the link of PTSD development with loss of consciousness, peritraumatic amnesia, or GCS<13 following the crash was not significant at one year. Remaining injury factors continued to show no link with PTSD at the one year interview.

The effects of prior assault and the occurrence of a fatality during the crash increased in magnitude in the one year regression analyses. However, the effects of female gender, history of depression, and being culpable for the crash were reduced in magnitude and were not significant by the one year interview (Table 4).

The differences in emotional, behavioral and cognitive changes between the PTSD and no PTSD group became more accentuated by the one year interview. Similar effects were seen for pain and limitation of ambulation. The difference in return to full employment (among those previously employed) between the PTSD and non PTSD groups was also larger by the one year interview. While the percentage of individuals with PTSD who were able to return to full employment was the same at 6 months and 1 year, the percentages of those able to return to work among the non PTSD patients increased from 53% at 6 month to 71% at one year. Furthermore, the number of unemployed (among those employed prior to the injury) among those with PTSD increased from 28% at 6 months to 45% at one year and decreased among those without PTSD from 19% to 12%.

DISCUSSION

Our population of moderately and severely injured vehicular crash patients screened positive for PTSD in 27.5% of the cases at 6 months and 24% at one year, in the middle range of the reported literature. (Blanchard and Hickling, 1997; Kessler et al, 1995; Norris, 1992).

Among demographic factors, we found female gender to be a risk factor for PTSD in both the univariate and multivariate analyses. This finding is consistent with many others in the literature (Zatzick et al, 2002; Holbrook et al., 1998; Breslau and Davis,1992; Kessler, 1995). However, while age appeared to be associated with PTSD in the univariate analyses, it was not associated with PTSD after adjustment for other factors.

We found history of depression and prior history of traumatic events to be also linked to PTSD risk in both univariate and multivariate analyses. Other authors have previously reported similar findings in trauma center- based studies (Zatzick et al, 2002; Irish et al, 2008). In a study of trauma patients by Ursano et al. (1999), it was noted that previous trauma was not a risk factor for motor vehicle crash-related PTSD, but previous PTSD was a risk factor (Ursano et al., 1999). This is consistent with our findings of a strong association of PTSD development with prior assault or violent injury, often a precursor for the development of PTSD, and the lack of association with prior motor vehicle crash.

None of the vehicular crash factors (fixed object versus other vehicle, delta V and occupant position) nor the use of alcohol or other substance of abuse appeared to influence the development of PTSD in our study. However, in agreement with another report (Blanchard et al, 1996) we found that the occurrence of a fatality in the crash was a predisposing factor. Our study indicated an even stronger effect for the development of PTSD when a family member died in the crash.

As reported by others (Mayou, Bryand and Duthie, 1993; Green et al, 1993) we also encountered a very strong association of fear of dying with PTSD development, although this variable was not added as a covariate in the regression analysis because less than 50% of the population had been asked this question. Blanchard et al. reported that the patient’s perception of threat to life was associated with early onset PTSD symptoms (Blanchard et al., 1995). Similar findings were also noted by Michaels et al (1999), who showed, by path analysis, that perceived threat to life was an integral component in the development of PTSD.

A paradoxical finding was the protective effect of being responsible for the crash at 6 months. This finding has been previously reported by Delhaunty et al (1997) and Hickling et al. (1999). Whether this association is mediated through a personality disorder or a coping style will require further study. Nevertheless, “antisocial” types, often engaging in risky behavior, typically have higher rates of prior trauma and PTSD incidence (Storr et al, 2007; Jang et al, 2003). Another possibility may be that not being responsible for the crash is linked to feelings of loss of control over what happened, which may predispose one to develop PTSD symptoms. This effect was no longer significant at the 1 year interview.

Heart rate on admission, another variable felt to be associated with the development of PTSD by others (Zatzik et al, 2005; Shalev et al, 1998b), was not associated with PTSD development at 6 months or at one year. Whether this lack of association is related to the injury severity of our population or other factors needs to be further explored.

The protective effect of LOC described by others (Mayou, Bryand and Duthie 1993; Sbordone and Liter, 1995; Gil et al., 2005) was also found in our study in the univariate analysis. Nevertheless, it is important to mention that while LOC/PTA decreases the odds of PTSD at 6 months by 45% (OR 0.55, CI 0.34–0.86), 50% of the PTSD patients had LOC/PTA. Hence, PTSD at 6 months and LOC/PTA frequently occur in the same patients. Of interest is the fact that by the one year follow up, the effect of LOC/PTA was no longer protective (OR 0.87, CI 0.52–1.47), perhaps due to mediating factors resulting from the injuries, as discussed below for lower extremity injuries.

We found that ISS and the number of days spent in the ICU or on mechanical ventilation were not related to PTSD. Malt and Blikra (1993) found the degree of physical injury to be an important variable. However, others have reported a lack of association between injury severity and subsequent PTSD (Zatzick et al, 2002; Holbrook et al, 1998) in support of our findings. We did find an association between lower extremity injury and PTSD development; however, this association was no longer significant following adjustment with other patient, crash, and injury factors.

We believe that the subgroup of lower extremity fractures with poor functional outcomes (especially those involving articular surfaces of joints affecting mobility and encompassing pain), by preventing the return to pre-injury activities and lifestyle (Read et al, 2004; Dischinger et al, 2005), is responsible for the association of lower extremity injuries and PTSD. The association between PTSD and the inability to recover ambulation supports this belief. Further analysis should focus on injuries with high likelihood to affect long term mobility and their association with PTSD.

An important risk factor noted in our study was the presence of persistent and severe pain among PTSD patients at both 6 months and one year. While acute pain after tissue damage is both adaptive and protective, chronic pain is not, and often evolves into a self- perpetuating cycle of psychological distress, suffering and deconditioning (Wald, Taylor and Fedoroff, 2004). Co-occurrence of chronic pain and PTSD has been reported between 20% and 75% of cases with chronic pain following a traumatic incident (Wald, Taylor and Fedoroff,2004). We cannot postulate a causal link, because pain perception could be modulated by the hyperarousal cluster symptoms present among the PTSD patients. Nevertheless, the stress response to chronic pain may have a place in the neurobiological processes responsible for the development or perpetuation of PTSD. Treatment of prolonged or chronic pain may be an important factor in the prevention or alleviation of PTSD symptoms.

The presence of depression and cognitive/behavioral changes in the group that developed PTSD underlines the degree of functional disability brought by this condition. This finding is consistent with many studies that have demonstrated a significant decrease in the quality of life and poorer psychosocial functioning post trauma among those MVC survivors with PTSD as compared to non-PTSD survivors. (Kuhn Blanchard and Hickling, 2003; Blanchard, et al, 2004; Read et al, 2004; Donahue, 2006; Fitzharris et al, 2007). The very high prevalence of depressive symptoms among the patients developing PTSD (92% at 6 months and 95% at one year) is consistent with prior reports that document PTSD and depression genetic liability (Koenen et al 2008) and co-morbidity (Shalev et al 1998a) as well as symptom overlap between these two conditions (Brady et al 2000).

While we reported similar unemployment rates at 6 months in those with and without PTSD, the PTSD patients were less likely to regain full time employment. Matthews (2005) reported PTSD patients to be highly motivated to return to work but also noted that high levels of depression, reduced time management ability and anxiety with physical injuries represent barriers to employability in this population. The percentage of PTSD patients that are unemployed in our study increased from the 6 month to the one year interview. Meanwhile, the non-PTSD patients experienced an increase in the employment rates.

While female gender and history of depression effects in the multivariate analysis at 6 months disappeared at the one year interview, the effects of prior assault injury and fatality during the crash remained strong throughout the entire study. This finding is of interest since it allows differentiating factors that may be linked to early but short duration PTSD versus factors that affect persistent PTSD. Another explanation may be related to symptom overlap between post-traumatic-depression and PTSD. The early gender effect may potentially be related to an overdiagnosis of PTSD among women secondary to their ability to express or share their symptoms more easily or their greater insight to their emotional experiences.

Of interest were the changes over time in the population that experienced PTSD symptoms. Not only did we find that 35% of the patients who screened positive for PTSD at 6 months had screened negative at one year, but also we found that 23% of the patients who screened positive at one year had screened negative for PTSD at 6 months. Similar fluidity in the recovery from PTSD prior to 6 months was described by others. Ursano et al (1999) reported that 34% of MVC victims met the criteria for PTSD one month after the crash and that nearly half had recovered by 6 months (Ursano et al, 1999). While a natural improvement over time appears to occur, our finding of new cases diagnosed after 6 months suggests that populations at risk may require longer term surveillance. This finding also suggests that other factors of the “after trauma” experience may have an influence in the development, establishment or resolution of PTSD. Future research should elucidate these factors. This is consistent also with the decreased influence of constitutional factors (i.e. gender and history of depression) among those that screen positive for PTSD at one year points and the increased influence of factors related to traumatic experiences (both history or current) and possibly the experience of the injury and its aftermath (i.e. pain and disability).

Given the fact that the main goal of the CIREN project is to determine the biomechanical aspects of motor vehicle-related injuries, we did not employ a standardized diagnostic instrument for a definitive diagnosis of PTSD [ie. the Clinician Administrated PTSD Scale (CAPS) (Weathers et al, 1992)]. However, a strength of our data is that all interviews were conducted by the same person, who is a clinical social worker and has been involved with the CIREN project for many years.

While the questions asked during the interview included the PTSD diagnostic symptom criteria categories outlined in the DSM-III-R (APA 1987), our definition of positive screen did not completely overlap with PTSD diagnosis. Table 6 shows the prevalence of symptom clusters (as per DSM III-R criteria) and their association with positive PTSD screen at 6 and 12 months. The DSM III-R criteria require individuals that were exposed to a traumatic event (criteria A) to experience re-experiencing symptoms (criteria B), avoidance or numbing symptoms (criteria C) and increased arousal symptoms (criteria D) for at least a month (criteria E).

Table 6

DSM III-R criteria prevalence^* (%) at 6 and 12 months.

	Screened at 6 months (n=367)	Positive PTSD screen at 6 months (n=101)	Screened at one year (n=317)	Positive PTSD screen at one year (n=77)
B (re-experiencing)	26	82	20	75
C (avoidance and numbing)	34	93	33	94
D (increased arousal)	64	100	52	96
B + C	20	75	17	70
B + D	25	82	19	73
C + D	32	97	29	92
All (B + C + D)	20	75	17	69

Open in a separate window

^*Our screening tool did not include nor completely overlap with all symptoms described in the DSM III-R.

Our screening tool did not include, nor completely overlap with all symptoms described in the DSM III-R. For table 6 we reclassified the symptoms in our screening questions matching DSM III-R criteria the following way: (1) frequent, recurrent, re-experiencing of the event by intrusive, intense thoughts, flashbacks; physical reactions when thinking or reminded of the crash, B; (2) persistent avoidance and fear of stimuli associated with the trauma, C; (3) hyper-alertness, hyper-vigilance, being on guard, angry outbursts, D; (4) diminished, and serious loss of interest/emotional numbing, inability to have loving feelings because of crash, C; (5) significant sleep disturbances, D; (6) excessive cognitive difficulties in memory, concentration, avoiding thoughts of crash, D; and (7) survivor’s guilt; not a DSM III-R criteria.

As shown in the table most of the patients fulfilling our case definition criteria for PTSD also fulfill the DSM III-R definition at 6 months (75%) and one year (69%). Strictly speaking, the remaining patients not fulfilling the DSM III- R definition but positive in our screen should be considered to have sub-syndromal PTSD instead of PTSD diagnosis.

Nevertheless, our PTSD assessment may not be reproduced in the typical clinical or research setting in the absence of a dedicated Clinical social worker, psychologist or psychiatrist. Other limitations to our study include the fact that our cohort is a select group of patients where the male to female ratio and the socioeconomic status of the enrolled participants differs from the general population of trauma patients.

Because of the study purpose, we also did not assess coping styles, avoidance, psychological stress subsequent to hospital discharge, dissociation, delayed peak emotional reactions to traumatic events and psychiatric co-morbidities other than depression and anxiety. The inability to assess these factors, known to be of importance in the development and maintenance of PTSD symptomatology (Fitzharris et al, 2006; Batten, Orsillo and Walser, 2005; Zatzick et al, 1997; Griffin, Resick and Mechanic, 1997; Fullerton et al, 2000; Brewin, Andrews and Valentine, 2000; Cardena and SPiegel,1993; Koopman, Classen and Spiegel, 1994; Gilboa-Schectman and Foa, 2001), is another limitation of this study.

Finally, another limitation in our report is a potentially large degree of colinearity in the assessment of the cognitive, emotional and behavioral outcomes (table 5) introduced by the behavioral, emotional and cognitive components of the questions used to define the presence of PTSD.

CONCLUSIONS

PTSD was seen to develop in approximately 28% and 24% at 6 and 12 months, respectively, of this sample of moderately and severely injured victims of motor vehicular crashes (20% and 17% if a stricter DSM III-R based definition is used). The most important predictors of PTSD at 6 months are female gender, history of depression, prior assault/trauma and the occurrence of a fatality during the crash. Being responsible for the crash, through an unknown mechanism, appears to confer a protective effect although not after adjustment of other characteristics. At one year, prior assault/trauma and the occurrence of a fatality during the crash become stronger predictors of PTSD while the other mentioned factors’ association with PTSD decrease in magnitude and became not significant.

Patients with one or more of these factors should be closely watched for the development of PTSD and would represent good candidates to study preventive interventions.

Acknowledgments

This work was performed for the Crash Injury Research and Engineering Network (CIREN) Project at the University of Maryland, National Study Center for Trauma and Emergency Medical Systems in cooperation with the United States Department of Transportation/National Highway Safety Administration (USDOT/NHTSA). Funding has been provided by NHTSA under Cooperative Agreement Number DTNH22-05-H-01001. Views expressed are those of the authors and do not represent the views of the NHTSA.

REFERENCES

American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders, Third Edition - revised (DSM-III-R), Washington, 1987
Batten S, Orsillo, Walser . Acceptance- and mindfulness-based approaches to the treatment of posttraumatic stress disorder. In: Orsillo SM, Roemer L, editors. Acceptance and Mindfulness-Based Approaches to Anxiety: Conceptualization and Treatment. New York: Plenum; 2005. pp. 241–269. [Google Scholar]
Blanchard EB, Hickling EJ, Freidenberg BM, et al. Two studies of psychiatric morbidity among motor vehicle accident survivors 1 year after the crash. Behaviour Research and Therapy. 2004;42(5):569–583. [PubMed] [Google Scholar]
Blanchard EB, Hickling EJ. After the crash. American Psychological Association. 1997 [Google Scholar]
Blanchard EB, Hickling EJ, Barton KA, et al. One-year prospective follow-up of motor vehicle accident victims. Behav Res Ther. 1996;34(10):775–86. [PubMed] [Google Scholar]
Blanchard EB, Hickling EJ, Taylor AE, et al. Psychiatric morbidity associated with motor vehicle accidents. J Nervous and Mental Disease. 1995;183:495–504. [PubMed] [Google Scholar]
Brady KT, Killen TK, Brewerton T, et al. Comorbidity of Psychiatric Disorders and PTSD. J Clin Psychiatry. 2000;61:22–32. [PubMed] [Google Scholar]
Breslau N, Davis GC. Posttraumatic stress disorder in an urban population of young adults: risk factors for chronicity. Am J Psychiatry. 1992;149:671–675. [PubMed] [Google Scholar]
Brewin CR, Andrews B, Valentine JD. Meta-analysis of risk factors for posttraumatic stress disorder in trauma-exposed adults. J Consult Clin Psychol. 2000;68(5):748–66. [PubMed] [Google Scholar]
Bryant RA, Harvey AG. Initial posttraumatic stress responses following motor vehicle accidents. J Trauma Stress. 1996;9:223–234. [PubMed] [Google Scholar]
Bryant R, Harvey A. Psychological impairment following motor vehicle accidents. Aust J Public Health. 1995;19:185–188. [PubMed] [Google Scholar]
Cardeña E, Spiegel D. Dissociative reactions to the San Francisco Bay Area earthquake of 1989. Am J Psychiatry. 1993;150(3):474–8. [PubMed] [Google Scholar]
Delahanty DL, Herberman HB, Craig KJ, et al. Acute and chronic distress and posttraumatic stress disorder as a function of responsibility for serious motor vehicle accidents. Journal of Consulting and Clinical Psychology. 1997;65:560–567. [PubMed] [Google Scholar]
Dischinger PC, Read KM, Kufera JA, et al.Consequences and costs of lower-extremity injuries. DOT HS 809 871, 2005 [PMC free article] [PubMed]
Donahue RG. Predictors of posttraumatic psychological distress in young adult motor vehicle accident survivors: gender similarities and differences (dissertation) State University of New York at Albany. 2006.
Feinstein A, Dolan R. Predictors of post-traumatic stress disorder following physical trauma: an explanation of the stressor criterion. Psychol Med. 1991;21:85–91. [PubMed] [Google Scholar]
Fitzharris M, Fildes B, Charlton J, et al. General health status and functional disability following injury in traffic crashes. Traffic Inj Prev. 2007;8(3):309–320. [PubMed] [Google Scholar]
Fitzharris M, Fildes B, Charlton J. Anxiety, acute- and post-traumatic stress symptoms following involvement in traffic crashes. Annu Proc Assoc Adv Automot Med. 2006;50:283–301. [PMC free article] [PubMed] [Google Scholar]
Fullerton CS, Ursano RJ, Epstein RS, et al. Peritraumatic dissociation following motor vehicle accidents: relationship to prior trauma and prior major depression. J Nerv Ment Dis. 2000;188(5):267–72. [PubMed] [Google Scholar]
Gentilello L. Alcohol interventions in trauma centers: the opportunity and the challenge. J Trauma. 2005;59:S18–S20. [PubMed] [Google Scholar]
Gil S, Caspi Y, Ben-Ari IZ, et al. 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(5):963–9. [PubMed] [Google Scholar]
Gilboa-Schechtman E, Foa EB. Patterns of recovery from trauma: the use of intraindividual analysis. J Abnorm Psychol. 2001;110(3):392–400. [PubMed] [Google Scholar]
Green BL. Psychological research in traumatic stress: An update. J Traumatic Stress. 1994;7:343–361. [PubMed] [Google Scholar]
Green M, McFarlane W, Hunter C, et al. Undiagnosed posttraumatic stress disorder following motor vehicle accidents. Med J Aust. 1993;159:529–534. [PubMed] [Google Scholar]
Griffin MG, Resick PA, Mechanic MB. Objective assessment of peritraumatic dissociation: psychophysiological indicators. Am J Psychiatry. 1997;154(8):1081–8. [PMC free article] [PubMed] [Google Scholar]
Hickling EJ, Blanchard EB, Buckley TC, et al. Effects of attribution of responsibility for motor vehicle accidents on severity of PTSD symptoms, ways of coping, and recovery over six months. J Trauma Stress. 1999 Apr;12(2):345–53. [PubMed] [Google Scholar]
Holbrook TL, Anderson JP, Sieber WJ, et al. Outcome after major trauma: 12-month and 18-month follow-up results from the Trauma Recovery Project. J Trauma. 1999 May;46(5):765–71. [PubMed] [Google Scholar]
Holbrook TL, Anderson JP, Sieber WJ, et al. Outcome after major trauma: discharge and 6-month follow-up results from the Trauma Recovery Project. J Trauma. 1998 Aug;45(2):315–23. [PubMed] [Google Scholar]
Irish L, Ostrowski SA, Fallon W, et al. Trauma history characteristics and subsequent PTSD symptoms in motor vehicle accident victims. J Traumatic Stress. 2008;21(4):377–384. [PubMed] [Google Scholar]
Jang KL, Stein MB, Taylor S, et al. Exposure to traumatic events and experiences: aetiological relationships with personality function. Psychiatry Res. 2003;120(1):61–9. [PubMed] [Google Scholar]
Joint Commission on Accreditation of Healthcare Organizations (JCAHO) Pain Assessment and Management, an Organizational Approach, 2000
Kessler RC, Sonnega A, Bromet E, et al. Posttraumatic stress disorder in the National Comorbidity Survey. Arch Gen Psychiatry. 1995;52:1048–1060. [PubMed] [Google Scholar]
Koenen KC, Fu QJ, Ertel K, et al. Common genetic liability to major depression and posttraumatic stress disorder in men. J Affect Disord. 2008;105(1–3):109–15. [PMC free article] [PubMed] [Google Scholar]
Koopman C, Classen C, Spiegel D. Predictors of posttraumatic stress symptoms among survivors of the Oakland/Berkeley, Calif., firestorm. Am J Psychiatry. 1994;151(6):888–94. [PubMed] [Google Scholar]
Kuch K, Cox BJ, Evans RJ. Posttraumatic Stress Disorder and motor vehicle accidents: A multidisciplinary overview. Canadian J Psych. 1996;41:429–434. [PubMed] [Google Scholar]
Kuhn E, Blanchard EB, Hickling EJ. Posttraumatic stress disorder and psychosocial functioning within two samples of MVA survivors. Behav Res Ther. 2003 Sep;41(9):1105–12. [PubMed] [Google Scholar]
Kupchik M, Strous RD, Erez R, et al. Demographic and clinical characteristics of motor vehicle accident victims in the community general health outpatient clinic: a comparison of PTSD and non-PTSD subjects. Depression and Anxiety. 2007;24(4):244–250. [PubMed] [Google Scholar]
Landsman I, Baum C, Arnkoff D, et al. The psychosocial consequences of traumatic injury. J Behav Med. 1990;13:561–581. [PubMed] [Google Scholar]
MacKenzie EJ, McCarthy ML, Ditunno JF, et al. Using the SF-36 for characterizing outcome after multiple trauma involving head injury. J Trauma. 2002;52(3):527–534. [PubMed] [Google Scholar]
Malt UF, Blikra G. Psychosocial consequences of road accidents. Eur Psychiatry. 1993;8:227–228. [Google Scholar]
Malt U. The long-term psychiatric consequences of accidental injury: a longitudinal study of 107 adults. Br J Psychiatry. 1988;153:810–818. [PubMed] [Google Scholar]
Matthews LR. Work potential of road accident survivors with post-traumatic stress disorder. Behaviour Research and Therapy. 2005;43:475–483. [PubMed] [Google Scholar]
Mayou R, Bryand B, Duthie R. Psychiatric consequences of road traffic accidents. BMJ. 1993;307:647–651. [PMC free article] [PubMed] [Google Scholar]
McHugh PR, Treisman G. PTSD: A problematic diagnostic category. J Anxiety Disorders. 2007;21(2):211–222. [PubMed] [Google Scholar]
Michaels AJ, Michaels CE, Moon CH, et al. Posttraumatic stress disorder after injury: impact on general health outcome and early risk assessment. J Trauma. 1999;47:460–466. [PubMed] [Google Scholar]
Norris FH. Epidemiology of trauma: frequency and impact of different potentially traumatic events on different demographic groups. J Consult Clin Psychol. 1992;60:409–418. [PubMed] [Google Scholar]
Perry S, Difede J, Musngi G, et al. Predictors of posttraumatic stress after burn injury. Am J Psychiatry. 1992;149:931–935. [PubMed] [Google Scholar]
Read KM, Kufera JA, Dischinger PC, et al. Life-altering outcomes after lower extremity injury sustained in motor vehicle crashes. J Trauma. 2004;57(4):815–823. [PubMed] [Google Scholar]
Ryb GE, Dischinger PC, Kufera JA, et al.J Trauma 2007November6351000-5. Delta V, principal direction of force, and restraint use contributions to motor vehicle crash mortality. [PubMed] [Google Scholar]
Ryb GE, Dischinger PC. Injury severity and outcome of overweight and obese patients after vehicular trauma: a crash injury research and engineering network (CIREN) study. J Trauma. 2008 Feb;64(2):406–11. [PubMed] [Google Scholar]
Satel S. Post-traumatic stress disorder and Iraq veterans. Testimony to House Committee on Veterans Affairs. American Enterprise Institute for Public Policy Research. 2004 Mar 11; [Google Scholar]
Sbordone RJ, Liter JC. Mild traumatic brain injury does not produce post traumatic stress disorder. Brain Inj. 1995;9:405–412. [PubMed] [Google Scholar]
Shalev AY, Freedman S, Peri T, et al.Prospective study of posttraumatic stress disorder and depression following trauma Am J Psychiatry 1555630–7.1998. (a) [PubMed] [Google Scholar]
Shalev AY, Sahar T, Freedman S, et al.A prospective study of heart rate response following trauma and the subsequent development of posttraumatic stress disorder Arch Gen Pscychiatry 55553–559.1998. (b) [PubMed] [Google Scholar]
Shalev A, Peri T, Canetti L, et al. Predictors of PTSD in injured trauma survivors: a prospective study. Am J Psychiatry. 1996;153:219–225. [PubMed] [Google Scholar]
Storr CL, Ialongo NS, Anthony JC, et al. Childhood antecedents of exposure to traumatic events and posttraumatic stress disorder. Am J Psychiatry. 2007;164(1):119–25. [PubMed] [Google Scholar]
Ursano RJ, Fullerton CS, Epstein RS, et al. Acute and chronic posttraumatic stress disorder in motor vehicle accident victims. Am J Psychiatry. 1999;156:589–595. [PubMed] [Google Scholar]
Wald J, Taylor S, Fedoroff IC.The challenge of treating PTSD in the context of chronic pain “Advances in treatment of posttraumatic stress disorder: cognitive-behavioral perspectives”, Chapter 3, 197–222.Taylor Steven.Springer NY, NY: 2004 [Google Scholar]
Ware JJ, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). 1. Conceptual framework and item selection. Med Care. 1992;30:473–483. [PubMed] [Google Scholar]
Weathers F, Blake D, Krinsley K, et al.The clinician administered PTSD scale. National Center for PTSD, Behavioral Science Division.
Whooley MA, Avins AL, Miranda J, et al. Case-finding instruments for depression: two questions are as good as many. J Gen Intern Med. 1997;12:439–445. [PMC free article] [PubMed] [Google Scholar]
Zatzick DF, Kan S, Muller H, et al. Predicting posttraumatic distress in hospitalized trauma survivors with acute injuries. Am J Psychiatry. 2002;159(6):941–946. [PubMed] [Google Scholar]
Zatzick DF, Russo J, Pitman RK, et al. Reevaluating the association between emergency department heart rate and the development of posttraumatic stress disorder: a public health approach. Biol Psychiatry. 2005;57:91–95. [PubMed] [Google Scholar]

Articles from Annals of Advances in Automotive Medicine / Annual Scientific Conference are provided here courtesy of Association for the Advancement of Automotive Medicine