Chronic pain and head injury following motor vehicle collisions: a double whammy or different sides of a coin.

Chronic pain and head injury are common and burdensome sequelae of motor vehicle collisions. The aim of this study was to compare differences in physical injury and functional impairment, psychological distress and pain coping in head injured and non-head injured chronic pain persons subsequent to motor vehicle collisions. Two groups of 54 participants matched in terms of age, gender, and years of formal education underwent a psychological-legal assessment. As part of the assessment, participants completed the Multidimensional Pain Inventory, Sickness Impact Profile, Minnesota Multiphasic Personality Inventory-2, and Coping Strategies Questionnaires. Select scales from questionnaires were combined and underwent multivariate analyses of covariance to test the effects of pain sites at the time of psychological-legal assessment (low, high) and head injury status (head injured and non-head injured chronic pain). Overall, some differences between the two groups were noted but the results did not strongly support the hypothesis that head injured chronic pain participants have a greater physical or psychological burden than non-head injured chronic pain participants. The results suggest the import of assessing and managing pain sites and pain severity in persons injured in motor vehicle collisions.

Keywords: injury; psychological distress; pain coping; Chronic pain

Motor vehicle collisions (MVCs) result in consequences that are significant and costly to both individuals and society. Every year in the US there are more than 40,000 persons killed and more than 5 million persons injured in MVCs (Goldberg, [16]). The total economic cost of MVCs in the US for the year 2000 was almost 231 billion dollars (Goldberg, [16]). The World Health Organization has estimated that by the year 2020 MVCs will rank second only heart disease and depression in terms of disability (Murray & Lopez, [32]).

Independently, chronic pain and head injury are significant trauma sequelae. One in five Americans suffers from chronic pain (Joranson & Leitman, [22]). The direct and indirect costs of treating chronic pain are considerable, with costs exceeding $125 billion per year (Turk, Okifuji, & Kaluaokalani, [43]). Following MVCs, traumatized chronic pain persons are more likely to experience physical and functional impairment, psychological distress, and maladaptive coping (Duckworth & Iezzi, [10]). In addition, more psychologically distressed chronic pain patients of various etiologies (e.g., MVCs, work-related injuries, and slip and falls) present with significantly more neurocognitive difficulties than less psychologically distressed chronic pain patients (Iezzi et al., [19]). Even after controlling for age, education, psychological distress, and pain severity, chronic pain patients experience significant difficulties with memory functioning (Iezzi et al., [20]).

MVCs are the leading cause of all head injuries (Krause & McArthur, [26]). Mild traumatic brain injuries comprise 90% of all head injuries (Thornhill et al., [42]). The costs associated with head injury have been estimated at $10 billion a year for new injuries (National Institute of Health Consensus Development Panel on Rehabilitation of Persons with Traumatic Brain Injury, [33]). Persons suffering from mild traumatic brain injury are more likely to have prolonged generalized disability and are less likely to return to work than other motor vehicle collision (MVC) samples (Bazarian et al., [5]; Friedland & Dawson, [13]; Thornhill et al., [42]). In addition to the burden of head injury, pain is a common consequence of head injury (Gerber & Schraa, [15]; Nicholson & Martelli, [34]).

Although chronic pain and head injury can significantly and independently impact physical, psychological, and neurocognitive functioning in persons involved in MVCs, the extent of repercussions incurred when both chronic pain and head injury are present in the same individual is not as clear (Richter et al., [35]). A study by Anderson et al. ([4]) noted that a brain injury could be obscured by the presence of chronic pain. They found seven cases in a sample of 67 consecutive cases referred to a chronic pain program had an undiagnosed brain injury. The authors noted that misdiagnosis of chronic pain and head injury occurs because of the similarity of symptoms between the two groups, health care providers are not used to thinking of these two conditions as co-occurring, and the traumatic effects of injury may be progressive and delayed.

A study by Iverson and McCracken ([21]) revealed the considerable overlap between chronic pain and postconcussive symptoms. The authors conducted an evaluation of physical, cognitive, and psychological symptoms in 170 chronic pain patients. The authors found that 80% of chronic pain patients endorsed three or more symptoms in Category C of the diagnostic criteria for postconcussive disorder and 39% met the definition of postconcussive disorder according the Diagnostic and Statistical Manual of Mental Disorders-IV (American Psychiatric Association, [3]).

Alexander ([1]) analysed the neurological and psychological functioning of 23 mild and 13 severe head injured cases. Mild head injured cases differed in their symptom profile from severe head injured cases. Patients with mild head injury had a higher frequency of periodic and chronic headaches and depression. The investigators concluded that cases with postconcussive syndrome were more likely to be suffering from chronic pain and depression than from a specific neurocognitive disability.

Uomoto and Esselman ([45]) examined the interaction of chronic pain and head injury in 104 head injured cases recruited from an outpatient brain injury rehabilitation programme. The authors indicated that 95% of mild head injured cases were suffering from a chronic pain condition of one kind or another, with posttraumatic headaches being the most common. Chronic neck and shoulder pain was identified in 51% of mild head injured cases. Of interest, 78% of cases with moderate or severe head injury did not experience a chronic pain problem. The authors concluded that excess disability in the mild head injured group was due, in part, to chronic pain.

Landy ([27]) examined 414 medical-legal cases identified as suffering a closed head injury. The main complaints reported by cases included headache, loss of concentration, and poor memory. A subgroup (n = 380) completed questionnaires post-settlement. Results indicated that the less severe the head or neck injury, the more likely these cases were to suffer symptoms of posttraumatic headaches or neck symptoms.

Although not examining the co-occurrence of chronic pain and head injury per se, a study by Curran et al. ([8]) compared coping strategies and emotional outcome in a head injured group (n = 88) with an orthopedically injured group (n = 40). All subjects had been involved in a MVC or work-related injury. Although there were significant levels of psychological distress and psychological distress was associated with maladaptive coping in both groups, there were surprisingly few differences between the two groups.

In sum, chronic pain and head injury are common sequelae of MVCs. Each condition can independently result in marked physical and psychological consequences and place a significant burden on coping; however, relatively little is known about the repercussions of MVCs when both chronic pain and head injury are present in the same sample. Compared to a chronic pain sample, this study specifically attempts to address if a combined presentation of chronic pain and head injury represents a 'double whammy' (twice the physical and psychological burden) or represents 'different sides of a coin' (i.e., similar but different for other reasons).

Participants were selected from a larger sample of more than 300 chronic pain patients referred for psychological-legal assessment to the first author over a ten-year period (1991 – 2001). Based on a review of medical briefs, head injured chronic pain (HICP) participants were identified as having a head injury according to criteria outlined by the American Congress of Rehabilitation Medicine ([2]): 1) reported loss of consciousness (LOC); 2) reported posttraumatic amnesia (PTA); or 3) Glasgow Coma Scale score (GCS; Teasdale & Jennett, [41]). In addition to American Congress Rehabilitation Medicine criteria, participants were included in the HICP group if medical briefs contained a neuropsychological evaluation documenting head injury. From a total of 59 head injured chronic pain (HICP) participants, 55 participants with complete assessment data were selected for inclusion in the current study. Among HICP participants, 40 participants (73%) were identified as having mild head injury, with an additional 12 participants (22%) identified as having moderate to severe head injury, and 3 participants (5%) having documentation that rendered head injury level (mild, moderate, and severe) inconclusive.

Fifty-five non-head injured chronic pain (CP) participants, matched to HICP participants on age, gender, and years of formal education, were also included in the current study. For all study participants, chronic pain symptoms were evidenced subsequent to involvement in one or more MVCs. Chronic pain sites were defined according to the International Association for the Study of Pain coding scheme (Merskey & Bogduk, [29]). The pain sites most commonly identified by participants at the time of MVC included neck (80.0%), upper back and shoulders (72.2%), head (70.0%), low back (45.5%), and lower limbs (44.6%). The pain sites most commonly identified by participants at the time of psychological-legal assessment included upper back and shoulders (70.9%), neck (69.1%), head (68.2%), low back (63.6%), and lower limbs (51.8%). Participants were excluded from study if they experienced chronic pain subsequent to an event other than a motor vehicle collision; if they experienced chronic but episodic pain (e.g., migraines); if they were not involved in litigation; and if they reported symptoms of psychosis or major depression not secondary to chronic pain.

A first match procedure was used to identify CP participants who matched HICP chronic pain participants on age, gender, and years of formal education. Univariate and chi-square analyses confirmed the adequacy of the matching procedure, these analyses revealing no significant differences between CP and HICP participants on age, gender, or formal years of education. These two groups of chronic pain participants also evidenced no differences in ethnic identity, marital status, employment status at the time of the MVC, employment status at the time of psychological-legal assessment, or source of referral for psychological-legal assessment.

The total sample of 110 CP and HICP participants ranged in age from 19 to 71 years (M = 39.7, SD = 11.8) and reported an average of 12.8 years of formal education (SD = 2.4). Referrals for psychological-legal assessment were received from lawyers (48.8%), rehabilitation consultants (17.3%), insurance companies (16.4%), medical specialists (5.5%), and general medical practitioners (5.5%). The majority of participants were Caucasian (90.0%), male (52.7%), and married (59.1%). At the time of the collision, 91.8% of the participants were working or studying full-time, 2.7% were working or studying part-time, 1.8% were unemployed, 0.9% had not returned to work but had a job waiting, and 2.7% were receiving disability due to a pre-existing medical condition. At the time of psychological-legal assessment, 16.4% of the participants were working or studying full-time, 18.2% were working or studying part-time, 57.3% were unemployed, 6.4% had not returned to work but had a job waiting, and 1.8% were receiving disability due to a pre-existing medical condition. For most non-working participants, the experience of MVC-related pain accounted for their unemployed status.

The Multidimensional Pain Inventory (MPI) is a 61-item inventory designed to assess subjective experience and perception of pain, the effects of pain on functioning, and responses of significant others (Kerns et al., [25]). Items are categorized into 13 scales (Pain severity, Interference, Life control, Affective distress, Support, Punishing responses, Solicitous responses, Distracting responses, Household chores, Outdoor work, Activities away from home, Social activities, and General activity level). Internal consistency estimates for the 13 scales ranges from.70 to.90 and test-retest reliability estimates for these scales range from.62 to.91 (Kerns et al., [25]).

The Sickness Impact Profile (SIP) is a 136-item measure of functional and lifestyle impairment across 12 dimensions of functioning (e.g., Mobility, Work, Household management, Social interactions, and Recreational pursuits; Bergner et al., [6]). The dimensions are combined to form Physical, Psychosocial, Other, and Overall Disability scales. The overall alpha is.94 in a host of medical presentations (Bergner et al., [6]) and.93 in head injured individuals (Smith et al., [39]). Test-retest reliability ranges from.69 to.87 over a three-week period (Debruin et al., [9]). The validity and utility for the SIP has been demonstrated in a chronic pain (Follick et al., [12]) and in head injury samples (Moore et al., [30]).

The Minnesota Multiphasic Personality Inventory-2 (MMPI-2) is a 567-item, true-false, self-report instrument designed to assess levels of emotional distress and long-standing patterns of behaviour. The MMPI-2 has three validity scales (Lie-L, Frequency-F, and Correction-K) and 10 clinical scales (Hypochondriasis-Hs, Depression-D, Hysteria-Hy, Psychopathic Deviate-Pd, Masculinity-Femininity-Mf, Paranoia-Pa, Psychasthenia-Pt, Schizophrenia-Sc, Mania-Ma, and Social Introversion-Si; Butcher et al., [7]). The MMPI-2 has been used extensively in the evaluation of chronic pain patients (Vendrig, [46]), with scales evaluating Hy, D, Hs, Pt, and Sc considered to be particularly relevant to the evaluation of psychological distress among chronic pain patients.

The Coping Strategies Questionnaire (CSQ) is a 48-item inventory designed to assess eight types of coping strategies (Diverting attention, Reinterpreting pain sensations, Ignoring pain sensations, Coping self-statements, Praying and hoping, Catastrophizing, Increasing behavioural activities, and Increasing pain behaviours; Rosenstiel & Keefe, [36]). Internal consistency coefficients range from 0.28 to 0.85 (Rosenstiel & Keefe, [36]). Adequate indices of test-retest reliability and concurrent validity have been reported in a number of studies (Haythornthwaite et al., [17]; Keefe et al., [24]).

A total of nine scales from the four assessment instruments were combined to represent physical injury and functional impairment, psychological distress, and pain coping domains, and were included in multivariate models testing the effect of pain sites at psychological-legal assessment and head injury status. The MPI Pain interference and Pain severity scales and SIP Physical scale comprised the physical injury and functional impairment domain. These scales were chosen based on empirical literature and clinical experience demonstrating utility in assessing differences in functional impairment in chronic pain (Follick et al., [12]; Iezzi et al., [18]; Kerns et al., [25]) and head injury samples (Jurkovich et al., [23]; Moore et al., [30]), and in differentiating traumatized from non-traumatized persons in chronic pain samples (Duckworth & Iezzi, [10]; Sherman et al., [38]).

The MPI Affective distress scale, MMPI-2 Pt scale, and SIP Psychosocial scale comprised the psychological distress domain. The MPI Affective distress and SIP Psychosocial scales were chosen based on similar reasons as for MPI Pain interference and MPI Pain severity and SIP Physical scales. In terms of the MMPI-2, only the Pt scale was included in the model testing the effects of pain sites at psychological-legal assessment and head injury status on psychological distress variables. The Pt scale is a good indicator of psychological turmoil (mostly in the form of anxiety) and its selection was based on 1) its moderate correlation with psychological distress scales taken from other measures, i.e., MPI Affective distress (r = .31) and SIP Psychosocial (r = .53); and 2) its relatively high intercorrelation with the other pain-relevant MMPI-2 scales, i.e., Hs (r = .55), D (r = .69), Hy (r = .49), and Sc (r = .83).

The CSQ Catastrophizing, Coping self-statements, and Praying and hoping scales comprised the pain coping domain. The CSQ Catastrophizing, Coping self-statements, and Praying and hoping scales were selected for inclusion in multivariate analyses testing the effects of pain sites at psychological-legal assessment and head injury status. The combination of pain coping variables was selected based on findings in the empirical literature that support the primary contribution of these particular coping strategies to the pain experience and functional status of chronic pain patients (Geisser et al., [14]; Haythornthwaite et al., [17]; Turner et al., [44]).

All participants underwent a two-hour semi-structured interview. The semi-structured interview covered collision details, pain symptoms, functional and lifestyle impairment, changes in psychological status, and pre-collision history. Participants completed psychological testing in three to four hours. Participants returned for a one-hour feedback session of test results, assessment conclusions, treatment recommendations, and prognosis within one month of their psychological assessment. All participant files were initially coded and reduced by the first author, with the second author coding 20% of participant files. An inter-rater reliability coefficient of.91 was obtained across a select group of assessment variables (see Duckworth & Iezzi, [10], for details regarding the calculation of inter-rater reliability).

The SAS System for Windows ([37] – [37]) was used for all analyses. Prior to conducting primary analyses, relevant demographic, MVC-related, physical injury and functional impairment, psychological distress, and pain coping variables were examined for accuracy of data entry, missing values, univariate outliers, and normality of distribution (Tabachnick & Fidell, [40]). Univariate analyses revealed significant skewness and kurtosis for two variables relevant to the primary analyses: years of formal education and MPI Affective distress. To avoid deletion of entire cases while correcting for skewness and kurtosis, these variables were first submitted to log linear and square root transformations. These transformation strategies did not adequately correct skewness or kurtosis observed for years of formal education or MPI Affective distress. Raw scores for these two variables were then transformed to z-scores and two univariate outliers were identified. Deletion of univariate outliers resulted in correction of skewness and kurtosis for the two variables. The final sample was comprised of 54 CP and 54 HICP participants. Variables representing elapsed time since MVC and pain sites at time of MVC were also observed to display significant skewness and/or kurtosis. However, because these variables were not part of any a priori hypotheses and because neither variable was significantly associated with variables representing physical injury and functional impairment, psychological distress, and pain coping, uncorrected scores were used in providing descriptive information related to these two variables.

Multivariate analyses of covariance (MANCOVAs) were performed to test the effects of pain sites at psychological-legal assessment (low, high) and head injury status (CP, HICP) on physical injury and functional impairment, psychological distress, and pain coping domains. Age and ethnicity were included as covariates in models testing the effects of pain sites at psychological-legal assessment and head injury status on physical injury and functional impairment, psychological distress, and pain coping domains based on the examination of correlations between demographic variables and variables representing physical injury and functional impairment, psychological distress, and pain coping. Bivariate correlations between pain sites at the time of psychological-legal assessment variable and variables representing physical injury and functional impairment, psychological distress, and pain coping suggested the inclusion of this variable as a covariate in planned multivariate analyses. However, heterogeneity of regression was observed across CP and HICP participants, rendering pain sites at assessment variable unacceptable as a covariate. Rather than excluding this variable from multivariate analyses, the decision was made to transform this continuous variable into a grouping variable, using the sample-specific median value for pain sites at psychological-legal assessment (Median = 4.00) to categorize CP and HICP as experiencing low (4 or fewer) or high (greater than 4) pain sites. Because of the significant number of comparisons performed in establishing group differences on sociodemographic and MVC-related variables, and because of the number of associations tested among sociodemographic and MVC-related variables and variables representing physical injury and functional impairment, psychological distress, and pain coping, a more conservative significance level of α = .01 was applied in evaluating these comparisons.

Univariate analyses of variance and chi-square analyses were performed to compare CP and HICP participants on MVC-related variables. A significant group difference was observed in the total number of pain sites identified at the time of MVC, F(1, 109) = 8.77, p = .004, with HICP participants identifying more pain sites at the time of MVC than CP participants (4.1 pain sites versus 3.3 pain sites). Chi-square analyses revealed significant group differences in the proportions of CP and HICP participants reporting loss of consciousness at the time of MVC, χ2(108) = 56.51, p = .0001, memory loss consequent to MVC, χ2(108) = 59.11, p = .0001, hospitalization consequent to the MVC, χ2(108) = 23.16, p = .0001, and surgery consequent to the MVC, χ2(108) = 17.70, p = .0001. Approximately 90% of HICP participants reported loss of consciousness consequent to MVC involvement compared to 16.7% of CP participants. Memory loss was reported by 92.6% of HICP participants compared to 22.2% of CP participants. Hospitalization and surgery were also more frequent among HICP participants than CP participants. Results of analyses comparing CP and HICP participants on sociodemographic and MVC-related variables are presented in Table I.

Table I. Descriptive statistics for demographic and MVC variables by head injury group.

Correlations among sociodemographic and MVC-related variables and variables representing physical injury and functional impairment, psychological distress, pain coping revealed several significant associations. Age was significantly correlated with MMPI-2 Pt scores, with younger CP participants reporting more anxiety than do older CP participants. Ethnic identity was significantly correlated with SIP Physical and Psychosocial scale scores and with CSQ Catastrophizing and Praying and hoping scale scores. Non-Caucasian participants reported greater physical injury and functional impairment, greater psychological distress and psychosocial dysfunction, and greater use of catastrophizing and praying and hoping as pain coping strategies. The total number of pain sites identified at the time of psychological-legal assessment was correlated with SIP Physical and Psychosocial, MPI Interference and Pain severity, MMPI-2 Pt, and CSQ Catastrophizing scale scores. The report of more pain sites at psychological-legal assessment was associated with greater physical injury and functional impairment, greater psychosocial dysfunction and anxiety, and greater use of catastrophizing as a pain coping strategy. Results of correlational analyses testing associations among sociodemographic, MVC-related, physical injury and functional impairment, psychological distress, and pain coping variables are presented in Table II.

Table II. Correlations among demographic, MVC-related, physical injury and impairment, psychological distress, and pain coping variables.

Table II. Correlations among demographic, MVC-related, physical injury and impairment, psychological distress, and pain coping variables.

Table II. Correlations among demographic, MVC-related, physical injury and impairment, psychological distress, and pain coping variables.

A 2 × 2 between groups' multivariate analysis of covariance (MANCOVA) was performed to test the effects of pain sites at psychological-legal assessment (low, high) and head injury status (CP, HICP) on the combination of physical injury and functional impairment variables (MPI Pain interference, MPI Pain severity, and SIP Physical). Ethnicity was included as a covariate in the model due to significant bivariate associations between this variable and variables representing physical injury and functional impairment. After adjusting for the effect of ethnicity, a significant multivariate effect of pain sites at psychological-legal assessment was revealed but no significant effect of head injury status or the interaction between pain sites at psychological-legal assessment and head injury status was revealed. To investigate the unique contribution of physical injury and functional impairment variables to differences observed between participants reporting low and high pain sites at psychological-legal assessment, a Roy-Bergman stepdown analysis was performed on the prioritized physical injury and functional impairment variables, with the alpha level for each test set at.017. Adjusting for the effect of ethnicity, MPI Pain severity scale scores were observed to contribute uniquely to the prediction of differences between participants reporting low and high pain sites at psychological-legal assessment, stepdown F(1, 107) = 14.24, p = .0003. Consistent with univariate findings, stepdown analyses revealed no unique contribution of either SIP Physical scale scores or MPI Interference scale scores to the prediction of differences between participants reporting low and high pain sites at psychological-legal assessment after ethnicity and the pattern of differences measured by MPI Pain severity scale scores were entered. Results of univariate and stepdown analyses for the significant main effect of pain sites at psychological-legal assessment are presented in Table III, with variables presented in the table from highest to lowest priority.

Table III. Adjusted means and univariate and stepdown statistics for physical injury and impairment variables by injuries at psychological-legal assessment.

A 2 × 2 between groups MANCOVA was performed to test the effects of pain sites at psychological-legal assessment (low, high) and head injury status (CP, HICP) on the combination of psychological distress variables (MMPI-2 Pt, MPI Affective distress, and SIP Psychosocial), with age and ethnicity included as covariates in the model. After adjusting for the effects of age and ethnicity, a significant interactive effect of pain sites at psychological-legal assessment and head injury status on the combination of psychological distress variables was revealed, Λ = .90, F(3, 99) = 3.47, p = .02. No significant main effect of pain sites at psychological-legal assessment or head injury status was observed.

Simple contrast analyses were performed to examine differences in psychological distress across pain sites at psychological-legal assessment (low, high) and head injury status (CP, HICP) conditions. See Figures 1 – 3 for graphic representations of simple contrast effects across the three measures of psychological distress.

Graph: Figure 1. Anxiety evidenced in response to pain sites as a function of head injury status.

Graph: Figure 2. Affective distress evidenced in response to pain sites as a function of head injury status.

Graph: Figure 3. Psychosocial dysfunction evidenced in response to pain sites as a function of head injury status.

Simple contrasts revealed that, among participants reporting low pain sites at psychological-legal assessment, there was no difference between CP (71.95) and HICP participants (70.11) in reports of anxiety as measured by the MMPI-2 Pt scale, F(1, 51) = 0.00, p = .98. In contrast, among participants reporting high pain sites at psychological-legal assessment, HICP participants (80.43) reported significantly more anxiety than CP participants (71.16), F(1, 51) = 3.86, p = .05.

Among participants reporting low pain sites at psychological-legal assessment, CP participants (54.37) reported significantly greater affective distress (as measured by the MPI Affective distress scale) than HICP participants (47.14), F(1, 51) = 8.70, p = .01. In contrast, among participants reporting high pain sites at psychological-legal assessment, there was no difference between CP and HICP participants in reports of affective distress, F(1, 51) = 0.33, p = .57.

Among participants reporting low pain sites at psychological-legal assessment, HICP participants (41.02) reported significantly greater psychosocial dysfunction (as measured by the SIP Psychosocial scale) than CP participants (32.37), F(1, 51) = 2.46, p = .02. In contrast, among participants reporting high pain sites at psychological-legal assessment, there was no difference between CP (46.48) and HICP participants (44.02) in reports of psychosocial dysfunction, F(1, 51) = 0.14, p = .71.

A 2 × 2 between groups MANCOVA was performed to test the effects of pain sites at psychological-legal assessment (low, high) and head injury status (CP, HICP) on the combination of pain coping variables (CSQ Catastrophizing, Coping self-statements, and Praying and hoping). After adjusting for the effect of ethnicity, tests of the main effects of pain sites at psychological-legal assessment and head injury status and the interactive effect of pain sites at psychological-legal assessment by head injury status were conducted. Tests revealed no significant interactive or main effects of pain sites at psychological-legal assessment and head injury status on the combination of pain coping variables.

Concerns about the uneven split between ethnicity categories prompted the post hoc conduct of a series of multivariate analyses of variance and covariance using only the sub-sample of chronic pain patients who self-identified as Caucasian (n = 97) to test the effects of pain sites at psychological-legal assessment (low, high) and head injury status (CP, HICP) on physical injury and functional impairment, psychological distress, and pain coping domains. Without exception, the results of the post hoc multivariate analyses paralleled those of the planned MANCOVAs performed on the complete sample of 108 participants.

The purpose of the present study was to compare the physical and psychological burden imposed by chronic pain to that imposed by the combination of chronic pain and head injury in persons injured in MVCs. More specifically, the study attempted to determine whether persons with a combined presentation of chronic pain and head injury experience a double whammy (a significantly greater physical and psychological burden) or whether chronic pain and head injury are better viewed as different sides of the same coin (convey a similar physical and psychological burden). Although the results of the present study suggested the importance of the complexity of the pain experience in differentiating non-head injured and head injured chronic pain patients, the results did not unequivocally support either hypothesis.

The matching procedure employed ensured that non-head injured and head injured chronic pain patients did not differ on a number of pain-relevant sociodemographic variables, including age, formal years of education, and gender. Comparisons revealed these groups to be equally similar with respect to other sociodemographic variables, including ethnic identification and marital status. Of the sociodemographic variables examined, age and ethnic identification were significantly associated with measures of physical injury and impairment, psychological distress, and pain coping. Younger chronic pain patients evidenced significantly more anxiety as measured by the MMPI-2 Pt scale than older chronic pain patients. Patients of diverse ethnic backgrounds evidenced significantly more physical injury and functional impairment and psychological distress than Caucasian patients. Being of a diverse ethnic background was also associated with more frequent use of catastrophizing and praying and hoping pain coping strategies. Because of their association with physical injury and impairment, psychological distress, and pain coping, these variables were included as covariates in primary analyses for the current study. However, the findings related to ethnicity indicate there is a need for studies that directly examine the role of ethnic identity in determining: 1) the availability and access to health care resources; and 2) the quality of interactions had with medical, allied health, legal, and insurance systems.

With respect to MVC-related variables, head injured chronic pain patients incurred injuries that resulted in more pain sites, greater frequency of hospitalization, and more surgical interventions than did non-head injured chronic pain patients. Although head injured chronic pain patients incurred more of a physical burden shortly following MVC, they were referred for a psychological-legal assessment much later than were non-head injured chronic pain patients. This finding would be in keeping with impressions that the delay in referring for assessment might be related to chronic pain and head injury sharing some similar clinical features, health care providers not being used to recognizing these two conditions as co-existing, and the effects of head injury taking more time to be recognized (Anderson et al., [4]). In addition, health care providers tend to the more immediate needs of MVC persons, and the report of pain in a number of body areas will usually precede recognition of neurocognitive impairments consistent with head injury.

MANCOVAs were used to assess the effects of pain sites at psychological-legal assessment and head injury status on physical injury and functional impairment, psychological distress, and pain coping. Patients' scores on measures of physical injury and functional impairment did not contribute to the separation of non-head injured and head injured chronic pain patients. Physical injury and functional impairment scores did contribute to the separation of patients reporting low and high pain sites at the time of psychological-legal assessment, with MPI Pain severity scores contributing most uniquely to the separation of low and high pain sites groups. The report of pain sites at the time of psychological-legal assessment may capture some combination of pain associated with physical injuries sustained at the time of MVC, pain associated with surgical interventions subsequent to MVC, and pain associated with deconditioning consequent to injuries and surgical interventions. Taken together, information on current pain sites and pain severity may be more predictive of chronic pain patients' return to function than information pertaining to initial injury. The current study is the first study to evaluate and note the influence of both the number of pain sites and its influence on MVC-related sequelae in head injured and non-head injured chronic pain persons. Subsequent studies addressing the relation of pain sites to post-MVC variables such as surgery and deconditioning would allow for a better assessment of physical rehabilitation needs in injured individuals and more accurate predictions regarding the potential for recovery and disability.

Findings revealed a significant interactive effect of pain sites at psychological-legal assessment and head injury status on measures of psychological distress. A review of simple contrasts revealed a variable pattern of psychological distress across the four pain sites at psychological-legal assessment (low, high) by head injury status (non-head injured, head injured) conditions. Head injured chronic pain patients reporting a high number of pain sites experienced significantly more anxiety (MMPI-2 Pt) than head injured chronic pain patients reporting low pain sites and non-head injured chronic pain patients reporting low and high pain sites. Head injured chronic pain patients reporting low pain sites experienced significantly less affective distress (MPI Affective distress) than head injured chronic pain patients with high pain sites and non-head injured chronic pain patients with low and high pain sites. Non-head injured chronic pain patients reporting low pain sites experienced significantly less psychosocial dysfunction (SIP Psychosocial) than non-head injured chronic pain patients reporting high pain sites and head injured chronic pain patients reporting low and high pain sites. Although these data suggest a relationship between pain sites and psychological distress, the variable pattern of results is also likely due to differences in the content of the domains sampled by the various psychological distress measures.

Post hoc analyses revealed that 73% of head injured chronic pain patients sustained mild head injuries and 22% of head injured chronic pain sustained moderate to severe head injuries, and the extent of head injury in 5% remained inconclusive. Some literature suggests that severe head injuries are associated with greater levels of psychological distress (Alexander, [1]; Morton & Wehman, [31]). It is possible that inconsistent findings in terms of psychological distress in head injured chronic pain persons with low and high pain sites may be affected by level of head injury status. Investigations using comparable samples of non-head injured, mildly head injured, moderately head injured, and severely head injured chronic pain patients would allow for more accurate and reliable statements regarding the experience of affective distress across these different patient groups.

The finding that non-head injured chronic pain patients reporting a low number of pain sites experience less psychosocial dysfunction is in line with what would be expected for this patient group. Relative to the other groups, non-head injured chronic pain patients who report a low number of pain sites are likely to experience less pain, less psychological distress, and fewer disruptions in social functioning.

The results revealed no significant main or interactive effects of pain sites at psychological-legal assessment and head injury on measures of pain coping. This finding is consistent with evidence reported by Curran et al. ([8]). These researchers compared head injured patients who were without pain to orthopaedically injured patients with pain on measures of psychological distress and coping. Few differences in psychological distress or coping were revealed across the two patient groups. As an explanation for these findings, the authors suggested that pre-event psychological characteristics might be instrumental to the ability to cope and adapt to injuries. This interpretation is consistent with findings from the larger coping literature that suggests that, in response to all types of events and situations, there are people who will meet the challenge and cope well and people who will be overwhelmed by the challenge and cope poorly. Similarly, it is not the level of injury and pain but how the individual copes with injury and pain that determines post-collision adjustment (Duckworth & Iezzi, [10]).

Although the current study has a number of strengths (i.e., use of a clinical sample of non-head injured and head injured chronic pain patients undergoing MVC-related litigation, use of well-regarded self-report instruments, and sampling across physical, psychological, and coping response domains), certain limitations need to be acknowledged. The results of the current study are based on referrals to the first author and may be subject to systematic influences in the referral process that, if present, would limit the generalizability of study findings. Study findings are also based, in part, on retrospective and archival data. Causal inferences related to MVC-related variables, pain sites and pain severity, head injury status, and the psychological, social, and legal concomitants of motor vehicle collision are better addressed through longitudinal research designs. Studies following MVC injured persons from the emergency room on to post-litigation resolution would yield a more thorough understanding of the more complicated MVC experiences. It would also be important to include measures that document other relevant aspects of the MVC experience (e.g., the injured person's pre-MVC physical and psychological history; the injured person's perception of support from health care providers, legal service providers, insurance industry representatives, and significant others; and the injured person's acceptance of residual consequences of MVC).

The current findings have implications for the assessment and management of chronic pain patients who have sustained injuries in the context of an MVC. It is clear that MVC-injured persons experience physical injury and functional impairment, psychological distress, and coping challenges that are characteristic of the general chronic pain experience. Standard pain management interventions aimed at increasing functional status, pacing, developing adaptive pain coping strategies, reducing psychological distress, and returning the individual to work are indicated. However, chronic pain experienced in the context of an MVC requires assessment and clinical consideration of co-morbid physical and psychiatric conditions that substantially impact the clinician's choice and implementation of intervention strategies. In the context of an MVC, chronic pain is more likely to be accompanied by conditions such as head injury and/or posttraumatic stress disorder, conditions that may require direct therapeutic address along with chronic pain management. A previous study of patients with MVC-related chronic pain determined that traumatized chronic pain patients required more psychological management and pharmacotherapy (in the form of antidepressant and opioid medications) than did non-traumatized chronic pain patients (Duckworth & Iezzi, [10]).

It is also the case that the MVC context often implies involvement in litigation, a circumstance that can contribute directly to stress and distress, and through these increases in stress and distress, to poorer tolerance of pain. In addition, injured individuals pursuing litigation experience more frequent, more intense, and more persistent symptom reporting than injured individuals not pursuing litigation (Williams et al., [47]). The very act of pursuing litigation leads to an increased focus on symptoms and disability (Youngjohn et al. ([48]). There are generally no financial rewards for doing well following MVC-related injuries. Thus, in the MVC context, litigation stress serves as yet another factor to be included in an intervention protocol.

In conclusion, the MVC context is one in which the chronic pain experience takes on a complexity as a function of multiple injuries and pain sites and as a function of additional medical and psychiatric diagnoses. Findings from the current study highlighted the impact of multiple pain sites on the chronic pain patient's experience of physical injury and functional impairment. They also suggested the interactive effective of pain sites and head injury on the chronic pain patient's experience of psychological distress and psychosocial dysfunction. Much like the role of pain as the fifth vital sign in the hospital context, findings from the current study point to the importance of the number of identified pain sites and the severity of pain in the MVC context. These variables seem to be suggestive of the clinical course from MVC to psychological-legal assessment and are critically important in establishing medical and psychological treatment goals and predicting response to treatment.