BLAST INJURIES account for a vast majority of all combat wounds in the current conflicts in Iraq and Afghanistan.1–4 This is due, in part, to asymmetrical warfare techniques employed by the enemy, which include improvised explosive devices and other crude forms of blast weaponry. The preponderance of blast weaponry has resulted in an increased prevalence of concussion, or mild traumatic brain injury (MTBI) compared with previous 20th century military conflicts.4,5 Furthermore, improvements in body armor and field medical care have resulted in a greater survivability rate and as such more personnel are surviving their wounds and are faced with adverse postinjury health sequelae.5
An array of negative health complaints have been reported following MTBI, including headache, dizziness, tinnitus, muscle aches, sleep problems, and cognitive deficits.6–8 This cluster of complaints is often referred to as postconcussion syndrome (PCS), but because these symptoms are also known to occur among those with nonhead injuries,9–11 psychiatric problems,12,13 and even among healthy persons,14 there is some controversy surrounding what constitutes a PCS diagnosis.
Examining the association between combat-related MTBI and PCS is further complicated by the austere environment in which the injury occurs. Military personnel with combat experiences have high rates of posttraumatic stress disorder (PTSD) and depression, and both of these disorders are associated with PCS symptoms.13,15 Furthermore, persons with MTBI typically report higher rates of PTSD16,17 and depression,18,19 which presents the question as to whether PCS symptoms following MTBI are a product of the organic brain injury or subsequent mental health problems. Another potential mediator is blast injury mechanism, which has been suggested to have differential effects relative to nonblast mechanisms.8
Recent studies among military personnel have suggested that PCS symptoms following MTBI are strongly mediated by PTSD and depression.17,20 After adjusting for PTSD and depression, Hoge et al17 found that only headache was significantly higher among combat veterans with MTBI compared to those with other injuries. Similarly, Pietrzak et al20 extended Hoge's findings by examining general health ratings and psychosocial functioning after MTBI and identified a strong mediating effect of PTSD. Both studies are limited, however, by the use of self-reported screening instruments to retrospectively identify personnel with MTBI.21 Relying on patient recall can be problematic because MTBI is known to negatively affect memory.22 Civilian studies have found that most MTBI symptoms resolve after 1 year.23 Because most screening instruments used in previous military studies do not inquire about specific time of injury, many personnel in these studies may be screened for symptoms well after 1 year postinjury. Finally, the study by Hoge et al in part characterizes MTBI as an experience or event (eg, blast, motor vehicle accident) that left the person feeling “dazed or confused.” This symptom is not specific to MTBI, but can also occur as a natural reaction to the stress of combat.24
The objective of the present analysis was to identify the association between MTBI and PCS symptoms among military personnel while accounting for the effects of PTSD and depression. By using in-theatre clinical records to identify MTBI, many of the limitations inherent with self-reported MTBI status were addressed. Specifically, the authors aimed to (1) compare PCS symptom reporting rates between MTBI and nonhead injuries after adjusting for PTSD and depression; (2) assess the potential mediating effects of loss of consciousness (LOC) among those with MTBI; and (3) identify the health complaints that contributed to declines in self-rated health among those with MTBI.
The Expeditionary Medical Encounter Database (EMED; formerly the Navy and Marine Corps Combat Trauma Registry) was queried for all personnel injured during Operation Iraqi Freedom who completed a Post-Deployment Health Assessment (PDHA) and a Post-Deployment Health Re-Assessment (PDHRA). This study was approved through the institutional review board at Naval Health Research Center (NHRC), San Diego, California.
The EMED is a deployment health database maintained by NHRC and consists of documented clinical encounters of deployed military service members.25 Clinical EMED records are completed by health care providers stationed at forward-deployed Navy and Marine Corps military treatment facilities (eg, facilities located in Iraq to treat Operation Iraqi Freedom casualties). Unique aspects of the EMED include details of the injury event which are collected at or near the point of occurrence, as well as the inclusion of persons with milder injuries who are returned to duty. Clinical EMED records are then provided to NHRC where professional coders review the records and assign medical codes using the Abbreviated Injury Scale (AIS), Injury Severity Score (ISS), and International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM).26–28 The AIS ranges from 1 (minor) to 6 (unsurvivable) and is assigned to 9 body regions. The ISS is a composite score based on the AIS and ranges from 1 to 75.
The PDHA and PDHRA are self-administered screening questionnaires developed by the Department of Defense to identify personnel in need of medical referral for a variety of health reasons.29,30 The PDHA is given at the end of each deployment and the PDHRA is given after the service member has returned home, in part, to identify mental health symptoms that manifest on readjustment. Both surveys have been used in previous research to identify population level mental health screening rates.31,32
In this retrospective cohort study, the sample was identified from the EMED. The sample was further refined by restricting to service members who sustained minor to moderate injuries (ISS: 1–8) during Operation Iraqi Freedom between March 2004 and April 2008 and who completed a PDHA and PDHRA on completion of deployment. To ensure the effects of the MTBI were being measured, the sample was restricted to personnel completing both surveys within 1 year of the injury date. The final sample consisted of 992 injured personnel, with 334 sustaining MTBI and 658 sustaining a nonhead injury.
Demographic variables included in the analysis were age, military rank, gender, and branch of service at the time of injury and were abstracted from the EMED clinical record and validated with information from the Defense Manpower Data Center, which maintains administrative records for all military personnel. Military rank was categorized as junior enlisted (E1–E3), mid-level enlisted (E4–E6), and senior enlisted/officer. Branch of service was categorized into Marines, Army, and Navy/Air Force. Age was analyzed as a continuous variable.
Injury group, “MTBI” or “non-head injury,” was identified from the EMED clinical record. As per previous literature,33 MTBIs were identified through a 2-step process. First, all service members a TBI ICD-9-CM code in the range of 800.0 to 801.9, 803.0 to 804.9, or 850.0 to 854.1 were identified. Then, each TBI ICD-9-CM code was checked against the corresponding AIS severity score to ensure only TBIs with AIS values of 1 or 2 were included; only TBI ICD-9-CM codes with corresponding AIS values of 1 or 2 were considered MTBI. The nonhead injury comparison group consisted of any injury where the head region was not specifically indicated as per AIS coding.
Type of injury was categorized as battle injury as a result of hostile action, or nonbattle, defined an injury resulting from nonhostile action. Injury type and presence/absence of blast mechanism were used to define injury mechanism as “battle, blast,” “battle, nonblast,” and “nonbattle.” In logistic regression analysis, this variable was dichotomized into blast and nonblast. For persons with MTBI, LOC was identified from provider notes on the EMED clinical record.
Both the PDHA and EMED were used to approximate combat exposure. The PDHA contains questions that ask if the service member was exposed to dead bodies, discharged their weapon, or had a perceived threat to life. The specific questions are shown in Table 1. These 3 questions were incorporated together with presence/absence of battle injury (from the EMED record) to create a 4-level combat exposure measure. The final combat exposure variable was categorized into “0–1,” “2,” “3,” and “4” to indicate the total number of combat exposures. Those endorsing all 3 combat exposure questions and sustaining a battle injury would be classified as having 4 combat exposures.
Mental health symptoms were ascertained from the PDHRA, which contains validated PTSD and depression screening instruments shown in Table 1. The 4-item PTSD screening instrument is based on the Primary Care PTSD screen.34,35 Endorsing any 3 of the 4 symptoms indicates a positive screen for PTSD. The depression-screening tool was derived from the Patient Health Questionnaire and a positive screen is indicated by answering “more than half the days” or “nearly every day” to at least 1 of the 2 questions.36
Main outcome measures
Symptoms of PCS were identified from the PDHRA. On the PDHRA, the service member is presented a list of health complaints and asked to endorse whether they currently have a health concern or condition that they feel is related to their deployment. Current complaints of 9 PCS symptoms—headache, dizziness, memory problems, still feeling tired after sleeping (ie, sleep problems), back pain, joint and muscle pain, weakness, irritability and tinnitus—were abstracted from the PDHRA. A separate variable was created to indicate whether the service member endorsed 3 or more of these symptoms.
A measure of self-rated health was utilized to assess changes in the service member's perceived health status. On both the PDHA and PDHRA, the service member is asked to rate their current health as excellent, very good, good, fair, or poor. A decline in self-rated health was defined as endorsing excellent, very good, or good health on the PDHA and endorsing fair or poor health on the PDHRA.
All statistical analyses were performed using SAS version 9.2 (Cary, North Carolina). Demographics, injury-specific information, self-reported combat exposure, and mental health problems were described for the study sample by MTBI status. The association between MTBI and PCS symptoms was shown using unadjusted odds ratios and odds ratios adjusted for other potential confounders. Separate models were run with each PCS symptom as the dependent variable, as well as one model with 3 or more symptoms as the outcome. A separate analysis was performed among those with MTBI exploring the mediating effect of LOC on reporting of PCS symptoms. The analysis of decline in self-rated health was restricted to personnel with MTBI who endorsed good to excellent health on the PDHA. Predictors of a decline in health were examined using backward selection logistic regression modeling and assessed the PCS and mental health symptoms associated with MTBI. Model fit was assessed with the Hosmer-Lemeshow test for all final multivariate logistic regression models at an alpha level of 0.10. Chi-square and 2 sample t tests were used to examine categorical and continuous variables, respectively.
The study sample consisted of 334 service members with MTBI and 658 with nonhead injuries. Of those with MTBI, 92 had concussion with LOC of 30 minutes or less (ICD-9-CM 850.11), 1 had concussion with LOC of 31 to 59 minutes (850.12), 10 had concussion with LOC of unspecified duration (850.5), 150 had concussion with no LOC (ICD-9 850.0), and 81 had MTBI characterized as “concussion, unspecified”(850.9). Characteristics of the study sample by MTBI status are shown in Table 2. Compared with nonhead injuries, those with MTBI were more likely to be male, younger, junior enlisted and Marines. Blasts accounted for more than twice as many MTBIs (88.9%) than nonhead injuries (41.6%), and personnel sustaining an MTBI had more severe injuries overall. Response time to the PDHRA differed between groups, with means of 249.2 days and 258.6 days for MTBI and nonhead injuries, respectively (P = .019). Those with MTBI reported higher levels of combat exposure, with 29.9% and 49.7% reporting 3 and 4 exposures, respectively, compared with only 20.2% and 29.1% among those with nonhead injuries. There were increased rates of mental health symptoms among personnel with MTBI compared with nonhead injuries, with higher screen positive rates for both PTSD (27.8% vs 17.2%, P < .001) and depression (20.7% vs 13.2%, P = .002).
Table 3 details the rates of PCS symptoms for MTBI and nonhead injuries and shows the results of logistic regression analysis. In univariate analysis, all symptoms, with the exception of weakness and muscle/joint pain, were significantly higher among those with MTBI. These results were consistent after adjusting for age, ISS, blast mechanism, and combat exposure. After adjustment for PTSD and depression, significant associations remained for headache (odds ratio [OR] = 3.37; 95% confidence interval [CI] = 2.19–5.17), back pain (OR = 1.79; 95% CI = 1.23–2.60), memory problems (OR = 1.86; 95% CI = 1.20–2.88), tinnitus (OR = 1.63; 95% CI = 1.10–2.41), and dizziness (OR = 2.13; 95% CI = 1.06–4.29). Persons with MTBI were also more likely than those with nonhead injury to report 3 or more PCS symptoms after adjusting for PTSD, depression, and other covariates (OR = 1.67; 95% CI = 1.13–2.46).
A subgroup analysis was conducted for personnel with MTBI who had documented LOC information, and results are shown in Table 4. Overall, LOC status was determined for 75.7% of those with MTBI (253/334). Of the PCS symptoms, only memory problems (34.0% vs 20.7%, P = .018) and irritability (35.0% vs 23.3%, P = .043) were significantly higher in those with a confirmed LOC. Rates of depression and PTSD were not significantly different between LOC groups.
The analysis of decline in self-rated health was restricted to personnel that endorsed good, very good, or excellent health on the PDHA; those who endorsed fair or poor health on the PDHA or who did not respond to the self-rated health question on either the PDHA or PDHRA were excluded. Overall, 23.8% (69/290) of those with MTBI declined in self-rated health, which was significantly higher than the 17.3% (101/585) of nonhead injuries who experienced a decline (P = .022). The above-mentioned PCS symptoms associated with MTBI were assessed with PTSD and depression to determine the greatest contributors to a decline in self-rated health, and results are shown in Table 5. In univariate analysis, all PCS and mental health symptoms were associated with a decline in health. Following backward selection regression modeling, however, only the effects of depression (OR = 2.73; 95% CI = 1.27–5.87), memory problems (OR = 5.07; 95% CI = 2.56–10.02), and dizziness (OR = 10.60; 95% CI = 3.48–32.27) remained significantly associated with self-reported health decline.
Mild traumatic brain injury is a preeminent wound of the current wartime environment. Consistent with previous literature, we found that a majority of MTBIs are the result of blasts,1–4 and are associated with postdeployment psychological morbidity.17 Even after taking the psychological symptoms of PTSD and depression into account, other physical and neurological symptoms were found to be elevated in MTBI compared with nonhead injuries. This may reflect the manifestation of a distinct postconcussion syndrome that occurs in parallel with psychological symptoms. Identification and treatment of these symptoms, particularly those associated with lower self-rated health, is paramount in the rehabilitation of MTBI patients.
The primary finding of this study was the increased odds of headache, back pain, memory problems, tinnitus, and dizziness among combat veterans with MTBI after adjustment for PTSD and depression. This finding suggests that MTBI is an independent predictor of multiple PCS symptoms. This finding is in contrast to a recent study by Hoge et al that found only headache was elevated in MTBI compared with non-MTBI in a similar multivariate analysis.17 The lack of comparable findings may be a result of methodological differences between the 2 studies. In this study, we used provider-diagnosed MTBI and restricted follow-up time to 1-year postinjury, whereas the Hoge et al study used self-report MTBI with limited knowledge of when the injury actually occurred. Our findings did suggest a role of psychological symptoms in presentation of PCS, as some of the associations between MTBI and PCS symptoms were attenuated after adjustment for PTSD and depression. The mediating effect of psychological symptoms in the persistence of PCS symptoms beyond 1 year was not examined in this analysis and needs further investigation, although one recent study suggested PCS symptoms beyond 1-year were largely mediated by PTSD.37
Another finding of interest was the independent associations of depression, memory problems, and dizziness with a decline in self-rated health among MTBI. With the array of symptoms that can occur following MTBI, identification of the symptoms that most impact a service-member's health may be important in targeting clinical interventions to maximize rehabilitation efforts. Both dizziness and memory problems have been implicated previously as reasons for lower return to work rates and life satisfaction among those with MTBI.38,39 A recent study by Bombardier et al40 found that among those with MTBI, comorbid major depressive disorder was an independent predictor of lower quality-of-life scores.40 More research on these symptoms and how they affect long-term treatment and rehabilitation outcome is warranted.
A mediating effect of LOC was found with irritability and memory problems, but no other PCS symptoms. The effect of LOC has recently been debated in the literature, with some studies showing no exacerbating effects of LOC beyond the acute phase of MTBI, typically defined as 2 weeks postinjury.41 The role of LOC in MTBI outcome needs to be further defined, with future research focusing on long-term influence on patterns of mental health and neurological disorders.
This study had several strengths relative to previous research. The aforementioned methodological differences, particularly the use of clinical records to ascertain MTBI status, were a key aspect of the study and may have minimized the effects of misclassification and recall bias. In addition to the presumed reduction in bias, these clinical records contained point of injury indicators which allowed for the incorporation of and adjustment for injury severity, a variable previously found associated with adverse psychological outcomes in military personnel.42 Finally, knowledge of the actual date of injury permitted the one-year restriction on PCS symptom ascertainment, thus limiting the follow-up period to best approximate organic effects of the MTBI.
There were limitations that warrant mention. The EMED consists of data collected at Navy and Marine Corps medical facilities, thus members of other services are likely underrepresented. In addition, although all service members with MTBI in this study met the inclusion criteria, it is possible that some diagnosed with “concussion with loss of consciousness of unspecified duration” or “concussion, unspecified” may not meet other definitions of MTBI based on duration of LOC. Furthermore, the PDHRA ascertained symptoms with a simplistic presence/absence query, whereas some symptoms such as memory problems may be best evaluated using cognitive testing. Similarly, the instruments used to assess PTSD and depression were based on clinical screening tools not necessarily developed for population level screening, though the PTSD instrument was recently validated against the more widely used 17-item PTSD checklist.35 It is also important to note that the PDHA and PDHRA are voluntary, thus noncompliance may limit generalizability.
Service members with MTBI are at risk for an array of adverse PCS symptoms that may potentially affect clinical, rehabilitative, vocational, and quality-of-life outcomes. The long-term persistence of these symptoms needs to be investigated, possibly by incorporating repeated measurements of PCS symptoms over time. Because many of these symptoms are also associated with PTSD and depression, these conditions need to be studied concurrently. As MTBI continues to cause significant morbidity among US military personnel, research initiatives should focus on identifying target areas for neurological and psychological care to foster postdeployment readjustment.
1. Okie S. Traumatic brain injury in the war zone. N Engl J Med. 2005;352(20):2043–2047.
2. Warden D. Military
MTBI during the Iraq and Afghanistan wars. J Head Trauma Rehabil. 2006;21(5):398–402.
3. Martin EM, Lu WC, Helmick K, French L, Warden DL. Traumatic brain injuries sustained in the Afghanistan and Iraq wars. Am J Nurs. 2008;108(4):40–47.
4. Zeitzer MB, Brooks JM. In the line of fire—traumatic brain injury among Iraq War Veterans. AAOHN J. 2008;56(8):347–353.
5. Institute of Medicine. Gulf War and Health: Long-Term Consequences of Traumatic Brain Injury. Vol 7. Washington, DC: National Academies Press; 2009.
6. Iverson GL. Outcome from mild traumatic brain injury
. Curr Opin Psychiatry. 2005;18:301–317.
7. Alves W, Macciocchi SN, Barth JT. Postconcussive symptoms after uncomplicated mild head injury. J Head Trauma Rehabil. 1993;8(3):48–59.
8. Howe LS. Giving context to post-deployment post-concussive-like symptoms: blast-related potential mild traumatic brain injury
and comorbidities. Clin Neuropsychol. 2009;23:1315–1337.
9. Lees-Haley PR, Fox DD, Courtney JC. A comparison of complaints by mild brain injury claimants and other claimants describing subjective experiences immediately following their injury. Arch Clin Neuropsychol. 2001;16:689–695.
10. Frenisy MC, Benony H, Chahraoui K, et al. Brain injured patients versus multiple trauma patients: some neurobehavioral and psychopathological aspects. J Trauma. 2006;60:1018–1026.
11. Meares S, Shores EA, Taylor AJ, et al. Mild traumatic brain injury
does not predict acute postconcussion syndrome. J Neurol Neurosurg Psychiatry. 2008;79:300–306.
12. Fox DD, Lees-Haley PR, Earnest K, Dolezal-Wood S. Post-concussive symptoms: base rates and etiology in psychiatric patients. Clin Neuropsychol. 1995;9(1):89–92.
13. Suhr JA, Gunstad J. Postconcussive symptom report: the relative influence of head injury and depression
. J Clin Exp Neuropsychol. 2002;24(8):981–993.
14. Iverson GL, Lange RT. Examination of “postconcussion-like” symptoms in a healthy sample. Appl Neuropsychol. 2003;10(3):137–144.
15. Schneiderman AI, Braver ER, Kang HK. Understanding sequelae of injury mechanisms and mild traumatic brain injury
incurred during the conflicts in Iraq and Afghanistan: persistent postconcussive symptoms and posttraumatic stress disorder
. Am J Epidemiol. 2008;167(12):1446–1452.
16. Carlson KF, Nelson D, Orazem RJ, Nugent S, Cifu DX, Sayer NA. Psychiatric diagnoses among Iraq and Afghanistan war veterans screened for deployment-related traumatic brain injury. J Trauma Stress. 2010;23(1):17–24.
17. Hoge CW, McGurk D, Thomas JL, Cox AL, Engel CC, Castro CA. Mild traumatic brain injury
in U.S. soldiers returning from Iraq. New Engl J Med. 2008;358(5):453–463.
18. Jorge RE, Robinson RG, Moser D, et al. Major depression
following traumatic brain injury. Arch Gen Psychiatry. 2004;61:42–50.
19. Vanderploeg RD, Curtiss G, Luis CA, Salazar AM. Long-term morbidities following self-reported mild traumatic brain injury
. J Clin Exp Neuropsychol. 2007;29(6):585–598.
20. Pietrzak RH, Johnson DC, Goldstein MB, Malley JC, Southwick SM. Posttraumatic stress disorder
mediates the relationship between mild traumatic brain injury
and health and psychosocial functioning in veterans of Operations Enduring Freedom and Iraqi Freedom. J Nerv Ment Dis. 2009;197(10):748–753.
21. Iverson GL. Clinical and methodological challenges with assessing mild traumatic brain injury
in the military
. J Head Trauma Rehabil. 2010;25(5):313–319.
22. Ruff RM, Iverson GL, Barth JT, Bush SS, Broshek DK. Recommendations for diagnosing a mild traumatic brain injury
: a National Academy of Neuropsychology education paper. Arch Clin Neuropsychol. 2009;24:3–10.
23. Carroll LJ, Cassidy D, Peloso PM, et al. Prognosis for mild traumatic brain injury
: results of the WHO collaborating centre task force on mild traumatic brain injury
. J Rehabil Med. 2004;(suppl 43):84–105.
24. Bryant RA. Disentangling mild traumatic brain injury
and stress reactions. New Engl J Med. 2008;358(5):525–527.
25. Galarneau MR, Hancock WC, Konoske P, et al. The Navy-Marine Corps Combat
Trauma Registry. Mil Med. 2006;171(8):691–697.
26. Gennarelli T, Wodzon E. The Abbreviated Injury Scale—2005. Des Plaines, IL: Association for the Advancement of Automotive Medicine; 2005.
27. Baker SP, O'Neill B, Haddon W, Long WB: The injury severity score: a method for describing patients with multiple injuries and evaluating emergency care. J Trauma. 1974;14:187–196.
28. Commission on Professional Hospital Activities. International Classification of Diseases, 9th Revision, Clinical Modification. Ann Arbor, MI: Edwards Brothers; 1977.
31. Hoge CW, Auchterlonie JL, Milliken CS. Mental health problems, use of mental health services, and attrition from military
service after returning from deployment to Iraq or Afghanistan. JAMA. 2006;295(9):1023–1032.
32. Milliken CS, Auchterlonie JL, Hoge CW. Longitudinal assessment of mental health problems among active and reserve component soldiers returning from the Iraq war. JAMA. 2007;298(18):2141–2148.
33. Ommaya AK, Ommaya AK, Dannenberg AL, Salazar AM. Causation, incidence, and costs of traumatic brain injury in the U.S. Military
medical system. J Trauma. 1996;40(2):211–217.
34. Prins A, Quimette P, Kimerling R, et al. The primary care PTSD
): development and operating characteristics. Prim Care Psychiatr. 2004;9:9–14.
35. Bliese PD, Wright KM, Adler AB, Cabrera O, Castro CA, Hoge CW. Validating the primary care posttraumatic stress disorder
screen and the posttraumatic stress disorder
checklist with soldiers returning from combat
. J Consult Clin Psychol. 2008;76(2):272–281.
36. Kroenke K, Spitzer RL, Williams JBW. The patient health questionnaire-2: validity of a two-item depression
screener. Med Care. 2003;41:1284–1292.
37. Polusny MA, Kehle SM, Nelson NW, Erbes CR, Arbisi PA, Thuras P. Longitudinal effects of mild traumatic brain injury
and posttraumatic stress disorder
comorbidity on postdeployment outcomes in national guard soldiers deployed to Iraq. Arch Gen Psychiatry. 2011;68(1):79–89.
38. Drake AI, Gray N, Yoder S, Pramuka M, Llewellyn M. Factors predicting return to work following mild traumatic brain injury
: a discriminant analysis. J Head Trauma Rehabil. 2000;15(5):1103–1112.
39. Chamelian L, Feinstein A. Outcome after mild to moderate traumatic brain injury: the role of dizziness. Arch Phys Med Rehabil. 2004;85(10):1662–1666.
40. Bombardier CH, Fann JR, Temkin NR, Esselman PC, Barber J, Dikmen SS. Rates of major depressive disorder and clinical outcomes following traumatic brain injury. JAMA. 2010;303(19):1938–1945.
41. McCrea M. Mild Traumatic Brain Injury
and Postconcussion Syndrome: The New Evidence Base for Diagnosis and Treatment. New York, NY: Oxford University Press (Oxford Workshop Series); 2008.
42. MacGregor AJ, Shaffer RA, Dougherty AL, et al. Psychological correlates of battle and nonbattle injury among Operation Iraqi Freedom veterans. Mil Med. 2009;174(3):224–231.