Authors alone are responsible for opinions expressed in the contribution and for its clearance through their federal health agency, if required.

Transcription

1 ORIGINAL ARTICLES Authors alone are responsible for opinions expressed in the contribution and for its clearance through their federal health agency, if required. MILITARY MEDICINE, 180, 8:882, 2015 Post-traumatic Stress Disorder Among Navy Health Care Personnel Following Combat Deployment LCDR Andrew J. MacGregor, MSC USN; Amber L. Dougherty, MPH; Jonathan A. Mayo, MPH; Peggy P. Han, MPH; Michael R. Galarneau, MS ABSTRACT U.S. Navy health care personnel are exposed to an array of psychological stressors during combat deployment. This study compared rates of post-traumatic stress disorder (PTSD) among Navy health care personnel with nonhealth care personnel following single and repeated combat deployments. The study sample was identified from electronic records indicating deployment to Iraq, Kuwait, or Afghanistan, and included 3,416 heath care and 4,648 nonhealth care personnel. Health care personnel had higher PTSD rates and an increasing trend in PTSD rates across repeated deployments. After adjusting for combat exposure and other covariates, health care compared with nonhealth care personnel were more likely to be diagnosed with PTSD after one (odds ratio [OR] 2.02; 95% confidence interval [CI] ), two (OR 2.27; 95% CI ), and three deployments (OR 4.37; 95% CI ). Exposure to wounded/dead friendly forces was associated with higher PTSD rates in health care personnel (OR 1.53; 95% CI ). Health care personnel occupy a unique and essential role in current wartime operations, and are a high-risk group for PTSD. These findings suggest that further research is needed on the effects of caregiver stress, and refinements to postdeployment screening for health care personnel should be pursued. Naval Health Research Center, 140 Sylvester Road, San Diego, CA The views and opinions expressed herein are those of the authors and do not necessarily reflect the official policy or position of the Department of the Navy, Department of Defense, or the U.S. Government. Approved for public release; distribution is unlimited. This research has been conducted in compliance with all applicable federal regulations governing the protection of human subjects in research (protocol NHRC ). doi: /MILMED-D INTRODUCTION Following the September 11, 2001, terrorist attacks, the U.S. military initiated Operations Enduring Freedom and Iraqi Freedom (OEF/OIF), which resulted in a mass influx of military personnel to Kuwait, Iraq, and Afghanistan. Military health care personnel serve a significant and unique role during wartime, providing emergency care in the battlefield and rendering decisions that affect patient survival. 1 They are faced with a diverse array of stressors, including direct involvement in combat, treatment of severely injured military personnel, and exposure to ill and traumatically wounded civilians. 1 3 Other work-related stressors include long hours, 4 survival guilt, triage and treatment decisions (Jacob s dissertation, as cited in Street, 2009), 5 environmental challenges, 6 and disturbed sleep. 7,8 Post-traumatic stress disorder (PTSD), an anxiety disorder that can develop subsequent to a traumatic event, has been documented in civilian health care providers resulting from secondary traumatization However, detailed research on PTSD in military health care personnel is limited. Two recent studies examined psychological outcomes among OEF/OIF health care personnel. Jones et al found that medics in OIF were 30% more likely than their nonhealth care counterparts to report general psychological distress, but not more likely to screen for PTSD. 13 This study was among British medics, however, who may have different deployment experiences than U.S. health care personnel. A survey conducted among U.S. military personnel similarly found elevated levels of adverse psychological symptoms in health care compared with nonhealth care personnel. 14 This study, however, failed to account for deployment experiences and because of its anonymous design, was not able to validate deployment length, frequency, or location. Moreover, neither of these studies examined the effects of health care occupation across repeated deployments, a hallmark of the OEF/OIF conflicts, which have been linked to increased risk of PTSD MILITARY MEDICINE, Vol. 180, August 2015

2 Currently, military health care personnel play a critical role in wartime operations and are exposed to a variety of psychological stressors. Nevertheless, this occupational group has not been thoroughly examined for risk of stressrelated conditions such as PTSD. The aims of the present study were to (1) compare rates and odds of diagnosed PTSD among U.S. Navy health care personnel with that of nonhealth care personnel following single and repeated OEF/OIF deployments; and (2) examine the relationship between specific combat experiences and PTSD among U.S. Navy health care personnel. METHODS Sample Selection U.S. Navy personnel were identified from electronic deployment records maintained by the Defense Manpower Data Center (DMDC). For inclusion, personnel must have had one, two, or three deployments to OEF/OIF between September 2001 and November 2008, and have completed a postdeployment health assessment (PDHA) within 60 days of their most recent deployment end-date. 16 The PDHA is a screening questionnaire given to military service members at the end of their deployment, and it queries the individual on deployment-related exposures and current health complaints. A deployment was defined as an OEF/OIF deployment lasting longer than 1 month, but not more than 18 months. Excluded from the sample were personnel with a deployment to a non-oef/oif location over the study period, personnel who died during deployment, Reserve/National Guard personnel because of their potential for differential access to health care, women because of small cell sizes in the three deployment group, and those with a previous PTSD diagnosis before their most recent deployment. For personnel with multiple deployments, only those with the same occupation across all deployments were included. The final study sample consisted of 8,064 Navy personnel. Measures Outcome Ascertainment Inpatient and outpatient medical databases were used to identify new-onset diagnoses of PTSD. New-onset PTSD was defined as presence of the code from the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) 17 at any time over the course of the most recent deployment and up to 1 year following the end of deployment, and can be an acute, chronic, or delayed response to a traumatic event. Military Occupation Occupation was defined according to the Occupational conversion index: Enlisted/officer/civilian (DoD Instruction ), 18 and it was abstracted from DMDC electronic deployment records. Occupation groups were categorized as health care personnel and nonhealth care personnel. Health care personnel included the full range of medical specialties. Nonhealth care personnel consisted primarily of crafts workers, electric/mechanical equipment repairers, and functional support/administration. Combat Exposure Self-reported combat exposure was abstracted from the PDHA. The PDHA asks 5 questions related to combat exposure including whether or not the service member discharged their weapon, felt in great danger of being killed, or was exposed to wounded/dead civilians, enemy, or friendly forces. The number of combat exposures endorsed was summed and treated as a continuous variable for the first aim of this study. In the second aim, the specific combat exposures for health care personnel were examined separately. Deployment-Specific Variables Location for most recent deployment was categorized as Kuwait or Afghanistan/Iraq. Length of most recent deployment was calculated from DMDC records as the difference between deployment begin and end date, and this variable was termed current deployment time. For those with two or three deployments, total previous deployment time was the summation of all previous deployment lengths and was termed previous deployment time. The variable previous dwell time, or total time at home between deployments, was calculated for those with two and three deployments as the difference between end date of previous deployment and begin date of the next deployment, and those with three deployments had both their previous dwell times summed. Covariates Demographic variables were identified from electronic DMDC records and included military rank (enlisted or officer), age, education level (some college or no college/other), and marital status (married or not married). Because of the potential for health care personnel to have higher rates of health service utilization, a variable was created and included in the analyses to adjust for medical utilization bias. Health care visits for any reason during the follow-up period were identified from inpatient and outpatient medical databases. The criterion for classification of a health care visit was presentation to a medical treatment facility over the course of the observation period where an ICD-9-CM code was assigned in the primary position on the electronic medical record. This medical utilization variable was categorized into low, defined as 0 2 visits, moderate, 3 to 8 visits, and high, 9 or more visits, and was based on quartiles (first two quartiles combined) for the overall study sample. Data Analysis All analyses were conducted using SAS software, version 9.3 (SAS Institute, Cary, NC). Personnel with one, two, and MILITARY MEDICINE, Vol. 180, August

3 TABLE I. Descriptive Characteristics of U.S. Navy Personnel by Frequency of Deployments and Occupation Characteristic Health Care (n = 2,503) three deployments were analyzed separately. Demographicand deployment-specific characteristics were presented for the study sample. Rates of PTSD for health care personnel compared with nonhealth care personnel were presented, and the χ 2 test for trend was used to examine the linear trend of PTSD diagnosis over the course of multiple deployments. Multiple logistic regression was used to identify the association between health care occupation and PTSD. Multiple logistic regression with backward selection was utilized to examine the independent effects of specific combat stressors on PTSD among health care personnel. The Hosmer Lemeshow test was used to assess model fit in all logistic regression analyses. RESULTS The study sample consisted of 8,064 Navy personnel: 6,109 with one deployment (2,503 health care personnel and 3,606 nonhealth care personnel), 1,660 personnel with two deployments (768 health care personnel and 892 nonhealth care personnel), and 295 personnel with three deployments (145 health care personnel and 150 nonhealth care personnel). Descriptive characteristics of the study sample stratified by deployment frequency are shown in Table I. The only variables consistently different between health care and nonhealth care personnel were combat exposure and current deployment length, with health care personnel reporting significantly more combat exposures and having more days deployed. Figure 1 presents the rates of PTSD by occupation. Health care personnel had significantly higher rates of PTSD One Deployment Two Deployments Three Deployments Nonhealth Care (n = 3,606) p Health Care (n = 768) Nonhealth Care (n = 892) p Health Care (n = 145) Nonhealth Care (n = 150) p Age (years), M (SD) 29.2 (8.3) 28.6 (7.6) < (7.3) 28.7 (6.8) (7.0) 27.9 (5.7) <0.01 Rank, no. (%) < Enlisted 1,980 (79.1) 2,982 (82.7) 703 (91.5) 816 (91.5) 138 (95.2) 145 (96.7) Officer 523 (20.9) 624 (17.3) 65 (8.5) 76 (8.5) 7 (4.8) 5 (3.3) Married, no. (%) 1,420 (56.7) 1,964 (54.5) (62.6) 574 (64.4) (60.7) 97 (64.7) 0.48 Some College, no. (%) 566 (22.6) 767 (21.3) (11.7) 124 (13.9) (9.7) 15 (10.0) 0.92 Medical Utilization, < no. (%) Low 1,255 (50.1) 1,873 (51.9) 382 (49.7) 471 (52.8) 55 (37.9) 80 (53.3) Moderate 642 (25.7) 923 (25.6) 213 (27.7) 225 (25.2) 49 (33.8) 40 (26.7) High 606 (24.2) 810 (22.5) 173 (22.5) 196 (22.0) 41 (28.3) 30 (20.0) Deployment Location, 0.66 <0.01 <0.01 no. (%) Afghanistan/Iraq 1,745 (69.7) 2,495 (69.2) 628 (81.8) 498 (55.8) 117 (80.7) 94 (62.7) Kuwait 758 (30.3) 1,111 (30.8) 140 (18.2) 394 (44.2) 28 (19.3) 56 (37.3) Combat Exposures (number), M (SD) 1.5 (1.4) 0.5 (0.9) < (1.5) 0.4 (0.9) < (1.4) 0.5 (1.0) <0.01 Deployment Time (days), M (SD) Current (62.8) (73.6) < (60.9) (43.3) < (53.9) (47.7) <0.01 Previous (57.4) (46.4) < (81.9) (74.9) 0.53 Previous Dwell Time (Days), M (SD) (380.1) (417.4) (317.1) (327.9) 0.01 compared with nonhealth care personnel after one (6.0% vs. 1.9%; χ 2 = 70.94, p < 0.001), two (9.0% vs. 2.5%; χ 2 = 34.84, p < 0.001), and three deployments (17.9% vs. 3.3%; χ 2 = 16.71, p < 0.001). A χ 2 test for trend indicated a significant linear trend in PTSD diagnosis rate among health care personnel, ( χ 2 = 29.6, p < 0.001), but not nonhealth care personnel ( χ 2 = 1.81, p = 0.18). Results of a multivariate logistic regression model adjusting for other covariates are presented in Table II. FIGURE 1. Rates of post-traumatic stress disorder (PTSD) diagnosis among health care personnel and nonhealth care personnel following one, two, and three deployments. *p < **p = MILITARY MEDICINE, Vol. 180, August 2015

4 TABLE II. Predictors of Post-traumatic Stress Disorder among U.S. Navy Personnel Variable OR 95% CI p OR 95% CI p OR 95% CI p Health Care Occupation 2.02 ( ) < ( ) < ( ) 0.02 Age 0.97 ( ) < ( ) ( ) 0.18 Rank a < Enlisted Ref Ref Officer 0.35 ( ) 0.54 ( ) Married 0.93 ( ) ( ) ( ) 0.94 Some College 1.61 ( ) ( ) ( ) 0.44 Medical Utilization <0.01 <0.01 <0.01 Low Ref Ref Ref Moderate 6.64 ( ) 3.40 ( ) ( ) High ( ) ( ) ( ) Deployment Location Afghanistan/Iraq Ref Ref Ref Kuwait 0.78 ( ) 0.81 ( ) 1.32 ( ) No. of Combat Exposures 1.62 ( ) < ( ) < ( ) 0.12 Current Deployment Time 1.00 ( ) < ( ) ( ) 0.88 Previous Deployment Time 1.00 ( ) ( ) 0.72 Previous Dwell Time 1.00 ( ) ( ) 0.97 Health care personnel had significantly higher odds of PTSD diagnosis after one (odds ratio [OR] 2.02; 95% confidence interval [CI] ), two (OR 2.27; 95% CI ), and three deployments (OR 4.37; 95% CI ), compared with nonhealth care personnel. The Hosmer Lemeshow test indicated a good fit forallmodels(one,p = 0.64; two, p = 0.20; and three deployments, p =0.74). Health care personnel were examined for the independent effects of specific combat exposures, and results are shown in Table III. Although all exposures were significantly predictive of PTSD in univariate analysis, following backward selection modeling only felt in great danger of being killed (OR 3.44; 95% CI ), engaged in direct combat and discharged weapon (OR 1.67; 95% CI ), and exposed to dead/wounded friendly forces (OR 1.53; 95% CI ) were positively associated with PTSD diagnosis. The backward selection model was also adjusted fornumberofdeployments. DISCUSSION The results of the present analysis highlight that Navy health care personnel are a high-risk group for deployment-related One Deployment Two Deployments Three Deployments OR, odds ratio; CI, confidence interval. a Not adjusted for in three deployments model due to one of the cells having zero. PTSD, with rates increasing across repeated deployments. This is in contrast to a previous study among the British military, which found no elevated risk of PTSD among medical personnel compared with other occupations. 13 This may be due to lower casualty rates seen among British forces in both OEF and OIF compared with U.S. forces, and thus less exposure to the traumatically wounded. 19 Further, research is warranted to clarify the mechanism for this association, and to explore strategies to better identify and mitigate PTSD among health care personnel. One possible explanation for the above-mentioned association is the effect of cumulative stress. Recent studies suggest that chronic stress on the body can adversely affect allostasis, or the body s ongoing adaptive process of maintaining physiological stability, which can be quantified as allostatic load. 20,21 The concept of allostatic load is defined as the wear and tear the body experiences when repeated allostatic responses are activated during stressful situations. 22 Although no studies have specifically examined burnout in deployed military health care providers, the prevalence of burnout in civilian paramedics has been estimated between 16% and 30%. 23,24 As evidenced by the findings of the present study, TABLE III. Effect of Specific Combat Exposures on PTSD among Navy Health Care Personnel (n = 3,416) Combat Exposure Unadjusted OR (95% CI) Adjusted OR (95% CI) a Felt in Great Danger of Being killed b 4.14 ( ) 3.44 ( ) Engaged in Direct Combat and Discharged Weapon c 3.43 ( ) 1.67 ( ) Exposed to Wounded/Dead Civilians 1.51 ( ) Exposed to Wounded/Dead Friendly Forces 1.99 ( ) 1.53 ( ) Exposed to Wounded/Dead Enemy 1.79 ( ) PTSD, post-traumatic stress disorder; OR, odds ratio. a Adjusted using backward selection logistic regression modeling; all combat exposure variables placed in model along with total number of deployments. b Missing data n = 59. c Missing data n = 382. MILITARY MEDICINE, Vol. 180, August

5 Navy health care personnel experience a confluence of multiple stressors that are associated with PTSD, such as repeated deployments, exposure to wounded/dead friendly forces, and participation in combat. Chronic allostatic responses (or overload) resulting from multiple and repeated stressors can induce a domino effect that leaves an individual susceptible to stress-related diseases. 20 Studies involving pre- and postdeployment measurement of biomarkers for allostatic load would be required to further explore this hypothesis. 25,26 The increasing trend of PTSD diagnosis among health care specialists across repeated deployments is indicative of a dose response relationship. Based on this finding, future research on the psychological impact of multiple combat deployments should account for occupation. Although personnel may be mentally resilient to endure a first or even second combat deployment, eventually capabilities to cope with stress may diminish. Resiliency programs have recently been adopted that specifically target military health care personnel, 27,28 and additional efforts should be considered that focus on maintaining resiliency over the course of repeated deployments. Also noteworthy is the increased magnitude of the odds ratio following third deployment, which suggests a possible mediating effect of repeated deployment that needs to be explored in future studies. The findings of the present study also suggest potential operational implications. First, health care personnel are at risk for burnout, defined by Maslach as emotional exhaustion, depersonalization, and reduced personal accomplishment within one s work. 29 One recent study suggested a link between burnout and PTSD. 30 Burnout has been linked to increased risk of medical errors, 31,32 which in a wartime scenario may adversely impact the probability of patient survival as well as further stress medical personnel. Second, the current postdeployment screening instrument may not be accurately capturing the stressful experiences of health care personnel. The PDHA is focused more on combat exposure, and the only question that is related to health care work is whether or not the member was exposed to wounded/dead bodies. A separate screening should be considered for health care personnel that queries job-specific stressors, such as frequency and types of trauma treated, personal acquaintance with casualties, and irregular work schedules. There are limitations that warrant mention. Although we attempted to control for differential medical utilization, we could not account for differences in symptom awareness and stigma. Also, because we grouped all health care personnel together without further delineation of specific duties, results should be interpreted with caution. The dwell time variable used in the analysis has not been heavily studied, and there may be better ways of examining it rather than summation of total days. Further, we only examined Navy personnel and, as such, these results cannot be generalized to other services that likely have different deployment characteristics. Notable strengths of this study include the use of electronic deployment records, which allowed for a large sample size and the specification of deployment frequency. Furthermore, the use of electronic medical databases also allowed for the novel method of adjusting for medical utilization. Navy health care personnel have significantly higher rates of postdeployment PTSD compared with nonhealth care personnel, which increase linearly over the course of repeated deployments. These findings suggest that health care personnel may be adversely affected by cumulative stressors experienced during combat deployment. There are several areas for future research, including the use of allostatic load as a biomarker, occupational burnout and its effects on operational performance, and the utility of focused postdeployment screening of health care personnel using measures relevant to their exposures. Health care personnel serving in a combat environment occupy a unique and essential role. Focused attention on this potentially high-risk occupational group is necessary to further characterize their traumatic wartime experiences and mental health needs. ACKNOWLEDGMENTS We thank Science Applications International Corporation for its contributions to this work. REFERENCES 1. Stewart DW: Casualties of war: compassion fatigue and health care providers. Medsurg Nurs 2009; 18: Gibbons SW, Hickling EJ, Watts DD: Combat stressors and posttraumatic stress in deployed military healthcare professionals: an integrative review. J Adv Nurs 2011; 68(1): Kenny DJ, Hull MS: Critical care nurses experiences caring for the casualties of war evacuated from the front line: lessons learned and needs identified. Crit Care Nurs Clin North Am 2008; 20: Biedermann N, Usher K, Williams A, Hayes B: The wartime experience of Australian army nurses in Vietnam, J Adv Nurs 2001; 35: Street AE, Vogt D, Dutra L: A new generation of women veterans: stressors faced by women deployed to Iraq and Afghanistan. Clin Psychol Rev 2009; 29: Kraft HS: Rule number two: lessons I learned in a combat hospital. New York, Little Brown & Company, Hemmings L: Vietnam memories: Australian army nurses, the Vietnam war, and oral history. Nurs Inq 1996; 3: Paul EA: Wounded healers: a summary of Vietnam nurse veteran project. Mil Med 1985; 150: Beck CT: Secondary traumatic stress in nurses: a systematic review. Arch Psychiatr Nurs 2011; 25: de Boer JC, Lok A, van t Verlaat E, et al: Work-related critical incidents in hospital-based health care providers and the risk of post-traumatic stress symptoms, anxiety, and depression: a meta-analysis. Soc Sci Med 2011; 73: Meadors P, Lamson A, Swanson M, White M, Sira N: Secondary traumatization in pediatric healthcare providers: compassion fatigue, burnout, and secondary traumatic stress. Omega (Westport) ; 60: Palm KM, Polusny MA, Follette VM: Vicarious traumatization: potential hazards and interventions for disaster and trauma workers. Prehosp Disaster Med 2004; 19: Jones M, Fear NT, Greenberg N, et al: Do medical services personnel who deployed to the Iraq war have worse mental health than other deployed personnel? Eur J Public Health 2008; 18: MILITARY MEDICINE, Vol. 180, August 2015

6 14. Hickling EJ, Gibbons S, Barnett SD, Watts D: The psychological impact of deployment on OEF/OIF healthcare providers. J Trauma Stress 2011; 24: Reger MA, Gahm GA, Swanson RD, Duma SJ: Association between number of deployments to Iraq and mental health screening outcomes in US Army soldiers. J Clin Psychiatry 2009; 70: Under Secretary of Defense for Personnel and Readiness: Deployment health (DoD Instruction ). Washington, DC, Department of Defense, Available at pdf/649003p.pdf; accessed January 30, Hart AC, Stegman MS, Ford B (editors): ICD-9-CM professional for physicians, volumes 1 & 2: International Classification of Diseases, 9th Revision, Clinical Modification, Ed 6. Salt Lake City, UT, Igenix, Under Secretary of Defense for Personnel and Readiness: Occupational conversion index: Enlisted/officer/civilian (DoD Instruction ). Washington, DC, Department of Defense, Available at accessed January 30, icasualties.org: Iraq Coalition Casualty Count. Available at accessed January 30, McEwen BS: Stress, adaptation, and disease. Allostasis and allostatic load. Ann N Y Acad Sci 1998; 840: McEwen BS, Stellar E: Stress and the individual. Mechanisms leading to disease. Arch Intern Med 1993; 153: Juster RP, McEwen BS, Lupien SJ: Allostatic load biomarkers of chronic stress and impact on health and cognition. Neurosci Biobehav Rev 2010; 35: Stassen W, Van Nugteren B, Stein C: Burnout among advanced life support paramedics in Johannesburg, South Africa. Emerg Med J 2013; 30(4): Nirel N, Goldwag R, Feigenberg Z, Abadai D, Halpern P: Stress, work overload, burnout, and satisfaction among paramedics in Israel. Prehosp Disaster Med 2008; 23: McEwen BS, Seeman T: Protective and damaging effects of mediators of stress. Elaborating and testing the concepts of allostasis and allostatic load. Ann N Y Acad Sci 1999; 896: Seeman TE, Singer BH, Rowe JW, Horwitz RI, McEwen BS: Price of adaptation allostatic load and its health consequences. Arch Intern Med 1997; 157: Adams S, Camarillo C, Lewis S, McNish N: Resiliency training for medical professionals. US Army Med Dept J 2010; April June: Kremer V: Graphic novel helps corpsmen cope with combat-related stress. Navy News, Available at display.asp?story_id=56546; accessed January 30, Maslach C: Burnout: A Multidimensional Perspective. Philadelphia, PA, Taylor & Francis, Mealer M, Burnham EL, Goode CJ, Rothbaum B, Moss M: The prevalence and impact of post-traumatic stress disorder and burnout syndrome in nurses. Depress Anxiety 2009; 26: Prins JT, van der Heijden FM, Hoekstra-Weebers JE, et al: Burnout, engagement and resident physicians self-reported errors. Psychol Health Med 2009; 14: Shanafelt TD, Balch CM, Bechamps G: Burnout and medical errors among American surgeons. Ann Surg 2010; 251: MILITARY MEDICINE, Vol. 180, August