PTSD Elderly War Veterans:A Clinical Controlled Pilot Study

Pedro Gamito, Ph.D.,1 Jorge Oliveira, M.Sc.,1 Pedro Rosa, M.Sc.,2 Diogo Morais, M.Sc.,1

Nuno Duarte, B.Sc.(Hons),3 Susana Oliveira, M.Sc.,3 and Tomaz Saraiva, M.Sc.1

Abstract

Around 25,000 war veterans in Portugal suffer posttraumatic stress disorder (PTSD). This clinically controlledstudy evaluates virtual reality exposure therapy (VRET) as an alternative procedure to reduce PTSD symptoms.Ten patients were assigned to three groups: VRET, exposure in imagination (EI), and waiting list (WL). Thepatients were Portuguese veterans from a series of wars fought in former African colonies more than 30 yearsago. While the EI group participated in traditional imagination therapy, the VRET group was exposed to avirtual reality (VR) war scenario. Cues such as ambush, mortar blasting, and waiting for rescue were used in theVR. Patients enrolled in the VRET group showed statistical reduction of PTSD-associated disorders like de-pression and anxiety. Far from being conclusive, this pilot study nonetheless presents some promising data onthe use of VRET on old war veteran populations.

Introduction

The incidence of anxiety disorders in psychiatricpopulations is 18.1%, with a lifetime prevalence of 28.8%

being the most common type of psychiatric disorder.1 Despitenot being the most prevalent anxiety disorder, posttraumaticstress disorder (PTSD) is found in 5.2 million U.S. adultsranging in age from 18 to 54 in any given year.2 A largespectrum of traumatic events can lead to PTSD. Warfare is oneof the severest due to the physical, emotional, cognitive, andpsychological high demands of the combat environment.3 Arecent study of soldiers from the Iraq conflict revealed that15.6% to 17.1% meet the criteria for major depression, gener-alized anxiety, or PTSD.4 A similar study was conducted inthe Portuguese adult population, revealing a PTSD preva-lence of 7.87%. Concerning war PTSD, around 10% of warveterans developed this disorder.5

The cognitive-behavior therapy (CBT), a time-limited,present-oriented approach to psychotherapy improves pa-tients’ cognitive and behavioral competencies needed tofunction adaptively.6 The exposure in imagination and thein vivo exposure are the most common therapeutic tech-niques.7 Nevertheless, they both have some shortcomingswhere PTSD treatment is concerned. This might happen be-cause there is substantial difficulty inducing an anxiousmental scenario in some PTSD patients.8

The increasing accessibility to powerful personal comput-ers and 3D visualization techniques has made possible thedevelopment of virtual reality (VR) and its use on mentalhealth. This technological technique (tech-tech) is being usedto treat patients with several anxiety disorders.9 Virtual real-ity exposure therapy (VRET) might be a better technique dueto its realistic approximation of the real world.10 VRETinduces higher levels of immersion than does imaginationexposure.11–13 Hyperrealistic threatening stimuli provided byVRET leads to higher attention10 and subsequent encapsula-tion, indicating that once the fear system is activated, it isdifficult to control its response by verbal instructions orstimulus consciousness.14 According to Vincelli,10 VRET thenreduces the decalage between reality and imagination by di-minishing potential distraction or cognitive avoidance of thethreatening stimuli. These and other studies9,15–19 revealedthat VRET can be an alternative to in vivo and imaginationexposure. VRET enables patients to be immersed in the VRworld, creating the sense of being there. This sense, also calledpresence, provides the ability to interact with the VR world asif one were truly in the real environment.20

In VRET, therapists may also have full control of the VRworld. They can add or withdraw threatening cues accordingto the patient and treatment requirements. VRET allows thetherapist to decide the events and even change the envi-ronment via a screen console. Exposure can be repeated ad

1Universidade Lusofona de Humanidades e Tecnologias, Lisbon, Portugal.2Lisbon University Institute (ISCTE-IUL), Lisbon, Portugal.3APOIAR, Lisbon, Portugal.

CYBERPSYCHOLOGY, BEHAVIOR, AND SOCIAL NETWORKING

Volume 13, Number 1, 2010ª Mary Ann Liebert, Inc.DOI: 10.1089=cyber.2009.0237

43

infinitum without the patient leaving the therapist’s room.Particularly for patients with PTSD, it is possible to reproducecues of events that are not replicable in real-life situations.21 Thetherapeutic VRET potential and its advantages in war PTSDtreatment have been shown in several studies.13,22–28 Between1961 and 1974, Portugal was engaged in three wars in its formercolonies of Angola, Mozambique, and Guinea-Bissau. Duringthis period, around 1 million soldiers were drafted to fight in aguerrilla-type war. Many of them developed PTSD after re-turning home. Today, it is estimated that around 25,000 warveterans still meet the DSM-IV criteria for PTSD.31

The purpose of the present study is to investigate whetherVRET brings about a decrease of depression, anxiety, andPTSD symptoms in elderly war veterans when compared toveterans exposed to imagination and those enrolled on awaiting list.

Method

Participants

The study sample consisted of 10 male war veteran pa-tients (M¼ 63.5 years; SD¼ 4.43) with war PTSD. Eachpatient was diagnosed by his assistant psychiatrist througha clinician-administered PTSD scale (CAPS)32 structured in-terview from DSM-IV.33 These patients participated in thePortuguese colonial war between 1963 and 1970 in Africa.

The participants were randomly assigned to three studygroups: five to the VRET group, two to the exposure inimagination (EI) group, and three to the waiting list (WL)group. The VRET group suffered one drop-out.

Each patient’s assistant psychiatrist agreed to maintainstable medication regimens in order for the patient to par-ticipate in this study. Prior to enrollment, patients gave theirinformed consent. Exclusion criteria consisted of cardiovas-cular disease, epilepsy, and absence of informed consent fromthe assistant psychiatrist.

Materials=apparatus

Regarding the self-report measures, PTSD symptomsavoidance, intrusion, and hyperarousal were evaluated bythe Impact of Events Scale Revised (IES-R).34 Psychopathol-ogy and depressive symptoms were respectively assessed bythe Symptoms Checklist Revised (SCL-90-R)35 and the BeckDepression Inventory (BDI).36 The VRET group was also as-sessed with self-report measures in order to control VR-related variables like the sense of presence, immersion, andcybersickness (ITC SOPI).37

Patients were exposed through a Z800 head-mounteddisplay (e-Magin Corp.) plugged to a high-end computerwith dual core processor (4 GB RAM) and a 3D graphics unitwith DirectX 10 support (Nvidia 9600 GT). The hardwarewas installed in the Association for War Veterans in Lisbon,Portugal.

Procedure

The study design consisted of a controlled study (i.e.,VRET, EI, and WL groups). Both the VRET and EI conditionsmaintained the regular psychotherapy rationale based oncognitive desensitization with 12 graded exposure sessions ofVR for the VRET condition and 12 sessions of imaginationexposure for the EI condition. Patients in the WL condition(control group) were enrolled in the VRET group after thefinal assessment. These results will be discussed at a later datein a following study.

The PTSD clinical diagnosis was established in the firstsession along with the self-report measures for PTSD symp-toms (IES-R) and psychopathology (SCL-90-R). The lastassessment was conducted with the CAPS, IES-R, and SCL-90-R. Data for depression (BDI) was also collected for theVRET condition.

In the VRET group, exposure began in the second treatmentsession; self-report data for the sense of presence was alsocollected at the end of the session. For both experimentalgroups, VRET and EI, the first session was dedicated to pa-tients’ anamnesis and psychoeducation for anxiety manage-ment techniques. In order to better understand VRET, thepatients in the experimental condition were also assessedduring treatment (at the end of the fifth session) with self-report measures (IES and SCL-90-R). Being an elderly andchronic sample, one of the challenges of the study was to re-duce the chances of dropping out. In this way, the assessment’sevents and instruments were reduced to a minimum in ordernot to overload the patients and break the therapeutic flow.

The displayed VR world was developed using Hammergraphics editor (Valve Corp.) and consisted of a footpathsurrounded by dense vegetation, where the participants hadto follow a column of virtual soldiers (Fig. 1).

As seen in Figure 2, the activating episodes were composedof three typified cues. Each cue is described as follows: am-bush (A), sounds of gun firing (AK 47) and tracing bullets;mortar (M), sound of blasting amidst black smoke andspraying particles; waiting for an evacuation (E), waiting nearan injured soldier for an Alouette II to fly in and evacuatehim. According to this hierarchy, Figure 2 shows the in-creasing intensity and frequency from session to session.

FIG. 1. Activating cues in the virtual reality exposure therapy.

44 GAMITO ET AL.

Results

To evaluate the effect of the experimental group on thedependent variables for PTSD assessment and psychopa-thology, statistical analyses were carried out through a re-peated measures ANOVA with one within-participant factorwith two levels (pretreatment and posttreatment) and onebetween-participants factor with three levels (VRET, EI, andWL groups).

For the CAPS screening scores, the ANOVA showed nostatistically significant effects ( pi> 0.05). However, throughthe descriptive analyses. an 8% decrease was observed onPTSD symptoms for the posttreatment condition in VRETgroup as well as for EI (1%) and WL groups (6%). CriterionsB, C, and D are described in Table 1.

The same design was performed for the self-report mea-sures of PTSD. Descriptive analysis showed a reduction of theIES-R scores only for the VRET condition, whereas for the WLand EI, an increase was reported (Table 1).

Regarding the psychopathology dimensions, a reductionof SCL-90-R mean scores for the VRET group was observed.Inferential analysis (ANOVA) showed an effect of statisticalinteraction between factors for depression, F(1, 5)¼ 12.490;p¼ 0.011. This indicates a significant decrease in depressivesymptoms (Fig. 3) from the pretreatment to posttreatmentassessment for the VRET group.

Descriptive analysis for the sense of presence (ITC-SOPI)indicated that 65% of the patients agreed, or strongly agreed,that they were present in the VR world, and 62% reported nonegative effects after exposure.

The BDI for depression was assessed only for theVRET group. Repeated measures ANOVA with one within-participants factor with two levels (pretreatment andposttreatment) revealed significant differences between as-sessments for depression total score in the VRET condition,(F(1, 3)¼ 75.000; p¼ 0.003), reflecting a significant decreasefrom pretreatment to posttreatment assessment. As seen inFigure 4, the BDI total scores decreased about 40% frommoderate depression (M¼ 24.25; SD¼ 9.46) to mild depres-sion state (M¼ 14.25; SD¼ 7.67).36

Within group analysis, VRET group was evaluated withSCL-90-R and the IES-R in three assessments: pretreatment,during treatment (after the fifth session) and posttreatment(after the 12th session). SCL-90-R scores were analyzedthrough a repeated measures ANOVA with one within-participants factor with three levels (pretreatment, duringtreatment, and posttreatment). The results showed statisti-cally significant differences between assessments for the fol-lowing SCL-90R dimensions: somatization, F(2, 4)¼ 24.589;p< 0.01, and anxiety, F(2, 4)¼ 9.224; p< 0.05, with a signifi-cant decrease on psychopathology scores on the posttreat-ment assessment (Fig. 5). Estimated marginal means werecompared through post hoc tests with Bonferroni adjustment.The pretreatment assessment significantly differed only fromthe posttreatment assessment ( p< 0.05). No significant dif-ferences were observed for the other psychopathology di-mensions. Concerning IES scores related to PTSD symptoms,the same design was conducted. Despite the decrease of IESmean scores, the ANOVA revealed no statistically significantdifferences among the three assessments for the PTSD self-report measures.

Discussion

This study focused on a controlled trial that compared warveterans with PTSD symptoms using three conditions: VRET,EI, and WL. VRET participants were exposed to a VR world,where anxiety-propelling cues were displayed (ambush,

Table 1. Percentage Variation between

Pretreatment and Posttreatment Assessments

for VRET, EI, and WL Groups

VRET (%) EI (%) WL (%)

CAPS total severity score �8 �1 �6Intrusion (criterion B) �18 �1 �20Avoidance (criterion C) �38 4 �13Hyperarousal (criterion D) 10 �10 20IES: avoidance 9 11 24IES: intrusion �7 �5 �2IES: hyperarousal �4 0 8IES-R score �1 1 7

EI, exposure in imagination; VRET, virtual reality exposuretherapy; WL, waiting list.

FIG. 2. Virtual reality exposure therapy design.

VRET AND PTSD IN WAR VETERANS 45

FIG. 3. Effects of the experimental group on the SCL-90-R depression subscale for virtual reality exposure therapy, exposurein imagination, and waiting list groups.

FIG. 4. Beck depression analysis between pretreatment and posttreatment assessments for virtual reality exposure therapy.

FIG. 5. Somatization and anxiety analysis between treatment assessments (pretreatment, during treatment, and post-treatment) for virtual reality exposure therapy.

mortar=mine blasting, and waiting near an injured soldier forhelicopter rescue).

VRET seemed to deliver at least the same results as EI. Theresults obtained after the last assessment (12th session)showed a decrease on PTSD (CAPS) as well as on psycho-pathological (SCL-90-R) symptoms in the VRET group whencompared to EI and WL groups. This decrease was significantonly for depression scores. Despite not having many con-trolled studies available,24–30,38–41 the same trend was found.

VRET appears to have also caused a reduction in patients’depression and anxiety levels. After the 12 therapy sessions,the veterans who were exposed to VR cues reported lessanxious behavior and reduced depressive symptomatology.

Despite the difficulty of generalization due to a smallsample size, these results are encouraging. They suggest thatelderly war veterans with chronic PTSD can benefit fromVRET. However, a follow-up assessment should done 6months after the last session in order to corroborate thesefindings and account for the nature of PTSD.

In a further study, an enlargement of the sample sizewould be required to evaluate the reproducibility of this gain.

Also, psychophysiological and neuropsychophysiologicalassessments and monitoring should be considered. The firstcan provide a better insight of the fear response associatedwith the anxiety cues during VRET. The second can help inbetter understanding the anatomic and functional mecha-nisms of PTSD. Functional neuroimaging42 suggests thatPTSD patients differ cognitively and behaviorally while re-experiencing traumatic events. The authors suggest that dif-ferent neuronal mechanisms may generate these differentreactions of traumatic triggers. Regardless of the findings, thepathophysiologic nature underlying PTSD is not yet clear.Moreover, the effects of psychotherapeutic treatment proto-cols on the structure, and consequently on the functions ofrelated brain systems, are not yet known.

Another limitation concerns the number of evaluationsmade for the EI and WL groups, as stated in the methodsection. This design did not allow for group comparisonsamong the three assessments. Yet these preliminary resultssuggest that VR can be an option for the psychotherapeuticprocess. As demonstrated by other studies,13,22–28,38–41 it ispossible43 through VRET to force patients to relive and to beemotionally engaged in a traumatic war event—one of thecentral goals of psychotherapy.

Acknowledgments

The VR world was developed under a grant from thePortuguese Military Academy.

Disclosure Statement

No competing financial interests exist.

References

1. Kessler RC, Chiu W, Demler O, et al. Prevalence, severity,and comorbidity of 12-month DSM-IV disorders in the Na-tional Comorbidity Survey Replication. Archives of GeneralPsychiatry 2005; 62:617–27.

2. Kessler R, McGonagle K, Zhao S, et al. Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the

United States: results from the National Comorbidity Sur-vey. Archives of General Psychiatry 1994; 51:8–19.

3. Rizzo A, Pair J, Graap K, et al. (2006) A virtual reality ex-posure therapy application for Iraq war military personnelwith posttraumatic stress disorder: from training to toy totreatment. In Roy M, ed. NATO Advanced research workshopon novel approaches to the diagnosis and treatment of posttrau-matic stress disorder. Washington DC: IOS Press, pp. 235–50.

4. Hoge C, Castro C, Messer S, et al. Combat duty in Iraq andAfghanistan, mental health problems, and barriers to care.New England Journal of Medicine 2004; 351:13–22.

5. Albuquerque A, Soares C, Jesus P, et al. Perturbacao pos-traumatica do stress (PTSD): avaliacao da taxa de ocorrenciana populacao adulta Portuguesa. Acta Medica Portuguesa2003; 16:309–20.

6. Najavits L. (2007) Psychosocial treatments for posttraumaticstress disorder. In Nathan PE, Gorman JM, eds. A guide totreatments that work. 3rd ed. New York: Oxford UniversityPress, pp. 513–29.

7. Leahy RL. (2003) Cognitive therapy techniques: a practi-tioner’s guide. New York: Guilford Press.

8. Riva G, Alcaniz A, Anolli L, et al. The VESPY updatedproject: virtual environments in clinical psychology. Cy-berPsychology & Behavior 2001; 4:449–55.

9. Cote S, Bouchard S. Virtual reality exposure’s efficacy in thetreatment of specific phobias: a critical review. Journal ofCybertherapy & Rehabilitation 2008; 1:75–91.

10. Vincelli F, Riva G. Virtual reality as a new imaginative toolin psychotherapy. Studies in Health Technology & Infor-matics 2000; 70:356–8.

11. Riva G, Molinari E, Vincelli F. (2002). Interaction and pres-ence in the clinical relationship: virtual reality (VR) ascommunicative medium between patient and therapist.IEEE Transactions on Information Technology in Biomedi-cine 2002; 6:198–205

12. Botella C, Banos RM, Villa H, et al. Virtual reality in thetreatment of claustrophobic fear: a controlled, multiple-baseline design. Behavior Therapy 2000; 31:583–95.

13. Rothbaum B, Hodges L, Smith S, et al. A controlled study ofvirtual reality exposure therapy for fear of flying. Journal ofConsulting & Clinical Psychology 2000; 68:1020–6.

14. Hamm A, Weike A. The neuropsychology of fear-learningand fear regulation. International Journal of Psychophy-siology 2005; 57:5–14.

15. Gerardi M, Rothbaum B, Ressler K, et al. Virtual reality ex-posure therapy using a virtual Iraq: case report. Journal ofTraumatic Stress 2008; 21:209–13.

16. Stetz M, Thomas M, Russo M, et al. Stress, mental health,and cognition: a brief review of relationships and counter-measures. Aviation, Space, & Environmental Medicine 2007;78:252–60.

17. Gamito P, Oliveira J, Morais D, et al. Virtual reality therapycontrolled study for war veterans with PTSD: preliminaryresults. Studies in Health Technology & Informatics 2009;144:269–72.

18. Reger G, Gahm G. Virtual reality exposure therapy for activeduty soldiers. Journal of Clinical Psychology 2008; 64:940–6.

19. Wood D, Murphy J, McLay R, et al. Cost effectiveness ofvirtual reality graded exposure therapy with physiologicalmonitoring for the treatment of combat related posttrau-matic stress disorder. Studies in Health Technology & In-formatics 2009; 144:223–9.

20. Ditton T, Lombard M. At the heart of it all: the concept ofpresence. Journal of Computer-Mediated Communication

VRET AND PTSD IN WAR VETERANS 47

1997; 3. http:==jcmc.indiana.edu=vol3=issue2=lombard.html(accessed Oct. 2009).

21. Gamito P, Saraiva T, Oliveira J, et al. Physiological assess-ment of motor vehicle accident posttraumatic stress disorderpatient during virtual reality exposure: a clinical case study.Journal of Cybertherapy & Rehabilitation 2008; 1:258–67.

22. Rizzo A, Pair J, Parsons T, et al. (2006) A virtual reality therapyapplication for OEF=OIF combat-related posttraumatic stressdisorder. Marina del Rey, CA: USC Institute for CreativeTechnologies.

23. Rizzo A, Reger G, Gahm G, et al. (2008) Virtual reality expo-sure therapy for combat-related PTSD. In Shiromani P, KeaneT, LeDoux J, eds. Posttraumatic stress disorder: basic science andclinical practice. Totowa, NJ: Humana Press, pp. 93–122.

24. Miyahira S, Hoffman H, Folen R. Virtual reality in the treat-ment of combat-related posttraumatic stress disorder withwarfighters. CyberPsychology & Behavior 2006; 9:701–2.

25. Miyahira S, Hoffman H, Folen R, et al. Virtual reality ex-posure: efficacious treatment for combat PTSD? CyberPsy-chology & Behavior 2009; 12:89–90.

26. Wood D, Murphy J, Center K, et al. Combat-related post-traumatic stress disorder: a case report using virtual realitywith physiological monitoring. CyberPsychology & Beha-vior 2007; 10:309–15.

27. Wood D, Murphy J, Center K, et al. Combat related post-traumatic stress disorder: a multiple case report using vir-tual reality graded exposure therapy with physiologicalmonitoring. Studies in Health Technology & Informatics2008; 132:556–61.

28. Rizzo A, Graap K, Perlman K, et al. Virtual Iraq: initial re-sults from a VR exposure therapy application for combat-related PTSD. Studies in Health Technology & Informatics2008; 132:420–5.

29. Rizzo A, Graap K, McLay R, et al. Clinical results from avirtual Iraq PTSD exposure therapy application. Studies inHealth Technology & Informatics; in press.

30. Roy M, Francis J, Friedlander J, et al. Initial results from theViRTICo trial: virtual reality therapy and imaging in combatveterans. Annual Review of Cybertherapy and Telemedicine2008; 6:49–54.

31. Gamito P, Oliveira J, Morais D, et al. War PTSD: a VR pre-trial case study. Annual Review of Cybertherapy & Tele-medicine 2007; 5:191–8.

32. Blake A, Weathers F, Nagy L, et al. A clinician rating scalefor assessing current and lifetime PTSD: the CAPS-1. Beha-vior Therapist 1992; 13:187–8.

33. American Psychiatric Association. (2000). Diagnostic andstatistical manual of mental disorders. 4th ed. Washington, DC:Author.

34. Weiss DS, Marmar CR. (1996) The Impact of Event Scale–Revised. In Wilson J, Keane TM, eds. Assessing psychologicaltrauma and PTSD. New York: Guilford, pp. 399–411.

35. Derogatis LR. (1994). SCL-90-R: Administration, scoring andprocedures manual. Minneapolis MN: National ComputerSystems.

36. Beck AT, Steer RA, Garbin GM. Psychometric properties ofthe Beck Depression Inventory: twenty-five years of evalu-ation. Clinical Psychology Review 1988; 8:77–100.

37. Lessiter J, Freeman J, Keogh E, et al. A cross-media presencequestionnaire: the ITC-Sense of Presence Inventory. Presence:Teleoperators & Virtual Environments 2001; 10:282–97.

38. Rothbaum B, Hodges L, Alarcon R, et al. A virtual envi-ronment for the treatment of chronic combat-related post-traumatic stress disorder. Journal of Traumatic Stress 1999;12:263–71.

39. Rothbaum BO, Hodges L, Ready D, et al. Virtual real-ity exposure therapy for Vietnam veterans with posttrau-matic stress disorder. Journal of Clinical Psychiatry 2001; 62:617–22.

40. Walshe D, Lewis E, O’Sullivan K, et al. Virtually driving:Are the driving environments ‘‘real enough’’ for exposuretherapy with accident victims? An explorative study. Cyber-Psychology & Behavior 2005; 8:532–7.

41. Rizzo A, Difede J, Rothbaum B, et al. VR PTSD exposuretherapy results with active duty OIF=OEF combatants.Medicine Meets Virtual Reality 2009; 17.

42. Lanius RA, Williamson PC, Densmore M, et al. The natureof traumatic memories: a 4-T fMRI functional connectivityanalysis. American Journal of Psychiatry 2004; 161:36–44.

43. Rothbaum B, Hodges L, Alarcon R, et al. Virtual realityexposure therapy for PTSD Vietnam veterans: a case study.Journal of Traumatic Stress 2005; 12:263–71.

Address correspondence to:Dr. Pedro Gamito

Universidade Lusofona de Humanidades e TecnologiasCampo Grande, 376

LisbonPortugal 1749-024

E-mail: pedro.gamito@gmail.com

48 GAMITO ET AL.