Journal of Traumatic Stress December 2013, 26, 772–775
BRIEF REPORT
The Role of Military Social Support in Understanding the Relationship Between PTSD, Physical Health, and Healthcare Utilization in Women Veterans Keren Lehavot,1,2 Claudia Der-Martirosian,3 Tracy L. Simpson,1,2 Jillian C. Shipherd,4,5,6,7 and Donna L. Washington8,9 1
VA Puget Sound Health Care System, Seattle, Washington, USA Department of Psychiatry and Behavioral Sciences, University of Washington, Seattle, Washington, USA 3 Veterans Emergency Management Evaluation Center (VEMEC), Department of Veterans Affairs, North Hills, California, USA 4 National Center for PTSD, Women’s Health Sciences Division, Boston, Massachusetts, USA 5 VA Boston Healthcare System, Boston, Massachusetts, USA 6 Boston University School of Medicine, Boston, Massachusetts, USA 7 VA Patient Care Services LGBT Program, Washington, DC, USA 8 VA Greater Los Angeles Health Services Research and Development (HSR&D) Center for the Study of Healthcare Innovation, Implementation & Policy (CSHIIP), Sepulveda, California, USA 9 Department of Medicine, University of California Los Angeles (UCLA) David Geffen School of Medicine, Los Angeles, California, USA 2
Posttraumatic stress disorder (PTSD) is a significant predictor of both poorer physical health and increased health care utilization, whereas adequate social support is associated with better physical health and less health care utilization. However, research has not previously examined the simultaneous effects of PTSD and social support on health and health care utilization. This study examined both the independent and interactive effects of PTSD and a particular type of social support (postactive-duty social support from military friends) on self-reported physical health and number of Veterans Health Administration (VHA) visits in the last year. These relationships were examined in a representative, national sample of 3,524 women veterans who completed telephone interviews as part of the National Survey of Women Veterans in 2008–2009. Regression analyses were conducted using these cross-sectional data to examine main effects of PTSD and military social support on physical health and VHA utilization and their interaction. Screening positive for PTSD was associated with poorer health (B = −3.19, SE = 1.47) and increased VHA utilization (B = 0.98, SE = 0.16), whereas greater military social support was associated with better health (B = 0.97, SE = 0.44) and less frequent VHA utilization (B = −0.15, SE = 0.05). Neither moderation model was significant, such that military social support behaved in a similar way regardless of PTSD status.
Women veterans comprise a growing population with high levels of trauma (Zinzow, Grubauth, Monnier, SuffolettaMaierle, & Frueh, 2007), and consequently high prevalence rates of posttraumatic stress disorder (PTSD; Escalona, Achilles, Waitzkin, & Yager, 2004). PTSD symptoms have been found to be significantly associated with both postdeployment physical health (Smith et al., 2011) and with greater healthcare utilization among this population (Suris, Lind, Kashner, Borman, & Petty, 2004). Little research has examined protective factors that may attenuate these relationships. Social support may be one such factor, especially among women veterans as this group reports poorer social support than their male counterparts (Frayne et al., 2006). Lower social support has been independently associated with PTSD (Pietrzak, Johnson, Goldstein, Malley, & Southwick,
This study was funded by the Department of Veterans Affairs (VA) Women’s Health Services in the Office of Patient Care Services and the VA Health Services Research and Development (HSR&D) Service (#SDR-08-270). This material was also supported by resources from the VA Office of Academic Affiliations, Advanced Fellowship Program in Mental Illness Research and Treatment, VA Puget Sound Healthcare System, VA Greater Los Angeles HSR&D Center of Excellence, and VA Boston Healthcare System. The authors gratefully acknowledge Su Sun Mor, MPH and Michael Mitchell, PhD, for assistance with database construction, and Mark Canning for project management. The views expressed within are solely those of the authors, and do not necessarily represent the views of the Department of Veterans Affairs or of the United States government. Correspondence concerning this article should be addressed to Keren Lehavot, 1660 S. Columbian Way (S-116-POC), Seattle, WA 98108. E-mail: [email protected] Published 2013. This article is a US Government work and is in the public domain in the USA. View this article online at wileyonlinelibrary.com DOI: 10.1002/jts.21859
772
Social Support, PTSD, and Health in Women Veterans
2009), poorer physical health (Uchino, 2004), and increased health care utilization (Kouzis & Eaton, 1998). However, less is known about specific aspects of social support, such as support from military peers, and its role as a potential protective factor. One study found that male combat veterans distinguished between specific support providers (e.g., family, friends, military peers), and that support from military peers was associated with lower PTSD symptoms (Wilcox, 2010). Furthermore, it is unclear if social support predicts physical health and health care utilization when taking PTSD into account, and whether it may influence the relationships between PTSD and health or PTSD and utilization. On the one hand, the effect of social support on health and utilization may be more pronounced for those with PTSD relative to those without. On the other hand, a recent study found that social support was less protective for suicide risk for veterans screening positive for PTSD relative to those who did not screen positive (Jakupcak et al., 2010). This study examined the simultaneous independent effects of PTSD and postactive-duty social support from military friends on physical health and health care utilization in the Veterans Health Administration (VHA) among women veterans. We predicted that PTSD and military social support would each contribute unique variance to these outcomes. Also, the interactive effects of PTSD and military social support were evaluated. These examinations were exploratory given the existence of competing hypotheses about how social support may operate in the presence of PTSD.
Method Participants and Procedure Participants in the National Survey of Women Veterans represent a national, cross-sectional, population-based, stratified random sample (Washington, Sun, & Canning, 2010). Data were collected by telephone in 2008–2009 from all veterans including National Guard or Reserves who had been called to duty, but excluding active-duty military, VA employees, hospitalized veterans, and nursing home residents. Sampled veterans were mailed an information packet and then screened for eligibility by study interviewers. This study was approved by the VA Greater Los Angeles Institutional Review Board and the survey was also approved by the U.S. Office of Management and Budget. Eligible women veterans (N = 4,535) were identified with 3,611 participants. This sample includes the 3,524 women with complete data on key variables. The mean age was 55.99 years (SD = 17.61). Using weighted percentages, 56.1% (n = 1,070) reported a household income of $50,000 or more, 76.9% were non-Hispanic Whites, 11.0% non-Hispanic Blacks, 4.6% Hispanic, and 7.5% other. Fourteen percent (14.2%) served prior to the Vietnam war, 4.8% were Operation Enduring Freedom/Operation Iraqi Freedom veterans, and 81.0% served during all other military service periods.
773
Measures Presence of probable lifetime PTSD was ascertained via a screening scale based on the Diagnostic and Statistical Manual of Mental Disorders (4th ed., DSM-IV; American Psychiatric Association, 1994; Breslau, Peterson, Kessler, & Schultz, 1999). If participants indicated a Criterion A event (e.g., combat exposure, military sexual trauma, or nonmilitary traumas), they were asked if they had experienced each of seven symptoms (five avoidance and numbing items, two arousal items) as a result of the traumatic event. A total score of 4 or more is considered PTSD positive, and had 80.3% sensitivity and 97.3% specificity for PTSD (Breslau et al., 1999). In the current sample, 69.9% (n = 2,478) reported trauma-exposure, and of these 19.1% (n = 751) screened positive for PTSD. Postactive-duty military social support was assessed with the item: “You have been able to maintain the social support of your military friends.” Participants indicated whether they strongly agree, somewhat agree, somewhat disagree, or strongly disagree with this statement, rated on a continuous Likert scale from 4 to 1. The SF-12 Health Survey version 2 (SF-12v2; Ware, Kosinski, & Keller, 1996) is composed of 12 items. The Physical Component Summary (PCS) score is a composite t score about perceived physical health functioning that was calculated via a standard scoring algorithm based on responses to all 12 items, with scores ranging from 0 = worst health to 100 = best health. The summary score has good test-retest reliability and is highly correlated with the well-validated SF-36 (Ware, Kosinski, Turner-Bowker, & Gandek, 2002). In the current sample, the mean PCS score was 46.74 (SD = 10.39). Participants were asked to think about all of the outpatient health care visits they had during the past 12 months and indicate how many were at the VHA or paid for by the VHA. They were instructed not to include visits they had while they were patients in the hospital. In the current sample, the mean number of visits was 5.00 (SD = 13.74). Data Analysis Hierarchical linear regression analysis was conducted with the PCS as a dependent, continuous variable. A different analytic approach was taken with health care utilization because it represents a count variable (i.e., number of visits) and was positively skewed; therefore, a zero-inflated negative binomial regression was used to predict utilization. For each model, the following background variables were entered into Block 1: age, whether the participant reported sexual assault during military service, and whether the participant had ever served in a combat or war zone. Age was entered due to its known association with worsening physical health (Centers for Disease Control, 2011). Years since military service was strongly correlated with age (r = .91), and therefore was not included in the models. Sexual assault and combat exposure were entered to examine whether PTSD, rather than trauma experiences, was associated with the outcomes. We specifically adjusted for trauma experiences that
Journal of Traumatic Stress DOI 10.1002/jts. Published on behalf of the International Society for Traumatic Stress Studies.
774
Lehavot et al.
Table 1 Linear Regression of Physical Health and Zero-Inflated Negative Binomial Regression of VHA Healthcare Utilization Physical health Predictor variable Block 1 Age Sexual assault Combat exposure Block 2 Age Sexual assault Combat exposure PTSD Military social support Block 3 Age Sexual assault Combat exposure PTSD Military social support PTSD × Military social support
VHA health care utilization SE
Wald’s χ2
0.01 1.01*** 0.18
0.01 0.19 0.13
29.17***
.15***
0.01 0.53** 0.16 0.98*** − 0.15**
0.01 0.19 0.16 0.16 0.05
55.06***
.15***
0.01 0.54** 0.15 0.71** − 0.17** 0.11
0.01 0.19 0.15 0.27 0.06 0.10
61.04***
B
SE
R2
− 0.21*** − 3.05 − 0.40
0.02 1.85 1.13
.13***
− 0.23*** − 1.17 − 0.33 − 3.19* 0.97*
0.02 1.90 1.17 1.47 0.44
− 0.23*** − 1.08 − 0.44 − 5.53 0.84 1.00
0.02 1.88 1.15 2.96 0.48 1.12
B
Note. PTSD = posttraumatic stress disorder; VHA = Veterans Health Administration; SE = standard error. *p < .05. **p < .01. ***p < .001.
occurred in the military given the focus on military social support as one of the independent variables. In Block 2, PTSD and military social support were entered. Finally, Block 3 included the PTSD × Military Social Support interaction (see Table 1). All analyses applied weights to account for disproportional allocation of the population by strata (Washington, Bean-Mayberry, Riopelle, & Yano, 2011). The resulting estimates are therefore representative of the U.S. women veteran population. All statistical analyses were conducted using Stata v.12 (StataCorp, 2011). Results The results of the two regression models are shown in Table 1. In the first step for the PCS, only age (but not sexual assault during military service or combat exposure) was significantly associated with poorer physical health. In the second step, both screening positive for PTSD and military social support were independently associated with physical health, with PTSD status predicting poorer health and social support predicting better health. In the third step, the interaction term was not associated with poorer health, suggesting that military social support did not differentially protect against poorer health in veterans screening positive or negative for PTSD. In the first step for VHA utilization, age was marginally positively associated with greater number of visits (p = .062) as well as having endorsed sexual assault while in the military (p < .001). Similar to the previous model, both screening pos-
itive for PTSD and military social support independently predicted VHA visits, with PTSD status predicting greater number of visits and social support predicting the opposite. The interaction term did not significantly predict utilization when entered in the final step. Discussion Screening positive for PTSD and social support from one’s military friends were each uniquely predictive of physical health and VHA utilization in this national, representative sample of women veterans. These relationships held while adjusting for age, which was associated with worse physical health but not VHA utilization; and sexual assault during military service, which was associated with increased VHA utilization but not physical health. We did not find support for either of the moderation models, and instead found that military social support behaved in a similar way regardless of PTSD status. The fact that military social support exerted influence on both outcomes over and above effects of PTSD, and that its beneficial effects were not diminished for women screening positive for PTSD, is an encouraging finding. Findings from this study should be interpreted in light of several limitations. Data were based on self-report, and thus did not include clinician assessment of PTSD or objective information detailing physical health functioning and actual number of VHA visits. Self-reported physician visits may be biased toward underreporting at higher number of visits (Bhandari &
Journal of Traumatic Stress DOI 10.1002/jts. Published on behalf of the International Society for Traumatic Stress Studies.
Social Support, PTSD, and Health in Women Veterans
Wagner, 2006). Because individuals with PTSD tend to have a greater number of health visits, our findings, though significant, may underestimate this association. Assessment of health care utilization only covered care received through VHA and this may also have led to an underestimation of this association. Assessment of military social support was limited to one item on perceptions of social support of military friends. Social support is complex and multifaceted (Shumaker & Brownell, 1984), and future research should use comprehensive measures that account for different sources (e.g., employers and coworkers during deployment, spousal, family, nonmilitary friends) and types of social support. For example, one study found that military social support was protective for men against mental health sequelae, whereas civilian social support was more protective for female Marine recruits (Smith et al., 2013). Additionally, our data are based on a cross-sectional survey that precludes us from making conclusions about causal relationships. Longitudinal studies may provide more nuanced understanding of how and under what circumstances PTSD and military social support influence health and health care visits. In sum, women veterans who screen positive for PTSD are more likely to report poorer physical health and greater VHA utilization, whereas those who report greater maintenance of military social support also report more favorable health and less frequent VHA utilization. Military social support was equally protective regardless of women’s PTSD status, and its associative role with health should be further examined. References American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author. Bhandari, A., & Wagner, T. (2006). Self-reported utilization of health care services: Improving measurement and accuracy. Medical Care Research and Review, 63, 217–235. doi:10.1177/1077558705285298 Breslau, N., Peterson, E. L., Kessler, R. C., & Schultz, L. R. (1999). Short screening scale for DSM-IV posttraumatic stress disorder. American Journal of Psychiatry, 156, 908–911. Centers for Disease Control. (2011). Enhancing use of clinical preventive services among older adults: Closing the gap. Retrieved from www.aarp.org/healthpros Escalona, R., Achilles, G., Waitzkin, H., & Yager, J. (2004). PTSD and somatization in women treated at a VA primary care clinic. Psychosomatics, 45, 291–296. doi:10.1176/appi.psy.45.4.291 Frayne, S. M., Parker, V. A., Christiansen, C. L., Loveland, S., Seaver, M. R., Kazis, L. W., & Skinner, K. M. (2006). Health status among 28,000 women veterans. Journal of General Internal Medicine, 21(S3), S40–S46. doi:10.1111/j.1525-1497.2006.00373.x Jakupcak, M., Vannoy, S., Immel, Z., Cook, J. W., Fontana, A., Rosenheck, R., & McFall, M. (2010). Does PTSD moderate the relationship be-
775
tween social support and suicide risk in Iraq and Afghanistan war veterans seeking mental health treatment? Depression and Anxiety, 27, 1001–1005. doi:10.1002/da.20722 Kouzis, A. C., & Eaton, W. W. (1998). Absence of social networks, social support and health services utilization. Psychological Medicine, 28, 1301– 1310. doi:10.1017/S0033291798007454 Pietrzak, R. H., Johnson, D. C., Goldstein, M. B., Malley, J. C., & Southwick, S. M. (2009). Psychological resilience and postdeployment social support protect against traumatic stress and depressive symptoms in soldiers returning from Operations Enduring Freedom and Iraqi Freedom. Depression and Anxiety, 26, 745–751. doi:10.1002/da.20558 Shumaker, S. A., & Brownell, A. (1984). Toward a theory of social support: Closing conceptual gaps. Journal of Social Issues, 40, 11–36. doi:10.1111/j.1540-4560.1984.tb01105.x Smith, B. N., Shipherd, J. C., Schuster, J. L., Vogt, D. S., King, L. A., & King, D. W. (2011). Posttraumatic stress symptomatology as a mediator of the association between military sexual trauma and post-deployment physical health in women. Journal of Trauma and Dissociation, 12, 275– 289. doi:10.1080/15299732.2011.551508 Smith, B. N., Vaughn, R. A., Vogt, D. S., King, L. A. & King, D. W., & Shipherd, J. C. (2013). Main and interactive effects of social support in predicting mental health symptoms in men and women following military stressor exposure. Anxiety, Stress, and Coping, 26, 52–69. doi:10.1080/10615806.2011.634001 StataCorp. (2011). Stata statistical software (Release 12) [Computer Software]. College Station, TX: Author. Suris, A., Lind, L., Kashner, T. M., Borman, P. D., & Petty, F. (2004). Sexual assault in women veterans: An examination of PTSD risk, health care utilization, and cost of care. Psychosomatic Medicine, 66, 749–756. doi:10.1097/01.psy.0000138117.58559.7b Uchino, B. N. (2004). Social support and physical health: Understanding the health consequences of our relationships. New Haven, CT: Yale University Press. Ware, J. E., Kosinski, M., & Keller, S. D. (1996). A 12-item short-form health survey: Construction of scales and preliminary tests of reliability and validity. Medical Care, 34, 220–233. doi:10.1097/00005650-199603000-00003 Ware, J. E., Kosinski, M., Turner-Bowker, D. M., & Gandek, B. (2002). How to score version 2 of the SF-12 Health Survey (with a supplement documenting version 1). Lincoln, RI: QualityMetric Inc. Washington, D. L., Bean-Mayberry, B., Riopelle, D., & Yano, E. M. (2011). Access to care for women veterans: Delayed healthcare and unmet need. Journal of General Internal Medicine, 26, 655–661. doi:10.1007/s11606011-1772-z Washington, D. L., Sun, S., & Canning, M. (2010). Creating a sampling frame for population-based veteran research: Representativeness and overlap of VA and Department of Defense databases. Journal of Rehabilitation Research & Development, 47, 763–772. doi:10.1682/JRRD.2009.08.0127 Wilcox, S. (2010). Social relationships and PTSD symptomatology in combat veterans. Psychological Trauma: Theory, Research, Practice, and Policy, 2, 175–182. doi:10.1037/a0019062 Zinzow, H. M., Grubaugh, A. L., Monnier, J., Suffoletta-Maierle, S., & Frueh, B. C. (2007). Trauma among female veterans: A critical review. Trauma, Violence, and Abuse, 8, 384–400. doi:10.1177/1524838007307295
Journal of Traumatic Stress DOI 10.1002/jts. Published on behalf of the International Society for Traumatic Stress Studies.
BRIEF REPORT
The Role of Military Social Support in Understanding the Relationship Between PTSD, Physical Health, and Healthcare Utilization in Women Veterans Keren Lehavot,1,2 Claudia Der-Martirosian,3 Tracy L. Simpson,1,2 Jillian C. Shipherd,4,5,6,7 and Donna L. Washington8,9 1
VA Puget Sound Health Care System, Seattle, Washington, USA Department of Psychiatry and Behavioral Sciences, University of Washington, Seattle, Washington, USA 3 Veterans Emergency Management Evaluation Center (VEMEC), Department of Veterans Affairs, North Hills, California, USA 4 National Center for PTSD, Women’s Health Sciences Division, Boston, Massachusetts, USA 5 VA Boston Healthcare System, Boston, Massachusetts, USA 6 Boston University School of Medicine, Boston, Massachusetts, USA 7 VA Patient Care Services LGBT Program, Washington, DC, USA 8 VA Greater Los Angeles Health Services Research and Development (HSR&D) Center for the Study of Healthcare Innovation, Implementation & Policy (CSHIIP), Sepulveda, California, USA 9 Department of Medicine, University of California Los Angeles (UCLA) David Geffen School of Medicine, Los Angeles, California, USA 2
Posttraumatic stress disorder (PTSD) is a significant predictor of both poorer physical health and increased health care utilization, whereas adequate social support is associated with better physical health and less health care utilization. However, research has not previously examined the simultaneous effects of PTSD and social support on health and health care utilization. This study examined both the independent and interactive effects of PTSD and a particular type of social support (postactive-duty social support from military friends) on self-reported physical health and number of Veterans Health Administration (VHA) visits in the last year. These relationships were examined in a representative, national sample of 3,524 women veterans who completed telephone interviews as part of the National Survey of Women Veterans in 2008–2009. Regression analyses were conducted using these cross-sectional data to examine main effects of PTSD and military social support on physical health and VHA utilization and their interaction. Screening positive for PTSD was associated with poorer health (B = −3.19, SE = 1.47) and increased VHA utilization (B = 0.98, SE = 0.16), whereas greater military social support was associated with better health (B = 0.97, SE = 0.44) and less frequent VHA utilization (B = −0.15, SE = 0.05). Neither moderation model was significant, such that military social support behaved in a similar way regardless of PTSD status.
Women veterans comprise a growing population with high levels of trauma (Zinzow, Grubauth, Monnier, SuffolettaMaierle, & Frueh, 2007), and consequently high prevalence rates of posttraumatic stress disorder (PTSD; Escalona, Achilles, Waitzkin, & Yager, 2004). PTSD symptoms have been found to be significantly associated with both postdeployment physical health (Smith et al., 2011) and with greater healthcare utilization among this population (Suris, Lind, Kashner, Borman, & Petty, 2004). Little research has examined protective factors that may attenuate these relationships. Social support may be one such factor, especially among women veterans as this group reports poorer social support than their male counterparts (Frayne et al., 2006). Lower social support has been independently associated with PTSD (Pietrzak, Johnson, Goldstein, Malley, & Southwick,
This study was funded by the Department of Veterans Affairs (VA) Women’s Health Services in the Office of Patient Care Services and the VA Health Services Research and Development (HSR&D) Service (#SDR-08-270). This material was also supported by resources from the VA Office of Academic Affiliations, Advanced Fellowship Program in Mental Illness Research and Treatment, VA Puget Sound Healthcare System, VA Greater Los Angeles HSR&D Center of Excellence, and VA Boston Healthcare System. The authors gratefully acknowledge Su Sun Mor, MPH and Michael Mitchell, PhD, for assistance with database construction, and Mark Canning for project management. The views expressed within are solely those of the authors, and do not necessarily represent the views of the Department of Veterans Affairs or of the United States government. Correspondence concerning this article should be addressed to Keren Lehavot, 1660 S. Columbian Way (S-116-POC), Seattle, WA 98108. E-mail: [email protected] Published 2013. This article is a US Government work and is in the public domain in the USA. View this article online at wileyonlinelibrary.com DOI: 10.1002/jts.21859
772
Social Support, PTSD, and Health in Women Veterans
2009), poorer physical health (Uchino, 2004), and increased health care utilization (Kouzis & Eaton, 1998). However, less is known about specific aspects of social support, such as support from military peers, and its role as a potential protective factor. One study found that male combat veterans distinguished between specific support providers (e.g., family, friends, military peers), and that support from military peers was associated with lower PTSD symptoms (Wilcox, 2010). Furthermore, it is unclear if social support predicts physical health and health care utilization when taking PTSD into account, and whether it may influence the relationships between PTSD and health or PTSD and utilization. On the one hand, the effect of social support on health and utilization may be more pronounced for those with PTSD relative to those without. On the other hand, a recent study found that social support was less protective for suicide risk for veterans screening positive for PTSD relative to those who did not screen positive (Jakupcak et al., 2010). This study examined the simultaneous independent effects of PTSD and postactive-duty social support from military friends on physical health and health care utilization in the Veterans Health Administration (VHA) among women veterans. We predicted that PTSD and military social support would each contribute unique variance to these outcomes. Also, the interactive effects of PTSD and military social support were evaluated. These examinations were exploratory given the existence of competing hypotheses about how social support may operate in the presence of PTSD.
Method Participants and Procedure Participants in the National Survey of Women Veterans represent a national, cross-sectional, population-based, stratified random sample (Washington, Sun, & Canning, 2010). Data were collected by telephone in 2008–2009 from all veterans including National Guard or Reserves who had been called to duty, but excluding active-duty military, VA employees, hospitalized veterans, and nursing home residents. Sampled veterans were mailed an information packet and then screened for eligibility by study interviewers. This study was approved by the VA Greater Los Angeles Institutional Review Board and the survey was also approved by the U.S. Office of Management and Budget. Eligible women veterans (N = 4,535) were identified with 3,611 participants. This sample includes the 3,524 women with complete data on key variables. The mean age was 55.99 years (SD = 17.61). Using weighted percentages, 56.1% (n = 1,070) reported a household income of $50,000 or more, 76.9% were non-Hispanic Whites, 11.0% non-Hispanic Blacks, 4.6% Hispanic, and 7.5% other. Fourteen percent (14.2%) served prior to the Vietnam war, 4.8% were Operation Enduring Freedom/Operation Iraqi Freedom veterans, and 81.0% served during all other military service periods.
773
Measures Presence of probable lifetime PTSD was ascertained via a screening scale based on the Diagnostic and Statistical Manual of Mental Disorders (4th ed., DSM-IV; American Psychiatric Association, 1994; Breslau, Peterson, Kessler, & Schultz, 1999). If participants indicated a Criterion A event (e.g., combat exposure, military sexual trauma, or nonmilitary traumas), they were asked if they had experienced each of seven symptoms (five avoidance and numbing items, two arousal items) as a result of the traumatic event. A total score of 4 or more is considered PTSD positive, and had 80.3% sensitivity and 97.3% specificity for PTSD (Breslau et al., 1999). In the current sample, 69.9% (n = 2,478) reported trauma-exposure, and of these 19.1% (n = 751) screened positive for PTSD. Postactive-duty military social support was assessed with the item: “You have been able to maintain the social support of your military friends.” Participants indicated whether they strongly agree, somewhat agree, somewhat disagree, or strongly disagree with this statement, rated on a continuous Likert scale from 4 to 1. The SF-12 Health Survey version 2 (SF-12v2; Ware, Kosinski, & Keller, 1996) is composed of 12 items. The Physical Component Summary (PCS) score is a composite t score about perceived physical health functioning that was calculated via a standard scoring algorithm based on responses to all 12 items, with scores ranging from 0 = worst health to 100 = best health. The summary score has good test-retest reliability and is highly correlated with the well-validated SF-36 (Ware, Kosinski, Turner-Bowker, & Gandek, 2002). In the current sample, the mean PCS score was 46.74 (SD = 10.39). Participants were asked to think about all of the outpatient health care visits they had during the past 12 months and indicate how many were at the VHA or paid for by the VHA. They were instructed not to include visits they had while they were patients in the hospital. In the current sample, the mean number of visits was 5.00 (SD = 13.74). Data Analysis Hierarchical linear regression analysis was conducted with the PCS as a dependent, continuous variable. A different analytic approach was taken with health care utilization because it represents a count variable (i.e., number of visits) and was positively skewed; therefore, a zero-inflated negative binomial regression was used to predict utilization. For each model, the following background variables were entered into Block 1: age, whether the participant reported sexual assault during military service, and whether the participant had ever served in a combat or war zone. Age was entered due to its known association with worsening physical health (Centers for Disease Control, 2011). Years since military service was strongly correlated with age (r = .91), and therefore was not included in the models. Sexual assault and combat exposure were entered to examine whether PTSD, rather than trauma experiences, was associated with the outcomes. We specifically adjusted for trauma experiences that
Journal of Traumatic Stress DOI 10.1002/jts. Published on behalf of the International Society for Traumatic Stress Studies.
774
Lehavot et al.
Table 1 Linear Regression of Physical Health and Zero-Inflated Negative Binomial Regression of VHA Healthcare Utilization Physical health Predictor variable Block 1 Age Sexual assault Combat exposure Block 2 Age Sexual assault Combat exposure PTSD Military social support Block 3 Age Sexual assault Combat exposure PTSD Military social support PTSD × Military social support
VHA health care utilization SE
Wald’s χ2
0.01 1.01*** 0.18
0.01 0.19 0.13
29.17***
.15***
0.01 0.53** 0.16 0.98*** − 0.15**
0.01 0.19 0.16 0.16 0.05
55.06***
.15***
0.01 0.54** 0.15 0.71** − 0.17** 0.11
0.01 0.19 0.15 0.27 0.06 0.10
61.04***
B
SE
R2
− 0.21*** − 3.05 − 0.40
0.02 1.85 1.13
.13***
− 0.23*** − 1.17 − 0.33 − 3.19* 0.97*
0.02 1.90 1.17 1.47 0.44
− 0.23*** − 1.08 − 0.44 − 5.53 0.84 1.00
0.02 1.88 1.15 2.96 0.48 1.12
B
Note. PTSD = posttraumatic stress disorder; VHA = Veterans Health Administration; SE = standard error. *p < .05. **p < .01. ***p < .001.
occurred in the military given the focus on military social support as one of the independent variables. In Block 2, PTSD and military social support were entered. Finally, Block 3 included the PTSD × Military Social Support interaction (see Table 1). All analyses applied weights to account for disproportional allocation of the population by strata (Washington, Bean-Mayberry, Riopelle, & Yano, 2011). The resulting estimates are therefore representative of the U.S. women veteran population. All statistical analyses were conducted using Stata v.12 (StataCorp, 2011). Results The results of the two regression models are shown in Table 1. In the first step for the PCS, only age (but not sexual assault during military service or combat exposure) was significantly associated with poorer physical health. In the second step, both screening positive for PTSD and military social support were independently associated with physical health, with PTSD status predicting poorer health and social support predicting better health. In the third step, the interaction term was not associated with poorer health, suggesting that military social support did not differentially protect against poorer health in veterans screening positive or negative for PTSD. In the first step for VHA utilization, age was marginally positively associated with greater number of visits (p = .062) as well as having endorsed sexual assault while in the military (p < .001). Similar to the previous model, both screening pos-
itive for PTSD and military social support independently predicted VHA visits, with PTSD status predicting greater number of visits and social support predicting the opposite. The interaction term did not significantly predict utilization when entered in the final step. Discussion Screening positive for PTSD and social support from one’s military friends were each uniquely predictive of physical health and VHA utilization in this national, representative sample of women veterans. These relationships held while adjusting for age, which was associated with worse physical health but not VHA utilization; and sexual assault during military service, which was associated with increased VHA utilization but not physical health. We did not find support for either of the moderation models, and instead found that military social support behaved in a similar way regardless of PTSD status. The fact that military social support exerted influence on both outcomes over and above effects of PTSD, and that its beneficial effects were not diminished for women screening positive for PTSD, is an encouraging finding. Findings from this study should be interpreted in light of several limitations. Data were based on self-report, and thus did not include clinician assessment of PTSD or objective information detailing physical health functioning and actual number of VHA visits. Self-reported physician visits may be biased toward underreporting at higher number of visits (Bhandari &
Journal of Traumatic Stress DOI 10.1002/jts. Published on behalf of the International Society for Traumatic Stress Studies.
Social Support, PTSD, and Health in Women Veterans
Wagner, 2006). Because individuals with PTSD tend to have a greater number of health visits, our findings, though significant, may underestimate this association. Assessment of health care utilization only covered care received through VHA and this may also have led to an underestimation of this association. Assessment of military social support was limited to one item on perceptions of social support of military friends. Social support is complex and multifaceted (Shumaker & Brownell, 1984), and future research should use comprehensive measures that account for different sources (e.g., employers and coworkers during deployment, spousal, family, nonmilitary friends) and types of social support. For example, one study found that military social support was protective for men against mental health sequelae, whereas civilian social support was more protective for female Marine recruits (Smith et al., 2013). Additionally, our data are based on a cross-sectional survey that precludes us from making conclusions about causal relationships. Longitudinal studies may provide more nuanced understanding of how and under what circumstances PTSD and military social support influence health and health care visits. In sum, women veterans who screen positive for PTSD are more likely to report poorer physical health and greater VHA utilization, whereas those who report greater maintenance of military social support also report more favorable health and less frequent VHA utilization. Military social support was equally protective regardless of women’s PTSD status, and its associative role with health should be further examined. References American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author. Bhandari, A., & Wagner, T. (2006). Self-reported utilization of health care services: Improving measurement and accuracy. Medical Care Research and Review, 63, 217–235. doi:10.1177/1077558705285298 Breslau, N., Peterson, E. L., Kessler, R. C., & Schultz, L. R. (1999). Short screening scale for DSM-IV posttraumatic stress disorder. American Journal of Psychiatry, 156, 908–911. Centers for Disease Control. (2011). Enhancing use of clinical preventive services among older adults: Closing the gap. Retrieved from www.aarp.org/healthpros Escalona, R., Achilles, G., Waitzkin, H., & Yager, J. (2004). PTSD and somatization in women treated at a VA primary care clinic. Psychosomatics, 45, 291–296. doi:10.1176/appi.psy.45.4.291 Frayne, S. M., Parker, V. A., Christiansen, C. L., Loveland, S., Seaver, M. R., Kazis, L. W., & Skinner, K. M. (2006). Health status among 28,000 women veterans. Journal of General Internal Medicine, 21(S3), S40–S46. doi:10.1111/j.1525-1497.2006.00373.x Jakupcak, M., Vannoy, S., Immel, Z., Cook, J. W., Fontana, A., Rosenheck, R., & McFall, M. (2010). Does PTSD moderate the relationship be-
775
tween social support and suicide risk in Iraq and Afghanistan war veterans seeking mental health treatment? Depression and Anxiety, 27, 1001–1005. doi:10.1002/da.20722 Kouzis, A. C., & Eaton, W. W. (1998). Absence of social networks, social support and health services utilization. Psychological Medicine, 28, 1301– 1310. doi:10.1017/S0033291798007454 Pietrzak, R. H., Johnson, D. C., Goldstein, M. B., Malley, J. C., & Southwick, S. M. (2009). Psychological resilience and postdeployment social support protect against traumatic stress and depressive symptoms in soldiers returning from Operations Enduring Freedom and Iraqi Freedom. Depression and Anxiety, 26, 745–751. doi:10.1002/da.20558 Shumaker, S. A., & Brownell, A. (1984). Toward a theory of social support: Closing conceptual gaps. Journal of Social Issues, 40, 11–36. doi:10.1111/j.1540-4560.1984.tb01105.x Smith, B. N., Shipherd, J. C., Schuster, J. L., Vogt, D. S., King, L. A., & King, D. W. (2011). Posttraumatic stress symptomatology as a mediator of the association between military sexual trauma and post-deployment physical health in women. Journal of Trauma and Dissociation, 12, 275– 289. doi:10.1080/15299732.2011.551508 Smith, B. N., Vaughn, R. A., Vogt, D. S., King, L. A. & King, D. W., & Shipherd, J. C. (2013). Main and interactive effects of social support in predicting mental health symptoms in men and women following military stressor exposure. Anxiety, Stress, and Coping, 26, 52–69. doi:10.1080/10615806.2011.634001 StataCorp. (2011). Stata statistical software (Release 12) [Computer Software]. College Station, TX: Author. Suris, A., Lind, L., Kashner, T. M., Borman, P. D., & Petty, F. (2004). Sexual assault in women veterans: An examination of PTSD risk, health care utilization, and cost of care. Psychosomatic Medicine, 66, 749–756. doi:10.1097/01.psy.0000138117.58559.7b Uchino, B. N. (2004). Social support and physical health: Understanding the health consequences of our relationships. New Haven, CT: Yale University Press. Ware, J. E., Kosinski, M., & Keller, S. D. (1996). A 12-item short-form health survey: Construction of scales and preliminary tests of reliability and validity. Medical Care, 34, 220–233. doi:10.1097/00005650-199603000-00003 Ware, J. E., Kosinski, M., Turner-Bowker, D. M., & Gandek, B. (2002). How to score version 2 of the SF-12 Health Survey (with a supplement documenting version 1). Lincoln, RI: QualityMetric Inc. Washington, D. L., Bean-Mayberry, B., Riopelle, D., & Yano, E. M. (2011). Access to care for women veterans: Delayed healthcare and unmet need. Journal of General Internal Medicine, 26, 655–661. doi:10.1007/s11606011-1772-z Washington, D. L., Sun, S., & Canning, M. (2010). Creating a sampling frame for population-based veteran research: Representativeness and overlap of VA and Department of Defense databases. Journal of Rehabilitation Research & Development, 47, 763–772. doi:10.1682/JRRD.2009.08.0127 Wilcox, S. (2010). Social relationships and PTSD symptomatology in combat veterans. Psychological Trauma: Theory, Research, Practice, and Policy, 2, 175–182. doi:10.1037/a0019062 Zinzow, H. M., Grubaugh, A. L., Monnier, J., Suffoletta-Maierle, S., & Frueh, B. C. (2007). Trauma among female veterans: A critical review. Trauma, Violence, and Abuse, 8, 384–400. doi:10.1177/1524838007307295
Journal of Traumatic Stress DOI 10.1002/jts. Published on behalf of the International Society for Traumatic Stress Studies.
