Adverse Childhood Experiences, Military Service, and Adult Health Jodie G. Katon, PhD, MS,1,4,7 Keren Lehavot, PhD,2,6 Tracy L. Simpson, PhD,2,3,6 Emily C. Williams, PhD,1,4 Sarah Beth Barnett, MA,4 Joel R. Grossbard, PhD,1 Mark B Schure, PhD,1,4 Kristen E. Gray, PhD,1,4 Gayle E. Reiber, PhD1,4,5 Introduction: Prevalence of adverse childhood experiences (ACE) and associations with adult health may vary by gender and military service. This study compares the gender-specific prevalence of ACE by military service and determines the associations of ACE with adult health risk factors and health-related quality of life (HRQOL). Methods: This 2014 analysis used data from the 2011 and 2012 CDC Behavioral Risk Factor Surveillance System. Total ACE was operationalized as the number of reported ACE. Associations of total ACE with adult health risk factors were estimated using general linear models; associations with HRQOL were estimated using negative binomial regression. All analyses adjusted for age and race/ ethnicity.
Results: Those with military service had more total ACE than civilians. Higher ACE was associated with poorer HRQOL among women (physical health, military service, relative risk [RR]¼1.20, 95% CI¼1.09, 1.33; civilians, RR¼1.18, 95% CI¼1.17, 1.20; mental health, military service, RR¼1.21, 95% CI¼1.12, 1.32; civilians, RR¼1.25, 95% CI¼1.23, 1.26). Among men, these associations were somewhat attenuated in those with military service relative to civilians (physical health, military service, RR¼1.13, 95% CI¼1.09, 1.18; civilians, RR¼1.20, 95% CI¼1.17, 1.24; mental health, military service, RR¼1.21, 95% CI¼1.16, 1.27; civilians, RR¼1.30, 95% CI¼1.27, 1.34). Conclusions: Relative to civilians, men and women with military service report more ACE, but associations of ACE with adult HRQOL are weaker among men with military service relative to civilians. There is a need to implement and disseminate evidence-based programs to prevent ACE and for research on the long-term health consequences of ACE in military populations. (Am J Prev Med 2015;49(4):573–582) Published by Elsevier Inc. on behalf of American Journal of Preventive Medicine
Introduction From the 1Health Services Research and Development, Veterans Affairs Puget Sound Health Care System, Seattle, Washington; 2Mental Illness Research, Education, and Clinical Center, Veterans Affairs Puget Sound Health Care System, Seattle, Washington; 3Center of Excellence in Substance Abuse Treatment and Education, Veterans Affairs Puget Sound Health Care System, Seattle, Washington; 4Department of Health Services, School of Public Health, University of Washington, Seattle, Washington; 5 Department of Epidemiology, School of Public Health, University of Washington, Seattle, Washington; 6Department of Psychiatry and Behavioral Sciences, School of Medicine, University of Washington, Seattle, Washington; and 7the Veterans Affairs Office of Patient Care Services, Office of Women’s Health Services, Department of Veterans Affairs, Washington, District of Columbia Address correspondence to: Jodie G. Katon, PhD, MS, 1660 South Columbian Way S-152, Seattle WA 98108. E-mail: [email protected]. 0749-3797/$36.00 http://dx.doi.org/10.1016/j.amepre.2015.03.020
A
dverse childhood experiences (ACE), including household dysfunction, neglect, and physical and sexual abuse, are associated with increased risk of a host of adult health risk factors (e.g., smoking, obesity); poor health outcomes; and reduced healthrelated quality of life (HRQOL), which ultimately may lead to early death.1–6 Numerous studies have examined associations between specific types of ACE (e.g., childhood physical abuse) and adult health outcomes; however, multiple types of ACE frequently co-occur.4 Among those with any individual ACE, the probability of exposure to additional ACE ranges from 65% to 93%.4 As prior work indicates a cumulative effect of ACE on adult health,7 it is important to consider the total burden of ACE when investigating the association
Published by Elsevier Inc. on behalf of American Journal of Preventive Medicine
Am J Prev Med 2015;49(4):573–582 573
574
Katon et al / Am J Prev Med 2015;49(4):573–582 3,4,8,9
between ACE and adult health. Additionally, prevalence of ACE and associations with adult health vary by gender.10–12 Compared with civilians, those with military service have higher rates of childhood physical and sexual abuse.13,14 Less is known about the prevalence of household dysfunction or neglect, including exposure to domestic violence, substance use, and mental illness, among those with military service.10,14–16 One recent study17 found that those with military service were more likely than civilians to report nearly all forms of ACE, including household dysfunction and neglect, and that, among men, this association was strongest in those eligible for service during the all-volunteer era. Prevalence of ACE is even higher among non–population based samples of Veterans (e.g., Veterans Affairs [VA] users).16,18–20 Montgomery and colleagues21 found that ACE may not be as strongly associated with physical health problems in those with military service compared with civilians but did not examine variations by gender. Therefore, the aim of this study is to extend prior research findings by comparing the gender-specific prevalence of ACE, including household dysfunction, neglect, and physical and sexual abuse by military service, in a national sample. We also examine the association of total ACE with adult health risk factors and HRQOL among men and women with and without military service.
Methods Study Population This 2014 analysis used data from the 2011 and 2012 CDC Behavioral Risk Factor Surveillance System (BRFSS). BRFSS is a cross-sectional telephone survey conducted by CDC in conjunction with state health departments, which includes a representative national sample of non-institutionalized civilian residents aged 418 years.22 BRFSS median response rates for the national core questions across states ranged from 28% to 53% in 2011 and 35% to 49% in 2012.22 An optional ACE module was administered in ten states in 2011 (Minnesota, Montana, Vermont, Washington, Wisconsin, California, Maine, Nebraska, Nevada, Oregon) and six states in 2012 (Iowa, North Carolina, Tennessee, Oklahoma, Wisconsin, Nebraska). The present analysis was limited to respondents in these states who completed the ACE module with non-missing data on military service, age, and race/ethnicity. For individual analyses, respondents missing any of the outcomes or covariates specific to that analysis were excluded. This study was deemed exempt from full IRB review by the VA Puget Sound IRB.
Measures Gender was assessed by self-report. Military service was assessed by the question: Have you ever served on active duty in the United States Armed Forces, either in the regular military or in a National
Guard or military reserve unit? Active duty does not include training for the Reserves or National Guard, but DOES include activation, for example, for the Persian Gulf War.23–26 The ACE module consists of questions that solicit information on adverse experiences occurring before age 18 years.4 Specific ACE included living with someone who was mentally ill or depressed, abused drugs or alcohol, or who went to jail; having parents who are separated or divorced; witnessing domestic violence; and experiencing physical, verbal, or sexual abuse. Responses to questions regarding domestic, physical, or verbal abuse were only considered positive if a participant reported experiencing it more than once. A single experience of sexual abuse was considered a positive response (Appendix Table 1).4,21,27 Total ACE was summarized with a single score, ranging from 0 to 8, with a point for each of the eight reported ACE (Table 2).4,21,27 Health risk factors included obesity, ever smoking, and binge drinking. Obesity was defined as BMI Z30, based on calculations from self-reported height and weight. Ever smokers included current and former smokers. Binge drinking was defined as four or more and five or more drinks in one sitting in the past 30 days for women and men, respectively.28 Respondents were asked to rate their self-perceived health as excellent, very good, good, fair, or poor, which we dichotomized as fair/poor versus excellent/very good/good. We included two additional measures of HRQOL, that is, the number of days in the past 30 in which a respondent’s (1) physical health was not good and (2) mental health was not good. These two outcomes were treated as separate continuous variables (range¼0–30 days). Days of poor physical and mental health are accepted and well-validated measures of an individual’s perceived health, which is related to both health risk factors and chronic disease.29 Demographic descriptors included age; race/ethnicity (white, African American, other, multiracial, Hispanic); education (high school or less, at least some college); annual household income (o$15,000, $15,000–$24,999, $25,000–$34,999, $35,000–$49,999, Z$50,000); and current employment (employed, unemployed, homemaker, student, retired).
Statistical Analysis Characteristics of participants were examined, stratified by military service and gender. Given that prevalence of ACE was expected to be 410%, general linear models with a log link were used to estimate the prevalence of individual ACE and differences in prevalence by military service. Models were stratified by gender with military service as the exposure and adjusted for age and race/ ethnicity, as both vary by military service and are risk factors for ACE. Stratifying by gender, the total burden of ACE was compared between those with and without military service and adjusted for age and race/ethnicity using multivariable linear regression. General linear models with a log link were employed to directly obtain estimates of relative risk [RR] for the association between total ACE and (1) adult health risk factors and (2) selfreported health status.30 Separate models were run for men and women. Owing to evidence of overdispersion of zeroes, negative binomial regression was used to examine the association of total ACE with HRQOL as measured by days of poor physical health and mental health stratified by gender. Interaction terms for military service and total ACE were introduced to test whether associations differed by military service. Adjusted mean days per www.ajpmonline.org
Katon et al / Am J Prev Med 2015;49(4):573–582 month of poor physical and mental health were calculated by gender and military service. All models adjusted for age and race/ethnicity. All analyses used appropriate sample weights to account for the survey design and were conducted using STATA, version 13 with a two-sided α-value of 0.05.
Results In total, 102,381 BRFSS respondents completed the ACE module. Forty-four individuals with missing data on military service (women, eight; men, 36) and 721 individuals missing data on race/ethnicity were excluded (women, nine military service, 365 civilians; men, 129 military service, 218 civilians). The final analytic sample included 101,616 respondents (women, 1,077 military service, 59,887 civilians; men, 12,244 military service, 28,408 civilians), of whom o10% were missing data on health risk factors, health status, or HRQOL. Among women, those with military service tended to be more likely to be aged between 25 and 54 years; be African American; have at least some college; earn Z$50,000 per year; and were slightly more likely to be employed, smoke, or binge drink compared with civilians (Table 1). Among men, those with military service were more likely to be aged Z65 years, have some college education, and be obese compared with civilians. Among women, those with military service had a higher prevalence of nearly all individual ACE after adjusting for age and race/ethnicity compared with civilians (Table 2). Similarly, after adjustment, men with military service generally reported higher prevalence of specific ACE compared with civilians. Small but consistently positive associations were observed between increasing ACE score and nearly all health risk factors across the four groups (Table 3). Differences in the association of total ACE with ever smoking between those with and without military service were detected for women (p for interactiono0.01) and men (p for interactiono0.001). Among men and women with military service, the association of ACE with ever smoking was attenuated relative to that observed among civilians. Among women with military service, a 1-point increase in ACE was associated with a 7% increased risk of ever smoking (RR¼1.07, 95% CI¼1.03, 1.12), but this association was stronger among female civilians (RR¼1.14, 95% CI¼1.13, 1.15). Similarly, among men with military service, a 1-point increase in ACE was associated with a 6% increased risk of ever smoking (RR¼1.06, 95% CI¼1.05, 1.07), but the association was stronger among male civilians (RR¼1.12, 95% CI¼1.11, 1.13). Higher total ACE was associated with increased days of poor physical and mental health in all groups.
October 2015
575
Statistically significant differences in these associations were detected among men with and without military service. For example, among men with military service, a 1-point increase in ACE was associated with a 21% increase in the number of days of poor mental health (RR¼1.21, 95% CI¼1.16, 1.27) but a 30% increase in male civilians (RR¼1.30, 95% CI¼1.27, 1.34). Figure 1 illustrates differences in mean days of poor physical health and mental health for men and women with and without military service across levels of total ACE, after adjusting for age and race/ethnicity. These differences were most apparent among men with the highest total ACE. For example, among men with seven total ACE, after adjusting for age and race/ethnicity, those with military service reported 8.1 days of poor mental health in the past 30 days on average, whereas civilians reported 11.5 days.
Discussion Compared with their civilian counterparts, men and women with military service had a greater total burden of ACE in this national population-based study. Although no formal statistical test was conducted comparing all four groups, women with military service reported the highest total ACE (2.2 compared with 1.7 for women civilians, 1.6 for men with military service, and 1.3 for men civilians), supporting the hypothesis that individuals, particularly women, may join the military as a means to escape chaotic or dysfunctional home environments.13 Among all groups, there were small but consistent positive associations between higher total ACE and (1) adult health risk factors and (2) poorer HRQOL. The association between ACE and smoking was weaker among men and women with military service relative to civilians. Among men, the association between ACE and HRQOL was also weaker in those with military service relative to civilians, although it is not clear whether these differences are clinically significant. In the current study, the prevalence of ACE among civilians was similar to that reported in earlier studies.4,10 Estimates for the prevalence of physical and sexual abuse among women with military service were approximately 50% lower than previously reported among VA-using populations, where rates ranged from 35% for physical abuse to as high as 49% for sexual abuse.16,18–20 As poor health and low SES are positively associated with ACE history4,31 and use of VA health care,32,33 the prevalence of ACE likely would be higher in these populations using VA care. However, our findings are consistent with those of Blosnich et al.,17 which used a population-based sample including both VA users and non-users.
Katon et al / Am J Prev Med 2015;49(4):573–582
576
Table 1. Characteristics of Women and Men With and Without Military Service, BRFSS 2011–2012 Women Military service (n¼1,077)
Men Civilians (n¼59,887)
Military service (n¼12,244)
Civilians (n¼28,408)
Demographics Age (years)a 18–24
4.8
11.9
2.3
16.4
25–34
18.0
15.7
7.0
19.2
35–44
20.4
16.4
9.7
19.6
45–54
26.1
18.7
14.2
20.9
55–64
14.7
16.9
23.8
14.9
15.6
20.4
43.0
9.0
74.9
81.1
84.0
77.9
14.3
7.6
7.4
7.2
Other
3.5
5.4
2.7
7.7
Multi-racial
5.5
4.3
3.4
5.4
Hispanic
1.8
1.6
2.6
1.9
High school or less
20.7
40.2
38.7
45.4
Some college
45.1
35.0
37.2
29.4
College graduate
34.2
24.8
24.1
25.2
8.7
11.6
6.7
9.1
15,000–24,000
11.8
20.2
16.9
17.3
25,000–34,000
11.8
13.0
14.1
11.9
35,000–49,000
16.9
15.2
19.7
14.9
50.8
40.0
42.6
46.7
53.6
51.9
43.4
69.7
Z65 Race/ethnicity
a
White African American
Educationa,b
Income ($)
a,b
o15,000
Z50,000 Currently employed
a,b
Health risk factors Obesitya,b
26.0
28.3
31.6
29.1
a,b
51.6
40.8
65.7
47.5
Current smoker
21.8
18.6
19.2
22.5
Former smoker
29.7
22.3
46.5
25.0
Never smoker
48.4
59.2
34.3
52.5
a,b
13.0
12.6
15.5
25.6
16.5
16.3
20.3
15.0
Smoking
Binge drinking
Health-related quality of life Poor/fair self-perceived healtha,b
(continued on next page)
www.ajpmonline.org
Katon et al / Am J Prev Med 2015;49(4):573–582
577
Table 1. Characteristics of Women and Men With and Without Military Service, BRFSS 2011–2012 (continued) Women
Men
Military service (n¼1,077)
Civilians (n¼59,887)
Military service (n¼12,244)
Civilians (n¼28,408)
Days of poor physical healtha,c
5.1 (0.59)
4.0 (0.06)
4.7 (0.16)
3.2 (0.08)
Days of poor mental healtha,c
5.7 (0.61)
4.0 (0.06)
2.8 (0.13)
3.0 (0.08)
a
Weighted percent. Missing data: race/ethnicity (women: military service n¼9, civilians n¼365; men: military service n¼129, civilians n¼218); education (women: military service n¼2, civilians n¼95; men: military service n¼22, civilians n¼39); income (women: military service n¼109, civilians n¼8,628; men: military service n¼1,120, civilians n¼2,463); currently employed (women: military service n¼1, civilians n¼151; men: military service n¼21, civilians n¼72); obesity (women: military service n¼63, civilians n¼4,099; men: military service n¼83, civilians n¼354); current smoking (women: military service n¼0, civilians n¼211; men: military service n¼67, civilians n¼111); binge drinking(women: military service n¼10, civilians n¼586; men: military service n¼162, civilians n¼404); self-perceived health (women: military service n¼3, civilians n¼169; men: military service n¼41, civilians n¼80); days of poor physical health (women: military service n¼18, civilians n¼1,268; men: military service n¼225, civilians n¼408); days of poor mental health (women: military service n¼12, civilians n¼902; men: military service n¼174, civilians n¼375). c Weighted mean (SD). BRFSS, Behavioral Risk Factor Surveillance System. b
Two other recent reports17,21 used the 2010 BRFSS data to examine ACE by military service, and although our present findings are largely in agreement with these earlier studies, there are several differences worth noting. Notably, the 2013 study by Montgomery and colleagues21 did not stratify by gender. Therefore, because women with military service remain a minority population, the estimates presented by Montgomery et al. for those with military service obscure the extremely high prevalence of ACE among women with military service. More recently, Blosnich and colleagues17 reported associations of Veteran status and history of ACE by service era using AORs. In contrast, the present analysis provides adjusted prevalence estimates, enabling meaningful assessment of the absolute, rather than relative, difference in prevalence of ACE among those with and without military service.34 Finally, the present analysis extends prior work by examining variations in associations of ACE with adult health risk factors and HRQOL by gender and military service. Similar associations between total ACE and adult health risk factors, health status, and HRQOL were observed in women with and without military service. These associations were also significant in men, but were weaker among men with military service relative to civilians. ACE may influence adult health through indirect and direct pathways that we cannot separate in our analyses. For example, exposure to childhood trauma may increase the likelihood of adult victimization, including military sexual trauma.35,36 ACE may also increase the probability of exposure to intensive combat situations, presumably through a selection mechanism related to military occupation.37–39 For example, those with ACE may have lower educational attainment prior October 2015
to joining the military and therefore lower rank.40 Additionally, early maltreatment is associated with dysregulation of the hypothalamic–pituitary–adrenal axis, potentially interfering with development of emotional learning and coping skills.41 Thus, those with ACE may be more likely to experience adult stressors and may lack key coping skills, making them more vulnerable to potential adverse health outcomes resulting from these experiences.41–43 However, because BRFSS did not contain data on adult trauma, it is only possible to speculate regarding these pathways, their relative importance, and the potential additive association of ACE and adult trauma on health and health risk factors (e.g., smoking). The aforementioned pathways between ACE and adult health may be modified by personal characteristics such as resilience or receipt of post-trauma interventions.44 The present analysis found that ACE score was consistently associated with poor HRQOL in all groups, although these associations were slightly weaker among men with military service. To join the military, men and women need to meet physical and educational eligibility requirements. Thus, although the prevalence of ACE is higher among those with military service, these men and women may have increased capacity for resilience compared with their similarly exposed civilian counterparts. Alternatively, among those who experience ACE, those who go on to join the military may have increased access to post-trauma interventions, which address and ameliorate the impact of ACE prior to or during their military service. Finally, the present findings may reflect lack of information on the severity or frequency of ACE, as those with the most severe ACE may be less likely to join the military because of severe injury or trauma.
Katon et al / Am J Prev Med 2015;49(4):573–582
578
Table 2. Prevalence of Reported Adverse Childhood Experiences by Gender and Military Service Military service
Civilians
%adja (95% CI)
%adja (95% CI)
28.0 (23.4, 32.6)
19.8 (19.2, 20.5)
37.4 (32.4, 42.3)
29.7 (29.0, 30.4)
32.7 (27.9, 37.6)
26.4 (25.7, 27.1)
15.6 (11.8, 19.3)
10.8 (10.3, 11.3)
Women Household dysfunction and neglect Lived with someone mentally illb,*** Lived with someone who abused alcohol or drugs Lived with someone with alcohol misuse Lived with someone who abused drugs
b,**
**
**
b
9.0 (5.5, 12.5)
7.5 (7.0, 8.0)
b,**
33.2 (28.2, 38.2)
26.3 (25.6, 27.0)
b
21.0 (17.0, 25.0)
17.7 (17.1, 18.3)
23.2 (18.8, 27.6)
16.2 (15.6, 16.8)
41.5 (36.4, 46.6)
27.4 (26.7, 28.1)
21.2 (17.2, 25.1)
16.3 (15.8, 16.9)
19.4 (15.6, 23.2)
14.5 (14.0, 15.1)
12.2 (9.0, 15.4)
10.1 (9.7, 10.6)
Lived with someone who served time or was sentenced to serve time Parents separated/divorced
Witnessed domestic violence Abuse
Childhood physical abuseb,*** Childhood verbal abuse
b,***
Any childhood sexual abuse
b,**
Touched sexually by anyone Z5 years older
**
Made to sexually touch anyone Z5 years older
**
Forced to have sex by someone Z5 years older
**
Total ACE***
9.7 (6.8, 12.6)
6.2 (5.9, 6.6)
2.2 (1.9, 2.4)
1.6 (1.6, 1.7)
15.1 (13.5, 16.7)
13.5 (12.8, 14.2)
31.2 (29.3, 33.0)
25.3 (24.4, 26.2)
26.6 (24.9, 28.3)
21.5 (20.6, 22.3)
13.0 (11.4, 14.6)
10.5 (9.8, 11.2)
9.6 (8.1, 11.0)
7.8 (7.1, 8.4)
31.0 (29.1, 32.9)
24.4 (23.5, 25.3)
20.0 (18.4, 21.6)
14.9 (14.2, 15.7)
20.2 (18.6, 21.8)
15.0 (14.3, 15.7)
31.8 (30.0, 33.6)
25.9 (25.0, 26.7)
7.9 (6.8, 8.9)
6.1 (5.6, 6.6)
6.0 (5.1, 7.0)
4.8 (4.3, 5.3)
5.7 (4.8, 6.6)
4.3 (3.8, 4.7)
2.3 (1.7, 2.9)
2.1 (1.8, 2.5)
Men Household dysfunction and neglect Lived with someone mentally illb Lived with someone who abused alcohol or drugs Lived with someone with alcohol misuse Lived with someone who abused drugs
b,***
***
**
Lived with someone who served time or was sentenced to serve time Parents separated or divorced Witnessed domestic violence
b,***
b,***
b,*
Abuse Childhood physical abuseb,*** Childhood verbal abuse
b,***
Any childhood sexual abuse
b,**
Touched sexually by anyone Z5 years older
*
Made to sexually touch anyone Z5 years older
**
Forced to have sex by someone Z5 years older Total ACE
***
1.7 (1.6, 1.7)
1.3 (1.3, 1.4) n
Note: Boldface indicates statistical significance. p-values comparing Veterans and civilians within gender strata, po0.05, a Adjusted for age (continuous) and race/ethnicity. b Used to calculate total ACE. ACE, adverse childhood experiences.
nn
po0.01, nnnpo0.001.
www.ajpmonline.org
Katon et al / Am J Prev Med 2015;49(4):573–582
579
validated in the literature, may not necessarily correlate Military service Civilians with the severity of one’s a,b a,b childhood experiences, and RR (95% CI) RR (95% CI) (95% CI) use of total ACE may Women obscure the impact of differences in relative prevalence Health risk factors of specific ACE (e.g., sexual Obese 1.07 (1.00, 1.15) 1.08 (1.07, 1.09) abuse) among the four 1.07 (1.03, 1.12) 1.14 (1.13, 1.15) Ever smoked** groups. We did not stratify by service era, as the draft Binge drinking 0.95 (0.84, 1.08) 1.05 (1.03, 1.07) only applied to men, and we Health related quality of life did not anticipate that the Poor/fair self-perceived health 1.14 (1.04, 1.26) 1.20 (1.18, 1.22) association of ACE with Days of poor physical health 1.20 (1.09, 1.33) 1.18 (1.17, 1.20) HRQOL would vary by service era. This may have led to Days of poor mental health 1.21 (1.12 1.32) 1.25 (1.23, 1.26) underestimation of differenMen ces in prevalence of ACE Health risk factors among men with and without military service. When Obese 1.05 (1.02, 1.07) 1.02 (1.00, 1.04) interaction terms for military 1.06 (1.05, 1.07) 1.12 (1.11, 1.13) Ever smoked*** service and service era were Binge drinking 1.05 (1.01, 1.09) 1.04 (1.02, 1.06) tested in models for ACE among men, statistically sigHealth-related quality of life nificant differences by service Poor/fair self-perceived health*** 1.10 (1.06, 1.14) 1.19 (1.16, 1.22) era were only detected for 1.13 (1.09, 1.18) 1.20 (1.17, 1.24) Days of poor physical health* living with someone with mental illness and childhood Days of poor mental health** 1.21 (1.16, 1.27) 1.30 (1.27, 1.34) physical abuse (data not Note: Boldface indicates statistical significance. p-values for interaction of veteran status and total ACE n nn nnn shown). In both service eras po0.001. within gender strata, po0.05, po0.01, a For dichotomous outcomes (e.g., obese, ever smoked), RR¼relative risk of change and corresponds to (draft versus all volunteer), the change in probability of health risk factor associated with a one-point increase in total ACE. For men with military service continuous outcomes (e.g., days of poor physical health), RR¼rate ratio and corresponds to the change were more likely to have in number of days of poor physical or mental health associated with a one-point increase in total ACE. b Adjusted for age and race/ethnicity. lived with someone with ACE, adverse childhood experiences. mental illness or have experienced childhood physical abuse, but differences Limitations between the two populations were smaller for those in the draft era. Finally, although less than 10% of the data on either This study has several limitations. These include the relatively ACE or HRQOL was missing, it is likely that these data are small number of women with military service, which limited not missing at random because of social desirability bias. If power in the analyses; inability to differentiate between active ACE is associated with poorer adult health and those with a duty service members and Veterans, which may have led to history of ACE were more likely to have missing data for underestimation of the association of ACE and HRQOL; selfACE and for adult health variables, then this analysis report of ACE and health variables, which may have been underestimated the association of ACE with adult health. subject to recall bias; lack of screening data for mental health (e.g., depression); and our inability to adjust for potential confounders, such as parental education level and income, Conclusions when comparing prevalence of ACE by military service. Also, Nevertheless, the current study has several strengths, variables on adult traumatic experiences such as sexual including the population-based study design, examinaassault were unavailable, and the association of ACE with tion of multiple ACE categories, assessment of total ACE, adult health outcomes may include the impact of adult and examination of the association of ACE with HRQOL. trauma.45 Additionally, the total ACE score, though well Table 3. Total ACE, Health Risk Factors, and Health-Related Quality of Life
October 2015
580
Katon et al / Am J Prev Med 2015;49(4):573–582
Figure 1. Relationship between total adverse childhood experiences and health-related quality of life stratified by gender and military service.
Our findings underscore the high prevalence of ACE in the U.S., particularly among women with military service. Given the observed association between ACE and health risk factors such as smoking and depression,4 it is important for clinicians to screen for ACE as part of routine medical practice and to consider these experiences when caring for patients.39 For example, incorporating psychotherapy into depression treatment and smoking-cessation programs may be particularly important for women with depression and a history of ACE.46,47 Longitudinal research is needed to examine specific mechanisms by which ACE may impact adult health, particularly among active duty servicemembers, to identify potential points of intervention. Furthermore, in line with a recent IOM report,48 there is a need to explore effective delivery systems to implement evidencebased programs to reduce child abuse and neglect. Our findings underscore the pressing need to implement and disseminate such programs to prevent ACE and its longterm health consequences.
This work was supported by the Denver-Seattle Veterans Affairs (VA) Health Services Research and Development (HSR&D) Center of Innovation and the VA Office Women’s Health Services. The authors declare no conflicts of interest. Drs. Lehavot and Williams are supported by VA Career Development Awards from CSR&D (CX000867) and HSR&D
(CDA 12-276), respectively. Dr. Williams is also an investigator with the Implementation Research Institute (IRI) at the George Warren Brown School of Social Work at Washington University. IRI is supported through an award from the National Institute of Mental Health (R25 MH080916-01A2) and the VA HSR&D, Quality Enhancement Research Initiative. Drs. Gray and Schure are supported by the Office of Academic Affiliations’ Associated Health Postdoctoral Fellowships (TPP 61-026 and TPP 61-028, respectively), and Dr. Reiber is supported by a VA HSR&D Senior Career Scientist Award (RCS 98-353). The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the VA. No financial disclosures were reported by the authors of this paper.
References 1. Walker EA, Gelfand A, Katon WJ, et al. Adult health status of women with histories of childhood abuse and neglect. Am J Med. 1999;107 (4):332–339. http://dx.doi.org/10.1016/S0002-9343(99)00235-1. 2. MacMillan HL, Fleming JE, Streiner DL, et al. Childhood abuse and lifetime psychopathology in a community sample. Am J Psychiatry. 2001;158(11):1878–1883. http://dx.doi.org/10.1176/appi.ajp.158.11. 1878. 3. Edwards VJ, Holden GW, Felitti VJ, Anda RF. Relationship between multiple forms of childhood maltreatment and adult mental health in community respondents: results from the Adverse Childhood Experiences Study. Am J Psychiatry. 2003;160(8):1453–1460. http://dx.doi. org/10.1176/appi.ajp.160.8.1453.
www.ajpmonline.org
Katon et al / Am J Prev Med 2015;49(4):573–582 4. Felitti VJ, Anda RF, Nordenberg D, et al. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults. The Adverse Childhood Experiences (ACE) Study. Am J Prev Med. 1998;14 (4):245–258. http://dx.doi.org/10.1016/S0749-3797(98)00017-8. 5. Ben-Shlomo Y, Kuh D. A life course approach to chronic disease epidemiology: conceptual models, empirical challenges and interdisciplinary perspectives. Int J Epidemiol. 2002;31(2):285–293. http://dx. doi.org/10.1093/ije/31.2.285. 6. Kuh D, Ben-Shlomo Y, Lynch J, Hallqvist J, Power C. Life course epidemiology. J Epidemiol Community Health. 2003;57(10):778–783. http://dx.doi.org/10.1136/jech.57.10.778. 7. Danese A, Moffitt TE, Harrington H, et al. Adverse childhood experiences and adult risk factors for age-related disease: depression, inflammation, and clustering of metabolic risk markers. Arch Pediatr Adolesc Med. 2009;163(12):1135–1143. http://dx.doi.org/10.1001/ archpediatrics.2009.214. 8. Rodgers CS, Lang AJ, Laffaye C, Satz LE, Dresselhaus TR, Stein MB. The impact of individual forms of childhood maltreatment on health behavior. Child Abuse Negl. 2004;28(5):575–586. http://dx.doi.org/ 10.1016/j.chiabu.2004.01.002. 9. Lange A, de Beurs E, Dolan C, Lachnit T, Sjollema S, Hanewald G. Long-term effects of childhood sexual abuse: objective and subjective characteristics of the abuse and psychopathology in later life. J Nerv Ment Dis. 1999;187(3):150–158. http://dx.doi.org/10.1097/00005053199903000-00004. 10. Finkelhor D, Turner H, Ormrod R, Hamby SL. Violence, abuse, and crime exposure in a national sample of children and youth. Pediatrics. 2009;124(5):1411–1423. http://dx.doi.org/10.1542/peds.2009-0467. 11. Fuller-Thomson E, Filippelli J, Lue-Crisostomo CA. Gender-specific association between childhood adversities and smoking in adulthood: findings from a population-based study. Public Health. 2013;127 (5):449–460. http://dx.doi.org/10.1016/j.puhe.2013.01.006. 12. Fuller-Thomson E, Dalton AD. Gender differences in the association between parental divorce during childhood and stroke in adulthood: findings from a population-based survey. Int J Stroke. 2012. http://dx. doi.org/10.1111/j.1747-4949.2012.00935.x. 13. Sadler AG, Booth BM, Mengeling MA, Doebbeling BN. Life span and repeated violence against women during military service: effects on health status and outpatient utilization. J Womens Health (Larchmt). 2004;13(7):799–811. http://dx.doi.org/10.1089/jwh.2004.13.799. 14. Scherrer JF, Xian H, Kapp JM, et al. Association between exposure to childhood and lifetime traumatic events and lifetime pathological gambling in a twin cohort. J Nerv Ment Dis. 2007;195(1):72–78. http: //dx.doi.org/10.1097/01.nmd.0000252384.20382.e9. 15. Briere J, Elliott DM. Prevalence and psychological sequelae of selfreported childhood physical and sexual abuse in a general population sample of men and women. Child Abuse Negl. 2003;27(10):1205–1222. http://dx.doi.org/10.1016/j.chiabu.2003.09.008. 16. Zinzow HM, Grubaugh AL, Monnier J, Suffoletta-Maierle S, Frueh BC. Trauma among female veterans: a critical review. Trauma Violence Abuse. 2007;8(4):384–400. http://dx.doi.org/10.1177/1524838007307295. 17. Blosnich JR, Dichter ME, Cerulli C, Batten SV, Bossarte RM. Disparities in adverse childhood experiences among individuals with a history of military service. JAMA Psychiatry. 2014;71(9):1041–1048. http://dx.doi.org/10.1001/jamapsychiatry.2014.724. 18. Clancy CP, Graybeal A, Tompson WP, et al. Lifetime trauma exposure in veterans with military-related posttraumatic stress disorder: association with current symptomatology. J Clin Psychiatry. 2006;67(9): 1346–1353. http://dx.doi.org/10.4088/JCP.v67n0904. 19. Kelly UA, Skelton K, Patel M, Bradley B. More than military sexual trauma: interpersonal violence, PTSD, and mental health in women veterans. Res Nurs Health. 2011;34(6):457–467. http://dx.doi.org/10.1002/nur.20453. 20. Lapp KG, Bosworth HB, Strauss JL, et al. Lifetime sexual and physical victimization among male veterans with combat-related post-traumatic stress disorder. Mil Med. 2005;170(9):787–790.
October 2015
581
21. Montgomery AE, Cutuli JJ, Evans-Chase M, Treglia D, Culhane DP. Relationship among adverse childhood experiences, history of active military service, and adult outcomes: homelessness, mental health, and physical health. Am J Public Health. 2013;103(suppl 2):S262–S268. http://dx.doi.org/10.2105/AJPH.2013.301474. 22. Behavioral Risk Factor Surveillance System. Annual survey data, 2011 and 2012. Atlanta, GA: USDHHS, CDC. http://www.cdc.gov/brfss/ annual_data/annual_data.htm. 23. Reiber GE, Koepsell TD, Maynard C, Haas LB, Boyko EJ. Diabetes in nonveterans, veterans, and veterans receiving Department of Veterans Affairs health care. Diabetes Care. 2004;27(suppl 2):B3–B9. http://dx. doi.org/10.2337/diacare.27.suppl_2.B3. 24. Grossbard JR, Lehavot K, Hoerster KD, Jakupcak M, Seal KH, Simpson TL. Relationships among veteran status, gender, and key health indicators in a national young adult sample. Psychiatr Serv. 2013;64 (6):547–553. http://dx.doi.org/10.1176/appi.ps.003002012. 25. Hoerster KD, Lehavot K, Simpson T, McFall M, Reiber G, Nelson KM. Health and health behavior differences: U.S. Military, veteran, and civilian men. Am J Prev Med. 2012;43(5):483–489. http://dx.doi.org/ 10.1016/j.amepre.2012.07.029. 26. Lehavot K, Hoerster KD, Nelson KM, Jakupcak M, Simpson TL. Health indicators for military, veteran, and civilian women. Am J Prev Med. 2012;42(5):473–480. http://dx.doi.org/10.1016/j.amepre.2012.01.006. 27. Andersen JP, Blosnich J. Disparities in adverse childhood experiences among sexual minority and heterosexual adults: results from a multistate probability-based sample. PLoS One. 2013;8(1):e54691. http://dx. doi.org/10.1371/journal.pone.0054691. 28. NIAAA. Helping Patients Who Drink Too Much: A Clinician’s Guide. Bethesda, MD: National Institute on Alcohol Abuse and Alcoholism, USDHHS, NIH; 2005. 29. Measuring Healthy Days. USDHHS, CDC, National Center for Chronic Disease Prevention and Health Promotion, Division of Adult and Community Health; 2000. 30. Lumley T, Kronmal R, Ma S. Relative Risk Regression in Medical Research: Models, Contrasts, Estimators, and Algorithms. UW Biostatistics Working Paper Series, 2006; Working Paper No. 293. 31. Currie J, Widom CS. Long-term consequences of child abuse and neglect on adult economic well-being. Child Maltreat. 2010;15(2):111– 120. http://dx.doi.org/10.1177/1077559509355316. 32. Mengeling MA, Sadler AG, Torner J, Booth BM. Evolving comprehensive VA women’s health care: patient characteristics, needs, and preferences. Womens Health Issues. 2011;21(4 suppl):S120–S129. http: //dx.doi.org/10.1016/j.whi.2011.04.021. 33. Washington DL, Yano EM, Simon B, Sun S. To use or not to use. What influences why women veterans choose VA health care. J Gen Intern Med. 2006;21(suppl 3):S11–S18. http://dx.doi.org/10.1111/j.1525-1497. 2006.00369.x. 34. Koepsell TD, Weiss NS. Epidemiologic Methods Studying the Occurrence of Illness. New York, NY: Oxford University Press; 2003. 35. Coid J, Petruckevitch A, Feder G, Chung W, Richardson J, Moorey S. Relation between childhood sexual and physical abuse and risk of revictimisation in women: a cross-sectional survey. Lancet. 2001;358 (9280):450–454. http://dx.doi.org/10.1016/S0140-6736(01)05622-7. 36. Merrill LL, Newell CE, Thomsen CJ, et al. Childhood abuse and sexual revictimization in a female Navy recruit sample. J Trauma Stress. 1999;12(2):211–225. http://dx.doi.org/10.1023/A:1024789723779. 37. Iversen AC, Fear NT, Simonoff E, et al. Influence of childhood adversity on health among male UK military personnel. Br J Psychiatry. 2007;191:506–511. http://dx.doi.org/10.1192/bjp.bp.107.039818. 38. Van Voorhees EE, Dedert EA, Calhoun PS, Brancu M, Runnals J, Beckham JC. Childhood trauma exposure in Iraq and Afghanistan war era veterans: implications for posttraumatic stress disorder symptoms and adult functional social support. Child Abuse Negl. 2012;36(5):423– 432. http://dx.doi.org/10.1016/j.chiabu.2012.03.004.
582
Katon et al / Am J Prev Med 2015;49(4):573–582
39. King DW, King LA, Foy DW, Gudanowski DM. Prewar factors in combat-related posttraumatic stress disorder: structural equation modeling with a national sample of female and male Vietnam veterans. J Consult Clin Psychol. 1996;64(3):520–531. http://dx.doi.org/10.1037/ 0022-006X.64.3.520. 40. Ye D, Reyes-Salvail F. Adverse childhood experiences among Hawai’i adults: Findings from the 2010 Behavioral Risk Factor Survey. Hawaii J Med Public Health. 2014;73(6):181–190. 41. Koenen KC, Moffitt TE, Poulton R, Martin J, Caspi A. Early childhood factors associated with the development of posttraumatic stress disorder: results from a longitudinal birth cohort. Psychol Med. 2007;37(2):181–192. http://dx.doi.org/10.1017/S00332 91706009019. 42. Van Voorhees E, Scarpa A. The effects of child maltreatment on the hypothalamic-pituitary-adrenal axis. Trauma Violence Abuse. 2004;5 (4):333–352. http://dx.doi.org/10.1177/1524838004269486. 43. Gillespie CF, Bradley B, Mercer K, et al. Trauma exposure and stressrelated disorders in inner city primary care patients. Gen Hosp Psychiatry. 2009;31(6):505–514. http://dx.doi.org/10.1016/j.genhosppsych.2009.05.003. 44. Maccani MA, Delahanty DL, Nugent NR, Berkowitz SJ. Pharmacological secondary prevention of PTSD in youth: challenges and
45.
46.
47.
48.
opportunities for advancement. J Trauma Stress. 2012;25(5):543–550. http://dx.doi.org/10.1002/jts.21731. Min MO, Minnes S, Kim H, Singer LT. Pathways linking childhood maltreatment and adult physical health. Child Abuse Negl. 2013;37 (6):361–373. http://dx.doi.org/10.1016/j.chiabu.2012.09.008. Nemeroff CB, Heim CM, Thase ME, et al. Differential responses to psychotherapy versus pharmacotherapy in patients with chronic forms of major depression and childhood trauma. Proc Natl Acad Sci U S A. 2003;100(24):14293–14296. http://dx.doi.org/10.1073/pnas.2336126100. Blalock JA, Minnix JA, Mathew AR, Wetter DW, McCullough JP, Cinciripini PM. Relationship of childhood trauma to depression and smoking outcomes in pregnant smokers. J Consult Clin Psychol. 2013;81(5):821–830. http://dx.doi.org/10.1037/a0033381. USDHHS. Child abuse and neglect. Child Welfare Information Gateway. www.childwelfare.gov/can/.
Appendix Supplementary data Supplementary data associated with this article can be found at, http://dx.doi.org/10.1016/j.amepre.2015.03.020.
www.ajpmonline.org
Results: Those with military service had more total ACE than civilians. Higher ACE was associated with poorer HRQOL among women (physical health, military service, relative risk [RR]¼1.20, 95% CI¼1.09, 1.33; civilians, RR¼1.18, 95% CI¼1.17, 1.20; mental health, military service, RR¼1.21, 95% CI¼1.12, 1.32; civilians, RR¼1.25, 95% CI¼1.23, 1.26). Among men, these associations were somewhat attenuated in those with military service relative to civilians (physical health, military service, RR¼1.13, 95% CI¼1.09, 1.18; civilians, RR¼1.20, 95% CI¼1.17, 1.24; mental health, military service, RR¼1.21, 95% CI¼1.16, 1.27; civilians, RR¼1.30, 95% CI¼1.27, 1.34). Conclusions: Relative to civilians, men and women with military service report more ACE, but associations of ACE with adult HRQOL are weaker among men with military service relative to civilians. There is a need to implement and disseminate evidence-based programs to prevent ACE and for research on the long-term health consequences of ACE in military populations. (Am J Prev Med 2015;49(4):573–582) Published by Elsevier Inc. on behalf of American Journal of Preventive Medicine
Introduction From the 1Health Services Research and Development, Veterans Affairs Puget Sound Health Care System, Seattle, Washington; 2Mental Illness Research, Education, and Clinical Center, Veterans Affairs Puget Sound Health Care System, Seattle, Washington; 3Center of Excellence in Substance Abuse Treatment and Education, Veterans Affairs Puget Sound Health Care System, Seattle, Washington; 4Department of Health Services, School of Public Health, University of Washington, Seattle, Washington; 5 Department of Epidemiology, School of Public Health, University of Washington, Seattle, Washington; 6Department of Psychiatry and Behavioral Sciences, School of Medicine, University of Washington, Seattle, Washington; and 7the Veterans Affairs Office of Patient Care Services, Office of Women’s Health Services, Department of Veterans Affairs, Washington, District of Columbia Address correspondence to: Jodie G. Katon, PhD, MS, 1660 South Columbian Way S-152, Seattle WA 98108. E-mail: [email protected]. 0749-3797/$36.00 http://dx.doi.org/10.1016/j.amepre.2015.03.020
A
dverse childhood experiences (ACE), including household dysfunction, neglect, and physical and sexual abuse, are associated with increased risk of a host of adult health risk factors (e.g., smoking, obesity); poor health outcomes; and reduced healthrelated quality of life (HRQOL), which ultimately may lead to early death.1–6 Numerous studies have examined associations between specific types of ACE (e.g., childhood physical abuse) and adult health outcomes; however, multiple types of ACE frequently co-occur.4 Among those with any individual ACE, the probability of exposure to additional ACE ranges from 65% to 93%.4 As prior work indicates a cumulative effect of ACE on adult health,7 it is important to consider the total burden of ACE when investigating the association
Published by Elsevier Inc. on behalf of American Journal of Preventive Medicine
Am J Prev Med 2015;49(4):573–582 573
574
Katon et al / Am J Prev Med 2015;49(4):573–582 3,4,8,9
between ACE and adult health. Additionally, prevalence of ACE and associations with adult health vary by gender.10–12 Compared with civilians, those with military service have higher rates of childhood physical and sexual abuse.13,14 Less is known about the prevalence of household dysfunction or neglect, including exposure to domestic violence, substance use, and mental illness, among those with military service.10,14–16 One recent study17 found that those with military service were more likely than civilians to report nearly all forms of ACE, including household dysfunction and neglect, and that, among men, this association was strongest in those eligible for service during the all-volunteer era. Prevalence of ACE is even higher among non–population based samples of Veterans (e.g., Veterans Affairs [VA] users).16,18–20 Montgomery and colleagues21 found that ACE may not be as strongly associated with physical health problems in those with military service compared with civilians but did not examine variations by gender. Therefore, the aim of this study is to extend prior research findings by comparing the gender-specific prevalence of ACE, including household dysfunction, neglect, and physical and sexual abuse by military service, in a national sample. We also examine the association of total ACE with adult health risk factors and HRQOL among men and women with and without military service.
Methods Study Population This 2014 analysis used data from the 2011 and 2012 CDC Behavioral Risk Factor Surveillance System (BRFSS). BRFSS is a cross-sectional telephone survey conducted by CDC in conjunction with state health departments, which includes a representative national sample of non-institutionalized civilian residents aged 418 years.22 BRFSS median response rates for the national core questions across states ranged from 28% to 53% in 2011 and 35% to 49% in 2012.22 An optional ACE module was administered in ten states in 2011 (Minnesota, Montana, Vermont, Washington, Wisconsin, California, Maine, Nebraska, Nevada, Oregon) and six states in 2012 (Iowa, North Carolina, Tennessee, Oklahoma, Wisconsin, Nebraska). The present analysis was limited to respondents in these states who completed the ACE module with non-missing data on military service, age, and race/ethnicity. For individual analyses, respondents missing any of the outcomes or covariates specific to that analysis were excluded. This study was deemed exempt from full IRB review by the VA Puget Sound IRB.
Measures Gender was assessed by self-report. Military service was assessed by the question: Have you ever served on active duty in the United States Armed Forces, either in the regular military or in a National
Guard or military reserve unit? Active duty does not include training for the Reserves or National Guard, but DOES include activation, for example, for the Persian Gulf War.23–26 The ACE module consists of questions that solicit information on adverse experiences occurring before age 18 years.4 Specific ACE included living with someone who was mentally ill or depressed, abused drugs or alcohol, or who went to jail; having parents who are separated or divorced; witnessing domestic violence; and experiencing physical, verbal, or sexual abuse. Responses to questions regarding domestic, physical, or verbal abuse were only considered positive if a participant reported experiencing it more than once. A single experience of sexual abuse was considered a positive response (Appendix Table 1).4,21,27 Total ACE was summarized with a single score, ranging from 0 to 8, with a point for each of the eight reported ACE (Table 2).4,21,27 Health risk factors included obesity, ever smoking, and binge drinking. Obesity was defined as BMI Z30, based on calculations from self-reported height and weight. Ever smokers included current and former smokers. Binge drinking was defined as four or more and five or more drinks in one sitting in the past 30 days for women and men, respectively.28 Respondents were asked to rate their self-perceived health as excellent, very good, good, fair, or poor, which we dichotomized as fair/poor versus excellent/very good/good. We included two additional measures of HRQOL, that is, the number of days in the past 30 in which a respondent’s (1) physical health was not good and (2) mental health was not good. These two outcomes were treated as separate continuous variables (range¼0–30 days). Days of poor physical and mental health are accepted and well-validated measures of an individual’s perceived health, which is related to both health risk factors and chronic disease.29 Demographic descriptors included age; race/ethnicity (white, African American, other, multiracial, Hispanic); education (high school or less, at least some college); annual household income (o$15,000, $15,000–$24,999, $25,000–$34,999, $35,000–$49,999, Z$50,000); and current employment (employed, unemployed, homemaker, student, retired).
Statistical Analysis Characteristics of participants were examined, stratified by military service and gender. Given that prevalence of ACE was expected to be 410%, general linear models with a log link were used to estimate the prevalence of individual ACE and differences in prevalence by military service. Models were stratified by gender with military service as the exposure and adjusted for age and race/ ethnicity, as both vary by military service and are risk factors for ACE. Stratifying by gender, the total burden of ACE was compared between those with and without military service and adjusted for age and race/ethnicity using multivariable linear regression. General linear models with a log link were employed to directly obtain estimates of relative risk [RR] for the association between total ACE and (1) adult health risk factors and (2) selfreported health status.30 Separate models were run for men and women. Owing to evidence of overdispersion of zeroes, negative binomial regression was used to examine the association of total ACE with HRQOL as measured by days of poor physical health and mental health stratified by gender. Interaction terms for military service and total ACE were introduced to test whether associations differed by military service. Adjusted mean days per www.ajpmonline.org
Katon et al / Am J Prev Med 2015;49(4):573–582 month of poor physical and mental health were calculated by gender and military service. All models adjusted for age and race/ethnicity. All analyses used appropriate sample weights to account for the survey design and were conducted using STATA, version 13 with a two-sided α-value of 0.05.
Results In total, 102,381 BRFSS respondents completed the ACE module. Forty-four individuals with missing data on military service (women, eight; men, 36) and 721 individuals missing data on race/ethnicity were excluded (women, nine military service, 365 civilians; men, 129 military service, 218 civilians). The final analytic sample included 101,616 respondents (women, 1,077 military service, 59,887 civilians; men, 12,244 military service, 28,408 civilians), of whom o10% were missing data on health risk factors, health status, or HRQOL. Among women, those with military service tended to be more likely to be aged between 25 and 54 years; be African American; have at least some college; earn Z$50,000 per year; and were slightly more likely to be employed, smoke, or binge drink compared with civilians (Table 1). Among men, those with military service were more likely to be aged Z65 years, have some college education, and be obese compared with civilians. Among women, those with military service had a higher prevalence of nearly all individual ACE after adjusting for age and race/ethnicity compared with civilians (Table 2). Similarly, after adjustment, men with military service generally reported higher prevalence of specific ACE compared with civilians. Small but consistently positive associations were observed between increasing ACE score and nearly all health risk factors across the four groups (Table 3). Differences in the association of total ACE with ever smoking between those with and without military service were detected for women (p for interactiono0.01) and men (p for interactiono0.001). Among men and women with military service, the association of ACE with ever smoking was attenuated relative to that observed among civilians. Among women with military service, a 1-point increase in ACE was associated with a 7% increased risk of ever smoking (RR¼1.07, 95% CI¼1.03, 1.12), but this association was stronger among female civilians (RR¼1.14, 95% CI¼1.13, 1.15). Similarly, among men with military service, a 1-point increase in ACE was associated with a 6% increased risk of ever smoking (RR¼1.06, 95% CI¼1.05, 1.07), but the association was stronger among male civilians (RR¼1.12, 95% CI¼1.11, 1.13). Higher total ACE was associated with increased days of poor physical and mental health in all groups.
October 2015
575
Statistically significant differences in these associations were detected among men with and without military service. For example, among men with military service, a 1-point increase in ACE was associated with a 21% increase in the number of days of poor mental health (RR¼1.21, 95% CI¼1.16, 1.27) but a 30% increase in male civilians (RR¼1.30, 95% CI¼1.27, 1.34). Figure 1 illustrates differences in mean days of poor physical health and mental health for men and women with and without military service across levels of total ACE, after adjusting for age and race/ethnicity. These differences were most apparent among men with the highest total ACE. For example, among men with seven total ACE, after adjusting for age and race/ethnicity, those with military service reported 8.1 days of poor mental health in the past 30 days on average, whereas civilians reported 11.5 days.
Discussion Compared with their civilian counterparts, men and women with military service had a greater total burden of ACE in this national population-based study. Although no formal statistical test was conducted comparing all four groups, women with military service reported the highest total ACE (2.2 compared with 1.7 for women civilians, 1.6 for men with military service, and 1.3 for men civilians), supporting the hypothesis that individuals, particularly women, may join the military as a means to escape chaotic or dysfunctional home environments.13 Among all groups, there were small but consistent positive associations between higher total ACE and (1) adult health risk factors and (2) poorer HRQOL. The association between ACE and smoking was weaker among men and women with military service relative to civilians. Among men, the association between ACE and HRQOL was also weaker in those with military service relative to civilians, although it is not clear whether these differences are clinically significant. In the current study, the prevalence of ACE among civilians was similar to that reported in earlier studies.4,10 Estimates for the prevalence of physical and sexual abuse among women with military service were approximately 50% lower than previously reported among VA-using populations, where rates ranged from 35% for physical abuse to as high as 49% for sexual abuse.16,18–20 As poor health and low SES are positively associated with ACE history4,31 and use of VA health care,32,33 the prevalence of ACE likely would be higher in these populations using VA care. However, our findings are consistent with those of Blosnich et al.,17 which used a population-based sample including both VA users and non-users.
Katon et al / Am J Prev Med 2015;49(4):573–582
576
Table 1. Characteristics of Women and Men With and Without Military Service, BRFSS 2011–2012 Women Military service (n¼1,077)
Men Civilians (n¼59,887)
Military service (n¼12,244)
Civilians (n¼28,408)
Demographics Age (years)a 18–24
4.8
11.9
2.3
16.4
25–34
18.0
15.7
7.0
19.2
35–44
20.4
16.4
9.7
19.6
45–54
26.1
18.7
14.2
20.9
55–64
14.7
16.9
23.8
14.9
15.6
20.4
43.0
9.0
74.9
81.1
84.0
77.9
14.3
7.6
7.4
7.2
Other
3.5
5.4
2.7
7.7
Multi-racial
5.5
4.3
3.4
5.4
Hispanic
1.8
1.6
2.6
1.9
High school or less
20.7
40.2
38.7
45.4
Some college
45.1
35.0
37.2
29.4
College graduate
34.2
24.8
24.1
25.2
8.7
11.6
6.7
9.1
15,000–24,000
11.8
20.2
16.9
17.3
25,000–34,000
11.8
13.0
14.1
11.9
35,000–49,000
16.9
15.2
19.7
14.9
50.8
40.0
42.6
46.7
53.6
51.9
43.4
69.7
Z65 Race/ethnicity
a
White African American
Educationa,b
Income ($)
a,b
o15,000
Z50,000 Currently employed
a,b
Health risk factors Obesitya,b
26.0
28.3
31.6
29.1
a,b
51.6
40.8
65.7
47.5
Current smoker
21.8
18.6
19.2
22.5
Former smoker
29.7
22.3
46.5
25.0
Never smoker
48.4
59.2
34.3
52.5
a,b
13.0
12.6
15.5
25.6
16.5
16.3
20.3
15.0
Smoking
Binge drinking
Health-related quality of life Poor/fair self-perceived healtha,b
(continued on next page)
www.ajpmonline.org
Katon et al / Am J Prev Med 2015;49(4):573–582
577
Table 1. Characteristics of Women and Men With and Without Military Service, BRFSS 2011–2012 (continued) Women
Men
Military service (n¼1,077)
Civilians (n¼59,887)
Military service (n¼12,244)
Civilians (n¼28,408)
Days of poor physical healtha,c
5.1 (0.59)
4.0 (0.06)
4.7 (0.16)
3.2 (0.08)
Days of poor mental healtha,c
5.7 (0.61)
4.0 (0.06)
2.8 (0.13)
3.0 (0.08)
a
Weighted percent. Missing data: race/ethnicity (women: military service n¼9, civilians n¼365; men: military service n¼129, civilians n¼218); education (women: military service n¼2, civilians n¼95; men: military service n¼22, civilians n¼39); income (women: military service n¼109, civilians n¼8,628; men: military service n¼1,120, civilians n¼2,463); currently employed (women: military service n¼1, civilians n¼151; men: military service n¼21, civilians n¼72); obesity (women: military service n¼63, civilians n¼4,099; men: military service n¼83, civilians n¼354); current smoking (women: military service n¼0, civilians n¼211; men: military service n¼67, civilians n¼111); binge drinking(women: military service n¼10, civilians n¼586; men: military service n¼162, civilians n¼404); self-perceived health (women: military service n¼3, civilians n¼169; men: military service n¼41, civilians n¼80); days of poor physical health (women: military service n¼18, civilians n¼1,268; men: military service n¼225, civilians n¼408); days of poor mental health (women: military service n¼12, civilians n¼902; men: military service n¼174, civilians n¼375). c Weighted mean (SD). BRFSS, Behavioral Risk Factor Surveillance System. b
Two other recent reports17,21 used the 2010 BRFSS data to examine ACE by military service, and although our present findings are largely in agreement with these earlier studies, there are several differences worth noting. Notably, the 2013 study by Montgomery and colleagues21 did not stratify by gender. Therefore, because women with military service remain a minority population, the estimates presented by Montgomery et al. for those with military service obscure the extremely high prevalence of ACE among women with military service. More recently, Blosnich and colleagues17 reported associations of Veteran status and history of ACE by service era using AORs. In contrast, the present analysis provides adjusted prevalence estimates, enabling meaningful assessment of the absolute, rather than relative, difference in prevalence of ACE among those with and without military service.34 Finally, the present analysis extends prior work by examining variations in associations of ACE with adult health risk factors and HRQOL by gender and military service. Similar associations between total ACE and adult health risk factors, health status, and HRQOL were observed in women with and without military service. These associations were also significant in men, but were weaker among men with military service relative to civilians. ACE may influence adult health through indirect and direct pathways that we cannot separate in our analyses. For example, exposure to childhood trauma may increase the likelihood of adult victimization, including military sexual trauma.35,36 ACE may also increase the probability of exposure to intensive combat situations, presumably through a selection mechanism related to military occupation.37–39 For example, those with ACE may have lower educational attainment prior October 2015
to joining the military and therefore lower rank.40 Additionally, early maltreatment is associated with dysregulation of the hypothalamic–pituitary–adrenal axis, potentially interfering with development of emotional learning and coping skills.41 Thus, those with ACE may be more likely to experience adult stressors and may lack key coping skills, making them more vulnerable to potential adverse health outcomes resulting from these experiences.41–43 However, because BRFSS did not contain data on adult trauma, it is only possible to speculate regarding these pathways, their relative importance, and the potential additive association of ACE and adult trauma on health and health risk factors (e.g., smoking). The aforementioned pathways between ACE and adult health may be modified by personal characteristics such as resilience or receipt of post-trauma interventions.44 The present analysis found that ACE score was consistently associated with poor HRQOL in all groups, although these associations were slightly weaker among men with military service. To join the military, men and women need to meet physical and educational eligibility requirements. Thus, although the prevalence of ACE is higher among those with military service, these men and women may have increased capacity for resilience compared with their similarly exposed civilian counterparts. Alternatively, among those who experience ACE, those who go on to join the military may have increased access to post-trauma interventions, which address and ameliorate the impact of ACE prior to or during their military service. Finally, the present findings may reflect lack of information on the severity or frequency of ACE, as those with the most severe ACE may be less likely to join the military because of severe injury or trauma.
Katon et al / Am J Prev Med 2015;49(4):573–582
578
Table 2. Prevalence of Reported Adverse Childhood Experiences by Gender and Military Service Military service
Civilians
%adja (95% CI)
%adja (95% CI)
28.0 (23.4, 32.6)
19.8 (19.2, 20.5)
37.4 (32.4, 42.3)
29.7 (29.0, 30.4)
32.7 (27.9, 37.6)
26.4 (25.7, 27.1)
15.6 (11.8, 19.3)
10.8 (10.3, 11.3)
Women Household dysfunction and neglect Lived with someone mentally illb,*** Lived with someone who abused alcohol or drugs Lived with someone with alcohol misuse Lived with someone who abused drugs
b,**
**
**
b
9.0 (5.5, 12.5)
7.5 (7.0, 8.0)
b,**
33.2 (28.2, 38.2)
26.3 (25.6, 27.0)
b
21.0 (17.0, 25.0)
17.7 (17.1, 18.3)
23.2 (18.8, 27.6)
16.2 (15.6, 16.8)
41.5 (36.4, 46.6)
27.4 (26.7, 28.1)
21.2 (17.2, 25.1)
16.3 (15.8, 16.9)
19.4 (15.6, 23.2)
14.5 (14.0, 15.1)
12.2 (9.0, 15.4)
10.1 (9.7, 10.6)
Lived with someone who served time or was sentenced to serve time Parents separated/divorced
Witnessed domestic violence Abuse
Childhood physical abuseb,*** Childhood verbal abuse
b,***
Any childhood sexual abuse
b,**
Touched sexually by anyone Z5 years older
**
Made to sexually touch anyone Z5 years older
**
Forced to have sex by someone Z5 years older
**
Total ACE***
9.7 (6.8, 12.6)
6.2 (5.9, 6.6)
2.2 (1.9, 2.4)
1.6 (1.6, 1.7)
15.1 (13.5, 16.7)
13.5 (12.8, 14.2)
31.2 (29.3, 33.0)
25.3 (24.4, 26.2)
26.6 (24.9, 28.3)
21.5 (20.6, 22.3)
13.0 (11.4, 14.6)
10.5 (9.8, 11.2)
9.6 (8.1, 11.0)
7.8 (7.1, 8.4)
31.0 (29.1, 32.9)
24.4 (23.5, 25.3)
20.0 (18.4, 21.6)
14.9 (14.2, 15.7)
20.2 (18.6, 21.8)
15.0 (14.3, 15.7)
31.8 (30.0, 33.6)
25.9 (25.0, 26.7)
7.9 (6.8, 8.9)
6.1 (5.6, 6.6)
6.0 (5.1, 7.0)
4.8 (4.3, 5.3)
5.7 (4.8, 6.6)
4.3 (3.8, 4.7)
2.3 (1.7, 2.9)
2.1 (1.8, 2.5)
Men Household dysfunction and neglect Lived with someone mentally illb Lived with someone who abused alcohol or drugs Lived with someone with alcohol misuse Lived with someone who abused drugs
b,***
***
**
Lived with someone who served time or was sentenced to serve time Parents separated or divorced Witnessed domestic violence
b,***
b,***
b,*
Abuse Childhood physical abuseb,*** Childhood verbal abuse
b,***
Any childhood sexual abuse
b,**
Touched sexually by anyone Z5 years older
*
Made to sexually touch anyone Z5 years older
**
Forced to have sex by someone Z5 years older Total ACE
***
1.7 (1.6, 1.7)
1.3 (1.3, 1.4) n
Note: Boldface indicates statistical significance. p-values comparing Veterans and civilians within gender strata, po0.05, a Adjusted for age (continuous) and race/ethnicity. b Used to calculate total ACE. ACE, adverse childhood experiences.
nn
po0.01, nnnpo0.001.
www.ajpmonline.org
Katon et al / Am J Prev Med 2015;49(4):573–582
579
validated in the literature, may not necessarily correlate Military service Civilians with the severity of one’s a,b a,b childhood experiences, and RR (95% CI) RR (95% CI) (95% CI) use of total ACE may Women obscure the impact of differences in relative prevalence Health risk factors of specific ACE (e.g., sexual Obese 1.07 (1.00, 1.15) 1.08 (1.07, 1.09) abuse) among the four 1.07 (1.03, 1.12) 1.14 (1.13, 1.15) Ever smoked** groups. We did not stratify by service era, as the draft Binge drinking 0.95 (0.84, 1.08) 1.05 (1.03, 1.07) only applied to men, and we Health related quality of life did not anticipate that the Poor/fair self-perceived health 1.14 (1.04, 1.26) 1.20 (1.18, 1.22) association of ACE with Days of poor physical health 1.20 (1.09, 1.33) 1.18 (1.17, 1.20) HRQOL would vary by service era. This may have led to Days of poor mental health 1.21 (1.12 1.32) 1.25 (1.23, 1.26) underestimation of differenMen ces in prevalence of ACE Health risk factors among men with and without military service. When Obese 1.05 (1.02, 1.07) 1.02 (1.00, 1.04) interaction terms for military 1.06 (1.05, 1.07) 1.12 (1.11, 1.13) Ever smoked*** service and service era were Binge drinking 1.05 (1.01, 1.09) 1.04 (1.02, 1.06) tested in models for ACE among men, statistically sigHealth-related quality of life nificant differences by service Poor/fair self-perceived health*** 1.10 (1.06, 1.14) 1.19 (1.16, 1.22) era were only detected for 1.13 (1.09, 1.18) 1.20 (1.17, 1.24) Days of poor physical health* living with someone with mental illness and childhood Days of poor mental health** 1.21 (1.16, 1.27) 1.30 (1.27, 1.34) physical abuse (data not Note: Boldface indicates statistical significance. p-values for interaction of veteran status and total ACE n nn nnn shown). In both service eras po0.001. within gender strata, po0.05, po0.01, a For dichotomous outcomes (e.g., obese, ever smoked), RR¼relative risk of change and corresponds to (draft versus all volunteer), the change in probability of health risk factor associated with a one-point increase in total ACE. For men with military service continuous outcomes (e.g., days of poor physical health), RR¼rate ratio and corresponds to the change were more likely to have in number of days of poor physical or mental health associated with a one-point increase in total ACE. b Adjusted for age and race/ethnicity. lived with someone with ACE, adverse childhood experiences. mental illness or have experienced childhood physical abuse, but differences Limitations between the two populations were smaller for those in the draft era. Finally, although less than 10% of the data on either This study has several limitations. These include the relatively ACE or HRQOL was missing, it is likely that these data are small number of women with military service, which limited not missing at random because of social desirability bias. If power in the analyses; inability to differentiate between active ACE is associated with poorer adult health and those with a duty service members and Veterans, which may have led to history of ACE were more likely to have missing data for underestimation of the association of ACE and HRQOL; selfACE and for adult health variables, then this analysis report of ACE and health variables, which may have been underestimated the association of ACE with adult health. subject to recall bias; lack of screening data for mental health (e.g., depression); and our inability to adjust for potential confounders, such as parental education level and income, Conclusions when comparing prevalence of ACE by military service. Also, Nevertheless, the current study has several strengths, variables on adult traumatic experiences such as sexual including the population-based study design, examinaassault were unavailable, and the association of ACE with tion of multiple ACE categories, assessment of total ACE, adult health outcomes may include the impact of adult and examination of the association of ACE with HRQOL. trauma.45 Additionally, the total ACE score, though well Table 3. Total ACE, Health Risk Factors, and Health-Related Quality of Life
October 2015
580
Katon et al / Am J Prev Med 2015;49(4):573–582
Figure 1. Relationship between total adverse childhood experiences and health-related quality of life stratified by gender and military service.
Our findings underscore the high prevalence of ACE in the U.S., particularly among women with military service. Given the observed association between ACE and health risk factors such as smoking and depression,4 it is important for clinicians to screen for ACE as part of routine medical practice and to consider these experiences when caring for patients.39 For example, incorporating psychotherapy into depression treatment and smoking-cessation programs may be particularly important for women with depression and a history of ACE.46,47 Longitudinal research is needed to examine specific mechanisms by which ACE may impact adult health, particularly among active duty servicemembers, to identify potential points of intervention. Furthermore, in line with a recent IOM report,48 there is a need to explore effective delivery systems to implement evidencebased programs to reduce child abuse and neglect. Our findings underscore the pressing need to implement and disseminate such programs to prevent ACE and its longterm health consequences.
This work was supported by the Denver-Seattle Veterans Affairs (VA) Health Services Research and Development (HSR&D) Center of Innovation and the VA Office Women’s Health Services. The authors declare no conflicts of interest. Drs. Lehavot and Williams are supported by VA Career Development Awards from CSR&D (CX000867) and HSR&D
(CDA 12-276), respectively. Dr. Williams is also an investigator with the Implementation Research Institute (IRI) at the George Warren Brown School of Social Work at Washington University. IRI is supported through an award from the National Institute of Mental Health (R25 MH080916-01A2) and the VA HSR&D, Quality Enhancement Research Initiative. Drs. Gray and Schure are supported by the Office of Academic Affiliations’ Associated Health Postdoctoral Fellowships (TPP 61-026 and TPP 61-028, respectively), and Dr. Reiber is supported by a VA HSR&D Senior Career Scientist Award (RCS 98-353). The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the VA. No financial disclosures were reported by the authors of this paper.
References 1. Walker EA, Gelfand A, Katon WJ, et al. Adult health status of women with histories of childhood abuse and neglect. Am J Med. 1999;107 (4):332–339. http://dx.doi.org/10.1016/S0002-9343(99)00235-1. 2. MacMillan HL, Fleming JE, Streiner DL, et al. Childhood abuse and lifetime psychopathology in a community sample. Am J Psychiatry. 2001;158(11):1878–1883. http://dx.doi.org/10.1176/appi.ajp.158.11. 1878. 3. Edwards VJ, Holden GW, Felitti VJ, Anda RF. Relationship between multiple forms of childhood maltreatment and adult mental health in community respondents: results from the Adverse Childhood Experiences Study. Am J Psychiatry. 2003;160(8):1453–1460. http://dx.doi. org/10.1176/appi.ajp.160.8.1453.
www.ajpmonline.org
Katon et al / Am J Prev Med 2015;49(4):573–582 4. Felitti VJ, Anda RF, Nordenberg D, et al. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults. The Adverse Childhood Experiences (ACE) Study. Am J Prev Med. 1998;14 (4):245–258. http://dx.doi.org/10.1016/S0749-3797(98)00017-8. 5. Ben-Shlomo Y, Kuh D. A life course approach to chronic disease epidemiology: conceptual models, empirical challenges and interdisciplinary perspectives. Int J Epidemiol. 2002;31(2):285–293. http://dx. doi.org/10.1093/ije/31.2.285. 6. Kuh D, Ben-Shlomo Y, Lynch J, Hallqvist J, Power C. Life course epidemiology. J Epidemiol Community Health. 2003;57(10):778–783. http://dx.doi.org/10.1136/jech.57.10.778. 7. Danese A, Moffitt TE, Harrington H, et al. Adverse childhood experiences and adult risk factors for age-related disease: depression, inflammation, and clustering of metabolic risk markers. Arch Pediatr Adolesc Med. 2009;163(12):1135–1143. http://dx.doi.org/10.1001/ archpediatrics.2009.214. 8. Rodgers CS, Lang AJ, Laffaye C, Satz LE, Dresselhaus TR, Stein MB. The impact of individual forms of childhood maltreatment on health behavior. Child Abuse Negl. 2004;28(5):575–586. http://dx.doi.org/ 10.1016/j.chiabu.2004.01.002. 9. Lange A, de Beurs E, Dolan C, Lachnit T, Sjollema S, Hanewald G. Long-term effects of childhood sexual abuse: objective and subjective characteristics of the abuse and psychopathology in later life. J Nerv Ment Dis. 1999;187(3):150–158. http://dx.doi.org/10.1097/00005053199903000-00004. 10. Finkelhor D, Turner H, Ormrod R, Hamby SL. Violence, abuse, and crime exposure in a national sample of children and youth. Pediatrics. 2009;124(5):1411–1423. http://dx.doi.org/10.1542/peds.2009-0467. 11. Fuller-Thomson E, Filippelli J, Lue-Crisostomo CA. Gender-specific association between childhood adversities and smoking in adulthood: findings from a population-based study. Public Health. 2013;127 (5):449–460. http://dx.doi.org/10.1016/j.puhe.2013.01.006. 12. Fuller-Thomson E, Dalton AD. Gender differences in the association between parental divorce during childhood and stroke in adulthood: findings from a population-based survey. Int J Stroke. 2012. http://dx. doi.org/10.1111/j.1747-4949.2012.00935.x. 13. Sadler AG, Booth BM, Mengeling MA, Doebbeling BN. Life span and repeated violence against women during military service: effects on health status and outpatient utilization. J Womens Health (Larchmt). 2004;13(7):799–811. http://dx.doi.org/10.1089/jwh.2004.13.799. 14. Scherrer JF, Xian H, Kapp JM, et al. Association between exposure to childhood and lifetime traumatic events and lifetime pathological gambling in a twin cohort. J Nerv Ment Dis. 2007;195(1):72–78. http: //dx.doi.org/10.1097/01.nmd.0000252384.20382.e9. 15. Briere J, Elliott DM. Prevalence and psychological sequelae of selfreported childhood physical and sexual abuse in a general population sample of men and women. Child Abuse Negl. 2003;27(10):1205–1222. http://dx.doi.org/10.1016/j.chiabu.2003.09.008. 16. Zinzow HM, Grubaugh AL, Monnier J, Suffoletta-Maierle S, Frueh BC. Trauma among female veterans: a critical review. Trauma Violence Abuse. 2007;8(4):384–400. http://dx.doi.org/10.1177/1524838007307295. 17. Blosnich JR, Dichter ME, Cerulli C, Batten SV, Bossarte RM. Disparities in adverse childhood experiences among individuals with a history of military service. JAMA Psychiatry. 2014;71(9):1041–1048. http://dx.doi.org/10.1001/jamapsychiatry.2014.724. 18. Clancy CP, Graybeal A, Tompson WP, et al. Lifetime trauma exposure in veterans with military-related posttraumatic stress disorder: association with current symptomatology. J Clin Psychiatry. 2006;67(9): 1346–1353. http://dx.doi.org/10.4088/JCP.v67n0904. 19. Kelly UA, Skelton K, Patel M, Bradley B. More than military sexual trauma: interpersonal violence, PTSD, and mental health in women veterans. Res Nurs Health. 2011;34(6):457–467. http://dx.doi.org/10.1002/nur.20453. 20. Lapp KG, Bosworth HB, Strauss JL, et al. Lifetime sexual and physical victimization among male veterans with combat-related post-traumatic stress disorder. Mil Med. 2005;170(9):787–790.
October 2015
581
21. Montgomery AE, Cutuli JJ, Evans-Chase M, Treglia D, Culhane DP. Relationship among adverse childhood experiences, history of active military service, and adult outcomes: homelessness, mental health, and physical health. Am J Public Health. 2013;103(suppl 2):S262–S268. http://dx.doi.org/10.2105/AJPH.2013.301474. 22. Behavioral Risk Factor Surveillance System. Annual survey data, 2011 and 2012. Atlanta, GA: USDHHS, CDC. http://www.cdc.gov/brfss/ annual_data/annual_data.htm. 23. Reiber GE, Koepsell TD, Maynard C, Haas LB, Boyko EJ. Diabetes in nonveterans, veterans, and veterans receiving Department of Veterans Affairs health care. Diabetes Care. 2004;27(suppl 2):B3–B9. http://dx. doi.org/10.2337/diacare.27.suppl_2.B3. 24. Grossbard JR, Lehavot K, Hoerster KD, Jakupcak M, Seal KH, Simpson TL. Relationships among veteran status, gender, and key health indicators in a national young adult sample. Psychiatr Serv. 2013;64 (6):547–553. http://dx.doi.org/10.1176/appi.ps.003002012. 25. Hoerster KD, Lehavot K, Simpson T, McFall M, Reiber G, Nelson KM. Health and health behavior differences: U.S. Military, veteran, and civilian men. Am J Prev Med. 2012;43(5):483–489. http://dx.doi.org/ 10.1016/j.amepre.2012.07.029. 26. Lehavot K, Hoerster KD, Nelson KM, Jakupcak M, Simpson TL. Health indicators for military, veteran, and civilian women. Am J Prev Med. 2012;42(5):473–480. http://dx.doi.org/10.1016/j.amepre.2012.01.006. 27. Andersen JP, Blosnich J. Disparities in adverse childhood experiences among sexual minority and heterosexual adults: results from a multistate probability-based sample. PLoS One. 2013;8(1):e54691. http://dx. doi.org/10.1371/journal.pone.0054691. 28. NIAAA. Helping Patients Who Drink Too Much: A Clinician’s Guide. Bethesda, MD: National Institute on Alcohol Abuse and Alcoholism, USDHHS, NIH; 2005. 29. Measuring Healthy Days. USDHHS, CDC, National Center for Chronic Disease Prevention and Health Promotion, Division of Adult and Community Health; 2000. 30. Lumley T, Kronmal R, Ma S. Relative Risk Regression in Medical Research: Models, Contrasts, Estimators, and Algorithms. UW Biostatistics Working Paper Series, 2006; Working Paper No. 293. 31. Currie J, Widom CS. Long-term consequences of child abuse and neglect on adult economic well-being. Child Maltreat. 2010;15(2):111– 120. http://dx.doi.org/10.1177/1077559509355316. 32. Mengeling MA, Sadler AG, Torner J, Booth BM. Evolving comprehensive VA women’s health care: patient characteristics, needs, and preferences. Womens Health Issues. 2011;21(4 suppl):S120–S129. http: //dx.doi.org/10.1016/j.whi.2011.04.021. 33. Washington DL, Yano EM, Simon B, Sun S. To use or not to use. What influences why women veterans choose VA health care. J Gen Intern Med. 2006;21(suppl 3):S11–S18. http://dx.doi.org/10.1111/j.1525-1497. 2006.00369.x. 34. Koepsell TD, Weiss NS. Epidemiologic Methods Studying the Occurrence of Illness. New York, NY: Oxford University Press; 2003. 35. Coid J, Petruckevitch A, Feder G, Chung W, Richardson J, Moorey S. Relation between childhood sexual and physical abuse and risk of revictimisation in women: a cross-sectional survey. Lancet. 2001;358 (9280):450–454. http://dx.doi.org/10.1016/S0140-6736(01)05622-7. 36. Merrill LL, Newell CE, Thomsen CJ, et al. Childhood abuse and sexual revictimization in a female Navy recruit sample. J Trauma Stress. 1999;12(2):211–225. http://dx.doi.org/10.1023/A:1024789723779. 37. Iversen AC, Fear NT, Simonoff E, et al. Influence of childhood adversity on health among male UK military personnel. Br J Psychiatry. 2007;191:506–511. http://dx.doi.org/10.1192/bjp.bp.107.039818. 38. Van Voorhees EE, Dedert EA, Calhoun PS, Brancu M, Runnals J, Beckham JC. Childhood trauma exposure in Iraq and Afghanistan war era veterans: implications for posttraumatic stress disorder symptoms and adult functional social support. Child Abuse Negl. 2012;36(5):423– 432. http://dx.doi.org/10.1016/j.chiabu.2012.03.004.
582
Katon et al / Am J Prev Med 2015;49(4):573–582
39. King DW, King LA, Foy DW, Gudanowski DM. Prewar factors in combat-related posttraumatic stress disorder: structural equation modeling with a national sample of female and male Vietnam veterans. J Consult Clin Psychol. 1996;64(3):520–531. http://dx.doi.org/10.1037/ 0022-006X.64.3.520. 40. Ye D, Reyes-Salvail F. Adverse childhood experiences among Hawai’i adults: Findings from the 2010 Behavioral Risk Factor Survey. Hawaii J Med Public Health. 2014;73(6):181–190. 41. Koenen KC, Moffitt TE, Poulton R, Martin J, Caspi A. Early childhood factors associated with the development of posttraumatic stress disorder: results from a longitudinal birth cohort. Psychol Med. 2007;37(2):181–192. http://dx.doi.org/10.1017/S00332 91706009019. 42. Van Voorhees E, Scarpa A. The effects of child maltreatment on the hypothalamic-pituitary-adrenal axis. Trauma Violence Abuse. 2004;5 (4):333–352. http://dx.doi.org/10.1177/1524838004269486. 43. Gillespie CF, Bradley B, Mercer K, et al. Trauma exposure and stressrelated disorders in inner city primary care patients. Gen Hosp Psychiatry. 2009;31(6):505–514. http://dx.doi.org/10.1016/j.genhosppsych.2009.05.003. 44. Maccani MA, Delahanty DL, Nugent NR, Berkowitz SJ. Pharmacological secondary prevention of PTSD in youth: challenges and
45.
46.
47.
48.
opportunities for advancement. J Trauma Stress. 2012;25(5):543–550. http://dx.doi.org/10.1002/jts.21731. Min MO, Minnes S, Kim H, Singer LT. Pathways linking childhood maltreatment and adult physical health. Child Abuse Negl. 2013;37 (6):361–373. http://dx.doi.org/10.1016/j.chiabu.2012.09.008. Nemeroff CB, Heim CM, Thase ME, et al. Differential responses to psychotherapy versus pharmacotherapy in patients with chronic forms of major depression and childhood trauma. Proc Natl Acad Sci U S A. 2003;100(24):14293–14296. http://dx.doi.org/10.1073/pnas.2336126100. Blalock JA, Minnix JA, Mathew AR, Wetter DW, McCullough JP, Cinciripini PM. Relationship of childhood trauma to depression and smoking outcomes in pregnant smokers. J Consult Clin Psychol. 2013;81(5):821–830. http://dx.doi.org/10.1037/a0033381. USDHHS. Child abuse and neglect. Child Welfare Information Gateway. www.childwelfare.gov/can/.
Appendix Supplementary data Supplementary data associated with this article can be found at, http://dx.doi.org/10.1016/j.amepre.2015.03.020.
www.ajpmonline.org
