Victimization in childhood: General and specific associations with physical health problems in young adulthood

PSR-09038; No of Pages 7 Journal of Psychosomatic Research xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Journal of Psychosomatic Research

Victimization in childhood: General and specific associations with physical health problems in young adulthood Laura E. Miller-Graff a,⁎, Åsa Källström Cater b, Kathryn H. Howell c, Sandra A. Graham-Bermann d a

Department of Psychology, Kroc Institute for International Peace Studies, University of Notre Dame, 107 Haggar Hall, Notre Dame, IN 46556, United States School of Law, Psychology and Social Work, Örebro University, 701 82 Örebro, Sweden Department of Psychology, University of Memphis, 202 Psychology Building, Memphis, TN 38152, United States d Department of Psychology, University of Michigan, 530 Church Street, Ann Arbor, MI 48109-1043, United States b c

a r t i c l e

i n f o

Article history: Received 26 February 2015 Received in revised form 30 June 2015 Accepted 3 July 2015 Available online xxxx Keywords: Childhood victimization Mental health Physical health Polyvictimization

a b s t r a c t Objective: The goal of the current study was to examine the direct relationship between diverse types of childhood victimization and physical health problems in early adulthood, controlling for other common factors that contribute to physical health problems, including psychopathology and health risk behaviors. The associations between types of victimization (e.g., physical assault) and specific health problems (e.g., pain) were also examined. Methods: 2500 Swedish young adults reported on their exposure to victimization in childhood and their current mental and physical health as adults. Results: Using multiple regression, results indicated that the amount of childhood victimization was a significant predictor of health problems in adulthood, controlling for the significant negative effects of health risk behaviors and mental health problems on physical health. Logistic regressions indicated that physical assaults and sexual abuse were associated with all types of health problems assessed. Sleep problems were associated with almost all types of victimization history. Conclusions: The long-term effects of childhood victimization on physical health in adulthood are serious and warrant significant attention. Primary care providers should include assessments of past victimization as one way of screening for health risk. Health providers should also consider multiple points of intervention that may help to reduce physical illness. For example, providing a mental health intervention or social service support related to victimization experiences may not only address these difficulties, but also more broadly impact physical health as well. © 2015 Elsevier Inc. All rights reserved.

Introduction Victimization in childhood, including maltreatment, neglect, community crime, and witnessed/indirect violence, is relatively common [1,2] and has been linked to increased short and long term psychopathology [3–9]. In keeping with modern definitions of health [10], many research studies have begun to more deeply consider the associations between physical health, mental health, and the environmental correlates of each. In studies of individuals exposed to different forms of violence, crime, and neglect, the presence of mental health difficulties has been repeatedly associated with physical health problems. As examples, one study found that rates for asthma, allergies, gastrointestinal problems, headaches and cold/flu were higher for children who had symptoms of traumatic stress [11]. Other research identified a relationship between Posttraumatic Stress Disorder (PTSD) and Type 2 Diabetes [12]. ⁎ Corresponding author. E-mail address: [email protected] (L.E. Miller-Graff).

Cross-sectional relationships between physical and mental health identified in past research have been generally strong, and both mental and physical health show relative stability over time. Mediating and reciprocal effects, however, have tended to be weaker and less consistent across models [13] and studies have suggested that mental health factors may both precede physical health problems and be exacerbated by them [14]. Further, the relationship between mental and physical health may be multiply determined by other social, environmental and behavioral factors. For example, childhood adversity has been related to several health risk behaviors, such as cigarette smoking and weight gain, which in turn have shown direct effects on inflammatory markers in the body, increasing vulnerability to infections and disease [15,16]. For this reason, cross-sectional data on mental and physical health, while being able to adequately control for concurrent associations, may not provide highly useful information about the processes underlying the concordance between mental and physical health over time. Physiological research has provided important insight into some of the underlying processes explaining the association between mental

http://dx.doi.org/10.1016/j.jpsychores.2015.07.001 0022-3999/© 2015 Elsevier Inc. All rights reserved.

Please cite this article as: L.E. Miller-Graff, et al., Victimization in childhood: General and specific associations with physical health problems in young adulthood, J Psychosom Res (2015), http://dx.doi.org/10.1016/j.jpsychores.2015.07.001

2

L.E. Miller-Graff et al. / Journal of Psychosomatic Research xxx (2015) xxx–xxx

and physical health. In the context of victimization, mental health problems, such as PTSD, are thought to relate to physical health difficulties due, in part, to disturbances in stress hormone regulation systems [17]. For example, when examining physical, sexual and emotional abuse together, studies found that these children displayed high cortisol during the day [18], especially in the afternoon [19], which is typically when cortisol levels should be at their lowest concentrations. Such variation in this established cortisol pattern was a marker of risk for poor health [20]. More recently, Kuhlman et al. [21] investigated relations between stress responses and exposure to physical abuse, emotional abuse, or non-intentional trauma. They found that nonintentional trauma was associated with elevated cortisol in the evening, physical abuse was associated with faster physiological reactivity to acute stress, and emotional abuse was associated with delayed recovery of cortisol following acute stress. This body of research suggests that dysregulation of physiological stress systems contributes to both overand under-activation of systems that control inflammation and immune responses, which may make individuals more vulnerable to illness [17]. Other studies, however, found no link between PTSD and somatic symptom severity [22] or have found that children exhibiting more posttraumatic stress symptoms display fewer total health problems [23]. Despite some discrepancies regarding the relationship between mental health problems and physical illness after childhood victimization, there has been growing evidence of a direct link between victimization in childhood and physical health problems in both the immediate aftermath of such adverse events [11,23,24] and over the life course [22,25]. Two large-scale studies of adverse childhood experiences (including rejection, abuse, harsh parenting, violence against the mother, parental substance abuse, mental illness and suicide) linked cumulative exposure to adversity to physical health problems in adulthood. The Dunedin Multidisciplinary Health and Development Study took place in New Zealand and showed that 33% of the sample was exposed to maltreatment as a child, which greatly increased the odds (1:181) of having a physical health problem in adulthood [26]. A second large-sample study by Felitti and colleagues, the Adverse Childhood Experience (ACE) study, assessed 7 categories of childhood experiences including violence and abuse, with 50% reporting at least one ACE and 25% reporting two or more ACEs [27]. Here, there was a dose–response relationship between the number of ACEs and physical health risks as an adult. Such physical health problems exceed direct injuries sustained from the violent event and include greater numbers of somatic complaints and symptom severity, e.g., headaches, chest pain, and nausea [22], as well as arthritis and cardiovascular disease [25]. Further, a history of childhood victimization has been found to increase the risk of mortality and has been associated with obesity [2,25,26,28]. Exposure to violent events and serious life stressors has also been linked to increased visits to the doctor and nights spent in the hospital [29]. A history of victimization has also been connected to a number of health risk behaviors, such as excessive drinking, alcohol or drug misuse, risky sexual behavior, and becoming a teen parent [1,2,5,28]. Despite emerging evidence for the link between victimization in childhood and later physical health problems, few studies have examined a broad range of adverse events (as in McCall-Hosenfield et al.'s work [22]) and even fewer have assessed how varying histories of childhood exposure are differentially related to specific health problems in adulthood (as in Afifi et al.'s work [25]). Further, while many studies have taken into consideration sociodemographic effects on health, only a small number adjust for the contribution of mental health problems to physical health after exposure to adversity (as in Gawronski et al.'s work [29]). As such, the examination of associations between victimization types and specific physical health problems and the assessment of the contribution of number of victimizations to these health problems (controlling for the role of mental health) represents a significant and unique contribution to this growing literature. The first objective of the current study, therefore, was to evaluate the association between types of childhood victimization and specific

physical health problems in early adulthood. Based on previous research, it was expected that childhood victimization would be associated with the total number of physical health problems in early adulthood, but relationships between specific types of victimization and physical health problems were regarded as exploratory. The second study objective was to examine the association between the total amount of victimization in childhood and total physical health problems, controlling for other factors commonly associated with physical health problems in victimized populations (i.e., mental health problems, health risk behaviors). It was hypothesized that the amount of victimization in childhood would be uniquely associated with the total number of physical health problems, beyond expected significant relationships between health risk behaviors, mental health, and physical health.

Methods Participants Participants were randomly selected from the Swedish national inhabitant register based on proportional draws from different geographic regions. A total of 20,827 individuals with a registered telephone number were drawn from the register. Of these, 4455 individuals could be reached and agreed to participate in the study. However, 1955 were not included because an interview appointment could not be made, the participant did not arrive for the interview, or the participant changed his/her mind about participation. The sample thus included 2500 young adults (52.6% female) between the ages of 20 and 24 (M = 22.1, SD = 1.38). At the time of interview, 69.4% of participants were employed, 58.3% of participants were enrolled in college, and 47% were financed by study assistance. Participants responded to an extensive battery of questionnaires regarding childhood exposure to violence and current psychosocial functioning. To investigate whether the non-participants differed from the participants, 30 randomly chosen non-participating men and women were asked a few questions from the interview/questionnaire through a telephone interview and were then compared with the total sample of 2500 participants [1]. The analyses showed no significant differences between non-participants and participants in level of education, subjective well-being, prevalence of psychiatric diagnoses, alcohol risk use, criminality, or physical victimization.

Procedure All study procedures were evaluated and approved by the ethics committee in Uppsala, Sweden to ensure protection of human subjects. Selected participants were contacted by telephone and given information about the study. If willing to participate, they were scheduled for an interview at a time and location of their choosing. Interviews were conducted by trained employees of a Swedish survey company. Basic demographic information was obtained using a brief interview (5– 10 min). Participants then completed a survey questionnaire electronically, which assessed their history of violence exposure and current psychosocial functioning. This self-report survey took approximately one hour to complete, and after completion, participants received a voucher for 400 Swedish Kronor (approximately 60 USD) as compensation for their time. Referral information for local mental health resources was provided during the post-survey de-briefing.

Measures Demographics Demographic information was collected during a brief face-to-face interview and included information about the participant's sex and age.

Please cite this article as: L.E. Miller-Graff, et al., Victimization in childhood: General and specific associations with physical health problems in young adulthood, J Psychosom Res (2015), http://dx.doi.org/10.1016/j.jpsychores.2015.07.001

L.E. Miller-Graff et al. / Journal of Psychosomatic Research xxx (2015) xxx–xxx

Childhood victimization All participants responded to items assessing childhood victimization across a variety of domains, including property crimes (4 items), physical assaults (11 items), verbal assaults (2 items), sexual abuse (7 items), neglect (5 items), and witnessed/indirect violence exposure (7 items). Some items were drawn from the Juvenile Victimization Questionnaire [30,31] and others were drawn from violence prevalence studies conducted in Europe [32]. Full details on the 33-item questionnaire and its modifications for the current study can be found in Cater et al. [1]. For each item, participants reported whether they had experienced the event not at all (0), once (1), twice (2) three times (3) four times (4) or five or more times (5). Participants' responses were recoded to reflect whether or not they had experienced each item (0 = no, 1 = yes) and then summed within the subscale to reflect the total number of events that each participant experienced within that domain. Physical health problems Participants responded to a series of questions regarding their physical health problems, including headaches, gastrointestinal problems, chest pain, back/joint/muscle pain, and sleeping problems. Participants indicated whether they had experienced each health problem within the past year. Prevalence rates for each health problem in this sample are reported in Table 1. On average, participants endorsed between one and two health problems in the past year (see Table 1). Physical health problem indicators In addition to self-reported physical health problems, data were gathered on participants' overall concern with their physical health and body mass index. Specifically, participants were asked “Do you worry about your physical health?” (1 = No to 4 = Yes, very often). Body mass index (BMI) was calculated by dividing mass (kg) by height in meters squared. Health risk behaviors and mental health problems Due to past research that indicates clear associations between health risk behaviors and mental health difficulties in the development of Table 1 Descriptive statistics. Physical health problems

(%)/M(SD)

Range

Back, joint, or muscle pain Chest pain Headaches Insomnia Stomachaches, nausea, diarrhea Total health problems

42.9% 32.2% 20.1% 31.2% 27.5% 1.15(1.29)

0–5

Health problem indicators

M(SD)

Range

General health concern BMI

1.75(0.72) 23.25(3.77)

1–4 13.59–58.54

Health risk behaviors and mental health problems

M(SD)

Range

Anxiety Depression Posttraumatic stress Illegal drug use Alcohol misuse Smoking Victimization Property crimes Physical assaults Verbal abuse Sexual abuse Neglect Witnessed/indirect exposure Total victimization

5.31(3.81) 2.91(2.67) 0.43(0.65) 0.07(.26) 7.42(4.55) 1.93(1.04) M(SD) 1.03(1.18) 2.00(2.47) 0.58(.70) 0.47(1.13) 0.21(0.61) 0.19(0.49) 4.49(4.38)

0–21 0–18 0–4 0–1 1–31 1–4 Range 0–4 0–11 0–2 0–7 0–5 0–4 0–28

3

physical health problems, the following variables were examined in this study. Health risk behaviors. Participants reported on three types of health risk behaviors, including smoking frequency (1 = None, 2 = Yes, socially, 3 = Yes, occasionally, and 4 = Yes, daily/almost daily), illicit drug use (yes/no), and alcohol misuse. Alcohol misuse was assessed with the Alcohol Use Disorders Identification Test (AUDIT [33]). This measure consists of 10 items that examine the amount of alcohol consumed and whether drinking led to negative consequences, of which the majority of items were rated on a 5-point scale (0 = Never, 1 = Less than once a month, 2 = Every month, 3 = Every week, 4 = Daily or almost daily). Mental health problems. Posttraumatic stress symptoms (PTSS) were evaluated using the Impact of Events Scale—Revised (IES-R [34]), a 22-item questionnaire that assesses the frequency of re-experiencing, hyperarousal, and avoidance symptoms (1—Not at all to 5—Extremely often). Reliability for the total scale was α = .96. Depression and anxiety were evaluated using the Hospital Anxiety and Depression Scale (HADS [35]). This scale consists of 14 items assessing symptoms of depression and anxiety during the last week. Participants reported how much they experienced each symptom on a 4-point scale (0—No, not at all to 3—Yes, definitely). Internal reliability for the anxiety subscale of the HADS was α = .78, and internal reliability for the depression subscale of the HADS was α = .66. Data analysis The first study objective aimed to explore the direct associations between specific types of victimization in childhood and specific physical health problems in early adulthood. In order to examine this hypothesis, five logistic regressions were conducted, one for each type of physical health problem. For these analyses, the absence of the symptom (no = 0) was the referent category (see Table 2). Logistic regressions controlled for all types of victimization, as well as for sex, health problem indicators (i.e., health concern and BMI), and health risk behaviors (i.e., alcohol misuse, illegal drug use, cigarette smoking). The second objective of the study aimed to examine the contribution of total childhood victimization to total endorsed health problems in early adulthood, with additional controls for the role of mental health difficulties (i.e., anxiety, depression, posttraumatic stress symptoms). This analysis was conducted using multiple regression analysis, with total health problems set as the dependent variable. Total health problems and reports of childhood victimization experiences both exhibited significant positive skew, so all analyses were run with both non-transformed and log-transformed variables. The direction and significance of effects were identical in both sets of analyses. Given that the results were the same, and the use of untransformed variables permits easier interpretability of path coefficients, the results presented here reflect analyses conducted without variable transformation. Results On average, young adults in this sample reported 5 types of childhood victimization, although victimization varied considerably (SD = 4.38, see Table 1). The most commonly reported type of victimization was physical assault (57.6%), and the least commonly reported type of victimization was neglect (11.0%). On average, participants endorsed between one and two physical health problems over the past year, the most common being joint, muscle or back pain (42.9%) and the least common being headaches (20.1%). Descriptive statistics for all study variables are presented in Table 1. Participants' body mass indices revealed that 2.56% were underweight (BMI b 18) and 5.32% were obese (BMI N 30). Childhood victimization exposure showed significant, positive correlations across types of victimization, suggesting that victimization in one domain was related to experiences of victimization

Please cite this article as: L.E. Miller-Graff, et al., Victimization in childhood: General and specific associations with physical health problems in young adulthood, J Psychosom Res (2015), http://dx.doi.org/10.1016/j.jpsychores.2015.07.001

4

L.E. Miller-Graff et al. / Journal of Psychosomatic Research xxx (2015) xxx–xxx

Table 2 Correlations between childhood victimization types. 1. 1. Verbal abuse 2. Property offenses 3. Physical assault 4. Sexual abuse 5. Neglect 6. Witnessed violence

1

.23⁎⁎⁎ .37⁎⁎⁎ .24⁎⁎⁎ .32⁎⁎⁎ .17⁎⁎⁎

2. 1

.43⁎⁎⁎ .17⁎⁎⁎ .12⁎⁎⁎ .11⁎⁎⁎

3.

4.

1 .22⁎⁎⁎ .38⁎⁎⁎ .19⁎⁎⁎

1 .26⁎⁎⁎ .06⁎⁎

5.

Sex differences

1 .12⁎⁎⁎

t = 3.92, p b .001 t = −5.10, p b .001 t = −11.27, p b .001 t = 14.96, p b .001 t = 2.29, p b .05 t = −2.61, p b .01

Note: Negative t-values = average male exposure N average female exposure. ⁎ p b .05. ⁎⁎ p b .01. ⁎⁎⁎ p b .001.

in other domains. The size of these correlations, however, was low to moderate (see Table 2). There were significant differences between males and females in reports of victimization, with women reporting higher levels of past sexual abuse, verbal abuse, and neglect, and men reporting higher levels of past physical assaults, property offenses, and witnessed violence (see Table 2). To test the first hypothesis, logistic regressions were used to examine associations between physical health problems in early adulthood and types of childhood victimization (i.e., property crimes, physical assault, verbal abuse, sexual abuse, neglect, and witnessed/indirect violence). All models were highly significant, although variance explained was relatively low (see Table 3). An examination of the covariates indicated that women reported higher levels of gastrointestinal difficulties, headaches and back/joint/muscle pain than did men (RRRs 2.01–4.27; see Table 3). Smoking frequency was also significantly associated with all health problems, except gastrointestinal difficulties, and drug use was significantly associated with sleeping problems (RRR = 1.66). BMI was significantly associated with only back/joint/ muscle pain (RRR = 1.04).

Regarding the relationships between childhood victimization types and health problems in early adulthood, physical abuse and sexual abuse were consistently, significantly, and positively related to all five health problems assessed in this study (Table 3). In contrast, verbal abuse was related to sleeping problems (RRR = 1.33), gastrointestinal difficulties (RRR = 1.21) and chest pain (RRR = 1.27), whereas neglect was associated with only sleeping problems (RRR = 1.34). Exposure to property offenses was not related to any type of health problem. Exposure to witnessed violence was not significantly associated with sleeping problems (RRR = 1.18, p = .09), gastrointestinal difficulties (RRR = 1.20, p = .07), and back/muscle/ joint pain (RRR = 1.20, p = .07). The second hypothesis postulated that the amount of victimization in childhood would be associated with the total number of physical health problems endorsed in early adulthood, controlling for relevant health risk behaviors and mental health problems associated with higher levels of somatic complaints. This hypothesis was evaluated using multiple regression analysis and the overall model explained a significant portion of the variance in young adults' physical health

Table 3 Associations between specific types of childhood victimization and health problems. Sleep problems RRR (95% CI)

Stomachaches, nausea, vomiting, diarrhea RRR (95% CI)

Headaches RRR (95% CI)

Chest pain RRR (95% CI)

Back, joint, or muscle pain RRR (95% CI)

1.33⁎⁎⁎ (1.15, 1.54) .98 (.89, 1.07) 1.06⁎⁎

1.21⁎ (1.03, 1.40) 1.01 (.92, 1.11) 1.09⁎⁎

1.18 (.99, 1.40) .94 (.85, 1.05) 1.10⁎⁎

1.27⁎ (1.05, 1.56) .92 (.81, 1.04) 1.08⁎

.95 (.81, 1.12) 1.05 (.96, 1.16) 1.11⁎⁎

(1.01, 1.11) 1.10⁎ (1.01, 1.21) 1.34⁎⁎ (1.13, 1.43) 1.18 (.97, 1.43)

(1.04, 1.15) 1.11⁎ (1.01, 1.21) 1.09 (.92, 1.30) 1.20 (.98, 1.46)

(1.04, 1.17) 1.16⁎⁎ (1.06, 1.27) .98 (.81, 1.19) 1.11 (.89, 1.38)

(1.02, 1.15) 1.14⁎ (1.03, 1.27) 1.15 (.95, 1.39) 1.14 (.91, 1.44)

(1.06, 1.17) 1.12⁎ (1.02, 1.23) 1.16 (.99, 1.47) 1.04 (.97, 1.10)

χ2 Pseudo R2

(1.16, 2.37) 269.27⁎⁎⁎ 9.5%

2.94⁎⁎⁎ (2.31, 3.76) 1.00 (.97, 1.02) 1.87⁎⁎⁎ (1.62, 2.16) 1.07 (.97, 1.19) .99 (.97, 1.02) 1.16 (.78, 1.72) 324.02⁎⁎⁎ 12.0%

4.27⁎⁎⁎ (3.19, 5.70) 1.02 (.99, 1.05) 1.58⁎⁎ (1.36, 1.84) 1.16⁎⁎ (1.04, 1.30) .97 (.95, 1.00) 1.16 (.73, 1.84) 303.59⁎⁎⁎ 13.2%

1.30 (.95, 1.79) .99 (.96, 1.03) 1.55⁎⁎⁎ (1.30, 1.86) 1.15⁎ (1.01, 1.30) 1.02 (.99, 1.05) 1.31 (.83, 2.06) 131.90⁎⁎⁎ 7.9%

2.04⁎⁎⁎ (1.60, 2.61) 1.04⁎⁎

Drug use

1.14 (.92, 1.43) 1.01 (.98, 1.03) 1.61⁎⁎⁎ (1.41, 1.85) 1.22⁎⁎⁎ (1.11, 1.34) 1.01 (.99, 1.03) 1.66⁎⁎⁎

Victimization type Verbal abuse Property offenses Physical assault Sexual abuse Neglect Witnessed violence

Covariates Sex BMI Health concern Smoking Alcohol misuse

(1.01, 1.07) 1.68⁎⁎⁎ (1.46, 1.94) 1.12⁎ (1.01, 1.24) .99 (.96, 1.01) .89 (.59, 1.35) 230.29⁎⁎⁎ 9.1%

⁎ p b .05. ⁎⁎ p b .01. ⁎⁎⁎ p b .001.

Please cite this article as: L.E. Miller-Graff, et al., Victimization in childhood: General and specific associations with physical health problems in young adulthood, J Psychosom Res (2015), http://dx.doi.org/10.1016/j.jpsychores.2015.07.001

L.E. Miller-Graff et al. / Journal of Psychosomatic Research xxx (2015) xxx–xxx Table 4 Multiple regression analyses examining the association between number of types of childhood victimizations and physical health problems in young adulthood.

Total victimization Sex Anxiety Depression Posttraumatic stress Health concerns Illegal drug use Alcohol misuse Smoking BMI

B(SE)

t(p)

.04(.01) .49(.05) .05(.01) .04 (.01) .31(.05) .30(.04) .17(.09) −.01(.01) .10(.02) .01(.01)

6.14(b.001) 9.44(b.001) 6.51(b.001) 3.59(b.001) 6.76(b.001) 8.30(b.001) 1.88(.061) −1.44(.150) 4.08(b.001) 2.31(.021)

F = 96.99 (p b .001), R2 = .31.

problems (R2 = .31, F = 96.99, p b .001). As expected, childhood victimization was significantly associated with physical health problems (β = .04, p b .001), even after accounting for the role of current mental health problems (anxiety, depression, and PTSD; see Table 4), sex, smoking, BMI, illicit drug use, and alcohol misuse. Discussion The objective of the current study was to examine the association between childhood victimization and physical health problems in early adulthood. The rates of victimization in the current sample appeared to be higher than some studies, although it is difficult to compare studies that assess violence differently and occur in different contexts. Here, more than half (57.6%) were exposed to physical assault and 11% to neglect at some point during childhood, while the New Zealand study reported approximately one third (33%) exposed to maltreatment during the first 12 years of childhood [26]. These findings of the current study are more consistent with the Felitti et al. [27] ACEs study that showed at least 50% experienced at least on ACE during childhood and 25% experienced 2 or more ACEs. Although a link between victimization and health problems has been previously observed [11,23,24], few studies have evaluated a large breadth of childhood victimization experiences and their effects in adulthood, controlling for other known factors that contribute to physical health problems (e.g., mental illness). Due to this limitation in previous research, most studies have examined the cumulative effects of victimization on physical health without considering the unique influence of specific types of victimization experiences. The current study sought to address these gaps in the literature by evaluating both the specific associations between victimization typology and physical health problems, as well as the cumulative effect of victimization on physical health problems, controlling for mental health difficulties and health risk behaviors. In assessing the relationship between specific types of childhood victimization and physical health problems, physical and sexual victimization were consistently and significantly associated with all of the assessed health problems (see Table 3). The relationships between verbal abuse, neglect, and health problems, however, were less consistent. No relationships were evident between property crimes and health problems. Beyond the contribution of this information to the body of research on victimization and health, the investigation of specific associations between victimization and health reveals that individuals with particular trauma histories, especially physical and sexual victimization, may have greater health care needs than others. These individuals may require more comprehensive health and wellness interventions. Those experiencing other types of victimization, alternatively, may have more isolated health difficulties and require more targeted intervention (e.g., psychoeducation around sleep hygiene). The most commonly related health problem across types of victimization history was sleep problems (see Table 3, column 1), followed by gastrointestinal difficulties and chest pain. These findings reinforce other research demonstrating the cumulative effect of victimization in primary

5

insomnia [25] and gastrointestinal difficulties [26]. Headaches, conversely, were uniquely associated with physical and sexual abuse, as was back/ joint/muscle pain. The increased prevalence of headaches has been identified in other literature specific to sexual and physical abuse [36,37], but because previous studies did not assess a broad range of victimizations, until now, it has been difficult to establish whether headaches are associated with victimization more broadly. Of utmost importance, study results extend beyond previous research by indicating that physical health events are associated with the presence of a variety of typologies of victimization (as opposed to cumulative effects alone). In terms of clinical utility, these findings suggest that individuals with any history of victimization are likely to benefit from thorough assessment, psychoeducation, and remediation of sleep problems, especially given the emerging literature that connects sleep decrements to a host of other difficulties [38]. Notably, differences between males and females emerged both in regard to victimization history (Table 2) and endorsed health problems (Table 3). Women were significantly more likely than men to report histories of verbal abuse, sexual abuse and neglect, while males were more likely to report exposure to physical assaults, property offenses and witnessed violence. Generally speaking, these findings replicate similar studies of gender differences in lifetime exposure to adversity in the United States [39]. Regarding the assessed health problems, women were more likely than men to report higher levels of gastrointestinal difficulties, headaches, and back/joint/muscle pain, suggesting that women may need particular support in these domains, especially if they have a history of abuse. The main effect of total childhood victimization on health problems in early adulthood (Table 4) was also significant, even when controlling for other relevant health-related factors. This finding powerfully reinforces the direct effect of victimization on physical health identified in other studies [26,27,39,40] and extends such findings to individuals in Sweden, thus providing evidence for the cross-cultural impact of childhood victimization. This direct effect was present even after controlling for mental health factors that are common following exposure to violence (e.g., depression). Therefore, these study results suggest that it is not only the mental health ramifications or the increase in risky behaviors that solely drive physical health problems following victimization, rather the experience of being victimized in childhood is a strong, direct risk factor of young adult physical health problems. Importantly, anxiety, depression, and health risk behaviors were also significantly related to health problems. As such, healthcare providers should not discount their role in physical health problems, but rather consider that victimization history, mental health problems, and health risk behaviors represent a constellation of risk factors, each of which may hold unique explanatory power. A number of additional suggestions for healthcare providers and clinical practice emerge based on these findings. First, when conducting a comprehensive health evaluation, providers should attend not only to current individual and environmental factors typically associated with physical health, but also screen for exposure to past and present victimization, as well as psychopathology, that may potentially contribute to the expression of physical illness. Next, health providers should consider multiple points of intervention that may help to reduce physical health problems. For example, providing a mental health intervention or social service support related to victimization experiences may address these specific difficulties and the impact of physical health more broadly. Finally, study findings indicate that some physical health problems are linked to unique forms of victimization, therefore healthcare providers should tailor screenings to assess for these specific childhood experiences when examining conditions such as headache, pain, or weight concerns. Such information may help to develop the most appropriate and beneficial treatment. It is especially notable that the current study evaluated the relationships between victimization and health in Sweden for several reasons. First, this sample is overall, in extremely good health compared to samples from many other countries. For example, one study found that 30.3% of an American sample aged 20–39 was obese [41] whereas

Please cite this article as: L.E. Miller-Graff, et al., Victimization in childhood: General and specific associations with physical health problems in young adulthood, J Psychosom Res (2015), http://dx.doi.org/10.1016/j.jpsychores.2015.07.001

6

L.E. Miller-Graff et al. / Journal of Psychosomatic Research xxx (2015) xxx–xxx

problems would be useful in further identifying risk and protective factors. In addition, the role of family socioeconomic status in the relation between types of childhood victimization and physical health problems in Sweden requires further exploration.

the rate for the current study was 5.32%. That relationship between victimization and physical health is present here; even in the relative absence of other serious contributing factors to health problems, is notable. Further, although some social inequality related to health care has been noted in Sweden, only b 3% of Swedes reported having limited their use of or refrained from seeking health care in one national study [42]. Generally speaking, Sweden has achieved international acclaim for their universal health care system, which ranks among the top in the world in terms of access and outcomes [43]. In contrast, the United States continues to struggle with systematic sociodemographic disparities in health care [44]. Again, that the negative effects of victimization on health are evident even in the context of easy access to highly successful health care systems underscores the detrimental effects of childhood victimization.

Our study findings indicate that childhood victimization is a serious public health problem. To protect the health of young adults, early detection of victimization is essential. Thus, results highlight the importance of clinicians conducting a thorough evaluation of past victimization when completing health assessments for young adults. Health professionals should consider routinely asking about childhood victimization experiences when young adults seek care.

Limitations and future directions

Acknowledgments

While this study provides unique evidence linking victimization with physical health problems for Swedish young adults, a number of limitations should be taken into account when interpreting results. First, the sample consisted of young adults and thus results cannot be generalized to those in other age groups. Further, the measures used in this study relied on the self-report of participants who were asked to reflect on and to report about past victimization experiences. The construct validity and test–retest reliability of the JVQ has been established in US samples [30], but future studies would do well to further establish the validity and reliability of its use in combination with the additional items in new contexts. Although large national datasets in the United States have supported the utility of cumulative exposure scores in predicting distress [45], future studies might additionally examine the how other characteristics of abuse including duration, age of onset, and severity might affect physical health. To have additional information beyond self-report data, it would also be useful to consult police reports, public health records, or a parent's arrest for abuse. Thus, the study is limited in the extent to which participant's recall was accurate and complete. It should be noted, however, that when a measure of social desirability was used in previous, similar research, no bias was found in adult women's reports of various forms of past violence victimization [3]. The report of victimizations also might have been influenced by participants' mental health problems in adulthood. For example, childhood events may have been distorted when recalled through the lens of adult mental health problems, such as depression. In addition, the retrospective self-report questionnaire did not include questions about the socioeconomic status of the participants' family of origin. Such information may have enabled analyses of the role of socioeconomic adversity in the associations between childhood victimization and adult health. Even though the results from this and other studies consistently indicate that childhood victimization is related to physical health problems, the direction of associations is questionable given the crosssectional design, especially when only assessing childhood victimization and current young adulthood physical health, as in the present study. It may be that children with certain chronic conditions are at an increased risk for maltreatment [46] or identified associations are actually due to underlying factors that have not been controlled for in the present analysis. Nevertheless, the identified co-occurrence of childhood victimization and physical health problems is relevant when approaching young adults in need of support, notwithstanding the directions of causality. Given the cross-sectional nature of the present study, the use of a longitudinal design that identifies victimization experiences soon after they occur and then follows with physical and mental health assessments over time would be optimal. Further, studies that evaluate the biopsychosocial mechanisms that contribute to the association between victimizations during childhood and subsequent physical health

The data collection for the study was supported by funds from the National Board of Health and Welfare in Sweden (21-13275/2009).

Summary and conclusions

References [1] Å K Cater, AK Andershed, H Andershed, Youth victimization in Sweden: prevalence, characteristics and relation to mental health and behavioral problems in young adulthood, Child Abuse Negl. 38 (2014) 1290–1302, http://dx.doi.org/10.1016/j. chiabu.2014.03.002. [2] R Gilbert, C Spatz Widom, K Browne, D Fergusson, E Webb, S Janson, Burden and consequences of child maltreatment in high-income countries, Lancet 373 (2009) 68–81, http://dx.doi.org/10.1016/S0140-6736(08)61706-7. [3] SA Graham-Bermann, S Lynch, V Banyard, E Devoe, H Halabu, Community based intervention for children exposed to intimate partner violence: an efficacy trial, J. Consult. Clin. Psychol. 75 (2007) 199–209, http://dx.doi.org/10.1037/0022-006X. 75.2.199. [4] MR Holmes, Aggressive behavior of children exposed to intimate partner violence: an examination of maternal mental health, maternal warmth and child maltreatment, Child Abuse Negl. 37 (2013) 520–530, http://dx.doi.org/10.1016/j.chiabu. 2012.12.006. [5] JE Lansford, S Miller-Johnson, LJ Berlin, KA Dodge, JE Bates, G S Pettit, Early physical abuse and later violent delinquency: a prospective longitudinal study, Child Maltreat. 12 (2007) 233–245, http://dx.doi.org/10.1177/1077559507301841. [6] AA Levendosky, GA Bogat, C Martinez-Torteya, PTSD symptoms in young children exposed to intimate partner violence, Violence Against Women 19 (2013) 187–201, http://dx.doi.org/10.1177/1077801213476458. [7] KE Miller, A Rasmussen, War exposure, daily stressors, and mental health in conflict and post-conflict settings: bridging the divide between trauma-focused and psychosocial frameworks, Soc. Sci. Med. 70 (2010) 7–16, http://dx.doi.org/10.1016/j. socscimed.2009.09.029. [8] C Panter-Brick, A Goodman, W Tol, M Eggerman, Mental health and childhood adversities: a longitudinal study in Kabul, Afghanistan, J. Am. Acad. Child Adolesc. Psychiatry 50 (2011) 349–363, http://dx.doi.org/10.1016/j.jaac.2010.12.001. [9] L Sugaya, DS Hasin, M Olfson, KH Lin, BF Grant, C Blanco, Child physical abuse and adult mental health: a national study, J. Trauma. Stress. 25 (2012) 384–392, http://dx.doi.org/10.1002/jts.21719. [10] World Health Organization, WHO definition of healthRetrieved from: http://www. who.int/about/definition/en/print.html 2003. [11] SA Graham-Bermann, J Seng, Violence exposure and traumatic stress symptoms as additional predictors of health problems in high-risk children, J. Pediatr. 146 (2005) 349–354, http://dx.doi.org/10.1016/j.jpeds.2004.10.065. [12] K Lukaschek, J Baumert, J Kruse, R T Emeny, ME Lacruz, C Huth, KH Ladwig, Relationship between posttraumatic stress disorder and Type 2 Diabetes in a populationbased cross-sectional study with 2970 participants, J. Psychosom. Res. 74 (2013) 340–345, http://dx.doi.org/10.1016/j.jpsychores.2012.12.011. [13] RD Hays, GN Marshall, EYI Wang, CD Sherbourne, Four-year cross-lagged associations between physical and mental health in the Medical Outcomes Study, J. Consult. Clin. Psychol. 62 (1994) 441. [14] K Kroenke, et al., Reciprocal relationship between pain and depression: a 12-month longitudinal analysis in primary care, J. Pain 12 (2011) 964–973. [15] G E Miller, E Chen, K J Parker, Psychological stress in childhood and susceptibility to the chronic diseases of aging: moving toward a model of behavioral and biological mechanisms, Psychol. Bull. 137 (2011) 959. [16] E B Raposa, J E Bower, C L Hammen, J M Najman, P A Brennan, A developmental pathway from early life stress to inflammation the role of negative health behaviors, Psychol. Sci. 25 (2014) 1268–1274. [17] MA Gupta, Review of somatic symptoms in post-traumatic stress disorder, Int. Rev. Psychiatry 25 (2013) 86–99, http://dx.doi.org/10.3109/09540261.2012.736367. [18] D Cicchetti, F A Rogosch, The impact of child maltreatment and psychopathology on neuroendocrine functioning, Dev. Psychopathol. 13 (2001) 783–804. [19] K Bevans, A Cerbone, S Overstreet, Relations between recurrent trauma exposure and recent life stress and salivary cortisol among children, Dev. Psychopathol. 20 (2008) 257–272.

Please cite this article as: L.E. Miller-Graff, et al., Victimization in childhood: General and specific associations with physical health problems in young adulthood, J Psychosom Res (2015), http://dx.doi.org/10.1016/j.jpsychores.2015.07.001

L.E. Miller-Graff et al. / Journal of Psychosomatic Research xxx (2015) xxx–xxx [20] E Sjögren, P Leanderson, M Kristenson, Diurnal saliva cortisol levels and relations to psychosocial factors in a population sample of middle-aged Swedish men and women, Int. J. Behav. Med. 13 (2006) 193–200. [21] KR Kuhlman, EG Geiss, I Vargas, NL Lopez-Duran, Differential associations between childhood trauma subtypes and adolescent HPA-axis functioning, Psychoneuroendocrinology 54 (2015) 103–114. [22] J McCall-Hosenfeld, M Winter, T Heeren, JM Liebschutz, The association of interpersonal trauma with somatic symptom severity in a primary care population with chronic pain: exploring the role of gender and the mental health sequelae of trauma, J. Psychosom. Res. 77 (2014) 196–204, http://dx.doi.org/10.1016/j.jpsychores.2014.07.011. [23] KR Kuhlman, KH Howell, SA Graham-Bermann, Physical health in preschool children exposed to intimate partner violence, J. Fam. Violence 27 (2012) 499–510, http://dx.doi.org/10.1007/s10896-012-9444-2. [24] EM Annerbäck, L Sahlqvist, G Wingren, A cross-sectional study of victimisation of bullying among schoolchildren in Sweden: background factors and self-reported health complaints, Scand. J. Public Health 42 (2014) 270–277, http://dx.doi.org/10. 1177/1403494813514142. [25] TO Afifi, N Mota, HL MacMillan, J Sareen, Harsh physical punishment in childhood and adult physical health, Pediatrics 132 (2013) e333–e340, http://dx.doi.org/10. 1542/peds.2012-4021. [26] A Danese, TE Moffitt, H Harrington, BJ Milne, G Polanczyk, CM Pariante, A Caspi, Adverse childhood experiences and adult risk factors for age-related disease: depression, inflammation, and clustering of metabolic risk markers, Arch. Pediatr. Adolesc. Med. 163 (2009) 1135–1143. [27] V J Felitti, R F Anda, D Nordenberg, D F Williamson, A M Spitz, V Edwards, J S Marks, 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. 14 (1998) 245–258. [28] JM Hussey, JJ Chang, JB Kotch, Child maltreatment in the United States: prevalence, risk factors, and adolescent health consequences, Pediatrics 118 (2006) 933–942, http://dx.doi.org/10.1542/peds.2005-2452. [29] KAB Gawronski, E S Kim, LE Miller, Potentially traumatic events and serious life stressors are prospectively associated with frequency of doctor visits and overnight hospital visits, J. Psychosom. Res. 77 (2014) 90–96, http://dx.doi.org/10.1016/j. jpsychores.2014.05.009. [30] D Finkelhor, R Ormrod, H Turner, SL Hamby, The victimization of children and youth: a comprehensive, national survey, Child Maltreat. 10 (2005) 5–25, http:// dx.doi.org/10.1177/1077559504271287. [31] SL Hamby, D Finkelhor, RK Ormrod, HA Turner, The Juvenile Victimization Questionnaire (JVQ): Administration and Scoring Manual, Crimes Against Children Research Center, Durham, NH, 2004. [32] C May-Chahal, P Cawson, Measuring child maltreatment in the United Kingdom: a study of the prevalence of child abuse and neglect, Child Abuse Negl. 29 (2005) 969–984, http://dx.doi.org/10.1016/j.chiabu.2004.05.009.

7

[33] JP Allen, RZ Litten, JB Fertig, T Babor, A review of research on the alcohol use disorders identification test (AUDIT), Alcohol. Clin. Exp. Res. 21 (1997) 613–619, http://dx.doi.org/10.1111/j.1530-0277.1997.tb03811.x. [34] DS Weiss, The Impact of Event Scale: Revised, in: P J P Wilson, P C S Tang (Eds.), Cross-Cultural Assessment of Psychological Trauma and PTSD, Springer US 2007, pp. 219–238. [35] AS Zigmond, RP Snaith, The hospital anxiety and depression scale, Acta Psychiatr. Scand. 67 (1983) 361–370. [36] K Bader, V Schäfer, M Schenkel, L Nissen, J Schwander, Adverse childhood experiences associated with sleep in primary insomnia, J. Sleep Res. 16 (2007) 285–296, http://dx.doi.org/10.1111/j.1365-2869.2007.00608.x. [37] RD Goodwin, CW Hoven, R Murison, M Hotopf, Association between childhood physical abuse and gastrointestinal disorders and migraine in adulthood, Am. J. Public Health 93 (2003) 1065–1067, http://dx.doi.org/10.2105/AJPH.93.7.1065. [38] CA Czeisler, Duration, timing and quality of sleep are each vital for health, performance and safety, Sleep Health 1 (2015) 5–8. [39] SR Dube, VJ Felitti, M Dong, WH Giles, RF Anda, The impact of adverse childhood experiences on health problems: evidence from four birth cohorts dating back to 1900, Prev. Med. 37 (2003) 268–277, http://dx.doi.org/10.1016/S0091-7435(03)00123-3. [40] M Felitti, J Vincent, M Anda, F Robert, M Nordenberg, 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. 14 (1998) 245–258. [41] CL Ogden, MD Carroll, BK Kit, KM Flegal, Prevalence of childhood and adult obesity in the United States, 2011–2012, JAMA 311 (2014) 806–814, http://dx.doi.org/10. 1001/jama.2014.732. [42] A Molarius, et al., Social inequalities in self-reported refraining from health care due to financial reasons in Sweden: health care on equal terms? BMC Health Serv. Res. 14 (2014) 605. [43] Organisation for Economic Co-operation and Development OECD, Health Statistics 2014Retrieved from: http://www.oecd.org/els/health-systems/health-data.htm 2014. [44] BB Strickland, JR Jones, RM Ghandour, MD Kogan, PW Newacheck, The medical home: health care access and impact for children and youth in the United States, Pediatrics 127 (2011) 604–611. [45] D Finkelhor, RK Ormrod, HA Turner, Poly-victimization: a neglected component in child victimization, Child Abuse Negl. 31 (2007) 7–26. [46] S Svensson, CG Bornehag, S Janson, Chronic conditions in children increase the risk for physical abuse — but vary with socio-economic circumstances, Acta Paediatr. 100 (2011) 407–412, http://dx.doi.org/10.1111/j.1651-2227.2010.02029.x.

Please cite this article as: L.E. Miller-Graff, et al., Victimization in childhood: General and specific associations with physical health problems in young adulthood, J Psychosom Res (2015), http://dx.doi.org/10.1016/j.jpsychores.2015.07.001