General Hospital Psychiatry 64 (2020) 80–86

Contents lists available at ScienceDirect

General Hospital Psychiatry

journal homepage: www.elsevier.com/locate/genhospsych

Prevalence of adverse childhood experiences (ACEs) and associated physical T

and mental health problems amongst hospital patients: Results from a cross-
sectional study
⁎
David Riedla, Astrid Lampea, , Silvia Exenbergera, Tobias Nolteb,c, Iris Trawögera, Thomas Becka
a
University Clinic of Medical Psychology, Medical University of Innsbruck, Innsbruck, Austria
b
Wellcome Department of Imaging Neuroscience, University College London, London, United Kingdom
c
Anna Freud National Centre for Children and Families, London, United Kingdom

A R T I C LE I N FO A B S T R A C T

Keywords: Background: Adverse childhood experiences (ACEs) can have severe detrimental effects on physical and mental
Childhood maltreatment health. This study aimed to present prevalence rates of ACEs amongst a mixed sample of hospital patients.
Adverse childhood experiences Methods: In- and outpatients at seven departments of the University Hospital of Innsbruck (Austria) participated
Abuse in the study. They completed questionnaires regarding retrospective assessments of ACEs, physical and mental
Prevalence
health and experience of domestic violence. The impact of ACEs on patients' health was evaluated by calculation
Health impairment
of odds ratios (OR) in binary logistic regressions.
Results: A total of n = 2392 (74.3% of all approached patients) were included in the analyses. The results
showed that 36.1% of them reported at least one form of ACEs, and 6.3% were polyvictimized (i.e. they reported
≥4 forms of ACEs). Most frequent forms of ACE were emotional abuse (18.3%), peer abuse (14.2%), and neglect
(12.3%). ACEs were significantly associated with increased ORs for various physical diseases, mental health
problems and domestic violence.
Conclusion: Retrospectively assessed ACEs are highly prevalent amongst hospital patients and exposure to high
numbers of ACEs is associated with decreased physical and mental health. The identification of patients with
symptoms following ACEs and referral to appropriate treatment is a crucial challenge for health-care profes-
sionals.

1. Background risk for depression, anxiety and suicide attempts, but also physical
health impairment and ultimately premature mortality [2,4,12]. Stu-
Adverse childhood experiences (ACEs) are a worldwide problem dies have reported a significantly higher hazard ratio for common
with serious life-long health consequences [1–4]. ACEs include various physical diseases and mortality for polyvictimized individuals (i.e. > 4
types of physical and/or emotional ill-treatment, sexual abuse, neglect, ACEs), thus indicating a dose-response relationship of ACEs and phy-
negligence and commercial or other exploitation that occurs to children sical health [4,12]. Also, research has indicated potentially sensitive
under 18 years of age [3]. Other forms of ACEs that are less in- periods in infant brain development, associated with major depression
vestigated, include peer abuse or the witnessing of interparental vio- and other forms of psychological disorders in adult life [13–15]. The
lence and violence against siblings [4]. evidence on sensitive periods is still scarce and does not allow final
ACEs may impair the development of the nervous, endocrine and conclusions yet.
immune systems, leading to long-lasting alterations in those systems A recent meta-analysis [4] has found substantial differences in ACE
[5–7]. ACEs were also associated with patterns of dysfunctional health- prevalence rates depending on the population and country: 43–74% of
harming behaviors in adult life, including substance abuse, sexual risk the participants in the general population of western countries reported
taking and physical inactivity, and an increased risk to live in a violent at least one form of ACE and 7–21% reported at least 4 different types
relationship [4,8]. The cumulative effects of altered biophysiological of ACEs (polytraumatized). In retrospective epidemiological studies in
processes and maladaptive coping strategies harm later cognitive, social the general population of Germany, 31–68% of respondents reported at
and emotional functioning [9–11], ultimately leading to an increased least one form of ACE and 9% four or more types of ACEs [16,17].

⁎
Corresponding author at: University Clinic of Medical Psychology, Medical University of Innsbruck, Speckbacherstraße 23/4, 6020 Innsbruck, Austria.
E-mail address: [email protected] (A. Lampe).

https://doi.org/10.1016/j.genhosppsych.2020.03.005
Received 10 October 2019; Received in revised form 11 March 2020; Accepted 12 March 2020
0163-8343/ © 2020 Elsevier Inc. All rights reserved.
D. Riedl, et al. General Hospital Psychiatry 64 (2020) 80–86

Patients in primary care facilities reported slightly higher ACE pre- psychological distress. In this study, gender-specific cut-offs for de-
valence rates with 58–88% of patients having at least one ACE and pression (men: BSI-depr ≥4.29; women: BSI-depr ≥4.67), anxiety
18–38% with four or more ACEs [18,19]. In one further study, 48% of (men: BSI-anx ≥ 3.65; women: BSI-anx ≥ 4.28) and somatization
non-psychiatric hospital patients reported at least one form of ACEs [20]. (men: BSI-som ≥ 3.12; women: BSI-som ≥3.40) were applied to define
Thus, research indicates that ACEs might be more prevalent in patients clinically relevant symptomatology, as recommended in the BSI-18
than in the general population. However, only a very limited number of manual [25]. Good reliability and validity for the subscales and total
studies have been conducted in patients and previous studies included score have been reported [26,27].
either low numbers of participants or were conducted in high-risk po- Domestic violence was assessed using an adapted German version of
pulations. Also, in those studies, witnessing of intrafamilial violence or the Hurt-Insult-Threaten-Scream (HITS) scale [28], which consists of four
peer abuse have not been considered so far. Moreover, research has items asking whether the participant was ever (1) screamed at or in-
indicated that the patients' current age might influence the retro- sulted, (2) threatened, (3) cursed at, talked down to or (4) physically
spectively reported ACEs [17,21], which is specifically relevant as hurt by a partner. A weighted cut-off was chosen: if patients reported
hospital patients tend to be of higher age. physical violence (item 4) at least “rarely” or psychological violence
The aim of this study was to provide data on the prevalence of ex- (items 1–3) at least “sometimes”, we considered them domestic vio-
posure to retrospectively assessed ACEs in a sample of German-speaking lence cases.
hospital patients in Austria, including less investigated phenomena such
as peer abuse or the witnessing of intrafamilial violence and data at 2.3. Statistical procedures
which age the abuse took place. Secondly, the association of ACEs with
the patients' current physical and mental health was investigated. We Analyses were limited to individuals with complete data on ACEs,
specifically hypothesized, that (I) higher prevalence rates of ACEs will age and sex. Group differences were analyzed using chi-square tests, t-
be found amongst hospital patients than in the general population, (II) tests, analyses of variance (ANOVA) and Pearson and Spearman cor-
reported ACEs will differ amongst age groups, (III) peer abuse and relation coefficients. To investigate the distribution of ACEs amongst
witnessing of intrafamilial violence are significantly associated with age-cohorts, patients were split into four age groups: 18–30, 30–50,
patients physical and mental health, and (IV) a high ACE-score (4+) 50–70 and ≥ 70 years. Since previous research has indicated that
will be positively correlated with higher odds for physical and mental higher levels of current psychological distress may bias the recall of
health problems and domestic violence in adulthood. ACEs [29], Bonferroni-corrected analyses of variance (ANOVAs) were
calculated to determine whether patients in the different age groups
2. Methods differed in regard to anxiety or depression.
The association between the different ACE types and patients'
2.1. Sample and procedure physical health (defined as the sum of self-reported physical diseases),
mental health (depression, anxiety and somatization) and experiences
In this cross-sectional study, data was collected from inpatients and of domestic violence (HITS total score) were investigated by calculation
outpatients of seven departments at the University Hospital of of linear regression models. The seven ACE clusters were entered as
Innsbruck, Austria, (Otolaryngology, Trauma Surgery, Neurosurgery, independent and the health-outcomes and dependent variables. All
Neurology, Gynecology, Internal Medicine, Radiology) between models were adjusted for patients' age and sex.
October 2015 and March 2017. Data collection took place in waiting To investigate the dose-response relationship between ACEs and
areas at the hospital. Patients completed a paper-and-pencil ques- health-outcomes proposed in the literature, binary logistic regression
tionnaire in private – wherever possible, in a separate part of the analyses were performed for each health-outcome (self-reported pre-
waiting area. Research assistants were available to patients for any sence of disease: yes vs. no). Previous research has indicated a strongly
questions and to assure privacy. All patients received a contact address increased risk for physical and mental health problems for patients with
for professional psychological support free of charge. The study design four or more ACEs (i.e. polyvictimized patients) [4]. Thus, consistent
was approved by the ethics committee of the Medical University of with previous study methodology, the number of ACE trauma clusters
Innsbruck (AN2015–0175 351/4.18). each individual reported was counted and patients were then divided
into three groups: 0 ACEs, 1–3 ACEs, and 4+ ACEs (polyvictimization).
2.2. Measures The ACE groups were added as one categorical variable with 0 ACEs as
the reference category. All models were adjusted for patients age and
ACEs were assessed with the German version of the Maltreatment sex. Odds ratios (OR) are presented with 95% confidence intervals. P-
and Abuse Chronology of Exposure Scale (MACE) [KERF; 22, 23]. It values < .05 were considered statistically significant. All statistical
consists of 75 items that retrospectively gauge the severity of exposure analyses were performed with IBM SPSS (v22.0).
to different types of maltreatment up to age 18. For each year sepa-
rately, participants are asked to endorse whether maltreatment oc- 3. Results
curred during that particular year of their life or not. Thus, onset and
cumulative exposure are measured. The items were grouped into seven 3.1. Sociodemographic data
ACE clusters: emotional abuse (verbal or non-verbal), physical abuse,
neglect (emotional and physical), witnessing violence (between parents Of all the patients approached (n = 3220), 84.1% (n = 2708) were
or towards siblings), physical peer abuse, emotional peer abuse, and willing to participate in the study. Individuals with missing data re-
sexual abuse. The MACE has good test-retest reliability and validity lating to ACEs, age or sex (n = 316) were excluded from the analyses,
[22,23]. resulting in a final sample of n = 2392 patients. Patients with poly-
Health: We used a comprehensive self-report health-checklist of victimization were significantly more often single and were living alone
diseases derived from the German Pain Questionnaire [24] to retro- more often, had a lower level of education, were living in urban en-
spectively assess lifetime prevalence of physical diseases. The health- vironments and were of female gender. For details on socio-
checklist covered eleven major physical disease clusters. Apart from demographic data, see Table 1.
presence of disease, patients could also rate associated impairment.
Psychological distress was assessed with the Brief Symptom 3.2. Clinical data
Inventory (BSI-18) [25]. Three subscale scores (depression, anxiety,
somatization) and a global score can be computed to measure global Most frequently reported physical diseases were chronic pain

81
D. Riedl, et al. General Hospital Psychiatry 64 (2020) 80–86

Table 1
Sociodemographic data for patients with no ACEs, 1–3 ACEs and 4 or more ACEs.
0 ACEs 1–3 ACEs 4+ ACEs
(n = 1528) (n = 713) (n = 151)

mean/n (SD/%) mean/n (SD/%) mean/n (SD/%) p-value

Age 44.5 (16.3) 41.1 (15.7) 38.6 (15.1) < 0.001

18–30 years 396 (25.9%) 246 (34.5%) 56 (37.1%)
30–50 years 556 (36.4%) 249 (34.9%) 58 (36.1%)
50–70 years 471 (30.8%) 185 (25.9%) 33 (21.9%)
≥70 years 105 (6.9%) 33 (4.6%) 4 (2.6%)
Sex < 0.001
Male 745 (48.8%) 303 (42.5%) 52 (34.4%)
Female 782 (51.2%) 410 (57.5%) 99 (65.6%)
Relationship status 0.003
Married/long-term relationship 1039 (68.0%) 438 (61.4%) 85 (56.3%)
Single 330 (21.6%) 193 (27.1%) 48 (31.8%)
Divorced 90 (5.9%) 55 (7.7%) 14 (9.3%)
Widowed 22 (1.4%) 9 (1.3%) 1 (0.7%)
Missing data 47 (3.1%) 18 (2.5%) 3 (2.0%)
Level of education < 0.001
School not finished 11 (0.7%) 19 (2.7%) 8 (5.3%)
Compulsory school 116 (7.6%) 54 (7.6%) 20 (13.2%)
Compulsory school and apprenticeship 558 (36.5%) 239 (33.5%) 45 (29.8%)
Higher education 426 (27.9%) 217 (30.4%) 43 (28.5%)
University degree 281 (18.4%) 144 (20.2%) 25 (16.6%)
Missing data 136 (8.9%) 40 (5.6%) 10 (6.6%)
Living situation < 0.001
Living alone 269 (17.6%) 153 (21.5%) 42 (27.8%)
Living with partner/family 932 (61.0%) 387 (54.3%) 72 (47.7%)
Living with family of origin 110 (7.2%) 53 (7.4%) 6 (4.0%)
Living in shared apartment 100 (6.5%) 68 (9.5%) 19 (12.6%)
Missing data 117 (7.7%) 52 (7.3%) 12 (8.0%)
Parenthood 590 (38.6%) 359 (50.4%) 53 (35.1%) 0.38
Missing data 214 (14.0%) 99 (13.9%) 29 (19.2%)
Living environment < 0.001
Rural 857 (56.1%) 353 (49.5%) 65 (43.0%)
Urban 570 (37.3%) 325 (45.6%) 79 (52.3%)
Missing data 101 (6.6%) 35 (4.9%) 7 (4.6%)
Disability 76 (5.0%) 52 (7.3%) 11 (7.3%) 0.09
Missing data 61 (4.0%) 16 (2.2%) 3 (2.0%)
Frequency hospital usage (inpatient) in the last 5 years < 0.001
0–3× 790 (52.6%) 353 (50.1%) 50 (33.3%)
4–7× 335 (22.3%) 159 (22.6%) 47 (31.3%)
> 8× 378 (25.1%) 193 (27.4%) 53 (35.3%)
Missing data 25 (1.6%) 8 (1.1%) 1 (0.7%)
Frequency hospital usage (outpatient) in the last 5 years
0–3× 1219 (84.0%) 565 (83.1%) 110 (78.6%) 0.53
4–7× 129 (8.9%) 62 (9.1%) 18 (12.9%)
> 8× 104 (7.2%) 53 (7.8%) 12 (8.6%)
Missing data 76 (5.0%) 33 (4.6%) 11 (7.3%)
Medical service1 0.59
Emergency ward (n = 443) 276 (62.3%) 131 (29.6%) 36 (8.1%)
Gynecology (n = 176) 109 (61.9%) 51 (29.0%) 16 (9.1%)
Neuro-Surgery (n = 469) 310 (66.1%) 134 (28.6%) 25 (5.3%)
Otolaryngology (n = 1035) 652 (63.0%) 322 (31.1%) 61 (5.9%)
Neurology (n = 210) 141 (67.1%) 57 (27.1%) 12 (5.7%)
Internal Medicine (n = 16) 11 (68.8%) 5 (31.3%) 0 (0.0%)
Radiology (n = 13) 8 (61.5%) 5 (38.5%) 0 (0.0%)
Missing data 21 (1.4%) 8 (1.1%) 1 (0.7%)
Setting1
Outpatient (n = 2055) 1291 (62.8%) 629 (30.6%) 135 (6.6%) 0.054
Inpatient (n = 309) 216 (69.9%) 77 (24.9%) 16 (5.2%)
Missing data 21 (1.4%) 7 (1.0%) – –

ACEs = adverse childhood experiences; SD = standard deviation; distribution per row.

(n = 744; 31.1%), followed by respiratory (n = 383; 16.0%), muscu- domestic violence in at least one point of their adult life.
loskeletal (n = 376; 15.7%), neurological (n = 365; 15.3%), metabolic
(n = 335; 14.0%), and gastrointestinal diseases (n = 320; 13.4%), 3.3. Prevalence of ACEs
cancer (n = 317; 13.3%), cardiological (n = 308; 12.9%), skin
(n = 288; 12.0%), urogenital (n = 247; 10.3%), and gynecological 3.3.1. Type of ACEs
diseases (n = 221; 9.2%). The most frequent psychological mental Overall, 63.9% reported no ACEs at all, 29.8% reported 1–3 ACEs,
health problem was somatization (n = 706; 32.2%), followed by an- and the remaining 6.3% reported ≥4 ACEs. The most frequently re-
xiety (n = 527; 21.4%), and depression (n = 352; 16.1%). ported ACEs were emotional abuse (18.3%) and peer abuse (14.2%),
Additionally, 16.6% (n = 369) of the patients reported exposure to followed by emotional neglect (11.3%), witnessing intrafamilial

82
D. Riedl, et al. General Hospital Psychiatry 64 (2020) 80–86

18

16

Mean number of ACEs per year

14

12

10

0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Age
1-3 ACEs 4+ ACEs

Fig. 1. Abuse trajectories of patients reporting 1–3 and 4+ ACEs: mean number and standard deviation (SD) of positive MACE categories for each age group.

violence (9.7%), physical violence (7.1%), sexual violence (4.9%), and domestic violence. For details, see Table 2.
physical neglect (3.2%). Fig. 1 shows the mean number and standard Regarding the dose-response relationship of ACEs and health out-
deviation of ACEs reported by the patients for each life year in their comes, the highest OR for all assessed physical disorders (except for
childhood and adolescence. Polyvictimized patients reported a steady cancer and neurological disorders) were found for polyvictimized pa-
increase of ACEs, with a steep increase from the age of 5 years and a tients. Highest odds amongst those patients were found for chronic pain
peak between 10 and 14 years. After the age of 14 the mean number of (OR = 3.8), urogenital disorders (OR = 2.8), gastrointestinal and
ACEs decreased, while still remaining more than twice as high as for musculoskeletal disorders (both OR = 2.30). The same pattern was
patients with 1–3 ACEs. Patients with no ACEs are not displayed in the found for mental health issues: polyvictimized patients had 8.9-times
figure. increased odds of depression and 6.4-times odds of anxiety when
compared to patients without ACEs. Also, in our sample polyvictimized
3.3.2. ACEs across age-cohorts patients had 10.4-times increased odds of exposure to domestic violence
Overall, younger patients reported a higher mean number of ACEs in their adult life. For details see Table 3.
than older patients (r = − 0.09, p < .001). Thus, when comparing age
groups, lower rates of polyvictimized patients were found with in- 4. Discussion
creasing age (18–30 years: 8.0%; 30–50 years: 6.7%; 50–70 years:
4.8%; ≥ 70 years: 2.8%; χ2 = 9.33, p = .025). Patients in different age The aim of this cross-sectional single-center study was to present
groups reported significantly different values for emotional abuse data on the prevalence of retrospectively assessed levels of ACEs in a
(χ2 = 14.60, p = .002), physical abuse (χ2 = 9.14, p = .028) and broad sample of mixed hospital patients. Additionally, the association
neglect (χ2 = 7.89, p = .048), witnessing violence (χ2 = 8.72, of ACEs with physical and mental health in adulthood was investigated.
p = .033) and peer-abuse (χ2 = 90.29, p < .001), while no overall In our sample, approximately 36% of the patients reported some
significant differences were found for emotional neglect and sexual form of relevant abuse during childhood and 6% of all patients reported
abuse (both p < .05). While the frequency of reported emotional and polyvictimization (i.e. ≥4 forms of ACEs). Compared to two previous
peer abuse declined with increasing age, higher rates of physical ne- studies conducted in the health-care setting, patients reported generally
glect were reported by older patients. For physical abuse, there was a fewer ACEs and fewer patients met criteria for polyvictimization
clear ascending trend, apart from the oldest age group who reported the [18,19]. Yet, the prevalence rates of ACEs in our sample were com-
lowest values. In the case of intrafamilial violence, similar values were parable to the general population of Germany [16,17]. Thus, ACEs were
found for patients between 18 and 70 years, while patients older than not as prevalent as hypothesized based on previous research. This dis-
70 years reported a much lower prevalence (for details see Fig. 2). crepancy may be due to the higher age range in our sample, since older
In order to control for a reporting bias of ACEs, the association of people reported less ACEs in general. Also, ACE prevalence rates in
psychological distress amongst the age groups were investigated: pa- hospital patients may have been previously overestimated since both
tients below 30 years reported significantly higher values of depression previous studies were conducted in either rather small or high-risk
(F = 6.31, p < .001) and anxiety (F = 9.1, p < .001) than patients samples (i.e. low-income populations) [18,19]. Most prevalent forms of
above 50 years. ACEs in our sample were emotional abuse, peer abuse, and emotional
neglect. While the prevalence of most of the assessed abuse forms was
3.4. Physical and mental health comparable to previously reported rates in the general population of
Germany, the relative numbers in our sample were higher for emotional
A higher number of ACEs was significantly associated with more abuse (18% vs. 5–7%) and lower for physical neglect (3.2% vs.
physical disorders (r = 0.16, p < .001), depression (r = 0.32, 22–29%) [16,17,30].
p < .001), anxiety (r = 0.30, p < .001), somatization (r = 0.18, Previous research has shown an association of witnessing in-
p < .001) and experiences of domestic violence (r = 0.42, p < .001). trafamilial violence and peer abuse to a wide range of social, emotional,
Amongst the assessed forms of ACEs, emotional abuse and peer abuse behavioral and academic problems [31–33]. As hypothesized, in our
were most consistently associated with physical and mental health. sample both peer-abuse and being a witness of intrafamilial violence
Being a witness to intrafamilial violence was associated with a higher were independently associated with physical and mental health im-
number of physical diseases, as well as higher depression scores and pairment in adulthood, even when other ACEs, such as sexual or

83
D. Riedl, et al. General Hospital Psychiatry 64 (2020) 80–86

Self-reported abuse across age-groups

30%

24.6%
25%

Percentage of paents above cut-off

21.6%

19.1%
20%
15.6%
15% 13.8%
11.9% 13.3%
11.6% 11.3%
10.6%
9.9% 9.9% 10.3%
9.1%
10% 7.8%
7.5%
5.9% 5.5% 5.0%
5.2% 4.9%
4.1% 4.4% 4.6% 4.3%
5% 3.0% 3.0%
2.1%

0%
Emoonal Physical Emoonal Physical Witnessing Peer Sexual
abuse abuse neglect neglect violence abuse abuse

18-30 years 30-50 years 50-70 years > 70 years

(29.2%) (36.1%) (28.8%) (5.9%)

Fig. 2. Self-reported abuse above cut-off for all assessed abuse types across four age groups.

Table 2
Association of ACE types with physical and mental health in adulthood.
Emotional Abuse Physical Violence Neglect Witnessing Violence Peer-Abuse Sexual Abuse

β p β p β p β p β p β p

Physical diseasesa 0.09 < 0.001 0.13 0.90 0.05 0.029 0.06 0.021 0.05 0.023 0.03 0.19
Depression 0.16 < 0.001 0.07 0.005 0.03 0.14 0.08 0.001 0.12 0.008 0.06 0.008
Anxiety 0.18 < 0.001 0.05 0.041 0.05 0.033 0.03 0.15 0.10 < 0.001 0.06 0.002
Somatization 0.11 < 0.001 0.04 0.08 0.01 0.51 0.02 0.42 0.07 0.001 0.01 0.53
Domestic violence 0.26 < 0.001 0.02 0.51 0.10 < 0.001 0.12 < 0.001 0.10 < 0.001 0.04 0.08

a
Sum of self-reported physical diseases; β = standardized regression coefficient; p = p-value (level of significance).

Table 3
Odds ratio for physical and mental health problems for patients with increasing numbers of ACEs.
1–3 ACEs 4+ ACEs
(n = 713) (n = 151)

pa AOR CI 95% pb AOR CI 95% pb

Physical health
Chronic pain < 0.001 1.67 1.37–2.05 < 0.001 3.76 2.64–5.35 < 0.001
Gastrointestinal disorders < 0.001 1.41 1.09–1.83 0.010 2.30 1.51–3.49 < 0.001
Metabolic diseases 0.001 1.08 0.82–1.40 0.59 2.26 1.49–3.43 < 0.001
Musculoskeletal disorders < 0.001 1.45 1.13–1.86 0.004 2.30 1.51–3.51 < 0.001
Gynecological disorders 1 0.020 1.54 1.08–2.19 0.018 1.83 1.03–3.25 0.039
Respiratory diseases 0.003 1.38 1.08–1.75 0.009 1.79 1.18–2.72 0.006
Urogenital disorders < 0.001 1.35 1.00–1.82 0.049 2.83 1.80–4.44 < 0.001
Skin diseases 0.12 1.11 0.84–1.46 0.45 1.61 1.02–2.54 0.041
Cardiovascular diseases 0.032 1.05 0.79–1.38 0.76 1.86 1.17–2.97 0.009
Cancer 0.55 0.89 0.67–1.17 0.40 1.16 0.70–1.92 0.57
Neurological disorders 0.039 1.34 1.05–1.71 0.020 1.41 0.89–2.21 0.14

Mental health
Depression < 0.001 2.57 1.99–3.32 < 0.001 8.93 6.15–12.95 < 0.001
Anxiety < 0.001 2.29 1.85–2.84 < 0.001 6.38 4.46–9.13 < 0.001
Somatization < 0.001 1.71 1.41–2.08 < 0.001 2.46 1.74–3.49 < 0.001

Domestic violence
DV < 0.001 3.63 2.81–4.69 < 0.001 10.36 7.02–15.28 < 0.001

a
ACEs = adverse childhood experiences; AOR = adjusted odds ratio, corrected for sex and age; p refers to the overall significance of association between the outcome
measure and ACE counts; pb refers to the significance of association between the outcome measure and individual ACE categories, with 0 ACEs as reference category;
1
adjusted odds rations were calculated in the female sample only.

84
D. Riedl, et al. General Hospital Psychiatry 64 (2020) 80–86

physical abuse were taken into account. Thus, our results underscore associated health behavior. However, previous research has shown that
the importance of considering these less investigated forms of ACEs in especially lifestyle associated risk factors such as smoking, obesity,
future studies. physical activity, alcohol use are directly or indirectly influenced by
Older patients reported fewer ACEs and the rates of reported poly- ACEs [4]. Thus, the presented results should be interpreted with cau-
victimization decreased with increasing age. In our sample we also tion as the design of our study does not allow for causal conclusions.
found significant age-specific differences for some of the assessed abuse The assessment of physical and mental health was patient-rated,
types. There was a pronounced decrease of self-reported emotional and which is a subjective measure. However, as this approach assessed
peer abuse with increasing age. Regarding physical violence, it is pos- subjective impairment due to disease, it provided a picture of the health
sible that a shift in the type of intrafamilial abuse took place: as num- problems that are most important to the individual patient. Prevalence
bers of physical abuse steadily decrease since physical punishment is rates for ACEs in our sample were partially smaller than in previous
forbidden by law while a higher number of patients experience emo- epidemiological studies. A possible reason might be that in our study
tional violence within their families. Interestingly, patients of at least psychiatric clinics were not sampled, which may have led to higher
70 years (i.e. born in or before 1946) reported the lowest numbers of overall ACE prevalence rates. Also, data was collected from patients in
physical abuse and being a witness to intrafamilial violence. One ex- the waiting areas of the participating departments, usually with other
planation might be that amongst that cohort a higher level of fathers patients present. While several strategies were applied to guarantee best
were absent from their family during World-War II and had been possible privacy and confidentiality, we cannot rule out that, due to
prisoners of war in the following years. Furthermore, as indicated by concerns about privacy, some patients may have not participated or
Witt, Brown, Plener, Brahler and Fegert [17] due to the hardship and downplayed their experiences. In our study patients from internal
socio-economic conditions of this time the importance of and awareness medicine clinics were relatively underrepresented, which was caused
for physical violence might have been lower. For the next generation by logistical reasons (especially shorter waiting periods, less patients
(i.e. those born between 1946 and 1966) reports of physical abuse available). Data on ACEs has been collected retrospectively. A recent
markedly rose, which may be a result of war-traumatization and a meta-analysis has found only limited agreement between retrospective
general acceptance of physical violence as a means of education. and prospective reports of ACEs [39]. While the authors of the meta-
The age-specific differences may also be influenced by different analysis argue, that retrospective assessment may be more sensitive for
types of recall-biases, such as higher levels of mild cognitive impair- ACEs than prospective measures, retrospective assessments are always
ment in older patients, who may have more difficulties to correctly prone to memory bias.
retrieve episodic memory from their distant past as suggested by
Hanninen and Soininen [34]. Maercker [35] described a “selective” 6. Conclusion
trauma memory in older traumatized patients, who often exclude bio-
graphic episodes in which traumata had occurred. Norby [36] stated A considerable number of hospital patients reported ACEs, which
that, in comparison with younger adults, elderly people are more likely were associated with severe detrimental effects on physical and mental
to remember positive events over negative ones. Similarly, Hardt and health. Our results strongly suggest that early identification and ap-
Rutter [37] concluded in their review that older people with good propriate psychosocial support of patients with symptoms following
personal functioning tend to forget negative events concerning their early traumatization, such as anxiety, depression and living in a violent
parents. A kind of “forgiving and forgetting” seems to be present in relationship, is a crucial task for health-care professionals. Previously
older patients especially with good levels of personal functioning and neglected aspects of ACEs, such as peer abuse, are highly prevalent and
this may contribute to more positive feelings. Also, it is not clear may strongly contribute to the detrimental effect of polyvictimization
whether there is an increased recall bias for specific ACEs, i.e. peer on physical and mental health. Women were affected more severely by
abuse not being retained or remembered as long as sexual or physical ACEs and we found several age-related differences. Thus, when con-
abuse. To our knowledge, this aspect of ACE recall bias has not been ducting epidemiological research aiming at the retrospective assess-
investigated yet. Another explanation might be, that we found higher ment of ACEs, the participants' sex and age should be considered.
levels of anxiety and distress in younger patients, which have been
associated with an increased likelihood to report ACEs [29]. Author statement
In our sample, polyvictimization was associated with higher odds
for several physical disorders: in line with previous research [1,2,4,12], David Riedl: Conceptualization, Methodology, Formal analysis,
we found an higher prevalence for disorders such as chronic pain, ur- Writing (original draft), Data curation; Astrid Lampe:
ogenital, gastrointestinal or musculoskeletal disorders, but also clini- Conceptualization, Methodology, Supervision, Writing (review &
cally relevant self-reported symptoms of depression or anxiety. Poly- editing), Project administration; Silvia Exenberger: Writing (review &
victimized patients had been treated more frequently as inpatients than editing), Data curation; Tobias Nolte: Methodology, Writing (review &
patients without or lower numbers of ACEs, which also highlighted the editing); Iris Trawöger: Writing (review & editing), Data curation;
economical burden associated with the long-term effects of childhood Thomas Beck: Writing (review & editing), Data curation.
abuse. As described previously [8], polyvictimized patients also showed
a strongly increased likelihood to be living in a violent household. This Declaration of competing interest
is especially troubling, since in a previous analysis, we found an in-
creased cumulative likelihood for physical diseases in patients with The authors declare no conflicts of interest.
ACEs and experiences of domestic violence [38]. We found a peak of
ACEs between 10 and 14 years, which has been discussed as a highly References
sensitive phase in brain development, associated with increased risk for
depression and onset of other psychopathologies in adult life [13]. [1] Brown DW, Anda RF, Tiemeier H, Felitti VJ, Edwards VJ, Croft JB, et al. Adverse
childhood experiences and the risk of premature mortality. Am J Prev Med
5. Limitations of the study 2009;37:389–96.
[2] Felitti VJ, Anda RF, Nordenberg D, Williamson DF, Spitz AM, Edwards V, et al.
Relationship of childhood abuse and household dysfunction to many of the leading
Due to the length of the included questionnaires it was not feasible causes of death in adults. The adverse childhood experiences (ACE) study. Am J
to assess additional potential confounders for the relationship of ACEs Prev Med 1998;14:245–58.
[3] World Health Organization. Global status report on violence prevention 2014.
on the patients' health in adulthood, such as family history of medical Geneva: World Health Organization; 2014.
and mental health problems, socioeconomic status or lifestyle

85
D. Riedl, et al. General Hospital Psychiatry 64 (2020) 80–86

[4] Hughes K, Bellis MA, Hardcastle KA, Sethi D, Butchart A, Mikton C, et al. The effect hospital setting. Ir Med J 2006;99:42–5.
of multiple adverse childhood experiences on health: a systematic review and meta- [21] Lampe A. The prevalence of childhood sexual abuse, physical abuse and emotional
analysis. Lancet Public Health 2017;2:e356–66. neglect in Europe. Z Psychosom Med Psychother 2002;48:370–80.
[5] Danese A, McEwen BS. Adverse childhood experiences, allostasis, allostatic load, [22] Isele D, Teicher MH, Ruf-Leuschner M, Elbert T, Kolassa I-T, Schury K, et al.
and age-related disease. Physiol Behav 2012;106:29–39. KERF—an instrument for measuring adverse childhood experiences: construction
[6] Barboza Solis C, Kelly-Irving M, Fantin R, Darnaudery M, Torrisani J, Lang T, et al. and psychometric evaluation of the German MACE (maltreatment and abuse
Adverse childhood experiences and physiological wear-and-tear in midlife: findings chronology of exposure) scale. Zeitschrift für Klinische Psychologie und
from the 1958 British birth cohort. Proc Natl Acad Sci U S A 2015;112:E738–46. Psychotherapie: Forschung und Praxis 2014;43:121–30.
[7] McCrory E, De Brito SA, Viding E. The impact of childhood maltreatment: a review [23] Teicher MH, Parigger A. The ‘Maltreatment and abuse chronology of Exposure’
of neurobiological and genetic factors. Front Psych 2011;2:48. (MACE) scale for the retrospective assessment of abuse and neglect during devel-
[8] Kimber M, Adham S, Gill S, McTavish J, MacMillan HL. The association between opment. Plos One 2015;10:e0117423.
child exposure to intimate partner violence (IPV) and perpetration of IPV in [24] Association GP. German Pain Questionnaire. 2015.
adulthood-a systematic review. Child Abuse Negl 2018;76:273–86. [25] Franke GH. Mini-symptom-checklist (BSI-18). Manual. Göttingen: Hogrefe; 2017.
[9] Pechtel P, Pizzagalli DA. Effects of early life stress on cognitive and affective [26] Franke GH, Jaeger S, Glaesmer H, Barkmann C, Petrowski K, Braehler E.
function: an integrated review of human literature. Psychopharmacology Psychometric analysis of the brief symptom inventory 18 (BSI-18) in a re-
2011;214:55–70. presentative German sample. BMC Med Res Methodol 2017;17:14.
[10] Shonkoff JP, Boyce WT, McEwen BS. Neuroscience, molecular biology, and the [27] Spitzer C, Hammer S, Lowe B, Grabe HJ, Barnow S, Rose M, et al. The short version
childhood roots of health disparities: building a new framework for health pro- of the brief symptom inventory (BSI -18): preliminary psychometric properties of
motion and disease prevention. Jama 2009;301:2252–9. the German translation. Fortschr Neurol Psychiatr 2011;79:517–23.
[11] Schubert C. Psychoneuroimmunology of the life span: impact of childhood stress on [28] Sherin KM, Sinacore JM, Li XQ, Zitter RE, Shakil A. HITS: a short domestic violence
immune dysregulation and inflammatory disease in later life. Psychother screening tool for use in a family practice setting. Fam Med 1998;30:508–12.
Psychosom Med Psychol 2014;64:171–80. [29] Colman I, Kingsbury M, Garad Y, Zeng Y, Naicker K, Patten S, et al. Consistency in
[12] Bellis MA, Hughes K, Leckenby N, Hardcastle KA, Perkins C, Lowey H. Measuring adult reporting of adverse childhood experiences. Psychol Med 2016;46:543–9.
mortality and the burden of adult disease associated with adverse childhood ex- [30] OECD. Results from PISA 2015 (volume III): Students’ well-being - country note
periences in England: a national survey. Journal of public health (Oxford, England) Germany. OECD; 2017.
2015;37:445–54. [31] Takizawa R, Maughan B, Arseneault L. Adult health outcomes of childhood bullying
[13] Andersen SL, Teicher MH. Stress, sensitive periods and maturational events in victimization: evidence from a five-decade longitudinal British birth cohort. Am J
adolescent depression. Trends Neurosci 2008;31:183–91. Psychiatry 2014;171:777–84.
[14] Schalinski I, Teicher MH. Type and timing of childhood maltreatment and severity [32] Lereya ST, Copeland WE, Costello EJ, Wolke D. Adult mental health consequences
of shutdown dissociation in patients with schizophrenia spectrum disorder. PLoS of peer bullying and maltreatment in childhood: two cohorts in two countries.
One 2015;10:e0127151. Lancet Psychiatry 2015;2:524–31.
[15] Fujisawa TX, Shimada K, Takiguchi S, Mizushima S, Kosaka H, Teicher MH, et al. [33] Teicher MH, Vitaliano GD. Witnessing violence toward siblings: an understudied
Type and timing of childhood maltreatment and reduced visual cortex volume in but potent form of early adversity. PLoS One 2011;6:e28852.
children and adolescents with reactive attachment disorder. NeuroImage Clinical [34] Hanninen T, Soininen H. Age-associated memory impairment. Normal aging or
2018;20:216–21. warning of dementia? Drugs Aging 1997;11:480–9.
[16] Hauser W, Schmutzer G, Brahler E, Glaesmer H. Maltreatment in childhood and [35] Maercker A. Psychology of aging and clinical Gerontopsychology. Berlin,
adolescence: results from a survey of a representative sample of the German po- Heidelberg: Springer; 2015.
pulation. Deutsches Arzteblatt international 2011;108:287–94. [36] Norby S. Why forget? On the adaptive value of memory loss. Perspectives on psy-
[17] Witt A, Brown RC, Plener PL, Brahler E, Fegert JM. Child maltreatment in Germany: chological science: a journal of the Association for Psychological Science
prevalence rates in the general population. Child Adolesc Psychiatry Ment Health 2015;10:551–78.
2017;11:47. [37] Hardt J, Rutter M. Validity of adult retrospective reports of adverse childhood ex-
[18] LaNoue M, Graeber D, de Hernandez BU, Warner TD, Helitzer DL. Direct and in- periences: review of the evidence. J Child Psychol Psychiatry 2004;45:260–73.
direct effects of childhood adversity on adult depression. Community Ment Health J [38] Riedl D, Beck T, Exenberger S, Daniels J, Dejaco D, Unterberger I, et al. Violence
2012;48:187–92. from childhood to adulthood: the influence of child victimization and domestic
[19] Mouton CP, Hargreaves MK, Liu J, Fadeyi S, Blot WJ. Adult Cancer risk behaviors violence on physical health in later life. J Psychosom Res 2019;116:68–74.
associated with adverse childhood experiences in a low income population in the [39] Baldwin JR, Reuben A, Newbury JB, Danese A. Agreement between prospective and
southeastern United States. J Health Care Poor Underserved 2016;27:68–83. retrospective measures of childhood maltreatment: a systematic review and meta-
[20] Corcoran P, Gallagher J, Keeley HS, Arensman E, Perry IJ. Adverse childhood ex- analysis. JAMA Psychiat 2019;76:584–93.
periences and lifetime suicide ideation: a cross-sectional study in a non-psychiatric

86