Association of adverse childhood experiences with physical illness and self-rated health in the population-based Tromsø Study

Abstract

Adverse childhood experiences (ACEs) are highly prevalent and associated with life-long health consequences. Here, we investigated the relationship of ACEs to adult-onset health outcomes in general population. We conducted a cross-sectional analysis using the seventh survey of the Tromsø Study, including 20 843 participants. Main exposure variables were exposure to at least one ACE, rumination related to adverse experiences, cumulative ACEs, and two clusters of ACEs, i.e. interpersonal and impersonal ACEs. Logistic and ordinal regression models were fitted to estimate the risk of adult-onset physical illness and poor self-rated health adjusted for birth year, sex, smoking, education, and income. We found that exposure to at least one ACE together with rumination was associated with increased risk of hypertension, heart failure, atrial fibrillation, diabetes, obesity, kidney disease, chronic obstructive pulmonary disease, asthma, rheumatoid arthritis, arthrosis, migraine, chronic pain, and poor self-rated health. We observed dose–effect relationships of cumulative ACEs to the aggregated risk of 14 among 16 health outcomes. While increased risk of heart failure, kidney disease, and rheumatoid arthritis was only linked to the interpersonal ACEs, increased risk of coronary artery disease was associated only with the impersonal ACEs. Our findings demonstrate that exposure to ACEs increases the risk of adult-onset physical illness and poor self-rated health in a dose–effect relationship, and rumination related to adverse experiences is associated with an aggravated risk. Inquiry into exposure to ACEs might inform about health risks. Early intervention approaches to promote positive experiences and increase resilience might alleviate life-long health burden.

Introduction

Early life and adulthood experiences might have life-long effects on persons’ social, emotional, and health condition. Adverse childhood experiences (ACEs) are traumatic events that occur during childhood and can range from broad categories such as household dysfunction to more targeted experiences of bullying or exposure to crime [1, 2]. It is now established that the prevalence of ACEs is high [3]. Systematic reviews and meta-analyses have shown that individuals exposed to multiple ACEs were at increased risk of various health outcomes [4, 5]. The conditions associated with exposure to cumulative ACEs include lifestyle behaviors such as smoking or problematic alcohol use, mental health outcomes such as depressed mood or anxiety, and physical health conditions such as cardiovascular, respiratory and gastrointestinal diseases, diabetes, or cancer [4, 5]. Studies that examined clusters of ACEs showed that childhood abuse/maltreatment and neglect were linked to long-term health consequences [6], and household dysfunction was associated with the circulatory and endocrine-metabolic illness [7]. Additional items addressing potentially traumatic events were introduced frequently across the studies to expand the spectrum of ACEs. The other ACEs that were addressed included exposure to community violence, peer victimization, and loss of significant other [8, 9]. Certain patterns of co-occurrence and correlation among ACEs have been described [10, 11]. Network analysis of ACEs described in the Tromsø Study suggested that ACEs can be broadly categorized into interpersonal and impersonal ACEs [12–14].

Individual differences in cognitive responses to traumatic experiences and mental distress have been suggested as risk factors that might lead to persistent stress and mental health difficulties [15, 16]. One of these cognitive responses that is related with adverse experiences is repetitive, prolonged, and recurrent thinking about oneself, feelings, personal concerns, and experiences, also called rumination [16, 17]. Cognitive processes such as appraisals and rumination related to adverse experiences (abbreviated as rumination henceforth) have been associated with posttraumatic stress [18]. Rumination might be related with the severity of the adversity, and might contribute to maintenance of persistent stress, which can explain why some people have persistent traumatic stress while others recover without treatment after experiencing stressful events [16, 19, 20].

Health outcomes associated with certain ACEs, such as being exposed to serious illness or accident, bullying, natural disasters, or community violence has been less characterized. The data about the relationship between ACEs and adult-onset chronic diseases such as musculoskeletal, respiratory, and neurological illnesses are limited. Moreover, relationship between rumination and adult-onset physical illness and self-rated health has not been studied, and whether interpersonal and impersonal ACEs are associated with health outcomes remain unknown.

To address these knowledge gaps, we aimed to identify the risk of adult-onset physical illness and poor self-rated health associated with ACEs, rumination, and two core clusters of ACEs in a well-characterized population-based study.

Methods

Study design

We conducted our analysis using the cross-sectional data from the seventh survey of the population based Tromsø Study (2015–16) [21]. All residents aged 40 or older in the Tromsø municipality (N = 32 591) were invited to participate. The attendance rate was 65% (N = 21 083). Participants who responded to questionnaires and went through a health screening were included in our study. N = 20 843 participants (N = 240 had missing data), who answered the survey questionnaire, were included in the analysis. Main exposures were exposure to at least one ACE, reported rumination, number of ACEs, and the clusters of ACEs, i.e. interpersonal and impersonal ACEs. Main outcomes were lifetime prevalence of adult-onset physical illness and self-rated health. Covariates were birth year, sex, smoking, education, and household income.

Ethical approvals

This study has been performed in accordance with the Declaration of Helsinki and was approved by the Regional Committees for Medical and Health Research Ethics (REK#150757). Written informed consents were obtained from all participants.

Measurement of ACEs and exposure variables

Following items were used to measure ACEs in the Tromsø Study as described [13, 14]: (a) a life-threatening illness or a serious accident (e.g. fire, work accident, or car accident); (b) violence (e.g. being hit, kicked, beaten, robbed, or threatened with a firearm); (c) sexual abuse (i.e. sexual actions against one’s will); (d) bullying (e.g. been called negative things, marginalized, threatened, or bullied by schoolmates, fellow students, or coworkers over an extended period); (e) witnessed a loved one being exposed to violence or sexual abuse (e.g. hit, kicked, beaten, robbed, or threatened with a firearm); (f) something else frightening, dangerous, or violent (e.g. natural disaster, war, terror attack, being held captive); (g) severe grief after bereavement; (h) painful medical treatment when in hospital due to sickness or serious injury; (i) painful dental treatment; (j) a life-threatening illness or serious accident (e.g. fire, work accident, or car accident) of a loved one; (k) childhood neglect (e.g. not having received the necessary food, clothing, protection, and care/love from parents/caregivers). Items (a)–(j) had response options “no”, “yes, before the age of 18”, “yes, after the age of 18”, and “yes, in the previous year”. For item (k), the response options were “yes” and “no”. Items (a)–(k) were used to measure ACEs. Participants were grouped as exposed to ACE if they responded “yes, before the age of 18” to one or more of the questions. Number of ACEs were calculated by sum of the items responded “yes, before the age of 18”. N = 234 participants (1% of the included study population), who did not respond to any of the items about ACEs, were determined as not exposed to any ACEs.

The item to measure rumination addresses rumination reported at the survey date, and therefore, is related to all adverse experiences. Following item was used to measure rumination: participants were asked whether she/he still thinks a lot about what happened if she/he answered “yes” to at least one of the above questions. The response options were “yes” and “no”. Participants who reported being exposed to at least one ACE but had missing data about rumination (selected neither answer options for the item addressing rumination; N = 2043) were excluded from the analysis. Based on the answers three groups were defined: (1) individuals who were not exposed to any ACEs, (2) individuals who were exposed to at least one ACE and reported no rumination, and (3) individuals who were exposed to at least one ACE and reported rumination.

ACEs were grouped into two clusters based on previous network analysis [12]. Following items were used to measure the ACE clusters: participants were grouped as exposed to interpersonal ACEs if they responded “yes, before the age of 18” to one or more items (b), (c), (d), (e), and (k). Participants were grouped as exposed to impersonal ACEs if they responded “yes, before the age of 18” to one or more items (a), (f), (g), (h), (i), and (j).

Measurement of outcome variables and covariates

Physical illness was determined if the individual confirmed having had or have high blood pressure (hypertension), heart attack, angina pectoris, heart failure, atrial fibrillation, cerebral stroke/brain hemorrhage (stroke), diabetes, kidney disease not including urinary tract infection, chronic bronchitis/emphysema/chronic obstructive pulmonary disease (COPD), asthma, cancer, rheumatoid arthritis, arthrosis, migraine, chronic pain (persistent or constantly recurring pain that has lasted for three months or more). Participants were grouped into two categories for each of the physical illnesses as either currently or previously diagnosed with the condition, or not having the condition currently or previously. For each positive answer (currently or previously), age at first time was registered. Coronary artery disease (CAD) in an individual was determined if the individual had a history of heart attack and/or angina pectoris. Following question was used to define self-rated health: how do you in general consider your own health to be? Response options were “excellent”, “good”, “neither good nor bad”, “bad”, and “very bad”. Responses were organized to indicate poor self-rated health in an ascending order with “excellent” as the lowest rank and “very bad” as the highest rank. Body mass index of the participants was measured at the health screening. Obesity was defined as having a body mass index 30 or greater.

Physical illnesses were additionally grouped by involved body systems according to the 10th revision of the International Statistical Classification of Diseases and Related Health Problems. Having any of hypertension, CAD, heart failure, atrial fibrillation, and stroke were grouped as the circulatory system illness; diabetes and obesity were grouped as the endocrine-metabolic illness; COPD and asthma were grouped as the respiratory system illness; rheumatoid arthritis and arthrosis were grouped as the musculoskeletal system illness; migraine and chronic pain were grouped as the nervous system illness.

Smoking status was defined as not smoking, previously smoking, and currently smoking. Educational attainment was grouped into three as having primary/partial secondary education, upper secondary education, and tertiary education. Household income was categorized into three groups as low (<350 000 NOK [∼30 000 EUR]), middle (350 000–1 000 000 NOK [∼30 000–86 000 EUR]), and high (>1 000 000 NOK [∼86 000 EUR]).

Statistical analysis

Association between categorical variables were assessed using Pearson’s chi-squared test. Differences between quantitative variables were analysed using independent-samples t-test for variables with equal variances, Welch’s t-test for variables with unequal variances, and analysis of variance (ANOVA) tests. Two-sided P-values were reported.

Following associations were tested as the main exposures: (1) being exposed to at least one ACE (N = 9334), which was defined as a categorical variable; (2) being exposed to at least one ACE and reported no rumination (N = 6509), and being exposed to at least one ACE and reported rumination (N = 782), which were defined as categorical variables; (3) being exposed to one (N = 5188), two (N = 2397), three (N = 992), four or more (N = 757) ACEs, which were defined as categorical variables; (4) being exposed to interpersonal ACEs (N = 5683), which was defined as a categorical variable; (5) Being exposed to impersonal ACEs (N = 6237), which was defined as a categorical variable. All categories were compared to the reference category including individuals reported no ACE exposure (N = 11 509).

Participants who did not answer the item for the physical illness or who reported age at onset for the physical illness before age 18 years were excluded. When illnesses were grouped by body systems, participants with age at onset for any of the physical illnesses in the group before age 18 years were excluded. Study populations included for each of the physical illnesses are shown in Supplementary Tables S1 and S2. Risk of each physical illness was estimated using logistic regression. Ordinal regression was used to estimate the risk of poor self-rated health. Analyses were adjusted for birth year, sex, smoking, education, and household income. Risk estimates for the physical illnesses and poor self-rated health were reported as odds ratio (OR) with 95% confidence interval (CI). Analyses were performed using SPSS version 29 (IBM SPSS Statistics for Windows, Version 29 Armonk, NY: IBM Corp) and RStudio (R version 4.1.1) [22].

Results

Characteristics of the study population are shown in Table 1. The prevalence of exposure to at least one ACE was 45% and the prevalence of exposure to ≥4 ACEs was 4%. Individuals who were exposed to at least one ACE were younger than those reporting no ACEs, and females were more often exposed to ACE. Individuals exposed to at least one ACE were more often smokers currently or previously. Differences in age, sex, and smoking related to ACEs were further evident with cumulative ACEs; individuals exposed to more ACEs were younger, more often females, and more often current or previous smokers. Individuals exposed to at least one ACE had more often higher education, and educational attainment was higher gradually in individuals who were exposed to more ACEs. High income was less often reported by individuals exposed to cumulative ACEs.

Table 1.

Characteristics of the study population

	No ACE	At least one ACE	P-value	1 ACE	2 ACEs	3 ACEs	≥4 ACEs	P-value
N (%)	11509 (55)	9334 (45)		5188 (25)	2397 (11)	992 (5)	757 (4)
Age mean (STD)	58.6 (11.5)	55.7 (11.0)	<.001	56.9 (11.2)	54.9 (10.7)	54.1 (10.2)	52.0 (9.2)	<.001
Female N (%)	5936 (51.6)	5018 (53.8)	.002	2721 (52.4)	1283 (53.5)	549 (55.3)	465 (61.4)	<.001
Smoking N (%)
Never	5049 (44.3)	3585 (38.7)	<.001	2081 (40.4)	915 (38.5)	355 (36.2)	234 (31.1)	<.001
Currently	1453 (12.7)	1411 (15.2)	<.001	731 (14.2)	377 (15.9)	163 (16.6)	140 (18.6)	<.001
Previously	4900 (43.0)	4259 (46.0)	<.001	2333 (45.3)	1084 (45.6)	464 (47.3)	378 (50.3)	<.001
Education N (%)
Primary	2867 (25.4)	1831 (19.9)	<.001	1089 (21.3)	446 (18.9)	168 (17.1)	128 (17.0)	<.001
Secondary	3153 (28.0)	2544 (27.6)	.608	1401 (27.4)	668 (28.3)	275 (27.9)	200 (26.6)	.863
Tertiary	5258 (46.6)	4831 (52.5)	<.001	2617 (51.2)	1249 (52.9)	541 (55.0)	424 (56.4)	<.001
Income N (%)
Low	1449 (13.2)	1107 (12.3)	.054	593 (11.9)	284 (12.2)	124 (12.8)	106 (14.4)	.105
Middle	6781 (61.8)	5662 (62.8)	.126	3107 (62.2)	1451 (62.6)	621 (64.4)	483 (65.5)	.169
High	2750 (25.0)	2245 (24.9)	.820	1293 (25.9)	584 (25.2)	220 (22.8)	148 (20.1)	.007

Prevalence of physical illnesses after age at 18 years grouped on ACEs are summarized in Supplementary Tables S1 and S2. Being exposed to at least one ACE was associated with adult-onset circulatory, endocrine-metabolic, respiratory, musculoskeletal, and nervous system illness; ORs ranged from 1.14 (95% CI, 1.07–1.22) to 1.54 (95% CI, 1.45–1.63) (Supplementary Table S3). Individuals exposed to at least one ACE was at increased risk of adult-onset physical illness, specifically, hypertension, atrial fibrillation, diabetes, obesity, COPD, asthma, arthrosis, migraine, and chronic pain, and poor self-rated health; ORs ranged from 1.11 (95% CI, 1.04–1.19) to 1.58 (95% CI, 1.49–1.69).

Risk of circulatory, endocrine-metabolic, respiratory, musculoskeletal, and nervous system illness was increased in individuals who were exposed to at least one ACE and reported no rumination and was further increased in individuals who were exposed to at least one ACE and reported rumination (Supplementary Fig. S1). Similarly, adjusted risk estimates for physical illnesses and poor self-rated health were larger in individuals who were exposed to at least one ACE and reported rumination compared to individuals who were exposed to at least one ACE and reported no rumination (Fig. 1). Importantly, when risk of physical illness and poor self-rated health in individuals who were exposed to at least one ACE and reported rumination was compared to individuals who reported no ACE exposure, risk of 13 among 15 health outcomes were increased; ORs ranged from 1.29 (95% CI, 1.07–1.54) to 2.46 (95% CI, 1.71–3.54) (Fig. 1).

Figure 1.

Adjusted risk estimates for adult-onset physical illness and poor self-rated health associated with exposure to at least one ACE and grouped on reported rumination. CAD, coronary artery disease; COPD, chronic obstructive pulmonary disease.

Association of cumulative ACEs with health outcomes was evident as demonstrated by the dose–effect relationship observed in physical illnesses grouped by systems (Supplementary Fig. S2) and specific health outcomes (Table 2). Individuals exposed to one or more ACEs had a graded increased risk of having arthrosis, migraine, chronic pain, and poor self-rated health. Exposure to two or more ACEs was associated with a cumulative increased risk of atrial fibrillation and asthma. Exposure to three or more ACEs was associated with an increased risk of hypertension, diabetes, obesity, COPD, kidney disease, and rheumatoid arthritis whereas exposure to four or more ACEs was associated with increased risk of having CAD and heart failure (Table 2).

Table 2.

Adjusted risk estimates for adult-onset physical illness and poor self-rated health associated with exposure to cumulative ACEs

Condition	1 ACE OR (95% CI)	1 ACEs OR (95% CI)	3 ACEs OR (95% CI)	≥ 4 ACEs OR (95% CI)
Hypertension	1.06 (0.99–1.15)	1.10 (0.99–1.23)	1.26 (1.08–1.48)	1.34 (1.12–1.61)
CAD	1.06 (0.89–1.25)	1.00 (0.78–1.28)	1.32 (0.94–1.86)	1.78 (1.22–2.61)
Heart failure	1.13 (0.87–1.46)	1.21 (0.84–1.74)	0.97 (0.52–1.81)	2.12 (1.20–3.75)
Atrial fibrillation	1.10 (0.95–1.27)	1.42 (1.17–1.71)	1.61 (1.23–2.10)	1.93 (1.44–2.61)
Stroke	0.85 (0.67–1.07)	1.11 (0.81–1.51)	1.30 (0.84–2.01)	1.51 (0.91–2.51)
Diabetes	1.15 (0.98–1.35)	0.92 (0.73–1.17)	1.82 (1.39–2.38)	1.73 (1.25–2.39)
Obesity	1.20 (1.04–1.39)	1.15 (0.95–1.39)	1.49 (1.16–1.92)	1.95 (1.51–2.52)
Kidney disease	0.90 (0.79–1.23)	1.31 (0.99–1.73)	1.51 (1.03–2.23)	1.94 (1.28–2.92)
COPD	1.18 (0.96–1.46)	1.20 (0.90–1.61)	2.17 (1.54–3.07)	2.43 (1.65–3.57)
Asthma	1.10 (0.97–1.26)	1.37 (1.16–1.62)	1.74 (1.39–2.17)	1.82 (1.42–2.33)
Rheumatoid arthritis	1.09 (0.92–1.30)	1.06 (0.83–1.36)	1.40 (1.02–1.93)	1.45 (1.01–2.09)
Arthrosis	1.22 (1.11–1.34)	1.49 (1.31–1.69)	1.42 (1.18–1.71)	1.69 (1.37–2.07)
Migraine	1.25 (1.12–1.40)	1.33 (1.15–1.53)	1.74 (1.42–2.12)	1.57 (1.25–1.96)
Chronic pain	1.40 (1.30–1.50)	1.67 (1.52–1.85)	2.06 (1.79–2.37)	2.23 (1.90–2.61)
Cancer	0.99 (0.87–1.13)	0.87 (0.72–1.06)	1.22 (0.94–1.58)	1.01 (0.72–1.42)
Self-rated health	1.22 (1.15–1.31)	1.41 (1.29–1.54)	1.73 (1.52–1.97)	2.32 (2.00–2.69)

CAD, coronary artery disease; COPD, chronic obstructive pulmonary disease.

Both ACE clusters, i.e., interpersonal and impersonal ACEs, were associated with increased risk of the circulatory, endocrine-metabolic, respiratory, musculoskeletal, and nervous system illness with similar effect sizes (Supplementary Fig. S3). Interpersonal and impersonal ACEs were associated with increased risk of hypertension, atrial fibrillation, diabetes, obesity, COPD, asthma, arthrosis, migraine, chronic pain, and poor self-rated health; ORs for the associations with the interpersonal ACEs ranged from 1.12 (95% CI, 1.03–1.21) to 1.63 (95% CI, 1.51–1.75), and ORs for the associations with the impersonal ACEs ranged from 1.16 (95% CI, 1.08–1.25) to 1.66 (95% CI, 1.54–1.78) (Fig. 2). The associations of physical illness with the interpersonal ACEs had generally larger adjusted risk estimates compared to the associations with the impersonal ACEs. Furthermore, increased risk of heart failure, kidney disease, and rheumatoid arthritis was associated only with the interpersonal ACEs, whereas increased risk of CAD was associated only with the impersonal ACEs (Fig. 2).

Figure 2.

Adjusted risk estimates for adult-onset physical illness and poor self-rated health associated with exposure to interpersonal and impersonal ACEs. CAD, coronary artery disease; COPD, chronic obstructive pulmonary disease.

Discussion

This study has three main findings: (1) exposure to ACEs is associated with an increased and graded risk of adult-onset physical illness and poor self-rated health in a dose–effect relationship; (2) rumination contributes to an aggravated risk of physical illness and poor self-rated health; and (3) Both interpersonal and impersonal ACEs are associated with increased risk of multiple physical illnesses and poor self-rated health whereas increased risk of certain physical illnesses might be specifically associated with interpersonal ACEs.

Our study included a broad range of ACEs, e.g., being exposed to accident, bullying, painful medical treatment, loss of a close one, and being witness to stressful events, which were not characterized before in relation to health outcomes. Chronic diseases which have been relatively understudied in relation to ACEs, such as musculoskeletal, and respiratory illnesses, were included in our analysis to show accumulated risk of multiple physical illnesses associated with exposure to increasing number of ACEs. Various events have been used in studies to define ACEs, and there is no standardized instrument [23]. Heterogeneity in instruments to define ACEs therefore presents challenges in comparison of studies. The prevalence of ACEs and at least four ACEs measured in different countries have ranged 33%–88% and 1%–38%, respectively [5]. The pooled prevalence of exposure to at least one ACE and at least four ACEs has been estimated as 60.2% and 16.1% [3]. Differences in the prevalence of ACEs among the studies might be due to measured ACE types, instruments to measure ACEs, and demographics of the study population. The prevalence of ACEs in this study is in line with the findings from national surveys [5, 24]. While analysis of specific ACE types and grouping by demographic features might improve the prevalence estimations, the findings altogether indicate that ACEs are prevalent across the world in all socioeconomic domains and present a major risk for health throughout life.

Inclusion of ACE items addressing various adversities has led to identification of two clusters of ACEs, specifically interpersonal and impersonal ACEs [12]. While we found that both groups of ACEs are associated with multiple adult-onset physical illnesses and poor self-rated health, the group of interpersonal ACEs was associated with the most illnesses and was the only group that was associated with heart failure, kidney disease, and rheumatoid arthritis. It is also noteworthy that the group of interpersonal ACEs has the strongest relationships with the adult-onset physical illnesses and poor self-rated health, as measured by the effect sizes, compared to the group of impersonal ACEs. Therefore, our results emphasize that inclusion of a variety of ACE types in studies might improve association outcomes with physical illnesses.

Identification of factors that can modify the risk imposed by ACEs is important to develop intervention strategies. Our results indicate that rumination is associated with an aggravated risk of health conditions and might be considered as an additional risk factor. Various dimensions of adverse experiences such as frequency, severity, duration, and perception of adversity might have been culminated in the rumination response. Severity of adverse experiences reflected by rumination might increase the risk of physical illness by intense and prolonged stress. Rumination is correlated with mental health difficulties and tendency to health-harming behaviors, which are considered as important risk factors for physical illnesses and premature mortality [25–27]. Hence, it will be important to assess the moderating and/or mediating effects of mental health and health-harming behaviors in the relationship between ACEs and physical illness. Future research might also address whether synergistic additive effects between ACEs, lifestyle, and socioeconomic factors exist to increase the risk of physical illness.

Population-level surveys and epidemiological research on ACEs is recognized as a useful tool to understand the impact of adverse experiences on health at the population, and appropriate trauma-specific treatment for individuals with a history of childhood adversity is recommended [28]. However, screening of ACEs in clinical assessments is currently not suggested as part of a diagnostic assessment due to lack of clear evidence about the utility of ACE scoring in predicting health outcomes at the individual level [28, 29]. Although there are variations among the currently available ACE assessments methods, the assessment tools show consistent associations with poor health outcomes [30]. At the same time, patients might benefit from ACE screening if existing ACE items can provide additional information about health risks and can be evaluated as supplementary evidence similar to those provided by family medical history or smoking.

Our study has some limitations. We employed a retrospective study design and defined adverse experiences and physical illnesses based on self-reports. Using retrospective self-reports about ACEs is a common limitation in research for effects of ACEs [31, 32]. Important ACEs, such as emotional abuse and parental illicit use of drugs during childhood were not addressed as ACEs in the Tromsø Study and therefore could not be included in our study. However, those items are highly connected and often co-occur with the ACEs that are addressed in our study [33–37]. We also used self-reports to define physical illnesses in our analysis. Nevertheless, self-reported definitions of physician-diagnosed physical illnesses are consistent with the health records [38–40]. Exposure to ACEs is correlated with health-harming behaviors and poor mental health, which in turn are related to physical health. Therefore, it remains to be investigated whether lifestyle behaviors and psychosocial factors have a mediating role for the association between ACEs and physical illness. Smoking status, education, and income level were determined at the time of survey and used as covariates in our study due to cross-sectional study design. Although the large, population-based sample in our study and agreement of our results with previous studies demonstrate validity of our findings, it will be important to assess the association of ACEs with health conditions in prospective cohort studies.

Taken together, our study shows that ACEs are associated with an increased risk of multiple adult-onset physical illnesses and poor self-rated health in a dose–effect relationship. We also provide evidence that rumination is associated with an aggravated risk of health outcomes and that association of ACEs with certain physical illnesses might be dependent on the ACE type. It will be important to investigate whether screening of ACEs in children and young adults or improving rumination reduces the burden of health problems. Focusing on early intervention strategies such as reducing prevalence of ACEs, early follow-up, and prompt treatment by implementing protective factors might alleviate life-long effects of ACEs.

Author contributions

Safak Caglayan, Anne Høye, and Ole K. Grønli conceptualized and designed the study, and managed the project. Safak Caglayan performed the formal analysis. All authors contributed to interpretation of data. Safak Caglayan wrote the first draft of the article, and all authors commented on the previous versions of the article. All authors read and approved the final article.

Supplementary data

Supplementary data are available at EURPUB online.

Conflict of interest: None declared.

Funding

This study was supported by grants (HNF-156321) from the Northern Norway Regional Health Authority. The funder had no role in the design and conduct of the study, in the collection, analysis, and interpretation of the data, and in the preparation, review, or approval of the article, and in the decision to submit the article for publication.

Data availability

Data underlying the present study was retrieved from the Tromsø Study by permission. Data in the present study can be requested from the corresponding author with permission from the Tromsø Study.

Key points.

• Adverse childhood experiences are associated with adult-onset physical illness and poor self-rated health.

• Rumination related to adverse experiences contributes to an aggravated risk of poor health outcomes, indicating importance of improving resilience in preventing health burden.

• The risk of physical illness and poor self-rated health increases with exposure to increased number of adverse childhood experiences in a dose–response relationship, warranting an unmet need for early screening and intervention strategies.

• Both interpersonal and impersonal adverse childhood experiences are linked to adult-onset physical illness and poor self-rated health.