Skip to main content
Download PDF
  • Research article
  • Open access
  • Published:

Childhood maltreatment and risk of endocrine diseases: an exploration of mediating pathways using sequential mediation analysis

BMC Medicine volume 22, Article number: 59 (2024) Cite this article

Abstract

Background

Adverse childhood experiences (ACEs), including childhood maltreatment, have been linked with increased risk of diabetes and obesity during adulthood. A comprehensive assessment on the associations between childhood maltreatment and all major endocrine diseases, as well as the relative importance of different proposed mechanistic pathways on these associations, is currently lacking.

Methods

Based on the UK Biobank, we constructed a cohort including 151,659 participants with self-reported data on childhood maltreatment who were 30 years of age or older on/after January 1, 1985. All participants were followed from the index date (i.e., January 1, 1985, or their 30th birthday, whichever came later) until the first diagnosis of any or specific (12 individual diagnoses and 9 subtypes) endocrine diseases, death, or the end of follow-up (December 31, 2019), whichever occurred first. We used Cox models to examine the association of childhood maltreatment, treated as continuous (i.e., the cumulative number of experienced childhood maltreatment), ordinal (i.e., 0, 1 and ≥ 2), or binary (< 2 and ≥ 2) variable, with any and specific endocrine diseases, adjusted for multiple covariates. We further examined the risk of having multiple endocrine diseases using Linear or Logistic Regression models. Then, sequential mediation analyses were performed to assess the contribution of four possible mechanisms (i.e., suboptimal socioeconomic status (SES), psychological adversities, unfavorable lifestyle, and biological alterations) on the observed associations.

Results

During an average follow-up of 30.8 years, 20,885 participants received a diagnosis of endocrine diseases. We observed an association between the cumulative number of experienced childhood maltreatment and increased risk of being diagnosed with any endocrine disease (adjusted hazard ratio (HR) = 1.10, 95% confidence interval 1.09–1.12). The HR was 1.26 (1.22–1.30) when comparing individuals ≥ 2 with those with < 2 experienced childhood maltreatment. We further noted the most pronounced associations for type 2 diabetes (1.40 (1.33–1.48)) and hypothalamic–pituitary–adrenal (HPA)-axis-related endocrine diseases (1.38 (1.17–1.62)), and the association was stronger for having multiple endocrine diseases, compared to having one (odds ratio (95% CI) = 1.24 (1.19–1.30), 1.35 (1.27–1.44), and 1.52 (1.52–1.53) for 1, 2, and ≥ 3, respectively). Sequential mediation analyses showed that the association between childhood maltreatment and endocrine diseases was consistently and most distinctly mediated by psychological adversities (15.38 ~ 44.97%), while unfavorable lifestyle (10.86 ~ 25.32%) was additionally noted for type 2 diabetes whereas suboptimal SES (14.42 ~ 39.33%) for HPA-axis-related endocrine diseases.

Conclusions

Our study demonstrates that adverse psychological sequel of childhood maltreatment constitutes the main pathway to multiple endocrine diseases, particularly type 2 diabetes and HPA-axis-related endocrine diseases. Therefore, increased access to evidence-based mental health services may also be pivotal in reducing the risk of endocrine diseases among childhood maltreatment-exposed individuals.

Peer Review reports

Background

Adverse childhood experiences (ACEs), including childhood maltreatment, encapsulate a broad spectrum of traumatic and distressing events that occur during childhood, mainly including direct maltreatment and neglect, physical and sexual abuse, violence, and household dysfunction [1]. Despite some variation in findings, numerous studies have reported a considerably high prevalence of ACEs (20–72%) among individuals with diverse demographic and socioeconomic characteristics [2,3,4]. Furthermore, besides the well-documented impacts on psychopathology during this critical period of development, growing evidence demonstrate profound and long-lasting influence of ACEs on a wide spectrum of somatic adversities, such as respiratory diseases, cardiovascular disease, cancer, and premature mortality [4,5,6].

The incidence of many endocrine diseases, i.e. disturbances of multiple hormone-producing cells and organs throughout the body, particularly diabetes and obesity, have been increasing in virtually all regions of the world [7]. A recent meta-analysis of 49 observational studies demonstrated a positive association between ACEs and diabetes, yielding an odds ratio of 1.22 for any ACEs and 1.27 for neglect [8]. Similarly, ACEs have been consistently associated with increased odds of childhood [9] and adult [10] obesity. However, despite evidence from human and animal studies supporting a wide range of endocrine dysfunctions which possibly involve all endocrine-related axes after ACEs [11,12,13,14], to date, there is no comprehensive evaluation of the association of ACEs with multiple endocrine diseases. Also, existing data are mainly derived from cross-sectional studies with limited sample size [10, 15, 16] and little focus on mediation factors [8, 16], which precludes the application of these data for causal inference or further mechanistic exploration.

Proposed explanations underlying the links between ACEs and somatic adversities in later life have been summarized and introduced in prior reviews published by the US Centers for Disease Control and Prevention [17], primarily including behavioral impairment (e.g., impaired coping strategies and unfavorable lifestyle) [18, 19], and socioeconomic deprivation (e.g., lower educational attainment and income) [19,20,21]. Such a conceptual framework has been widely used to elucidate the associations of ACEs with cardiovascular diseases [6] and psychiatric disorders [22]. Using prospective data from a US cohort, one recent study indicated the possible mediating role of metabolic syndrome on the association of ACEs with self-reported type 2 diabetes [23]. Another prospective cohort of British civil servants also demonstrated depression and cardiometabolic dysregulations might be the pathways linking ACEs and diabetes [24]. In addition, elevated susceptibility to psychiatric disorders as well as the dysregulation of hypothalamic–pituitary–adrenal (HPA) axis, a vital biological pathway that mediates the effects of stressors by influencing numerous physiological processes [25, 26], have also been observed among the ACE-exposed population [5]. Other possible biological alternations indicated in experimental studies include epigenetic changes, disruption of central neural networks, chronic activation of inflammatory pathways, and immune dysfunction [25, 27,28,29]. Nevertheless, as most of these studies focus on a specific type [10, 15, 23] or a small set [19, 23] of mediators, the attributable proportions of these mechanistic pathways, and consequently their importance for disease intervention, remain rather unexplored.

Leveraging data from the UK Biobank, which provides wealth information on childhood maltreatment, socioeconomic status (SES), psychological factors, lifestyle, biological biomarkers, as well as longitudinal and complete disease-related outcomes, we conducted a cohort study to assess the association between childhood maltreatment and multiple endocrine diseases during adulthood, with a particular focus on specific diagnoses (e.g., diabetes) and subtypes of endocrine diseases (e.g., HPA-axis-related endocrine diseases) to shed light on potential pathways. Furthermore, we focused on elucidating the relative importance of four specific mechanistic pathways (i.e., suboptimal SES, psychological adversities, unfavorable lifestyles, and biological alterations) using sequential causal mediation analysis, which is a newly developed method for weighting the contribution of multiple causally related mediators after accommodating exposure-mediator interactions and various assumptions about the causal ordering of these mediators [30].

Methods

Data source and study design

The study was based on the UK Biobank, which enrolled 502,507 participants aged 40–69 years across England, Scotland, and Wales between 2006 and 2010. The UK Biobank is not representative of the general population in the UK, as it recruited only 5.5% of the invited population and the participants were predominately white (94.6%) [31]. At recruitment, data on sociodemographic characteristics, lifestyles and psychosocial factors, medical and family history, as well as physical measures were collected using touchscreen questionnaire [32]. Health-related outcomes including medical consequences and survival status were obtained through periodical linkages to multiple national datasets [33]. Information on inpatient hospital care was derived from Hospital Episode Statistics for England, Scottish Morbidity Record, and Patient Episode Database for Wales, covering almost all UK Biobank participants since 1997 [33]. Primary care data was obtained from numerous general practice systems and covers approximately 45% of the UK Biobank participants [33].

In the present study, we first identified 153,634 participants that responded to all five questions about ACEs (mainly childhood maltreatment) in the touchscreen questionnaire (see comparison on sociodemographic characteristics of participants with and without such data in Additional file 1: Table S1) and excluded 41 participants who withdrew their informed consent afterwards. Then, as data from primary care has been available since 1985 (most UK Biobank participants were above 58 at recruitment), we focused on endocrine diseases diagnosed at age 30 or after. Accordingly, after exclusion of 1934 participants with a history of endocrine diseases, we constructed a cohort including 151,659 participants who were 30 years of age or older on/after January 1, 1985. All participants were followed from the index date (i.e., January 1, 1985, or their 30th birthday, whichever came later), until the first diagnosis of any or specific endocrine diseases (12 individual diagnoses and 9 subtypes), death, or the end of follow-up (December 31, 2019), whichever occurred first.

Assessment of childhood maltreatment

In 2016, childhood maltreatment was assessed by five items (i.e., physical abuse, emotional abuse, sexual abuse, sexual neglect, and physical neglect) in an online mental health questionnaire survey, using the Childhood and Trauma Screener [34] (See Additional file 1: Table S2 for details). For instance, the online question for physical abuse is “When I was growing up: People in my family hit me so hard that it left me with bruises or marks”. Each item was dichotomized (yes or no) using cut-off points reported previously [35] (Additional file 1: Table S2) and the five items were then summed to generate the cumulative number of childhood maltreatment.

Ascertainment of endocrine diseases

We ascertained endocrine diseases based on diagnoses derived from the self-reported questionnaire, and linked data from primary care and hospital inpatient registers, using the International Classification of Diseases 10th edition (ICD-10: E00-E34). The mapping of the diagnoses from different data sources was performed through an expert peer review and consensus process, providing reliable information on the diagnosis date of the first occurrence of a set of specific health outcomes [36]. Besides considering all endocrine diseases as one outcome, we additionally identified 12 individual diagnoses of endocrine diseases, and 9 subtypes of endocrine diseases according to the involved gland (i.e., thyroid, pancreatic, parathyroid, hypothalamic-pituitary, adrenal, genital glands) or axis (i.e., HPA, hypothalamic-pituitary-thyroid [HPT] and hypothalamic-pituitary–gonadal [HPG] axis). All ICD-10 codes used for endocrine disease identification could be found in Additional file 1: Table S3.

Covariates and mediators

Given the well-established influences of demographic factors, childhood SES condition, and genetic susceptibility to disease on the studied associations [37, 38], we obtained information on multiple covariates, including sociodemographic characteristics (i.e., sex, birth year, ethnicity, and country of birth), childhood environment (i.e., number of siblings, maternal smoking, and being breastfed), and family history of diabetes, from the baseline survey (see Additional file 1: Table S2 for details).

Data on candidate mediators was collected at baseline. To test the importance of the four hypothesized pathways, we categorized them in four categories (i.e., SES, psychological factors, lifestyle factors, and biological biomarkers). Specifically, the SES category included Townsend Deprivation Index (TDI), educational level, household income, and employment status. Psychological factors included items related to self-rated mental problems [6] and social support (i.e., ability to confide in others, frequency of friend/family visits, and leisure/social activities) according to the baseline survey. Lifestyle factors indicated status of physical activity, smoking, alcohol, sleep pattern, diet, body mass index (BMI), systolic pressure, and diastolic pressure. In addition, as biomarkers of inflammation, nutrition, lipid, and glucose have been considered important to the associations between ACEs and the occurrence of diseases [27], we included C-reactive protein, total protein, Ca, Vitamin D, cholesterol, lipoprotein A, high-density lipoprotein (HDL), low-density lipoprotein (LDL), apolipoprotein A, apolipoprotein B, triglycerides, glucose, and glycosylated hemoglobin in the category of biomarkers.

For all variables, the answers “do not know” or “prefer not to answer” were considered as unknown. Details of the included mediators are listed in Additional file 1: Table S2 and Additional file 2: Supplementary method [6, 39,40,41,42,43].

Statistical analysis

To assess the association of childhood maltreatment, treated as continuous (i.e., the cumulative number of experienced childhood maltreatment), ordinal (i.e., 0, 1, and ≥ 2), or binary (< 2 and ≥ 2) variable, with any endocrine diseases diagnosed after age of 30, we applied Cox models, partly (models 1–3) or fully (model 4) adjusted for all covariates, calculating hazard ratios (HRs) with their 95% confidence intervals (CIs). We then performed separate analyses for different types of experienced childhood maltreatment, as well as for 12 individual diagnoses and 9 subtypes of endocrine diseases. Further, we evaluated the association of childhood maltreatment (as a binary variable: < 2 and ≥ 2) with multiple endocrine diseases (i.e., the number of any endocrine disease and the number of HPA-axis-related endocrine diseases), using Linear or Logistic Regression models.

Next, we performed the sequential mediation analyses to assess the contribution of four possible mechanisms (i.e., suboptimal SES, psychological adversities, unfavorable lifestyle, and biological alterations) to the associations of childhood maltreatment (as a binary variable: < 2 and ≥ 2) with any endocrine disease, type 2 diabetes, and HPA-axis-related endocrine diseases, in three steps. First, to ensure a temporal order between childhood maltreatment, mediators, and endocrine diseases, we restricted the analyses to participants with no censoring event (i.e., diagnosis of endocrine diseases, death, or loss of follow-up) before the recruitment to UK Biobank. The comparison between the new cohort (n = 138,498 for any endocrine disease cohort, Fig. 1) and the entire cohort of the present study showed a comparable distribution of baseline characteristics (Additional file 1: Table S4). Second, as we included several mediators in each hypothesized pathway, we conducted a simple mediation analysis to evaluate the contribution of each mediator. Specifically, using “CMAverse” package based on the counterfactual g-formula approach [44], the mediated proportion of each mediator was estimated by partitioning the total effect of exposure-outcome association into direct effect and indirect effect with causal mediation framework [45]. Parametric bootstrapping (100 times) was used to calculate 95% CIs and P value. We then chose the mediators with the greatest mediated proportion in each category for further analyses. Third, in the sequential mediation analyses, a multi-mediator model [30] was used to measure the mediation proportion of different pathways through assessing the path-specific effects [46], using Cox models adjusted for abovementioned covariates. Namely, by defining the selected variables for SES, psychological factors, lifestyle factors, and biomarkers as M1, M2, M3, and M4, respectively, we calculated the direct effect, the effect of M1-mediated pathways, the effect of M2-mediated pathways, the effect of M3-mediated pathways, and the effect of M4-mediated pathways, as components of the total effect. This calculation was done for all possible temporal orders of the suggested attributable mediators (e.g., M1 → M2 → M3 → M4 and M1 → M2 → M4 → M3, see proposed causal diagram in Fig. 3A and B). Particularly, as the multi-mediator models do not allow missing values, individuals with missing data of included mediators were excluded from the analysis. We then summarized the mediated proportions of each hypothesized pathway by the range (lowest-highest) of point estimates derived from different sequential models. Detailed explanations of the mediation analyses are provided in Additional file 2: Supplementary method [30, 46,47,48,49,50,51].

Fig. 1
figure 1

Study design

Full size image

In the sensitivity analyses, we applied a competing risk model to re-assess the studied associations, taking into account competing risk by death. To assess the soundness of the findings by the definition of childhood maltreatment exposure, we re-estimated the mediation proportions using an alternative grouping strategy (i.e., number of experienced childhood maltreatment ≥ 1 vs < 1), or after restricting the reference group to individuals without any childhood maltreatment experience (i.e., ≥ 2 vs 0). In addition, we assessed the influence of missingness in some mediators (0.13–14.02%, see Additional file 1: Table S5) to the obtained estimates by repeating the sequential mediation analyses after performing multiple imputation. All statistical analyses were conducted using R, version 4.0.2 (R Project for Statistical Computing). A two-sided P < 0.05 was considered statistically significant.

Results

Among 151,659 participants included in full cohort for the association analyses, the mean age at cohort entry was 34.6 years (standard deviation (SD) = 4.85) and 43.93% of them were male (Table 1). While there were few discrepancies regarding demographic characteristics between groups with a different exposure level of childhood maltreatment (0, 1, and ≥ 2), individuals exposed to a higher level of childhood maltreatment exposure were more likely to be women (55.79% for n = 0 [i.e., no childhood maltreatment] vs 53.59% for n = 1 and 59.78% for ≥ 2 group), have ≥ 3 siblings (21.94% vs 28.29% and 35.06%), have a family history of diabetes (20.35% vs 21.17% and 23.74%), and have reported maternal smoking (22.64% vs 25.43% and 31.06%), but less likely to have been breastfed (60.69% vs 58.56% and 54.94%) (Table 1).

Table 1 Baseline characteristics of the study population
Full size table

During an average follow-up of 30.8 years (SD = 6.53), we identified 20,885 individuals who developed endocrine diseases. The number of cases was 10,394, 5312, and 5179 among individuals with 0, 1, or ≥ 2 childhood maltreatment, corresponding to a crude incidence rate (IR) of 4.12, 4.44, and 5.45 per 1000 person-years, respectively. As shown in Table 2, we found a positive association between the cumulative number of experienced childhood maltreatment and any endocrine disease (birth year- and sex-adjusted HR (model 1) = 1.12 (95% CI 1.11–1.13); and fully adjusted HR (model 4) = 1.10 (95% CI 1.09–1.12)). Compared with individuals without childhood maltreatment, the HR of developing any endocrine diseases was 1.07 (95% CI 1.04–1.11) and 1.29 (95% CI 1.25–1.33) for individuals exposed to 1 or ≥ 2 childhood maltreatment experiences, respectively, according to the fully adjusted model. The HR was 1.26 (95% CI 1.22–1.30) when comparing individuals with ≥ 2 to those with < 2 childhood maltreatment experiences. Also, we observed comparable estimates for different types of childhood maltreatment, with fully adjusted HRs ranging from 1.17 (95% CI 1.14–1.21) for emotional neglect to 1.23 (95% CI 1.19–1.27) for physical neglect (Table 2).

Table 2 Hazard ratios (HRs) with 95% confidence intervals (CIs) for the associations between childhood maltreatment and any endocrine disease
Full size table

Among the 12 studied individual diagnoses of endocrine diseases, five (i.e., hypothyroidism, hyperthyroidism, type 2 diabetes, hyperparathyroidism, and hypofunction of teh pituitary gland) showed significant associations with prior childhood maltreatment exposure, as either ordinal (Additional file 1: Table S6) or binary (Fig. 2) variable, with the most pronounced associations noted for type 2 diabetes. In the analyses of subtypes, we observed excess risks of developing endocrine diseases in most studied glands, particularly adrenal (1.54 (95% CI 1.22–1.95)), pancreatic (1.35 (95% CI 1.29–1.42)), and hypothalamic-pituitary (1.31 (95% CI 1.06–1.61)) glands, among childhood maltreatment-exposed individuals (Fig. 2). In the analysis of the involved neuroendocrine axis, the highest HR was found for HPA-axis-related endocrine diseases (1.38 (95% CI 1.17–1.62)) (Fig. 2). The association was stronger for having multiple endocrine diseases, compared to having one disease only (odds ratio = 1.24 (95% CI 1.19–1.30), 1.35 (95% CI 1.27–1.44), 1.52 (95% CI 1.52–1.53) for having 1, 2, ≥ 3 endocrine diseases, respectively) (Additional file 1: Table S7). Similar results were found for the number of HPA-axis-related endocrine diseases.

Fig. 2
figure 2

Hazard ratios (HRs) with 95% confidence intervals (CIs) for the association between childhood maltreatment (≥ 2 vs < 2) and specific (individual or subtypes) endocrine diseases. * Cox models were adjusted for sex, birth year, ethnicity, country of birth, number of siblings, maternal smoking, being breastfed, and family history of diabetes

Full size image

We performed the mediation analyses for any endocrine disease, type 2 diabetes, and HPA-related endocrine diseases. First, we demonstrated that the observed associations remained valid in the cohorts constructed for mediation analyses (i.e., HR = 1.30 (95% CI 1.24–1.37), 1.38 (95% CI 1.29–1.47), and 1.44 (95% CI 1.17–1.79) for any endocrine disease, type 2 diabetes, and HPA-related endocrine diseases, respectively (Additional file 1: Table S8)). The results of simple mediation analyses for all candidate mediators are shown in Additional file 1: Table S9. For any endocrine disease, we accordingly selected the mediator with the greatest mediated proportion in each hypothesized pathway (i.e., TDI (mediated proportion = 9.14%) for SES, self-rated mental problem (mediated proportion = 20.06%) for psychological factors, BMI (mediated proportion = 21.45%) for lifestyle factors, and HDL (mediated proportion = 11.87%) for biomarkers) in the sequential mediation analysis. Then, sequential mediation analyses on 24 possible temporal orders indicated a direct effect of 41.23 ~ 45.06% between childhood maltreatment and any endocrine disease, as well an indirect effect of 32.19 ~ 41.69%, 7.99 ~ 16.08%, 3.53 ~ 6.48%, and 0.85 ~ 6.71% mediated by psychological adversities, unfavorable lifestyle, suboptimal SES, and biological alterations, respectively (Fig. 3C). With the identical analytic process, we found smaller proportions of direct effect for type 2 diabetes (37.02 ~ 42.35%) and HPA-axis-related endocrine diseases (24.00 ~ 29.19%) (Additional file 1: Tables S10–S12 and Fig. 3C). In addition, although psychological adversities consistently showed the greatest mediated proportion, a notable mediating role of unfavorable lifestyle (10.86 ~ 25.32%) was observed for type 2 diabetes, and suboptimal SES (14.42 ~ 39.33%) was found for HPA-axis-related endocrine diseases (Fig. 3C).

Fig. 3
figure 3

The proposed casual diagram with possible temporal orders and the summarized mediated proportion of four hypothesized pathways from the sequential mediation analyses. * The selected mediators for each hypothesized pathway were Townsend Deprivation Index for suboptimal SES, self-rated mental problem for psychological morbidities, body mass index for unfavorable lifestyle, and high-density lipoprotein for biological alternations in the cohorts for mediation analyses of any endocrine diseases and type 2 diabetes, as well as employment, self-rated mental problem, alcohol and triglycerides in the cohort for mediation analyses of HPA-axis-related endocrine diseases, according to the results of simple mediation analyses

Full size image

The application of competing risk models resulted in largely identical estimates in the association analyses (Additional file 1: Table S13). In addition, neither using an alternative grouping strategy (i.e., childhood maltreatment ≥ 1 vs < 1) nor considering merely individuals without any childhood maltreatment as the reference group (i.e., ≥ 2 vs 0) changed the findings of mediation analyses (Additional file 1: Table S14). Also, in the sensitivity analysis where multiple imputation was applied to deal with the missing values of selected mediators, largely similar results were obtained (Additional file 1: Table S14).

Discussion

In the community-based cohort study of UK Biobank, we found that individuals exposed to childhood maltreatment were at increased risk of developing any and specific (e.g., type 2 diabetes and HPA-axis-related) endocrine diseases in adulthood. The sequential mediation analyses suggested a considerable contribution of studied hypothesized pathways on the observed associations (54.94–76.00%), with the most notable mediating effect by psychological adversities. The importance of suboptimal SES was additionally noted for HPA-axis-related endocrine diseases after exposure to childhood maltreatment, while unfavorable lifestyle was considered critical for the link from childhood maltreatment to type 2 diabetes. These findings shed light on potential targets for interventions, such as early psychological interventions, to reduce the subsequent risk of endocrine diseases in childhood maltreatment-exposed population.

Our findings of an association between childhood maltreatment and subsequent endocrine diseases are consistent with previous studies, most of which, however, focused on diabetes [15]. For instance, the results of two meta-analyses including a total of 20 observational studies concluded that individuals exposed to sexual abuse had 39% [52] and 25% [8] increased risk of diabetes, respectively, later in life. Our study extends the body of existing knowledge by demonstrating an association of childhood maltreatment with any and multiple endocrine diseases, in line with previous research demonstrating the effects of ACEs on various endocrine axes, such as the initial hyperactivity and the subsequent hypoactivity of HPA axis [11], modified basal HPT functions [12], faster sexual maturation [13], and the alternations of hormone secretion [53]. Moreover, the pronounced risk elevation observed for HPA-axis-related endocrine diseases [54] and the high mediated proportion by psychological adversities corroborate the prevailing notion that ACEs can be “biologically embedded” through its persistent or chronic impact on psychopathology and the accompanied HPA-axis dysfunction [27]. Indeed, using disease trajectory analyses, individuals with psychiatric disorders, such as depression [55], showed increased risks of various endocrine diseases. Also, human and animal studies have indicated that progressive dysfunction of HPA axis can lead to insulin resistance and visceral obesity [56] and thereby increase the possibility of developing diabetes, as well as endocrine diseases in all HPA-involved glands (e.g., Cushing syndrome) [57]. Nevertheless, prior efforts are scarce in exploring the association between childhood maltreatment and other endocrine diseases, apart from limited evidence on thyroid diseases or ovarian dysfunction in some specific populations [53, 58]. Our findings, however, corroborated with several studies reporting signals of hormone abnormalities, such as reduced thyroid-hormone T3 [59] and symptoms of polycystic ovary syndrome [58], among individuals exposed to childhood maltreatment.

The underlying mechanisms through which ACEs, including childhood maltreatment, can contribute to the pathogenesis of various somatic diseases, such as endocrine diseases, remain inconclusive. According to the existing literature, we placed our focus of mediation analyses on four proposed mechanistic pathways, i.e., psychological adversities, suboptimal SES, unfavorable lifestyle, and biological alterations. Psychological stress and corresponding lifelong mental morbidity is a well-documented consequence of ACEs [60], which could subsequently affect functions of HPA axis through changes of neuroendocrine cells, brain structures (e.g., amygdala and hippocampus) [61], and epigenetic factors [62]. The role of chronic inflammation and immune suppression, as a result of HPA axis dysfunction and insufficient glucocorticoid signaling, has been well-established in the pathogenesis of type 2 diabetes and endocrine-associated complications [63, 64] through crosstalk between inflammatory system and thyroid or reproductive function [64, 65]. Moreover, a recent study also highlighted the DNA methylation as the novel mechanism linking HPA axis and immune dysregulation [66]. Similarly, SES disadvantages (e.g., lower educational attainment, deprivation, and unemployment status) have consistently been reported among ACEs-exposed individuals [19], which could increase susceptibility to diseases via reduced healthcare seeking and limited access to the healthcare system [18]. Furthermore, as childhood is a sensitive period of neurodevelopment, ACEs may affect behavioral regulations [67] and vulnerabilities for addiction to, e.g., cannabis [67] and alcohol [4] to cope with the emotional pain following childhood maltreatment, while these addictions may contribute to subsequent health hazards, including to the development of endocrine disease. Finally, as biological changes, including chronic inflammation and undermined immune suppression, and other imbalances (e.g., autonomic imbalances), are likely induced by childhood maltreatment (and all abovementioned mediators), it is plausible that the status of biomarkers, particularly inflammatory biomarkers [27, 29, 62], could reflect the extent to which childhood maltreatment influence the pathophysiology towards endocrine diseases.

However, while relatively abundant attempts on clarifying the potential mediating effect of different mechanistic pathway, studies are currently lacking in understanding the joint effects as well as in quantifying the mediated proportions of these pathways. Therefore, our study filled this knowledge gap by demonstrating a considerable proportion (54.96–76.00%) of indirect effect on the associations between childhood maltreatment and endocrine diseases, particularly type 2 diabetes and HPA-axis-related endocrine diseases. In addition to the greatest mediating proportion observed for psychological adversities, as expected, our results also demonstrate a mediating role of unfavorable lifestyle for type 2 diabetes. This is in line with previous reports and further underscores the importance of behavior modifications for diabetes prevention in childhood maltreatment-exposed individuals. Likewise, the notable mediating effect of suboptimal SES on the link between childhood maltreatment and HPA-axis-related endocrine diseases is plausible, as lower SES in adulthood could act as an adverse stimulus on the stress response system. Individuals with lower SES might also have less social support and material resources to handle stressful events [68]. Finally, because the biological characteristics were measured only once and at an average age of 56 years in the UK Biobank, the smaller mediated proportion of biological pathways observed in the present study should not preclude the importance of examining biomarkers in future studies.

The strengths of the present study include the large study population and longitudinal data derived from the UK Biobank. For instance, the diagnoses of endocrine diseases were prospectively obtained through data linkage with nationwide healthcare registers, which minimizes the possibility of information bias. Also, with most candidate mediators measured at recruitment, the clear temporal order between childhood maltreatment, mediating factors, and the diagnosis of endocrine diseases provides a solid basis for the mediation analyses. In addition, in contrast to the traditional mediation analyses requiring strong and unverifiable assumptions, the application of sequential mediation analyses enables the measure of path-specific mediated proportion accounting for different sequential assumptions for all involved pathways. Consequently, the range of estimates obtained under 24 possible temporal orders could, from a global perspective, be used to elucidate the relative importance of several involved mediators. This analytic strategy has been successfully applied in projects with similar objectives, e.g., evaluating mediating factors important for the association between childhood maltreatment and cardiovascular diseases [6].

There are also limitations in the study. First, the status of childhood maltreatment was defined based on five self-rated questions collected many years after actual exposure (e.g., in 2016). Therefore, the concern of misclassification due to recall bias may to some extent distort the observed associations. However, several prior studies have demonstrated a certain consistency between retrospective and prospective measures of childhood maltreatment, although the use of retrospectively reported childhood maltreatment poses the risk of underestimating the impact of ACEs on objectively measured life adversities [69, 70]. Additionally, apart from childhood maltreatment, other types of ACEs were not measured or addressed in the present study, such as those ones related to childhood household dysfunctions (e.g., childhood household mental illness and childhood parental death) that have been linked with adverse health consequences [8]. Second, despite the enrichment of available data on candidate mediators in the UK Biobank, these factors were measured only once at recruitment which could not accurately reflect the long-term status. Exposure (other than childhood maltreatment) might also have influenced these mediators and cannot fully considered. Third, our mediation analyses were performed for four pre-defined pathways. Other mechanistic pathways might have also contributed to the observed association and need to be investigated in future explorations. Last, the UK Biobank sample is not representative of the general UK population, as the participants were more likely white, healthier, older, and less socioeconomically deprived than nonparticipants [31]. Also, we have noted a lower SES status for the participants who provided data on childhood maltreatment, compared to those without. The generalizability of our findings to the whole UK or other populations needs to be evaluated further.

Conclusions

In conclusion, based on the community-based prospective cohort of UK Biobank, our study demonstrated the association between childhood maltreatment and multiple endocrine diseases is primarily driven by the mediating role of psychological adversities. The importance of suboptimal SES was additionally noted for HPA-axis-related endocrine diseases and of unfavorable lifestyle was noted for type 2 diabetes. These findings therefore carry an important message for prevention, namely that the targeted interventions (such as early psychological interventions) may reduce the downstream risk of endocrine diseases in the childhood maltreatment-exposed population.

Availability of data and materials

Data from UK Biobank are available per the researchers request (https://www.ukbiobank.ac.uk/enable-your-research/apply-for-access). This research was done using the UK Biobank Resource under Application 54803 (approved on October 29, 2019).

Abbreviations

ACE:

Adverse childhood experience

BMI:

Body mass index

CI:

Confidence interval

HDL:

High-density lipoprotein

HPA:

Hypothalamic-pituitary-adrenal

HPG:

Hypothalamic-pituitary–gonadal

HPT:

Hypothalamic-pituitary-thyroid

HR:

Hazard ratio

ICD-10:

International Classification of Diseases 10th edition

LDL:

Low-density lipoprotein

SD:

Standard deviation

SES:

Socioeconomic status

TDI:

Townsend Deprivation Index

References

  1. Godoy LC, Frankfurter C, Cooper M, Lay C, Maunder R, Farkouh ME. Association of adverse childhood experiences with cardiovascular disease later in life: a review. JAMA Cardiol. 2021;6(2):228–35.

    Article  PubMed  Google Scholar 

  2. Perales J, Olaya B, Fernandez A, Alonso J, Vilagut G, Forero CG, San L, Alda JA, Haro JM. Association of childhood adversities with the first onset of mental disorders in Spain: results from the ESEMeD project. Soc Psychiatry Psychiatr Epidemiol. 2013;48(3):371–84.

    Article  PubMed  Google Scholar 

  3. Wiehn J, Hornberg C, Fischer F. How adverse childhood experiences relate to single and multiple health risk behaviours in German public university students: a cross-sectional analysis. BMC Public Health. 2018;18(1):1005.

    Article  PubMed  PubMed Central  Google Scholar 

  4. Hughes K, Ford K, Bellis MA, Glendinning F, Harrison E, Passmore J. Health and financial costs of adverse childhood experiences in 28 European countries: a systematic review and meta-analysis. Lancet Public Health. 2021;6(11):e848–57.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Hughes K, Bellis MA, Hardcastle KA, Sethi D, Butchart A, Mikton C, Jones L, Dunne MP. The effect of multiple adverse childhood experiences on health: a systematic review and meta-analysis. Lancet Public Health. 2017;2(8):e356–66.

    Article  PubMed  Google Scholar 

  6. Soares AG, Howe LD, Heron J, Hammerton G, Rich-Edwards J, Magnus MC, Halligan SL, Fraser A. How does childhood maltreatment influence cardiovascular disease? A sequential causal mediation analysis. Int J Epidemiol. 2022;51(2):555–66.

    Article  PubMed  Google Scholar 

  7. Maiden P. 5. Lifestyle management: standards of medical care in diabetes-2019. Diabetes Care. 2019;42(Suppl 1):S46–60.

    Google Scholar 

  8. Zhu S, Shan S, Liu W, Li S, Hou L, Huang X, Liu Y, Yi Q, Sun W, Tang K, et al. Adverse childhood experiences and risk of diabetes: a systematic review and meta-analysis. J Glob Health. 2022;12:04082.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Schroeder K, Schuler BR, Kobulsky JM, Sarwer DB. The association between adverse childhood experiences and childhood obesity: a systematic review. Obes Rev. 2021;22(7):e13204.

    Article  PubMed  PubMed Central  Google Scholar 

  10. Danese A, Tan M. Childhood maltreatment and obesity: systematic review and meta-analysis. Mol Psychiatry. 2014;19(5):544–54.

    Article  CAS  PubMed  Google Scholar 

  11. Dempster KS, O’Leary DD, MacNeil AJ, Hodges GJ, Wade TJ. Linking the hemodynamic consequences of adverse childhood experiences to an altered HPA axis and acute stress response. Brain Behav Immun. 2021;93:254–63.

    Article  CAS  PubMed  Google Scholar 

  12. Jaimes-Hoy L, Pérez-Maldonado A, NarváezBahena E, de la Cruz Guarneros N, Rodríguez-Rodríguez A, Charli JL, Soberón X, Joseph-Bravo P. Sex dimorphic changes in trh gene methylation and thyroid-axis response to energy demands in maternally separated rats. Endocrinology. 2021;162(8):bqab110.

    Article  PubMed  Google Scholar 

  13. Sun Y, Fang J, Wan Y, Su P, Tao F. Association of early-life adversity with measures of accelerated biological aging among children in China. JAMA Netw Open. 2020;3(9):e2013588.

    Article  PubMed  PubMed Central  Google Scholar 

  14. Tay CT, Teede HJ, Loxton D, Kulkarni J, Joham AE. Psychiatric comorbidities and adverse childhood experiences in women with self-reported polycystic ovary syndrome: an Australian population-based study. Psychoneuroendocrinology. 2020;116:104678.

    Article  PubMed  Google Scholar 

  15. Zhang Y, Yin Y, Zhang X, Ye J, Zhang J. Association of adverse childhood experiences with diabetes: a systematic review and meta-analysis. J Diabetes Complications. 2022;36(10):108289.

    Article  PubMed  Google Scholar 

  16. Strenth CR, Mo A, Kale NJ, Day PG, Gonzalez L, Green R, Cruz II, Schneider FD. Adverse childhood experiences and diabetes: testing violence and distress mediational pathways in family medicine patients. J Interpers Violence. 2022;37(23–24):Np23035–np23056.

    Article  PubMed  Google Scholar 

  17. Adverse Childhood Experiences. About the CDC-Kaiser ACE Study. https://www.cdc.gov/violenceprevention/aces/about.html. Accessed 2023.

  18. Metzler M, Merrick MT, Klevens J, Ports KA, Ford DC. Adverse childhood experiences and life opportunities: shifting the narrative. Child Youth Serv Rev. 2017;72:141–9.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Font SA, Maguire-Jack K. Pathways from childhood abuse and other adversities to adult health risks: the role of adult socioeconomic conditions. Child Abuse Negl. 2016;51:390–9.

    Article  PubMed  Google Scholar 

  20. Merrick MT, Ford DC, Ports KA, Guinn AS, Chen J, Klevens J, Metzler M, Jones CM, Simon TR, Daniel VM, et al. Vital signs: estimated proportion of adult health problems attributable to adverse childhood experiences and implications for prevention - 25 States, 2015–2017. MMWR Morb Mortal Wkly Rep. 2019;68(44):999–1005.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Houtepen LC, Heron J, Suderman MJ, Fraser A, Chittleborough CR, Howe LD. Associations of adverse childhood experiences with educational attainment and adolescent health and the role of family and socioeconomic factors: a prospective cohort study in the UK. PLoS Med. 2020;17(3):e1003031.

    Article  PubMed  PubMed Central  Google Scholar 

  22. Brewerton TD. Mechanisms by which adverse childhood experiences, other traumas and PTSD influence the health and well-being of individuals with eating disorders throughout the life span. J Eat Disord. 2022;10(1):162.

    Article  PubMed  PubMed Central  Google Scholar 

  23. Gaston SA, Riley NM, Parks CG, Woo JMP, Sandler DP, Jackson CL. Racial/ethnic differences in associations between traumatic childhood experiences and both metabolic syndrome prevalence and type 2 diabetes risk among a cohort of U.S. women. Diabetes Care. 2023;46(2):341–50.

    Article  CAS  PubMed  Google Scholar 

  24. Deschênes SS, Graham E, Kivimäki M, Schmitz N. Adverse childhood experiences and the risk of diabetes: examining the roles of depressive symptoms and cardiometabolic dysregulations in the Whitehall II cohort study. Diabetes Care. 2018;41(10):2120–6.

    Article  PubMed  PubMed Central  Google Scholar 

  25. Danese A, McEwen BS. Adverse childhood experiences, allostasis, allostatic load, and age-related disease. Physiol Behav. 2012;106(1):29–39.

    Article  CAS  PubMed  Google Scholar 

  26. Juruena MF, Eror F, Cleare AJ, Young AH. The role of early life stress in HPA axis and anxiety. Adv Exp Med Biol. 2020;1191:141–53.

    Article  CAS  PubMed  Google Scholar 

  27. Berens AE, Jensen SKG, Nelson CA 3rd. Biological embedding of childhood adversity: from physiological mechanisms to clinical implications. BMC Med. 2017;15(1):135.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Shonkoff JP. Capitalizing on advances in science to reduce the health consequences of early childhood adversity. JAMA Pediatr. 2016;170(10):1003–7.

    Article  PubMed  Google Scholar 

  29. Soares S, Rocha V, Kelly-Irving M, Stringhini S, Fraga S. Adverse childhood events and health biomarkers: a systematic review. Front Public Health. 2021;9:649825.

    Article  PubMed  PubMed Central  Google Scholar 

  30. Huang YT, Yang HI. Causal mediation analysis of survival outcome with multiple mediators. Epidemiology. 2017;28(3):370–8.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Fry A, Littlejohns TJ, Sudlow C, Doherty N, Adamska L, Sprosen T, Collins R, Allen NE. Comparison of sociodemographic and health-related characteristics of UK Biobank participants with those of the general population. Am J Epidemiol. 2017;186(9):1026–34.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Sudlow C, Gallacher J, Allen N, Beral V, Burton P, Danesh J, Downey P, Elliott P, Green J, Landray M, et al. UK biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med. 2015;12(3):e1001779. https://doi.org/10.1371/journal.pmed.1001779.

    Article  PubMed  PubMed Central  Google Scholar 

  33. UK Biobank. Data providers and dates of data availability. https://biobank.ndph.ox.ac.uk/showcase/exinfo.cgi?src=Data_providers_and_dates. Accessed 2023.

  34. Grabe HJ, Schulz A, Schmidt CO, Appel K, Driessen M, Wingenfeld K, Barnow S, Spitzer C, John U, Berger K, et al. A brief instrument for the assessment of childhood abuse and neglect: the childhood trauma screener (CTS). Psychiatr Prax. 2012;39(3):109–15.

    PubMed  Google Scholar 

  35. Soares ALG, Hammerton G, Howe LD, Rich-Edwards J, Halligan S, Fraser A. Sex differences in the association between childhood maltreatment and cardiovascular disease in the UK Biobank. Heart. 2020;106(17):1310–6.

    Article  PubMed  Google Scholar 

  36. Dirksen J, Pinilla-Roncancio M, Wehrmeister FC, Ferreira LZ, Vidaletti LP, Kirkby K, Koller TS, Schlotheuber A, Tapia H, Fuertes CV, et al. Correction: exploring the potential for a new measure of socioeconomic deprivation status to monitor health inequality. Int J Equity Health. 2022;21(1):100.

    Article  PubMed  PubMed Central  Google Scholar 

  37. Li Y, Shang Y, Yang Y, Hou C, Yang H, Hu Y, Zhang J, Song H, Zhang W. Association of childhood adversities with psychosocial difficulties among Chinese children and adolescents. Int J Epidemiol. 2023;52:1887–97.

    Article  PubMed  PubMed Central  Google Scholar 

  38. Yilmaz Z, Larsen JT, Nissen JB, Crowley JJ, Mattheisen M, Bulik CM, Petersen LV. The role of early-life family composition and parental socio-economic status as risk factors for obsessive-compulsive disorder in a Danish national cohort. J Psychiatr Res. 2022;149:18–27.

    Article  PubMed  PubMed Central  Google Scholar 

  39. Mackenbach JP, Kesteloot K, Defever M, Cleemput I. Health and deprivation. In: Townsend P, Phillimore P, Beattie A, editors. Inequality and the North. London: Croom Helm Ltd. 1987. p. 221. ISBN 0–7099–4352–0, [pound sign]8.95. 1988.

  40. Howe LD, Kanayalal R, Harrison S, Beaumont RN, Davies AR, Frayling TM, Davies NM, Hughes A, Jones SE, Sassi F, et al. Effects of body mass index on relationship status, social contact and socio-economic position: Mendelian randomization and within-sibling study in UK Biobank. Int J Epidemiol. 2020;49(4):1173–84.

    Article  PubMed  Google Scholar 

  41. IPAQ scoring protocol - International Physical Activity Questionnaire. https://sites.google.com/site/theiqaq/scoring-protocol. Accessed 2023.

  42. Patane S. Sleep patterns, genetic susceptibility, and incident cardiovascular disease: the role of miRNAs. Eur Heart J. 2020;41(26):2501.

    Article  CAS  PubMed  Google Scholar 

  43. Arthur RS, Wang T, Xue X, Kamensky V, Rohan TE. Genetic factors, adherence to healthy lifestyle behavior, and risk of invasive breast cancer among women in the UK biobank. J Natl Cancer Inst. 2020;112(9):893–901.

    Article  PubMed  PubMed Central  Google Scholar 

  44. Shi B, Choirat C, Coull BA, VanderWeele TJ, Valeri L. CMAverse: a suite of functions for reproducible causal mediation analyses. Epidemiology. 2021;32(5):e20–2.

    Article  PubMed  Google Scholar 

  45. Robins JM, Greenland S. Identifiability and exchangeability for direct and indirect effects. Epidemiology. 1992;3(2):143–55.

    Article  CAS  PubMed  Google Scholar 

  46. VanderWeele TJ, Vansteelandt S. Mediation analysis with multiple mediators. Epidemiol Methods. 2014;2(1):95–115.

    CAS  PubMed  PubMed Central  Google Scholar 

  47. VanderWeele TJ, Vansteelandt S. Conceptual issues concerning mediation, interventions and composition. Stat Interface. 2009;2(4):457–68.

    Article  MathSciNet  Google Scholar 

  48. Huang YT, Yang HI, Liu J, Lee MH, Freeman JR, Chen CJ. Mediation analysis of hepatitis B and C in relation to hepatocellular carcinoma risk. Epidemiology. 2016;27(1):14–20.

    Article  PubMed  Google Scholar 

  49. Lange T, Hansen JV. Direct and indirect effects in a survival context. Epidemiology. 2011;22(4):575–81.

    Article  PubMed  Google Scholar 

  50. Tai AS, Lin PH, Huang YT, Lin SH. Path-specific effects in the presence of a survival outcome and causally ordered multiple mediators with application to genomic data. Stat Methods Med Res. 2022;31(10):1916–33.

    Article  MathSciNet  PubMed  Google Scholar 

  51. VanderWeele TJ. Concerning the consistency assumption in causal inference. Epidemiology. 2009;20(6):880–3.

    Article  PubMed  Google Scholar 

  52. Huang H, Yan P, Shan Z, Chen S, Li M, Luo C, Gao H, Hao L, Liu L. Adverse childhood experiences and risk of type 2 diabetes: a systematic review and meta-analysis. Metabolism. 2015;64(11):1408–18.

    Article  CAS  PubMed  Google Scholar 

  53. Moog NK, Heim CM, Entringer S, Kathmann N, Wadhwa PD, Buss C. Childhood maltreatment is associated with increased risk of subclinical hypothyroidism in pregnancy. Psychoneuroendocrinology. 2017;84:190–6.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  54. Simon L, Admon R. From childhood adversity to latent stress vulnerability in adulthood: the mediating roles of sleep disturbances and HPA axis dysfunction. Neuropsychopharmacology. 2023;48(10):1425–35.

    Article  PubMed  Google Scholar 

  55. Han X, Hou C, Yang H, Chen W, Ying Z, Hu Y, Sun Y, Qu Y, Yang L, Valdimarsdottir UA, et al. Disease trajectories and mortality among individuals diagnosed with depression: a community-based cohort study in UK Biobank. Mol Psychiatry. 2021;26(11):6736–46.

    Article  PubMed  PubMed Central  Google Scholar 

  56. Rosmond R. Stress induced disturbances of the HPA axis: a pathway to type 2 diabetes? Med Sci Monit. 2003;9(2):RA35–39.

    PubMed  Google Scholar 

  57. Raff H, Carroll T. Cushing’s syndrome: from physiological principles to diagnosis and clinical care. J Physiol. 2015;593(3):493–506.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  58. Pringle D, Suliman S, Seedat S, van den Heuvel LL. The impact of childhood maltreatment on women’s reproductive health, with a focus on symptoms of polycystic ovary syndrome. Child Abuse Negl. 2022;133:105831.

    Article  PubMed  Google Scholar 

  59. Brindle RC, Pearson A, Ginty AT. Adverse childhood experiences (ACEs) relate to blunted cardiovascular and cortisol reactivity to acute laboratory stress: a systematic review and meta-analysis. Neurosci Biobehav Rev. 2022;134:104530.

    Article  CAS  PubMed  Google Scholar 

  60. Mehta D, Kelly AB, Laurens KR, Haslam D, Williams KE, Walsh K, Baker PRA, Carter HE, Khawaja NG, Zelenko O, et al. Child maltreatment and long-term physical and mental health outcomes: an exploration of biopsychosocial determinants and implications for prevention. Child Psychiatry Hum Dev. 2023;54(2):421–35.

    Article  PubMed  Google Scholar 

  61. Murphy F, Nasa A, Cullinane D, Raajakesary K, Gazzaz A, Sooknarine V, Haines M, Roman E, Kelly L, O’Neill A, et al. Childhood trauma, the HPA axis and psychiatric illnesses: a targeted literature synthesis. Front Psychiatry. 2022;13:748372.

    Article  PubMed  PubMed Central  Google Scholar 

  62. Morris G, Berk M, Maes M, Carvalho AF, Puri BK. Socioeconomic deprivation, adverse childhood experiences and medical disorders in adulthood: mechanisms and associations. Mol Neurobiol. 2019;56(8):5866–90.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  63. Donath MY, Meier DT, Boni-Schnetzler M. Inflammation in the pathophysiology and therapy of cardiometabolic disease. Endocr Rev. 2019;40(4):1080–91.

    Article  PubMed  PubMed Central  Google Scholar 

  64. Popovic M, Sartorius G, Christ-Crain M. Chronic low-grade inflammation in polycystic ovary syndrome: is there a (patho)-physiological role for interleukin-1? Semin Immunopathol. 2019;41(4):447–59.

    Article  PubMed  Google Scholar 

  65. Lasa M, Contreras-Jurado C. Thyroid hormones act as modulators of inflammation through their nuclear receptors. Front Endocrinol. 2022;13:937099.

    Article  Google Scholar 

  66. Palma-Gudiel H, Prather AA, Lin J, Oxendine JD, Guintivano J, Xia K, Rubinow DR, Wolkowitz O, Epel ES, Zannas AS. HPA axis regulation and epigenetic programming of immune-related genes in chronically stressed and non-stressed mid-life women. Brain Behav Immun. 2021;92:49–56.

    Article  CAS  PubMed  Google Scholar 

  67. Strathearn L, Giannotti M, Mills R, Kisely S, Najman J, Abajobir A. Long-term cognitive, psychological, and health outcomes associated with child abuse and neglect. Pediatrics. 2020;146(4):e20200438.

    Article  PubMed  Google Scholar 

  68. Baum A, Garofalo JP, Yali AM. Socioeconomic status and chronic stress. Does stress account for SES effects on health? Ann N Y Acad Sci. 1999;896:131–44.

    Article  CAS  PubMed  ADS  Google Scholar 

  69. Reuben A, Moffitt TE, Caspi A, Belsky DW, Harrington H, Schroeder F, Hogan S, Ramrakha S, Poulton R, Danese A. Lest we forget: comparing retrospective and prospective assessments of adverse childhood experiences in the prediction of adult health. J Child Psychol Psychiatry. 2016;57(10):1103–12.

    Article  PubMed  PubMed Central  Google Scholar 

  70. Patten SB, Wilkes TC, Williams JV, Lavorato DH, El-Guebaly N, Schopflocher D, Wild C, Colman I, Bulloch AG. Retrospective and prospectively assessed childhood adversity in association with major depression, alcohol consumption and painful conditions. Epidemiol Psychiatr Sci. 2015;24(2):158–65.

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

This research has been conducted using the UK Biobank Resource under Application 54803. This work uses data provided by patients and collected by the NHS as part of their care and support. This research used data assets made available by National Safe Haven as part of the Data and Connectivity National Core Study, led by Health Data Research UK in partnership with the Office for National Statistics and funded by UK Research and Innovation (grant ref: MC_PC_20029 and MC_PC_20058). We express our sincere thanks to all the sponsors and team members involved in West China Biomedical Big Data Center and Med-X Center for Informatics, Sichuan University.

Funding

This work was supported by 1.3.5 Project for Disciplines of Excellence, West China Hospital, Sichuan University (No. ZYYC21005 to HS) and the EU Horizon 2020 Research and Innovation Action Grant (No. 847776 to UV and FF). The funders had no role in study design, data collection and analysis, interpretation of the data, approval of the manuscript, decision to publish, or preparation of the manuscript.

Author information

Author notes

  1. Shu Wen and Jianwei Zhu contributed equally to this work.

Authors and Affiliations

  1. Mental Health Center and West China Biomedical Big Data Center West China Hospital, Sichuan University, Guo Xue Lane 37, Chengdu, China

    Shu Wen, Xin Han, Yuchen Li, Haowen Liu, Huazhen Yang, Can Hou, Shishi Xu, Junren Wang, Yao Hu, Yuanyuan Qu, Di Liu & Huan Song

  2. Med-X Center for Informatics, Sichuan University, Chengdu, China

    Shu Wen, Xin Han, Haowen Liu, Huazhen Yang, Can Hou, Junren Wang, Yao Hu, Yuanyuan Qu, Di Liu & Huan Song

  3. Department of Critical Care Medicine, West China Hospital, Sichuan University, Chengdu, China

    Shu Wen

  4. Department of Orthopedic Surgery, West China Hospital, Sichuan University, Chengdu, China

    Jianwei Zhu

  5. Mental Health Center, West China Hospital, Sichuan University, Chengdu, China

    Yuchen Li

  6. Division of Endocrinology & Metabolism, West China Hospital, Sichuan University, Chengdu, China

    Shishi Xu

  7. Sichuan University - Pittsburgh Institute, Sichuan University, Chengdu, China

    Di Liu

  8. Center of Public Health Sciences, Faculty of Medicine, University of Iceland, Reykjavík, Iceland

    Thor Aspelund, Unnur A. Valdimarsdóttir & Huan Song

  9. Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden

    Fang Fang & Unnur A. Valdimarsdóttir

  10. Department of Epidemiology, Harvard T H Chan School of Public Health, Boston, MA, USA

    Unnur A. Valdimarsdóttir

Authors
  1. Shu Wen
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Jianwei Zhu
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Xin Han
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Yuchen Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Haowen Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Huazhen Yang
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Can Hou
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Shishi Xu
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Junren Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Yao Hu
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Yuanyuan Qu
    View author publications

    You can also search for this author in PubMed Google Scholar

  12. Di Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  13. Thor Aspelund
    View author publications

    You can also search for this author in PubMed Google Scholar

  14. Fang Fang
    View author publications

    You can also search for this author in PubMed Google Scholar

  15. Unnur A. Valdimarsdóttir
    View author publications

    You can also search for this author in PubMed Google Scholar

  16. Huan Song
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

HS, SW, and JZ were responsible for the study’s concept and design. HY, YH, and YQ did the data and project management. XH, YL, and SW did the data cleaning and analysis. HL, CH, SX, JW, and DL were involved in methodology and supervision. SW, JZ, XH, TA, UAV, FF, and HS interpreted the data. SW, JZ, UAV, and HS drafted the manuscript. FF, UVA, and HS reviewed and revised the manuscript. All the authors read and approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

Authors’ Twitter handles

Twitter handles: @JoyLab14169974 (Joy Lab).

Corresponding author

Correspondence to Huan Song.

Ethics declarations

Ethics approval and consent to participate

The UK Biobank has full ethical approval from the NHS National Research Ethics Service (16/NW/0274), with consent obtained from all the participants.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: Table S1.

The distribution of sociodemographic factors of UK Biobank participants. Table S2. The definition of exposure, covariates and candidate mediators in the present study. Table S3. International Classification of Diseases 10th edition (ICD-10) codes used to ascertain endocrine diseases in this study. Table S4. Baseline characteristics of study population in the cohort for mediation analyses (n=138,498). Table S5. Distribution of candidate mediators among individuals with different levels of childhood maltreatment experience (≥ 2 vs <2) in the cohorts for mediation analyses. Table S6. Hazard ratios (HRs) with 95% confidence intervals (CIs) for the association between childhood maltreatment experience (=1 and ≥ 2 vs 0) and specific (individual diagnoses or subtypes) endocrine diseases. Table S7. The association of childhood maltreatment experience (≥2 vs <2) and the number of any or HPA-axis-related endocrine diseases. Table S8. Hazard ratios (HRs) with 95% confidence intervals (CIs) for the association between childhood maltreatment experience (≥2 vs <2) with any and specific (individual diagnoses or subtypes) endocrine diseases in the cohorts for mediation analyses. Table S9. Estimated mediated proportion of candidate mediators for the association of childhood maltreatment with any and specific endocrine diseases, using simple mediation analyses. Table S10. Estimated mediated proportion of selected mediators, by different temporal orders, for the association between childhood maltreatment and any endocrine diseases, using sequential mediation analyses. Table S11. Estimated mediated proportion of selected mediators, by different temporal orders, for the association between childhood maltreatment and type 2 diabetes, using sequential mediation analyses. Table S12. Estimated mediated proportion of selected mediators, by different temporal orders, for the association between childhood maltreatment and HPA-axis-related endocrine diseases, using sequential mediation analyses. Table S13. The competing risk of death in the relationship between childhood maltreatment and endocrine diseases. Table S14. Sensitivity analyses for the summarized mediated proportions of four hypothesized pathways from the sequential mediation analyses.

Additional file 2.

Supplementary method.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Wen, S., Zhu, J., Han, X. et al. Childhood maltreatment and risk of endocrine diseases: an exploration of mediating pathways using sequential mediation analysis. BMC Med 22, 59 (2024). https://doi.org/10.1186/s12916-024-03271-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12916-024-03271-9

Keywords

BMC Medicine

ISSN: 1741-7015

Contact us