Abstract
Background
South African women experience high rates of abuse and cardiometabolic diseases, but research on their relationship is less investigated. This study examined the associations of exposure to abuse in childhood and adulthood with cardiometabolic risk factors [body mass index (BMI), waist circumference (WC), systolic (SBP) and diastolic blood pressure (DBP), total cholesterol (TC), HbA1c] over three years among women aged 18–40 years.
Methods
The longitudinal Rape Impact Cohort Evaluation (RICE) study, conducted between 2014 and 2019, examined the associations of self-reported exposures to childhood abuse (CA) (any, sexual, physical, emotional CA and parental neglect < 18 years of age), lifetime intimate partner violence (IPV) (any, sexual, physical, emotional and economic IPV) and lifetime non-partner rape with cardiometabolic outcomes. These were explored using linear mixed-effects models, with the inclusion of the interaction terms, (1) ‘abuse/trauma*rape-exposed’; (2) ‘abuse/trauma*baseline-rape exposure*time’ to account for potential effects of baseline rape-exposure or time.
Results
Cardiometabolic data were collected at baseline (N = 1617; mean age 25.3 years), 12 (N = 1178), 24 (N = 925) and 36 months (N = 571). Exposures to any CA (β = 0.79; se = 0.27; p = 0.003), physical CA (β = 0.68; se = 0.27; p = 0.013) and greater frequency of physical CA (β = 0.73; se = 0.36; p = 0.042) were associated with rising BMI over 3 years of follow-up. There was no evidence of significant effects between baseline-rape exposure or time and most CA types in the associations examined (p > 0.05 for all interaction tests). An exception was found for sexual CA, where baseline-rape exposure influenced its association with WC (p = 0.010 for the interaction test). Time also affected the associations of emotional CA with both WC and BMI (ps ≤ 0.039 for the interaction tests). Exposures to IPV or lifetime non-partner rape were not significantly associated with increased changes in any cardiometabolic variables investigated, and no effects of baseline rape-exposure or time.
Conclusions
This prospective analysis demonstrated that childhood abuse experiences were associated with increased BMI levels over 3 years of follow-up in young South African women. Further research over a longer period is required to clearly delineate the effect of abuse exposure on cardiometabolic diseases.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12889-025-24049-5.
Keywords: Childhood abuse, Intimate partner violence, Rape, Non-partner sexual violence, Cardiometabolic diseases risk, Women, South Africa
Introduction
Abuse against women, whether occurring in childhood (sexual, physical, emotional and parental neglect) or adulthood as intimate partner violence (IPV) (sexual, physical, emotional and economic) or as non-partner rape, has been a public health concern. Globally, approximately one-third of women experience at least one form of abuse in their lifetime, with young women being particularly at risk [1, 2]. Data from 2000 to 2018 show that, on average, 27% of women aged 15–49 years have experienced IPV [3], and 7% have been subjected to non-partner rape [2]. Additionally, nearly 60% of women reported exposures to physical or emotional abuse during childhood, and 20% reported experiencing sexual childhood abuse.
In South Africa, the prevalence of abuse among women is particularly high: 12–68% reported experiencing childhood abuse [4], 26–65% experiencing IPV [5], and 12–25% reported non-partner rape [2, 6]. While each of these abuse types can take various forms, all have negative effects on short- and long-term health [7–10]. Together with poor mental health, abuse affects physical health, including cardiometabolic diseases such as obesity, hypertension and diabetes [11–13].
There are multiple pathways, both direct and indirect, whereby abuse is linked to cardiometabolic diseases and risk factors [13, 14]. Experiencing abuse, particularly during childhood, can lead to long-term stress responses that disrupt physiological functions [15, 16]. This can result in alterations in neural function and structure, as well as increased activation of neurobiological systems [16, 17]. These changes may contribute to the accumulation of fat and blood lipids, as well as dysregulation of glucose metabolism, which can accelerate the development of cardiometabolic disease [9, 17, 18]. Indirectly, the uptake of unhealthy behaviours (e.g., poor diet, sedentary behaviour, smoking and alcohol consumption) as coping strategies can lead to the development of cardiometabolic disease and its risk factors [13, 19–21].
Moreover, the effects of abuse may not be uniform as different abuse types, frequency and severity may activate biological mechanisms differently [13, 15]. Hence, recent systematic reviews showed mixed results for the association of abuse with selected cardiovascular disease risk (e.g., obesity, hypertension, diabetes and dyslipidaemia), attributable to outcome type and measure, and time of abuse [22–24]. These reviews further highlighted the lack of longitudinal studies investigating the association between adult abuse and cardiometabolic diseases [22–24].
The potential association of abuse with cardiometabolic diseases is pertinent in South Africa because stroke and ischaemic heart disease and their risk factors of hypertension and diabetes are among the top ten leading causes of mortality in the country [25, 26]. Understanding the relationship between these two burdens is imperative for early screening and interventions to prevent the development of cardiometabolic diseases among South Africans. However, there have been no prospective studies examining the link between childhood and adulthood abuse and cardiometabolic diseases in young South African women [27, 28]. Accordingly, this study examined, among 18-40-year-old women, the impact of exposures to abuse on cardiometabolic outcomes (body mass index (BMI), waist circumference (WC), blood pressure (BP), glycated haemoglobin (HbA1c), and lipids profiles) over a 3-year period. The abuse exposures examined included childhood abuse (sexual, physical and emotional childhood abuse and parental neglect); lifetime IPV (sexual, physical, emotional and economic IPV), and lifetime non-partner rape.
Methods
Participants and setting
This study analysed the longitudinal data from the Rape Impact Cohort Evaluation (RICE) study that followed women from an index rape episode (rape-exposed group) and an age-matched group of non-rape-exposed women (control group) for 3 years. The study was conducted in the greater Durban area, KwaZulu-Natal Province, South Africa. The study protocol, published previously, describes the detailed methods [29]. The rape-exposed cohort comprised women who experienced a rape event and presented to the sexual assault service centres within 20 days of the event, regardless of prior exposure to abuse. This time window was chosen as it allowed us to document HIV-infection at baseline. This was important as a main aim of the RICE study was to examine HIV acquisition after a rape. In the control group, women who had not experienced a rape (non-rape exposed) within 20 days were recruited from family planning and well-baby clinic services in the same locality as the rape services. Participants were enrolled if they were between 18 and 40 years of age. Women who were older than 40 years of age were excluded, as we were interested in assessing HIV acquisition and HIV incidence is much lower in this age group relative to younger women [29].
Although the RICE study has two arms, we used the data from the pooled group of women to maximise the statistical power to achieve the desired aims. More details are available in the statistical analysis section. The South African Medical Research Council Research Ethics Committee provided ethical approval for this study (SAMRC; EC019-10/2013), and all participants provided informed consent.
Data collection
Data were collected at four time-points i.e., baseline, 12 months, 24 months and 36 months between 2014 and 2019. The repeated measures at 12-month interval were selected to capture the real-time changes in clinical and biochemical parameters in post-rape, comparing rape-exposed and non-exposed women, which would give more power to the study [29]. Trained female staff who worked with the post-rape service providers introduced the RICE study to the women attending the services. Women who were interested in participating were invited to the RICE study clinics, where written informed consent was taken. Trained female staff administered the interview-questionnaires in the participant’s preferred language using an electronic data capturing and management tool (Bryant System).
Measures
Childhood abuse
The Childhood Trauma Questionnaire Short Form (CTQ-SF) scale explored a history of exposure to sexual, physical and emotional abuse/maltreatment, and parental neglect before 18 years of age [30, 31]. The presence or absence of each childhood abuse type was assessed, and overall childhood abuse was defined as exposure to any of the four types of childhood abuse. Exposure to multiple childhood abuse types was classified from zero (no exposure) to four (exposure to all four types of childhood abuse). The frequency of exposure to each childhood abuse type was categorised as never (1), sometimes (2), often (3), or very often (4). The sum of the frequency of exposure yielded the severity score for each childhood abuse type, with sexual childhood abuse ranging 4–16, each other three types ranging 3–12. Scores for all four childhood abuse types were summed to yield overall severity scores ranging from 13 to 52, with higher scores indicating greater severity of exposure to childhood abuse.
Lifetime IPV
IPV type: Four types of IPV by a current or previous partner: sexual abuse (four items), physical abuse (five items), emotional abuse (seven items) and economic abuse (four items) were ascertained using the WHO multi-country study questionnaire [32], previously validated in South Africa [33]. For each IPV type, answers to each of the items were summed. Any IPV was defined as exposure to any of the four types of IPV. The frequency of exposure to each type of IPV was categorised as never (1), once (2), few (3), or many (4). The sum of the frequency of exposure for each type yielded the severity scores, ranged 4–16 (sexual and economical IPV), 5–20 (physical IPV) and 7–28 (emotional IPV). Scores of the frequency of exposure for all four IPV types were summed to yield overall scores ranging from 20 to 80, with higher scores indicating greater severity of exposure to IPV. Exposure to multiple IPV types was classified from zero (no exposure) to four (exposure to all 4 types of IPV).
Lifetime non-partner rape exposure
Experiences of lifetime rape by non-partners were assessed using four questions with this history obtained from all study participants. A positive answer to any of the four questions defined sexual violence by a non-partner. All the frequencies were summed for the overall lifetime severity of non-partner rape exposure score.
Lifestyle behaviours
Current tobacco use was defined as currently smoking any tobacco products such as cigarettes, cigars or pipes daily or occasionally. Women who reported ‘ever consuming alcohol in the past 12 months’ were categorised as current alcohol use.
Clinical examination
Anthropometry was done using standardised techniques. Heights and weights were measured with the participants in light clothing and without shoes. Waist circumference (WC) was measured to the nearest millimeter at the level of umbilicus. BP measurements were taken on the right arm, using a digital BP monitor (Omron, M6 Comfort, Netherland) after seating the participant in a resting position for at least five minutes. Three BP measurements were taken three minutes apart, and the average of the 2nd and 3rd readings was used in the analysis.
Biochemical parameters
Fasting blood samples were drawn via venipuncture, and biochemical analyses were performed at an accredited pathology laboratory. All lipid profiles were measured with an autoanalyser, Beckman Coulter AU 500 spectrophotometer. HbA1c was measured using high-performance liquid chromatography (VARIANT II TURBO, EDTA tubes) in accordance with the National Glycohaemoglobin Standardisation Programme (NGSP).
Statistical analysis
Data were analysed using R® statistical software (The R Foundation for Statistical Computing Platform) version 3.6.0 (2019-04-26). We performed descriptive analysis on cardiometabolic risk variables in the study participants over 3 years that included WC, BMI, SBP, DBP, total cholesterol (TC) and HbA1c. The data are presented as mean (SD) at 4 time-points – baseline, 12 months, 24 months and 36 months in the overall sample and for each group of women. The significant differences across time-points were assessed with linear mixed-effect models (with and without the interaction term, timepoints*rape-exposed).
We used “lme4” package to perform linear mixed effect models examining the associations of exposures to various types of childhood abuse (any childhood abuse and each of 4 childhood abuse types), of IPV (any IPV and each of 4 IPV types) and of non-partner rape, and multiple abuse types with cardiometabolic risk variables across four time-points. All abuse exposures and baseline rape exposure were entered into the model as fixed effects, and timepoints as random effects. Visual inspection of residual plots did not reveal any obvious deviations from homoscedasticity or normality. We adjusted for age at baseline, employment, education, current tobacco use, past year alcohol use and baseline rape-exposure in all the models.
Since the present analysis aimed to examine associations of abuse exposures with cardiometabolic variables among both baseline rape-exposed and non-exposed women, and not to determine and/or compare between the two groups of women; this analysis pooled the data from both cohorts of women. The potential interaction effects of baseline rape-exposure with abuse predictors on the cardiometabolic outcomes were accounted for through the inclusion of the interaction term between baseline rape exposure and each of the predictor variables, ‘abuse/trauma*rape-exposed’ in all models that included the main effects of the ‘abuse/trauma’ of interest and baseline ‘rape-exposure’ status. Additionally, the 3-way interaction term, ‘abuse/trauma*rape-exposed*time’ was included in all models to check for the potential interaction effects of baseline rape-exposure and time with abuse predictors on the cardiometabolic outcomes. However, the analyses revealed no such significant interaction effects, supporting the decision to combine the two groups of women to increase statistical power for regression analyses.
In secondary analyses in a subset of women with data available across all 4 time-points we used repeated measures ANOVA to compare the mean WC, BMI, SBP, DBP, TC and HbA1c within time-points in (1) the women who experienced any childhood abuse and those without this experience; (2) any IPV experience and those without; and (3) any lifetime non-partner rape and those without. The potential interaction effects of baseline rape-exposure and time with abuse of interest were also examined.
Additionally, we repeated the main analysis using an outcome, the modified allostatic load (AL) which included the six cardiometabolic markers (WC, BMI, SBP, DBP, TC and HbA1c) in the algorithm. For each marker, we defined the high-risk threshold based on the distribution of that marker in the RICE study sample. Each participant who had a marker reading above the threshold (> 75th percentile) was assigned a point for that marker using the standard method [34, 35]. The points were summed to determine the allostatic load score, with a maximum score of 6 possible. In the RICE study sample, high-risk thresholds were as follows: WC, 96 cm; BMI, 32; SBP, 110 mmHg; DBP, 75 mmHg; total cholesterol, 4.4 mmol/L; HbA1c, 5.6%. We then generated three binary variables for a high score: (1) AL scores as ≥ 1; (2) AL scores as ≥ 2; and (3) AL scores as ≥ 3. We used logistic mixed effects models to examine the probability of having a high score as a function of exposure to various types and severity of abuse across four timepoints.
Furthermore, we conducted a secondary analysis by adding the interactions of abuse with timepoints and baseline-rape (timepoints*rape-exposed*abuse) and timepoints by subjects (timepoints | subjects) to the main models.
Results
Measurements for cardiometabolic risk profiles were available for 1617 women at baseline, 1178 women at the 12-month, 925 at the 24-month, and 571 at the 36-month. Clinical, biochemical and lifestyle characteristics of the study participants are summarised in Table 1.
Table 1.
Cardiometabolic characteristics among women over a 3-year follow-up period
| Overall | Control | Rape-exposed | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| T0 (baseline) N = 1617 |
T1 (12 months) N = 1178 |
T2 (24 months) N = 925 |
T3 (36 months) N = 571 |
p-value | T0 (baseline) N = 889 |
T1 (12 months) N = 717 |
T2 (24 months) N = 598 |
T3 (36 months) N = 355 |
p-value | T0 (baseline) N = 728 |
T1 (12 months) N = 461 |
T2 (24 months) N = 327 |
T3 (36 months) N = 216 |
p-value | |
| Age, years | 25.3 (5.4) | 26.6 (5.5) | 27.8 (5.6) | 28.6 (5.5) | < 0.001 | 25.5 (5.4) | 26.9 (5.6) | 28 (5.7) | 28.8 (5.6) | < 0.001 | 24.9 (5.4) | 26.2 (5.5) | 27.4 (5.5) | 28.2 (5.4) | < 0.001 |
| WC, cm | 84.4 (14.1) | 86.8 (14.9) | 87.8 (15.1) | 88.3 (15.5) | < 0.001 | 86.2 (14.9) | 87.8 (15.6) | 89.2 (15.9) | 89.4 (16.0) | < 0.001 | 82.2 (12.6) | 85.2 (13.6) | 85.4 (13.4) | 86.6 (14.4) | < 0.001 |
| BMI | 27 (6.5) | 28.1 (6.8) | 28.6 (7.1) | 28.9 (7.2) | < 0.001 | 27.9 (6.8) | 28.8 (6.9) | 29.4 (7.3) | 29.5 (7.5) | < 0.001 | 26 (5.9) | 27.1 (6.5) | 27.4 (6.5) | 28 (6.6) | < 0.001 |
| SBP | 105 (8.3) | 104 (7.5) | 105 (8.0) | 106 (7.7) | 0.020 | 106 (8.3) | 104 (7.4) | 105 (8.4) | 106 (7.4) | 0.545 | 104 (8.1) | 104 (7.7) | 105 (7.3) | 106 (8.2) | 0.006 |
| DBP | 70 (7.5) | 69 (7.3) | 71 (7.5) | 72 (7.4) | < 0.001 | 71 (7.5) | 69 (7.3) | 71 (7.7) | 72 (7.7) | 0.002 | 69 (7.5) | 69 (7.3) | 71 (7.7) | 72 (7.7) | < 0.001 |
| TC, mmol/l | 3.8 (0.8) | 3.9 (0.8) | 3.9 (0.8) | 4.0 (0.9) | < 0.001 | 3.9 (0.8) | 3.9 (0.8) | 4.0 (0.8) | 4.0 (0.8) | 0.006 | 3.8 (0.8) | 3.9 (0.9) | 3.9 (0.8) | 4.1 (1.0) | < 0.001 |
| HbA1c, % | 5.3 (0.5) | 5.3 (0.5) | 5.4 (0.5) | 5.5 (0.6) | < 0.001 | 5.3 (0.5) | 5.3 (0.6) | 5.4 (0.5) | 5.5 (0.7) | < 0.001 | 5.3 (0.5) | 5.3 (0.65 | 5.3 (0.5) | 5.4 (0.4) | 0.012 |
| Current smoking, n (%) | 200/1617 (12.4) | 114/1178 (9.7) | 79/925 (8.5) | 48/571 (8.4) | 0.003 | 96/889 (10.8) | 57/717 (7.9) | 43/598 (7.2) | 24/355 (6.8) | 0.030 | 104/728 (14.3) | 57/461 (12.4) | 36/327 (11) | 24/216 (11) | 0.398 |
| Current alcohol use, n (%) | 848/1616 (52.5) | 455/1177 (38.7) | 345/923 (37.4) | 215/571 (37.7) | < 0.001 | 424/888 (47.7) | 241/716 (33.7) | 197/597 (33) | 136/355 (38.3) | < 0.001 | 424/728 (58.2) | 214/461 (46.4) | 148/326 (45.4) | 79/216 (36.6) | < 0.001 |
Data are mean (SD) otherwise indicated; P-value for simple linear mixed effect models or chi-square tests; WC, waist circumference; SBP, systolic blood pressure; DBP, diastolic blood pressure; TC, total cholesterol; LDL, low-density lipoprotein cholesterol; HDL, high-density lipoprotein cholesterol; TG, triglycerides. Current smoking: currently smoking any tobacco products such as cigarettes, cigars or pipes. Current alcohol use: ever consuming alcohol in the past 12 months
As shown in Table 1, over a 3-year period, mean levels of WC, BMI, DBP, TC, HbA1c increased linearly. Mean SBP increased overall and in the rape-exposed women, but not in the control group. Current alcohol use decreased significantly in all groups while current smoking decreased overall (p = 0.003) and in the control (p = 0.030), but not the rape-exposed group (p = 0.398).
In the overall sample, 57% (n = 927) reported childhood abuse exposure, 59% (n = 957) experienced IPV and 18% (n = 285), non-partner rape (Table S1). The magnitudes and mean severity scores of individual abuse types are presented in Table S1.
Women exposed to childhood abuse had higher BMI levels compared with their counterparts (Table S2). However, women exposed to any IPV, compared to those with no exposure, had lower mean BMI, DBP, TC and HbA1c levels. Other cardiometabolic variables were similar across the groups including by non-partner rape status.
Associations of childhood abuse with cardiometabolic variables over three years
Overall, childhood abuse was not strongly associated with increasing changes in cardiometabolic variables over the three years (Tables 2, 3 and 4). Mean waist circumference (WC) was generally unaffected by childhood abuse exposure (Table 2). There was no evidence of interaction effects between baseline-rape exposure or time and most childhood abuse types in their associations with WC (p > 0.05 for all interaction tests). An exception was found for sexual childhood abuse, where baseline-rape exposure influenced its association with WC (p = 0.010 for the 2-way “sexual CA*baseline-raped” interaction test). Time also affected the associations of emotional CA with WC (p = 0.007 for the 3-way “emotional CA*baseline-raped*time” interaction test).
Table 2.
Linear mixed effect models for the associations of child abuse exposures with waist circumference and body mass index over a 3-year follow-up period
| Waist circumference | Body mass index | |||||||
|---|---|---|---|---|---|---|---|---|
| Exposure variables | p-interact (CA*baseline-rape-exposed) | p-interact (CA*baseline-rape-exposed*time) | B (se) | p-value | p-interact (CA*baseline-rape-exposed) | p-interact (CA*baseline-rape-exposed*time) | B (se) | p-value |
| Any CA | 0.095 | 0.009 | 0.87 (0.58) | 0.136 | 0.904 | 0.237 | 0.79 (0.27) | 0.003 |
| Sexual CA | 0.010 | 0.858 | −0.70 (1.05) | 0.504 | 0.057 | 0.490 | −0.14 (0.48) | 0.776 |
| Physical CA | 0.959 | 0.230 | 1.09 (0.59) | 0.067 | 0.451 | 0.943 | 0.68 (0.27) | 0.013 |
| Emotional CA | 0.150 | 0.007 | 0.56 (0.65) | 0.389 | 0.531 | 0.039 | 0.37 (0.30) | 0.219 |
| Parental neglect | 0.495 | 0.398 | −0.13 (0.70) | 0.846 | 0.895 | 0.584 | 0.22 (0.32) | 0.495 |
| Number of CA type | 0.236 | 0.168 | 0.607 | 0.317 | 0.931 | 0.925 | ||
| 0 | Ref | Ref | ||||||
| 1 | 0.582 | 0.361 | 1.63 (0.71) | 0.023 | 0.501 | 0.930 | 1.25 (0.33) | < 0.001 |
| 2 | 0.009 | 0.051 | −0.45 (0.83) | 0.583 | 0.176 | 0.176 | −0.12 (0.38) | 0.751 |
| 3 | 0.629 | 0.621 | 0.31 (0.99) | 0.756 | 0.628 | 0.797 | 0.39 (0.45) | 0.402 |
| 4 | 0.719 | 0.094 | 5.26 (2.16) | 0.015 | 0.566 | 0.173 | 2.72 (1.00) | 0.006 |
| Combined frequency and severity of childhood maltreatment | ||||||||
| Sexual CA | 0.039 | 0.640 | 0.602 | 0.147 | 0.103 | 0.057 | ||
| Never | Ref | Ref | ||||||
| Some | 0.009 | 0.568 | −1.22 (1.15) | 0.288 | 0.048 | 0.098 | −0.37 (0.53) | 0.479 |
| Often | 0.704 | 0.361 | 1.70 (2.40) | 0.480 | 0.930 | 0.044 | 0.96 (1.11) | 0.385 |
| Physical CA | 0.660 | 0.238 | 0.085 | 0.114 | 0.620 | 0.036 | ||
| Never | Ref | Ref | ||||||
| Some | 0.501 | 0.217 | 1.34 (0.72) | 0.064 | 0.581 | 0.921 | 0.63 (0.33) | 0.059 |
| Often | 0.524 | 0.803 | 0.78 (0.78) | 0.316 | 0.093 | 0.314 | 0.73 (0.36) | 0.042 |
| Emotional CA | 0.152 | 0.070 | 0.376 | 0.742 | 0.395 | 0.323 | ||
| Never | Ref | Ref | ||||||
| Some | 0.395 | 0.033 | 1.85 (0.77) | 0.017 | 0.501 | 0.166 | 0.74 (0.36) | 0.039 |
| Often | 0.063 | 0.068 | − 1.65 (0.97) | 0.090 | 0.543 | 0.017 | −0.26 (0.45) | 0.558 |
| Parental neglect | 0.435 | 0.812 | 0.769 | 0.833 | 0.049 | 0.763 | ||
| Never | Ref | Ref | ||||||
| Some | 0.937 | 0.816 | 0.79 (0.79) | 0.318 | 0.521 | 0.051 | 0.55 (0.36) | 0.135 |
| Often | 0.063 | 0.051 | −2.85 (1.27) | 0.025 | 0.296 | 0.025 | −0.73 (0.59) | 0.216 |
| Severity score | ||||||||
| CA severity score | 0.225 | 0.769 | −0.06 (0.13) | 0.617 | 0.970 | 0.139 | 0.03 (0.06) | 0.603 |
| Sexual CA score | 0.065 | 0.938 | −0.39 (0.76) | 0.606 | 0.129 | 0.167 | −0.08 (0.35) | 0.810 |
| Physical CA score | 0.886 | 0.823 | 0.16 (0.25) | 0.537 | 0.468 | 0.198 | 0.15 (0.12) | 0.210 |
| Emotional CA score | 0.205 | 0.137 | −0.27 (0.34) | 0.425 | 0.716 | 0.791 | 0.01 (0.16) | 0.938 |
| Parental neglect score | 0.402 | 0.936 | −0.58 (0.43) | 0.174 | 0.797 | 0.140 | −0.05 (0.19) | 0.798 |
B: beta co-efficient of linear mixed effect model. Models adjusted for age at baseline, employment, education, current smoking, current alcohol use and baseline rape-exposure. CA, childhood abuse. P-interact (CA*baseline-rape-exposed): p-value for the interaction terms, child abuse*baseline-rape-exposed, in each exposure type and outcome examined
Table 3.
Linear mixed effect models for the associations of child abuse exposures with blood pressure over a 3-year follow-up period
| Exposure variables | Systolic blood pressure | Diastolic blood pressure | ||||||
|---|---|---|---|---|---|---|---|---|
| p-interact (CA*baseline-rape-exposed) | p-interact (CA*baseline-rape-exposed*time) | B (se) | p-value | p-interact (CA*baseline-rape-exposed) | p-interact (CA*baseline-rape-exposed*time) | B (se) | p-value | |
| Any CA | 0.455 | 0.579 | −0.23 (0.31) | 0.465 | 0.339 | 0.250 | −0.35 (0.30) | 0.245 |
| Sexual CA | 0.942 | 0.270 | 0.37 (0.57) | 0.512 | 0.513 | 0.092 | −0.38 (0.54) | 0.481 |
| Physical CA | 0.625 | 0.376 | −0.38 (0.32) | 0.241 | 0.334 | 0.890 | −0.32 (0.31) | 0.298 |
| Emotional CA | 0.744 | 0.896 | −0.11 (0.35) | 0.974 | 0.081 | 0.687 | −0.55 (0.33) | 0.101 |
| Parental neglect | 0.196 | 0.511 | −0.78 (0.38) | 0.040 | 0.007 | 0.554 | −0.94 (0.36) | 0.009 |
| Multiple CA | 0.441 | 0.606 | 0.127 | 0.413 | 0.321 | 0.470 | ||
| 0 | Ref | Ref | ||||||
| 1 | 0.350 | 0.632 | −0.04 (0.39) | 0.910 | 0.741 | 0.234 | 0.07 (0.37) | 0.851 |
| 2 | 0.450 | 0.891 | −0.34 (0.45) | 0.461 | 0.450 | 0.644 | −0.51 (0.43) | 0.239 |
| 3 | 0.631 | 0.473 | −0.36 (0.54) | 0.507 | 0.120 | 0.601 | −0.81 (0.51) | 0.113 |
| 4 | 0.201 | 0.725 | −1.16 (1.18) | 0.327 | 0.035 | 0.427 | −2.08 (1.13) | 0.065 |
| Combined frequency and severity of childhood maltreatment | ||||||||
| Sexual CA | 0.894 | 0.286 | 0.404 | 0.536 | 0.114 | 0.692 | ||
| Never | Ref | Ref | ||||||
| Some | 0.771 | 0.359 | 0.46 (0.62) | 0.461 | 0.511 | 0.242 | −0.30 (0.59) | 0.615 |
| Often | 0.730 | 0.692 | −0.026 (1.30) | 0.984 | 0.671 | 0.718 | −0.78 (1.25) | 0.536 |
| Physical CA | 0.332 | 0.351 | 0.102 | 0.296 | 0.656 | 0.915 | ||
| Never | Ref | Ref | ||||||
| Some | 0.637 | 0.334 | −0.55 (0.39) | 0.161 | 0.544 | 0.995 | −0.17 (0.37) | 0.644 |
| Often | 0.210 | 0.842 | −0.165 (0.42) | 0.697 | 0.330 | 0.188 | −0.50 (0.40) | 0.217 |
| Emotional CA | 0.661 | 0.799 | 0.622 | 0.185 | 0.836 | 0.591 | ||
| Never | Ref | Ref | ||||||
| Some | 0.906 | 0.929 | 0.14 (0.42) | 0.742 | 0.040 | 0.729 | −0.59 (0.40) | 0.143 |
| Often | 0.792 | 0.381 | −0.27 (0.53) | 0.611 | 0.565 | 0.524 | −0.49 (0.50) | 0.328 |
| Parental neglect | 0.084 | 0.594 | 0.462 | 0.011 | 0.472 | 0.286 | ||
| Never | Ref | Ref | ||||||
| Some | 0.968 | 0.666 | −0.75 (0.43) | 0.081 | 0.029 | 0.422 | −0.71 (0.41) | 0.082 |
| Often | 0.033 | 0.656 | −0.89 (0.69) | 0.201 | 0.035 | 0.830 | −1.62 (0.66) | 0.014 |
| Severity score | ||||||||
| CA severity score | 0.919 | 0.370 | −0.031 (0.05) | 0.432 | 0.144 | 0.268 | −0.035 (0.04) | 0.188 |
| Sexual CA score | 0.915 | 0.219 | 0.16 (0.19) | 0.416 | 0.545 | 0.102 | −0.32 (0.39) | 0.422 |
| Physical CA score | 0.201 | 0.782 | −0.10 (0.10) | 0.335 | 0.319 | 0.896 | −0.06 (0.09) | 0.556 |
| Emotional CA score | 0.613 | 0.403 | −0.07 (0.11) | 0.520 | 0.445 | 0.681 | −0.12 (0.11) | 0.261 |
| Parental neglect score | 0.187 | 0.413 | −0.15 (0.15) | 0.311 | 0.050 | 0.845 | −0.24 (0.14) | 0.097 |
B: beta co-efficient of linear mixed effect model. Models adjusted for age at baseline, employment, education, current smoking, current alcohol use and baseline rape-exposure. CA, childhood abuse. P-interact (CA*baseline-rape-exposed): p-value for the interaction terms, child abuse*baseline-rape-exposed, in each exposure type and outcome examined
Table 4.
Linear mixed effect models for the associations of childhood abuse exposures with total cholesterol and glycated haemoglobin over a 3-year follow-up period
| Exposure variables | Total cholesterol | Glycated haemoglobin | ||||||
|---|---|---|---|---|---|---|---|---|
| p-interact (CA*baseline-rape-exposed) | p-interact (CA*baseline-rape-exposed*time) | B (se) | p-value | p-interact (CA*baseline-rape-exposed) | p-interact (CA*baseline-rape-exposed*time) | B (se) | p-value | |
| Any CA | 0.707 | 0.039 | −0.01 (0.03) | 0.692 | 0.629 | 0.181 | 0.01 (0.02) | 0.974 |
| Sexual CA | 0.790 | 0.797 | −0.05 (0.05) | 0.356 | 0.723 | 0.383 | 0.02 (0.03) | 0.656 |
| Physical CA | 0.412 | 0.287 | 0.01 (0.03) | 0.774 | 0.333 | 0.099 | −0.01 (0.02) | 0.934 |
| Emotional CA | 0.168 | 0.849 | −0.09 (0.04) | 0.008 | 0.795 | 0.781 | 0.01 (0.02) | 0.992 |
| Parental neglect | 0.736 | 0.121 | 0.01 (0.04) | 0.818 | 0.179 | 0.408 | −0.04 (0.02) | 0.107 |
| Multiple CA type | 0.239 | 0.196 | 0.522 | 0.076 | 0.387 | 0.143 | ||
| 0 | Ref | Ref | ||||||
| 1 | 0.469 | 0.094 | 0.01 (0.04) | 0.833 | 0.233 | 0.707 | 0.03 (0.02) | 0.253 |
| 2 | 0.144 | 0.095 | 0.01 (0.05) | 0.956 | 0.522 | 0.622 | −0.04 (0.02) | 0.219 |
| 3 | 0.045 | 0.545 | −0.06 (0.05) | 0.307 | 0.567 | 0.162 | −0.03 (0.03) | 0.343 |
| 4 | 0.377 | 0.757 | −0.19 (0.12) | 0.113 | 0.067 | 0.659 | 0.12 (0.07) | 0.099 |
| Combined frequency and severity of childhood abuse | ||||||||
| Sexual CA | 0.556 | 0.825 | 0.179 | 0.433 | 0.539 | 0.937 | ||
| Never | Ref | Ref | ||||||
| Some | 0.663 | 0.878 | −0.10 (0.065) | 0.113 | 0.576 | 0.569 | 0.01 (0.03) | 0.925 |
| Often | 0.093 | 0.180 | 0.17 (0.14) | 0.214 | 0.173 | 0.501 | 0.07 (0.08) | 0.373 |
| Physical CA | 0.575 | 0.086 | 0.264 | 0.564 | 0.068 | 0.607 | ||
| Never | Ref | Ref | ||||||
| Some | 0.456 | 0.203 | 0.01 (0.04) | 0.983 | 0.199 | 0.098 | 0.01 (0.02) | 0.909 |
| Often | 0.602 | 0.342 | 0.02 (0.04) | 0.645 | 0.798 | 0.921 | −0.01 (0.02) | 0.786 |
| Emotional CA | 0.242 | 0.971 | 0.013 | 0.589 | 0.998 | 0.986 | ||
| Never | Ref | Ref | ||||||
| Some | 0.186 | 0.954 | −0.08 (0.04) | 0.046 | 0.743 | 0.884 | −0.01 (0.02) | 0.591 |
| Often | 0.401 | 0.833 | −0.11 (0.05) | 0.037 | 0.375 | 0.874 | 0.03 (0.03) | 0.451 |
| Parental neglect | 0.605 | 0.191 | 0.967 | 0.569 | 0.453 | 0.567 | ||
| Never | Ref | Ref | ||||||
| Some | 0.630 | 0.167 | 0.02 (0.04) | 0.710 | 0.047 | 0.384 | −0.06 (0.02) | 0.017 |
| Often | 0.145 | 0.320 | −0.08 (0.07) | 0.240 | 0.485 | 0.659 | 0.03 (0.04) | 0.410 |
| Severity score | ||||||||
| CA severity score | 0.804 | 0.382 | −0.01 (0.01) | 0.269 | 0.621 | 0.335 | −0.01 (0.01) | 0.230 |
| Sexual CA score | 0.385 | 0.292 | −0.05 (0.02) | 0.024 | 0.506 | 0.935 | −0.01 (0.01) | 0.714 |
| Physical CA score | 0.509 | 0.068 | 0.01 (0.01) | 0.449 | 0.593 | 0.146 | −0.01 (0.01) | 0.503 |
| Emotional CA score | 0.235 | 0.850 | −0.01 (0.01) | 0.032 | 0.287 | 0.968 | −0.01 (0.01) | 0.929 |
| Parental neglect score | 0.849 | 0.551 | −0.01 (0.01) | 0.825 | 0.892 | 0.379 | −0.01 (0.01) | 0.436 |
B: beta co-efficient of linear mixed effect model. Models adjusted for age at baseline, employment, education, current smoking, current alcohol use and baseline rape-exposure. CA, childhood abuse. P-interact (CA*baseline-rape-exposed): p-value for the interaction terms (CA*baseline-rape-exposed) in each exposure type and outcome examined
Any childhood abuse, particularly physical childhood abuse, and a greater frequency of physical childhood abuse were associated with rising BMI over time (Table 2). There was an interaction effect of time on the emotional childhood abuse-BMI association (p = 0.039 for 3-way interaction effect, emotional CA*baseline-rape exposure*time) (Table 2).
Associations of IPV and non-partner rape exposures with cardiometabolic variables over three years
Neither IPV nor the exposure to multiple types of IPV were associated with increases in cardiometabolic variables over 3 years (Tables 5, 6 and 7). The greater severity of exposure to IPV, physical or economic IPV was associated with decreases in BMI and DBP levels over time (all p ≤ 0.034 for the main effect). More than one episode of non-partner rape was associated with decreasing WC or BMI over time (Table 8).
Table 5.
Linear mixed effect models for the associations of intimate partner violence (IPV) exposures with waist circumference and body mass index over a 3-year follow-up period
| Waist circumference | Body mass index | |||||||
|---|---|---|---|---|---|---|---|---|
| Exposure variables | p-interact (IPV*baseline-rape-exposed) | p-interact (IPV*baseline-rape-exposed*time) | B (se) | p-value | p-interact (IPV*baseline-rape-exposed) | p-interact (IPV*baseline-rape-exposed*time) | B (se) | p-value |
| Any IPV | 0.606 | 0.670 | 0.03 (0.66) | 0.960 | 0.956 | 0.180 | −0.14 (0.30) | 0.649 |
| Sexual IPV | 0.365 | 0.939 | −1.21 (1.11) | 0.278 | 0.533 | 0.193 | −0.56 (0.52) | 0.282 |
| Physical IPV | 0.601 | 0.654 | −0.07 (0.75) | 0.930 | 0.734 | 0.535 | −0.30 (0.35) | 0.394 |
| Emotional IPV | 0.885 | 0.980 | 0.21 (0.72) | 0.775 | 0.567 | 0.137 | −0.09 (0.33) | 0.789 |
| Economic IPV | 0.837 | 0.274 | −0.44 (0.95) | 0.643 | 0.796 | 0.095 | −0.32 (0.44) | 0.475 |
| Number of IPV type | 0.898 | 0.766 | −0.09 (0.27) | 0.753 | 0.761 | 0.163 | −0.11 (0.12) | 0.380 |
| 0 | ||||||||
| 1 | 0.477 | 0.534 | 1.02 (0.99) | 0.303 | 0.914 | 0.120 | 0.32 (0.46) | 0.490 |
| 2 | 0.721 | 0.209 | 0.46 (1.04) | 0.661 | 0.978 | 0.951 | 0.01 (0.49) | 0.989 |
| 3 | 0.689 | 0.884 | −0.37 (1.29) | 0.773 | 0.450 | 0.106 | −0.41 (0.60) | 0.501 |
| 4 | 0.779 | 0.169 | −0.94 (1.53) | 0.536 | 0.795 | 0.616 | −0.58 (0.71) | 0.410 |
| Combined frequency and severity of IPV | ||||||||
| Sexual IPV | 0.161 | 0.990 | 0.708 | 0.534 | 0.441 | 0.477 | ||
| Never | Ref | Ref | ||||||
| Some | 0.801 | 0.918 | −1.66 (1.58) | 0.294 | 0.559 | 0.897 | −0.48 (0.74) | 0.524 |
| Often | 0.719 | 0.984 | −0.14 (0.86) | 0.867 | 0.623 | 0.547 | −0.25 (0.40) | 0.547 |
| Physical IPV | 0.958 | 0.605 | 0.383 | 0.705 | 0.188 | 0.095 | ||
| Never | Ref | Ref | ||||||
| Some | 0.818 | 0.169 | −1.16 (1.01) | 0.251 | 0.615 | 0.047 | −0.79 (0.47) | 0.096 |
| Often | 0.952 | 0.016 | −0.51 (0.63) | 0.423 | 0.625 | 0.126 | −0.47 (0.30) | 0.116 |
| Emotional IPV | 0.285 | 0.863 | 0.872 | 0.237 | 0.093 | 0.545 | ||
| Never | Ref | |||||||
| Some | 0.930 | 0.392 | 0.94 (1.10) | 0.393 | 0.909 | 0.606 | 0.35 (0.51) | 0.491 |
| Often | 0.276 | 0.378 | −0.14 (0.60) | 0.817 | 0.229 | 0.612 | −0.19 (0.28) | 0.502 |
| Economic IPV | 0.335 | 0.238 | 0.336 | 0.435 | 0.019 | 0.241 | ||
| Never | Ref | Ref | ||||||
| Some | 0.501 | 0.119 | 1.07 (1.32) | 0.419 | 0.562 | 0.040 | 0.23 (0.62) | 0.718 |
| Often | 0.248 | 0.307 | −0.94 (0.79) | 0.239 | 0.348 | 0.605 | −0.48 (0.37) | 0.194 |
| Severity score | ||||||||
| IPV severity score | 0.326 | 0.910 | −0.05 (0.02) | 0.068 | 0.385 | 0.574 | −0.03 (0.01) | 0.026 |
| Sexual IPV score | 0.302 | 0.587 | −0.20 (0.16) | 0.195 | 0.582 | 0.067 | −0.13 (0.07) | 0.076 |
| Physical IPV score | 0.840 | 0.748 | −0.18 (0.08) | 0.029 | 0.892 | 0.536 | −0.10 (0.03) | 0.005 |
| Emotional IPV score | 0.313 | 0.747 | −0.09 (0.06) | 0.136 | 0.296 | 0.362 | −0.05 (0.02) | 0.082 |
| Economic IPV score | 0.213 | 0.215 | −0.12 (0.15) | 0.439 | 0.296 | 0.155 | −0.04 (0.07) | 0.567 |
B: beta co-efficient of linear mixed effect model. Models adjusted for age at baseline, employment, education, current smoking, current alcohol use and baseline rape-exposure. IPV, intimate partner violence. P-interact (IPV*baseline-rape-exposed): p-value for the interaction terms (IPV*baseline-rape-exposed) in each exposure type and outcome examined
Table 6.
Linear mixed effect models for the associations of intimate partner violence (IPV) exposures with blood pressure over a 3-year follow-up period
| Exposure variables | Systolic blood pressure | Diastolic blood pressure | ||||||
|---|---|---|---|---|---|---|---|---|
| p-interact (IPV*baseline-rape-exposed) | p-interact (IPV*baseline-rape-exposed*time) | B (se) | p-value | p-interact (IPV*baseline-rape-exposed) | p-interact (IPV*baseline-rape-exposed*time) | B (se) | p-value | |
| Any IPV | 0.163 | 0.789 | 0.19 (0.36) | 0.602 | 0.352 | 0.120 | −0.14 (0.34) | 0.684 |
| Sexual IPV | 0.112 | 0.936 | −0.51 (0.61) | 0.404 | 0.313 | 0.778 | −0.75 (0.57) | 0.189 |
| Physical IPV | 0.462 | 0.710 | 0.07 (0.41) | 0.871 | 0.178 | 0.414 | 0.02 (0.39) | 0.955 |
| Emotional IPV | 0.898 | 0.583 | 0.12 (0.39) | 0.770 | 0.790 | 0.147 | −0.15 (0.37) | 0.681 |
| Economic IPV | 0.617 | 0.607 | −0.94 (0.52) | 0.072 | 0.446 | 0.731 | −1.32 (0.49) | 0.007 |
| Number of IPV type | 0.699 | 0.791 | −0.06 (0.11) | 0.621 | 0.990 | 0.402 | −0.19 (0.11) | 0.083 |
| 0 | ||||||||
| 1 | 0.240 | 0.921 | 0.67 (0.54) | 0.216 | 0.410 | 0.274 | 0.31 (0.51) | 0.544 |
| 2 | 0.418 | 0.492 | 0.44 (0.57) | 0.439 | 0.210 | 0.669 | 0.60 (0.54) | 0.267 |
| 3 | 0.650 | 0.293 | −0.34 (0.70) | 0.629 | 0.558 | 0.158 | −0.78 (0.66) | 0.238 |
| 4 | 0.459 | 0.255 | −0.92 (0.84) | 0.274 | 0.717 | 0.749 | −1.34 (0.78) | 0.087 |
| Combined frequency and severity of IPV | ||||||||
| Sexual IPV | 0.144 | 0.480 | 0.703 | 0.151 | 0.965 | 0.291 | ||
| Never | Ref | Ref | ||||||
| Some | 0.068 | 0.909 | −1.45 (0.90) | 0.108 | 0.275 | 0.249 | −1.81 (0.85) | 0.033 |
| Often | 0.297 | 0.546 | 0.36 (0.49) | 0.463 | 0.242 | 0.106 | −0.30 (0.46) | 0.518 |
| Physical IPV | 0.835 | 0.551 | 0.252 | 0.899 | 0.369 | 0.069 | ||
| Never | Ref | Ref | ||||||
| Some | 0.030 | 0.730 | −0.46 (0.57) | 0.427 | 0.021 | 0.081 | −0.54 (0.54) | 0.318 |
| Often | 0.985 | 0.358 | −0.40 (0.36) | 0.269 | 0.923 | 0.148 | −0.61 (0.34) | 0.075 |
| Emotional IPV | 0.963 | 0.396 | 0.336 | 0.822 | 0.057 | 0.062 | ||
| Never | Ref | Ref | ||||||
| Some | 0.504 | 0.271 | 0.14 (0.62) | 0.817 | 0.081 | 0.332 | −0.56 (0.58) | 0.341 |
| Often | 0.991 | 0.437 | −0.34 (0.34) | 0.319 | 0.948 | 0.895 | −0.59 (0.32) | 0.068 |
| Economic IPV | 0.769 | 0.510 | 0.007 | 0.235 | 0.638 | 0.001 | ||
| Never | Ref | Ref | ||||||
| Some | 0.089 | 0.202 | −1.16 (0.75) | 0.123 | 0.285 | 0.861 | −0.96 (0.71) | 0.175 |
| Often | 0.948 | 0.250 | −1.09 (0.45) | 0.015 | 0.315 | 0.809 | −1.55 (0.42) | 0.001 |
| Severity score | ||||||||
| IPV severity score | 0.994 | 0.820 | −0.01 (0.01) | 0.266 | 0.308 | 0.928 | −0.03 (0.01) | 0.019 |
| Sexual IPV score | 0.488 | 0.687 | −0.03 (0.09) | 0.741 | 0.326 | 0.418 | −0.07 (0.08) | 0.409 |
| Physical IPV score | 0.928 | 0.585 | −0.06 (0.04) | 0.171 | 0.458 | 0.796 | −0.09 (0.04) | 0.034 |
| Emotional IPV score | 0.677 | 0.863 | −0.02 (0.03) | 0.491 | 0.624 | 0.547 | −0.06 (0.03) | 0.051 |
| Economic IPV score | 0.547 | 0.805 | −0.11 (0.08) | 0.212 | 0.117 | 0.373 | −0.24 (0.08) | 0.003 |
B: beta co-efficient of linear mixed effect model. Models adjusted for age at baseline, employment, education, current smoking, current alcohol use and baseline rape-exposure. IPV, intimate partner violence. P-interact (IPV*baseline-rape-exposed): p-value for the interaction terms (IPV*baseline-rape-exposed) in each exposure type and outcome examined
Table 7.
Linear mixed effect models for the associations of intimate partner violence (IPV) exposures with cardiometabolic profiles over a 3-year follow-up period
| Exposure variables | Total cholesterol | Glycated haemoglobin | ||||||
|---|---|---|---|---|---|---|---|---|
| p-interact (IPV*baseline-rape-exposed) | p-interact (IPV*baseline-rape-exposed*time) | B (se) | p-value | p-interact (IPV*baseline-rape-exposed) | p-interact (IPV*baseline-rape-exposed*time) | B (se) | p-value | |
| Any IPV | 0.955 | 0.181 | −0.04 (0.03) | 0.307 | 0.135 | 0.948 | −0.04 (0.02) | 0.075 |
| Sexual IPV | 0.287 | 0.135 | 0.05 (0.06) | 0.409 | 0.711 | 0.306 | −0.02 (0.04) | 0.628 |
| Physical IPV | 0.533 | 0.607 | −0.05 (0.04) | 0.173 | 0.075 | 0.925 | −0.08 (0.02) | 0.004 |
| Emotional IPV | 0.365 | 0.046 | −0.06 (0.04) | 0.131 | 0.043 | 0.455 | −0.04 (0.02) | 0.139 |
| Economic IPV | 0.995 | 0.306 | −0.05 (0.05) | 0.317 | 0.580 | 0.534 | −0.04 (0.03) | 0.209 |
| Number of IPV type | 0.923 | 0.569 | −0.02 (0.01) | 0.122 | 0.082 | 0.454 | −0.01 (0.01) | 0.197 |
| 0 | ||||||||
| 1 | 0.529 | 0.415 | −0.04 (0.05) | 0.431 | 0.770 | 0.809 | −0.02 (0.03) | 0.623 |
| 2 | 0.325 | 0.511 | −0.06 (0.05) | 0.276 | 0.171 | 0.179 | −0.06 (0.03) | 0.101 |
| 3 | 0.549 | 0.288 | −0.13 (0.07) | 0.084 | 0.119 | 0.807 | −0.07 (0.04) | 0.132 |
| 4 | 0.220 | 0.477 | 0.02 (0.08) | 0.780 | 0.622 | 0.619 | −0.06 (0.05) | 0.292 |
| Sexual IPV | 0.208 | 0.094 | 0.545 | 0.688 | 0.184 | 0.742 | ||
| Never | Ref | Ref | ||||||
| Some | 0.501 | 0.200 | 0.10 (0.09) | 0.304 | 0.422 | 0.290 | 0.01 (0.05) | 0.965 |
| Often | 0.203 | 0.822 | 0.02 (0.05) | 0.681 | 0.748 | 0.874 | −0.01 (0.03) | 0.727 |
| Physical IPV | 0.450 | 0.782 | 0.447 | 0.100 | 0.749 | 0.042 | ||
| Never | Ref | Ref | ||||||
| Some | 0.542 | 0.710 | −0.11 (0.05) | 0.052 | 0.140 | 0.894 | −0.07 (0.03) | 0.041 |
| Often | 0.447 | 0.460 | −0.02 (0.03) | 0.532 | 0.069 | 0.975 | −0.04 (0.02) | 0.056 |
| Emotional IPV | 0.325 | 0.116 | 0.708 | 0.158 | 0.283 | 0.674 | ||
| Never | Ref | Ref | ||||||
| Some | 0.161 | 0.163 | −0.02 (0.06) | 0.785 | 0.589 | 0.508 | −0.01 (0.03) | 0.842 |
| Often | 0.384 | 0.584 | −0.01 (0.03) | 0.718 | 0.165 | 0.946 | −0.01 (0.02) | 0.680 |
| Economic IPV | 0.652 | 0.196 | 0.475 | 0.983 | 0.735 | 0.283 | ||
| Never | Ref | |||||||
| Some | 0.838 | 0.771 | −0.03 (0.07) | 0.703 | 0.955 | 0.323 | −0.04 (0.04) | 0.412 |
| Often | 0.609 | 0.363 | −0.03 (0.04) | 0.509 | 0.994 | 0.302 | −0.03 (0.03) | 0.347 |
| Severity score | ||||||||
| IPV severity score | 0.692 | 0.284 | −0.01 (0.00) | 0.650 | 0.195 | 0.625 | −0.01 (0.00) | 0.177 |
| Sexual IPV score | 0.526 | 0.227 | 0.01 (0.01) | 0.392 | 0.548 | 0.378 | −0.01 (0.01) | 0.358 |
| Physical IPV score | 0.855 | 0.292 | −0.01 (0.01) | 0.443 | 0.094 | 0.816 | −0.01 (0.00) | 0.045 |
| Emotional IPV score | 0.308 | 0.141 | −0.01 (0.01) | 0.367 | 0.249 | 0.638 | −0.01 (0.01) | 0.732 |
| Economic IPV score | 0.555 | 0.077 | 0.01 (0.01) | 0.612 | 0.882 | 0.702 | −0.01 (0.01) | 0.096 |
B: beta co-efficient of linear mixed effect model. Models adjusted for age at baseline, employment, education, current smoking, current alcohol use and baseline rape-exposure. IPV, intimate partner violence. P-interact (IPV*baseline-rape-exposed): p-value for the interaction terms (IPV*baseline-rape-exposed) in each exposure type and outcome examined
Table 8.
Linear mixed effect models for the associations of non-partner rape exposure with cardiometabolic profile over a 3-year follow-up period
| Waist circumference | Body mass index | |||||||
|---|---|---|---|---|---|---|---|---|
| Exposure variables | p-interact (NPR*baseline-rape-exposed) | p-interact (NPR*baseline-rape-exposed*time) | B (se) | p-value | p-interact (NPR*baseline-rape-exposed) | p-interact (NPR*baseline-rape-exposed*time) | B (se) | p-value |
| Any NPSV | 0.326 | 0.957 | −2.62 (1.30) | 0.043 | 0.497 | 0.794 | −0.99 (0.60) | 0.100 |
| Multiple NPSV | 0.193 | 0.858 | 0.007 | 0.274 | 0.727 | 0.020 | ||
| None | Ref | Ref | ||||||
| 1 episode | 0.731 | 0.580 | −1.21 (1.70) | 0.475 | 0.946 | 0.810 | −0.23 (0.79) | 0.769 |
| >=2 episodes | 0.318 | 0.531 | −4.62 (1.98) | 0.020 | 0.291 | 0.747 | −2.05 (0.59) | 0.027 |
| Systolic blood pressure | Diastolic blood pressure | |||||||
| Any NPSV | 0.823 | 0.046 | −0.91 (0.71) | 0.200 | 0.248 | 0.635 | −1.56 (0.67) | 0.019 |
| Multiple NPSV | 0.640 | 0.030 | 0.056 | 0.265 | 0.324 | 0.147 | ||
| None | Ref | Ref | ||||||
| 1 episode | 0.384 | 0.102 | −1.97 (0.93) | 0.033 | 0.242 | 0.732 | −2.28 (0.87) | 0.009 |
| >=2 episodes | 0.170 | 0.706 | 0.70 (1.08) | 0.517 | 0.684 | 0.510 | −0.39 (1.02) | 0.700 |
| Total cholesterol | Glycated haemoglobin | |||||||
| Any NPSV | 0.822 | 0.910 | −0.04 (0.07) | 0.503 | 0.529 | 0.306 | 0.01 (0.04) | 0.952 |
| Multiple NPSV | 0.866 | 0.733 | 0.262 | 0.176 | 0.212 | 0.073 | ||
| None | Ref | Ref | ||||||
| 1 episode | 0.227 | 0.626 | −0.11 (0.09) | 0.240 | 0.987 | 0.286 | 0.01 (0.06) | 0.977 |
| >=2 episodes | 0.238 | 0.711 | 0.04 (0.11) | 0.749 | 0.158 | 0.965 | −0.01 (0.07) | 0.966 |
B: beta co-efficient of linear mixed effect model. Models adjusted for age at baseline, employment, education, current smoking, current alcohol use and baseline rape-exposure. NPR, non-partner partner rape. P-interact (non-partner rape*baseline-rape-exposed): p-value for the interaction terms, non-partner rape*baseline-rape-exposed, in each exposure type and outcome examined
Secondary analyses
When considering the modified allostatic load as an outcome, the adjusted logistic mixed-effects models showed that childhood abuse - but not IPV or non-partner rape was associated with a higher probability of having a high allostatic load score compared to those without a history of childhood abuse (Tables S3-5).
Subgroup analysis on a subset of the women who completed all four visits showed that exposure to childhood abuse was associated with higher mean BMI, DBP and HbA1c over time (Table S6). Means of WC, BMI, SBP, DBP, TC or HbA1c were not significantly different over time in IPV and non-partner rape groups (Tables S7 and S8).
Discussion
To our knowledge, this study is the first to examine the longitudinal relationships between exposures to various types of abuse and objectively determined cardiometabolic profiles over three years in a low- and middle-income country. This study demonstrated that exposure to any childhood abuse, physical childhood abuse and greater frequency of physical childhood abuse was associated with increasing BMI. The association of childhood abuse with WC showed a similar trend but did not reach statistical significance. However, we did not find significant associations of exposures to childhood abuse, IPV, or non-partner rape with changes in BP, TC, or HbA1c over time. There was at most a marginal modification effect of baseline rape-exposure status on some of the investigated associations indicating our observed outcomes to be independent of the impact of baseline rape-exposure, if any.
Possibly, the women in our study were too young and the follow-up period was too short to demonstrate the impact of abuse on cardiometabolic risk. Prolonged exposures to traditional behavioural risk factors, and the complex interplay with ageing, genetics and psychosocial stressors contribute to the development of cardiometabolic diseases and the risk factors [36–40]. Longer follow-up periods into middle age may be required to determine the cardiometabolic consequences in women exposed to abuse. For example, at least one study supports this explanation, with the suggestion that BP levels increase more rapidly in individuals with childhood abuse compared to those without childhood abuse after the age of 30 years [41]. While physiological effects such as sustained changes in BP, glucose and lipid levels, which have multifactorial aetiologies, may occur over a prolonged period, weight gain is directly related to unhealthy dietary behaviours and likely impacts BMI almost immediately. Unfortunately, we do not have histories of dietary habits and caloric intake in this study to verify this.
These study findings are consistent with a meta-analysis of studies in high-income nations which demonstrated that childhood maltreatment (i.e., sexual, physical, and emotional abuse, neglect, and family conflict) was associated with an increased risk of obesity (defined by BMI and WC) over a lifetime [9]. The latter meta-analysis was derived from five studies which included older women compared to our study (> 40 years old vs. 28 years old, respectively). Additionally, the lack of childhood abuse-WC association observed in the RICE study might be attributable to lesser changes in WC in the short-term compared to BMI, with prior studies reporting that Black African women have a greater amount of lower-body fat mass compared to men with the same BMI levels [42].
Evidence on the associations of childhood abuse with multiple cardiometabolic risk factors are largely from high-income countries [12, 43], with positive associations often based on outcomes that were self-reported or obtained from medical records, which are limitations [12]. A large longitudinal study in the UK (N = 3612) used objective measurements, similar to our study, for cardiometabolic risk assessments (BP, lipids, glucose, insulin, C-reactive protein, carotid intima-media thickness) [44]. It found no evidence that experiencing any psychosocial adversity (sexual, physical childhood abuse, parental neglect, and death) was associated with cardiometabolic risk factors in mid-adulthood women (mean age 51 years-old). However, the study was underpowered due to the low prevalence of psychosocial adversity in the study cohort.
Prior research investigating the relationships of exposures to IPV and non-partner rape with cardiometabolic showed inconsistent findings. Similar to our study findings, a 5-year follow-up study in young women (± 26 years old) in the USA showed no significant association between sexual and physical IPV and the change in BMI, over 5 years [8]. In a much older population ((± 47 years-old) with a longer follow-up period compared to our study, the Nurses’ Health Study II found physical and emotional IPV, but not sexual IPV, were modestly associated with the incidence of type 2 diabetes after 7 years of follow-up [45].
There are several plausible reasons for our findings of no association with BP, serum lipid or HbA1c levels apart from a young population and a short follow-up period. The RICE study did not collect data on the use of antihypertensive drugs which may have resulted in lower BP levels and an underestimation of hypertension prevalence and incidence. This also applies to the use of lipid lowering drugs which would have resulted in lower lipid levels and an underestimation of dyslipidaemia. This at least, in part, may explain the significant IPV-hypertension associations demonstrated in the cross-sectional analyses (at baseline) [28], but not over 3 years. This explanation is supported by Stene and colleagues who considered the use of anti-hypertensive medication as a proxy for hypertension in their study, and found physical and sexual IPV increased the risk of receiving antihypertensive medication [11].
It’s worth noting that literature from high-income countries has suggested that women suffering from exposure to severe abuse might lean towards food addiction, which was found to be associated with subsequent obesity and cardiometabolic risk [7]. Such behaviour may be limited among the women in our study due to poverty and the related high levels of food insecurity in their households; the majority were unemployed and depended on social grants [46, 47]. This would reduce their ability for potential increases in caloric intake leading to the lack of associations between abuse exposures and cardiometabolic diseases. Further studies need to be conducted in low-income settings considering the relevant socioeconomic milieu.
After all, we cannot indeed rule out the possibility that the high prevalence of childhood abuse and IPV (> 50%) in this study sample, could cause the sample to rather look homogenous for these exposures, and likely conceal associations, if any, with cardiovascular disease risk.
Strengths and limitations
The key strengths of our study include the longitudinal study design with cardiometabolic outcomes repeatedly, and objectively measured at four time-points. The study limitations include the large number of missing data at different time-points which are a source of attrition bias. Further, a young population with a short duration of follow-up might partly account for the insignificant changes in cardiometabolic risk in this study, given the long period and the contribution of older age to the development of cardiometabolic diseases. Dietary habits, the use of medication for hypertension, dyslipidaemia, and diabetes were not collected in this study and could not be considered in the analyses. Therefore, future research should include a longer follow-up period, perhaps into mid-life, together with the collection of the appropriate medication history, dietary habits and physical activity levels. Studies on the pathways through which exposure to abuse affects cardiometabolic outcomes may provide a better understanding of these relationships. Furthermore, the findings of this study from women who frequently used social and health services may not be generalisable to women in the general population who do not utilise such services. Finally, we also acknowledge the potential for Type 1 error in our analyses given the many comparisons, highlighting the importance of replicating these findings.
Conclusions
This prospective analysis demonstrated that exposures to childhood abuse were associated with increasing BMI levels over 3 years of follow-up in young South African women. The findings suggest that childhood abuse exposures potentially increase the risk for cardiometabolic diseases over time. Health promotions managing weight gain and healthy behavioural choices will likely benefit overall cardiometabolic profiles in women exposed to childhood abuse. Investigating the changes in cardiometabolic profiles in abused women over longer follow-up periods is warranted.
Supplementary Information
Acknowledgements
The authors acknowledge the contributions of the RICE study investigators: Prof. Naeemah Abrahams, Prof. Soraya Seedat, Prof. Carl Lombard, Prof. Bronwyn Myers, Prof. Claudia Garcia-Moreno and Prof. Rachel Jewkes, the late Prof. Gita Ramjee, and Ms Shibe Mhlongo.
Abbreviations
- AL
Allostatic load
- ANOVA
Analysis of variance
- BMI
Body mass index
- BP
Blood pressure
- CA
Childhood abuse
- DBP
Diastolic blood pressure
- HbA1c
Glycated haemoglobin
- IPV
Intimate partner violence
- NGSP
National Glycohaemoglobin Standardisation Programme
- RICE
Rape Impact Cohort Evaluation
- SBP
Systolic blood pressure
- TC
Total cholesterol
- WC
Waist circumference
Authors’ contributions
KAN led the analysis and drafted the paper. KAN, APK, and NP co-conceived and designed the paper. APK and NP co-conceived the broader study and funding acquisition. UN and APK assisted with the statistical analysis. KAN, UN, APK, and NP commented on all drafts and approved the final submission. APK is the guarantor.
Funding
The RICE study was funded by South African Medical Research Council: Flagships Awards Project SAMRC-RFA-IFSP-01-2013. The work reported herein was made possible through funding by the South African Medical Research Council through its Division of Research Capacity Development under the SAMRC Postdoctoral Programme from funding received from the South African National Treasury. The content hereof is the sole responsibility of the authors and do not necessarily represent the official views of the SAMRC or the funders.
Data availability
Data are available on request.
Declarations
Ethics approval and consent to participate
This study was conformed to the Declaration of Helsinki and the WHO ethical and safety guidelines on conducting research on gender-based violence and applies good clinical practice. The South African Medical Research Council Research Ethics Committee provided ethical approval for this study (SAMRC; EC019-10/2013), and all participants provided informed consent.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.WHO. Violence against women prevalence estimates. 2018. Global, regional and national prevalence estimates for intimate partner violence against women and global and regional prevalence estimates for non-partner sexual violence against women. Geneva: on behalf of the United Nations Inter-Agency Working Group on Violence Against Women Estimation and Data (UNICEF, UNFPA, UNODC, UNSD, UNWomen). 2021. Licence: CC BY-NC-SA 3.0 IGO.
- 2.Abrahams N, Devries K, Watts C, Pallitto C, Petzold M, Shamu S, et al. Worldwide prevalence of non-partner sexual violence: a systematic review. Lancet. 2014;383(9929):1648–54. [DOI] [PubMed] [Google Scholar]
- 3. WHO. Violence Against Women Prevalence Estimates. 2018: WHO, on behalf of the United Nations Inter-Agency Working Group on Violence Against Women Estimation and Data (VAW-IAWGED). 2021. https://www.who.int/publications/i/item/9789240026681.
- 4.Myers B, Bantjes J, Lochner C, Mortier P, Kessler RC, Stein DJ. Maltreatment during childhood and risk for common mental disorders among first year university students in South Africa. Soc Psychiatry Psychiatr Epidemiol. 2021;56(7):1175–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Machisa M, C L. K R, R J. The war at home: the Gauteng GBV indicators research study. Gender Links and the South African Medical Research Council. 2011.
- 6.Steele SJ, Abrahams N, Duncan K, Woollett N, Hwang B, O’Connell L, et al. The epidemiology of rape and sexual violence in the platinum mining district of Rustenburg, South Africa: prevalence, and factors associated with sexual violence. PLoS One. 2019;14(7): e0216449. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Mason SM, Flint AJ, Field AE, Austin SB, Rich-Edwards JW. Abuse victimization in childhood or adolescence and risk of food addiction in adult women. Obesity. 2013;21(12):E775-81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Mason SM, Ayour N, Canney S, Eisenberg ME, Neumark-Sztainer D. Intimate partner violence and 5-year weight change in young women: a longitudinal study. J Womens Health (Larchmt). 2017;26(6):677–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Danese A, Tan M. Childhood maltreatment and obesity: systematic review and meta-analysis. Mol Psychiatry. 2014;19(5):544–54. [DOI] [PubMed] [Google Scholar]
- 10.Felitti VJ, Anda RF, Nordenberg D, Williamson DF, Spitz AM, Edwards V, et al. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults. The adverse childhood experiences (ACE) study. Am J Prev Med. 1998;14(4):245–58. [DOI] [PubMed] [Google Scholar]
- 11.Stene LE, Jacobsen GW, Dyb G, Tverdal A, Schei B. Intimate partner violence and cardiovascular risk in women: a population-based cohort study. J Womens Health (Larchmt). 2013;22(3):250–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Basu A, McLaughlin KA, Misra S, Koenen KC. Childhood maltreatment and health impact: the examples of cardiovascular disease and type 2 diabetes mellitus in adults. Clin Psychol (New York). 2017;24(2):125–39. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Liu X, Logan J, Alhusen J. Cardiovascular risk and outcomes in women who have experienced intimate partner violence: an integrative review. J Cardiovasc Nurs. 2020;35(4):400–14. [DOI] [PubMed] [Google Scholar]
- 14.Wilson DK, Kliewer W, Sica DA. The relationship between exposure to violence and blood pressure mechanisms. Curr Hypertens Rep. 2004;6(4):321–6. [DOI] [PubMed] [Google Scholar]
- 15.Saba SK, Godwin J, Hong SH, Pan T, Chang Y, Brindle E, et al. Associations between childhood maltreatment and physiological dysregulation in adulthood: methodological decisions and implications. Child Abuse Negl. 2023;144: 106369. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Ladd CO, Huot RL, Thrivikraman KV, Nemeroff CB, Plotsky PM. Long-term adaptations in glucocorticoid receptor and mineralocorticoid receptor mRNA and negative feedback on the hypothalamo-pituitary-adrenal axis following neonatal maternal separation. Biol Psychiatry. 2004;55(4):367–75. [DOI] [PubMed] [Google Scholar]
- 17.Liu D, Diorio J, Tannenbaum B, Caldji C, Francis D, Freedman A, et al. Maternal care, hippocampal glucocorticoid receptors, and hypothalamic-pituitary-adrenal responses to stress. Science. 1997;277(5332):1659–62. [DOI] [PubMed] [Google Scholar]
- 18.Rutter M. Achievements and challenges in the biology of environmental effects. Proc Natl Acad Sci U S A. 2012;109(Suppl 2):17149–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Chida Y, Hamer M. An association of adverse psychosocial factors with diabetes mellitus: a meta-analytic review of longitudinal cohort studies. Diabetologia. 2008;51(12):2168–78. [DOI] [PubMed] [Google Scholar]
- 20.Rod NH, Grønbaek M, Schnohr P, Prescott E, Kristensen TS. Perceived stress as a risk factor for changes in health behaviour and cardiac risk profile: a longitudinal study. J Intern Med. 2009;266(5):467–75. [DOI] [PubMed] [Google Scholar]
- 21.Gasperin D, Netuveli G, Dias-da-Costa JS, Pattussi MP. Effect of psychological stress on blood pressure increase: a meta-analysis of cohort studies. Cad Saude Publica. 2009;25(4):715–26. [DOI] [PubMed] [Google Scholar]
- 22.Peer N, Abrahams N, Kengne AP. A systematic review of the associations of adult sexual abuse in women with cardiovascular diseases and selected risk factors. Glob Heart. 2020;15(1):65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Jakubowski KP, Murray V, Stokes N, Thurston RC. Sexual violence and cardiovascular disease risk: a systematic review and meta-analysis. Maturitas. 2021;153:48–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Stubbs A, Szoeke C. The effect of intimate partner violence on the physical health and health-related behaviors of women: A systematic review of the literature. Trauma Violence Abuse. 2022;23(4):1157–72. [DOI] [PubMed] [Google Scholar]
- 25.SA S. Statistical Release P0309.3. Mortality and causes of death in South Africa: Findings from death notification 2021 2021. Available from: https://www.statssa.gov.za/publications/P03093/P030932020.pdf.
- 26.Vos T, Lim SS, Abbafati C, Abbas KM, Abbasi M, Abbasifard M, et al. Global burden of 369 diseases and injuries in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet. 2020;396(10258):1204–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Nguyen KA, Kengne AP, Abrahams N, Jewkes R, Mhlongo S, Peer N. Associations of childhood maltreatment with hypertension in South African women: a cross-sectional study. BMJ Open. 2022;12(9):e057436. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Nguyen KA, Abrahams N, Jewkes R, Mhlongo S, Seedat S, Myers B, et al. The associations of intimate partner violence and non-partner sexual violence with hypertension in South African women. Int J Environ Res Public Health. 2022. 10.3390/ijerph19074026. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Abrahams N, Seedat S, Lombard C, Kengne AP, Myers B, Sewnath A, et al. Study protocol for a longitudinal study evaluating the impact of rape on women’s health and their use of health services in South Africa. BMJ Open. 2017;7(9): e017296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Bernstein DP, Stein JA, Newcomb MD, Walker E, Pogge D, Ahluvalia T, et al. Development and validation of a brief screening version of the childhood trauma questionnaire. Child Abuse Negl. 2003;27(2):169–90. [DOI] [PubMed] [Google Scholar]
- 31.Jewkes RK, Dunkle K, Nduna M, Jama PN, Puren A. Associations between childhood adversity and depression, substance abuse and HIV and HSV2 incident infections in rural South African youth. Child Abuse Negl. 2010;34(11):833–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. WHO. WHO Multi Country Study on Women’s Health and Domestic Violence Against Women. Initial results on prevalence, health outcomes and women’s responses. Geneva: World Health Organization; 2005.
- 33.Jewkes R, Nduna M, Levin J, Jama N, Dunkle K, Khuzwayo N, et al. A cluster randomized-controlled trial to determine the effectiveness of stepping stones in preventing HIV infections and promoting safer sexual behaviour amongst youth in the rural Eastern Cape, South Africa: trial design, methods and baseline findings. Trop Med Int Health. 2006;11(1):3–16. [DOI] [PubMed] [Google Scholar]
- 34.Seeman TE, Singer BH, Rowe JW, Horwitz RI, McEwen BS. Price of adaptation–allostatic load and its health consequences. MacArthur studies of successful aging. Arch Intern Med. 1997;157(19):2259–68. [PubMed] [Google Scholar]
- 35.Geronimus AT, Hicken M, Keene D, Bound J. Weathering and age patterns of allostatic load scores among Blacks and Whites in the united States. Am J Public Health. 2006;96(5):826–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Osborne MT, Shin LM, Mehta NN, Pitman RK, Fayad ZA, Tawakol A. Disentangling the links between psychosocial stress and cardiovascular disease. Circ Cardiovasc Imaging. 2020;13(8): e010931. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Seguin ML, Rangnekar A, Renedo A, Palafox B, McKee M, Balabanova D. Systematic review of frameworks used to conceptualise health pathways of individuals diagnosed with cardiovascular diseases. BMJ Glob Health. 2020. 10.1136/bmjgh-2020-002464. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Tsao CW, Aday AW, Almarzooq ZI, Alonso A, Beaton AZ, Bittencourt MS, et al. Heart disease and stroke Statistics-2022 update: A report from the American heart association. Circulation. 2022;145(8):e153–639. [DOI] [PubMed] [Google Scholar]
- 39.Frąk W, Wojtasińska A, Lisińska W, Młynarska E, Franczyk B, Rysz J. Pathophysiology of cardiovascular diseases: new insights into molecular mechanisms of atherosclerosis, arterial hypertension, and coronary artery disease. Biomedicines. 2022. 10.3390/biomedicines10081938. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Francula-Zaninovic S, Nola IA. Management of measurable variable cardiovascular disease’ risk factors. Curr Cardiol Rev. 2018;14(3):153–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Su S, Wang X, Pollock JS, Treiber FA, Xu X, Snieder H, et al. Adverse childhood experiences and blood pressure trajectories from childhood to young adulthood: the Georgia stress and heart study. Circulation. 2015;131(19):1674–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Stefan N. Causes, consequences, and treatment of metabolically unhealthy fat distribution. Lancet Diabetes Endocrinol. 2020;8(7):616–27. [DOI] [PubMed] [Google Scholar]
- 43.Li L, Pinto Pereira SM, Power C. Childhood maltreatment and biomarkers for cardiometabolic disease in mid-adulthood in a prospective British birth cohort: associations and potential explanations. BMJ Open. 2019;9(3):e024079. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Anderson EL, Fraser A, Caleyachetty R, Hardy R, Lawlor DA, Howe LD. Associations of adversity in childhood and risk factors for cardiovascular disease in mid-adulthood. Child Abuse Negl. 2018;76:138–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Mason SM, Wright RJ, Hibert EN, Spiegelman D, Jun HJ, Hu FB, et al. Intimate partner violence and incidence of type 2 diabetes in women. Diabetes Care. 2013;36(5):1159–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Sulla VZ. Precious. Overcoming poverty and inequality in South africa: an assessment of drivers, constraints and opportunities (English). Washington, D.C.: World Bank Group; 2022. [Google Scholar]
- 47.Abrahams N, Mhlongo S, Chirwa E, Lombard C, Dunkle K, Seedat S, et al. Rape survivors in South Africa: analysis of the baseline socio-demographic and health characteristics of a rape cohort. Glob Health Action. 2020;13(1): 1834769. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data are available on request.