Abstract
Objective.
Sexual and physical abuse predict cardiovascular disease (CVD) among women in the general population. Women living with HIV (WLWH) report more abuse and have higher CVD risk compared to other women, yet associations between abuse history and CVD have not been considered among WLWH. This study fills this gap, and describes possible pathways linking abuse to CVD risk among WLWH and women living without HIV (WLWOH).
Methods.
Using 25 years of data from the Women’s Interagency HIV Study (n=2734; WLWH n=1963; WLWOH n=771), we used longitudinal generalized estimating equations to test associations between sexual (SA) and physical abuse (PA) with CVD risk. Framingham (FRS-H) and the American College of Cardiology/American Heart Association-Pooled Cohort Equation (ACC/AHA-PCE) scores were examined. Analyses were stratified by HIV-serostatus.
Results.
Among WLWH, childhood SA was associated with higher CVD risk (βFRS-H=1.25, SE=1.08, p=0.005; βACC/AHA-PCE=1.14, SE=1.07, p=0.04) compared to no abuse. Adulthood SA was associated with higher CVD risk for WLWH (βFRS-H=1.39, SE=1.08, p<0.0001) and WLWOH (βFRS-H=1.58, SE=1.14, p=0.0006). Childhood PA was not associated with CVD risk for either group. Adulthood PA was associated with CVD risk for WLWH (βFRS-H=1.44, SE=1.07; p<0.0001, βACC/AHA-PCE=1.18, SE=1.06, p=0.002) and WLWOH (βFRS-H=1.68, SE=1.12, p<0.0001; βACC/AHA-PCE=1.24, SE=1.11, p=0.03). Several pathway factors were significant, including depression, smoking, and hepatitis-C infection.
Conclusions.
Life course abuse may increase CVD risk among WLWH and women at high risk of acquiring HIV. Some co-morbidities help explain the associations. Assessing abuse experiences in clinical encounters may help contextualize cardiovascular risk among this vulnerable population and inform intervention.
INTRODUCTION
Adverse childhood experiences (ACEs) are traumatic and stressful events occurring before age 18 years, including physical and sexual abuse, neglect, and household dysfunction[1,2]. ACEs are common; 61% of adults report at least one, and 16% report four or more ACEs[3,4]. ACEs are associated with a many poor physical, behavioral, and mental health outcomes across the life course[1,5], with sexual and physical abuse (the most severe ACEs) yielding the strongest associations with later life outcomes[3,4,6]. Increasingly, providers are screening for ACEs in pediatric and adult primary care settings[7,8], with California recently mandating insurance coverage for ACE screening statewide[9]. Thus, while ACEs are an important concern for children, ACEs also have far-reaching impacts for health and clinical practice for adults.
Studies show consistent associations between ACEs and adulthood cardiovascular disease (CVD) in the general population[1,10–12]. The landmark CDC Adverse Childhood Experiences Study (n=13,494) found those with ≥4 ACEs experienced 2-fold increased risk of stroke and ischemic heart disease compared to those with no ACEs[1]. Dose-response associations have been consistently observed with cardiometabolic risks, including hypertension, dyslipidemia, obesity, and diabetes[2,10]. Despite this evidence, researchers have not considered the ACEs and CVD association among people living with HIV (PLWH). This is a key omission in the literature because PLWH and those at risk for acquiring HIV, including sexual and gender minority populations, tend to experience more ACEs than people without HIV[3,13–16]. Moreover, people with ACEs are more likely to experience intimate partner violence and other abuses later in life[17], particularly women living with HIV[18,19]. In addition, with increased utilization of effective antiretroviral therapy, AIDS mortality has decreased and survival times have increased among PLWH[20]. Yet with this prolonged survival, PLHW experience a 2-fold increased risk of CVD compared to the general population[21–24], and CVD emerges earlier in life and is a leading cause of non-AIDS mortality[25–27]. Given that PLWH are more likely to experience adversity across the life course and also have a high degree of CVD morbidity, it is plausible that abuse may contribute to cardiovascular risk among PLWH. This possibility has not been tested.
The pathways linking abuse experience to CVD risk among the general population include structural, psychosocial, behavioral, and physiologic factors[10]. While similar pathways are likely at play for PLWH, there may also be important differences given the underlying inflammatory state associated with HIV[28], adverse cardiovascular effects of historical antiretroviral treatment[29], and higher prevalence of health risk behaviors and co-morbidities like smoking[30], injection drug use, and hepatitis-C virus (HCV) infection[31]. Thus, the identification of pathways linking abuse with CVD among PLWH is warranted and may reveal novel opportunities for intervention.
A recent Scientific Statement by the American Heart Association (AHA)[10] and expert consensus from the National Institutes of Health/Heart Lung and Blood Institute (NIH/NHLBI)[6] indicated that while the association between ACEs and CVD has been consistent across studies in the general population, gaps remain, including characterizing the association among vulnerable populations and describing pathways through which ACEs may influence adulthood CVD. Therefore, we examined the association between life course history of sexual and physical abuse with CVD risk using 25 years of prospectively collected data among women living with HIV (WLWH) and socio-demographically similar women living without HIV (WLWOH) but at high risk for acquiring HIV. We also examined potential pathways through which abuse may contribute to CVD risk in this vulnerable population.
METHODS
Study Population
The Women’s Interagency HIV Study (WIHS) was a longstanding multi-site, prospective cohort of United States women living with and without HIV. Women were recruited at 10 sites over 4 time periods starting in 1994, with follow up through 2019. WIHS methods and cohort characteristics have been described elsewhere[32]. Briefly, WLWH were eligible to participate if they were HIV-seropositive, assigned female sex at birth, and did not acquire HIV perinatally. WLWOH were eligible if she or her partner met at least one of the following criteria within 5 years prior to enrollment: diagnosis of >1 sexually transmitted infection, unprotected sex with ≥6 individuals, having sex for money, drugs, or shelter, sex with an HIV-positive individual, injection drug use, unprescribed drug use (crack, cocaine, heroin, methamphetamine). Information on HIV treatment during the study period was repeatedly assessed. At the last study visit included in the present analysis, 85.5% were receiving antiretroviral treatment, with combination antiretroviral therapy integrase inhibitors (39%), protease inhibitors (24.7%), and nucleoside/nucleotide analogue reverse transcriptase inhibitors (18.2%) being the most prevalent. Data were collected using structured in-person interviews and standardized physical and laboratory assessments; study visits occurred every six months. Beginning in 2019, follow up is continuing as part of the Multicenter AIDS Cohort Study/WIHS Combined Cohort Study (MWCCS)[33].
HIV seroconversions (n=21) and participants missing abuse and outcome information were excluded, resulting in n=2734 for the present analysis (n=771 WLWOH; n=1963 WLWH). Participants had a median of 23 study visits (IQR: 12–31; WLWOH: median 23, IQR: 13–31; WLWH: median 23, IQR: 12–31). The median date of the abuse history data collected at enrollment was 11/07/2006 (IQR: 05/05/1995–02/28/2007) for WLWH and 11/27/2006 (IQR: 03/13/1995–03/13/2007) for WLWOH. Outcomes were ascertained through 9/30/2019. Institutional review boards at the clinical research centers approved protocols, and participants provided written informed consent. This sub-study was additionally approved by the University at Albany.
Measures
Cardiovascular Risk Scores.
We calculated the Framingham Risk Score for hard events (FRS-H), which indicates the 10-year risk for myocardial infarction and coronary death[34–36]; and the American College of Cardiology/American Heart Association-Pooled Cohort Equations for atherosclerotic disease (ACC/AHA-PCE) which indicates the 10-year risk of hard atherosclerotic events, defined as coronary death, non-fatal myocardial infarction, and fatal or non-fatal stroke[37]. Scores were calculated using age, hypertension treatment (yes/no), smoking status (yes/no); measured systolic blood pressure (mmHg), total and HDL cholesterol (Roche Diagnostics Corporation, Indianapolis, IN). The ACC/AHA-PCE additionally included race (Black/African-American vs. other) and diabetes (fasting glucose ≥126 mg/dl, hemoglobin A1c ≥6.5%, self-report of diabetes, or taking diabetes medication at the visit). Risk scores were calculated for established valid age ranges (FRS-H: 30–79 years[34]; ACC/AHA-PCE=40–75 years[37]). Risk scores were natural log transformed due to skewed distributions.
Lifetime history of sexual and physical abuse.
At WIHS enrollment, participants answered the following questions on abuse history: “At any time in your life, has anyone ever pressured or forced you to have sexual contact? By sexual contact I mean them touching your sexual parts, you touching their sexual parts, or sexual intercourse.”; and “Have you ever experienced serious physical violence (physical harm by another person)? By that I mean were you ever hurt by a person using an object or were you ever slapped, hit, punched, kicked?” For “yes” responses to either item, follow-up questions determined the age when the abuse first occurred. Two three-category variables were created reflecting whether the abuse occurred in childhood (before 18 years), in adulthood, or the abuse did not occur. Physical abuse (PA) and sexual abuse (SA) were examined separately in analysis.
Covariates.
Potential confounders included study site, enrollment wave, education at enrollment (high school or less versus more than high school), and among WLWH, CD4 T cell count/uL, use of antiretroviral therapy, and HIV viral load below the lower limit of quantification (LLQ). Adjusted models for FRS-H additionally included race. Age is a component of the CVD risk outcome variables and was therefore not controlled for in analysis.
Pathway variables.
We considered several variables that could possibly link abuse history to CVD events and atherosclerotic disease[10]: depressive symptoms as measured by the Center for Epidemiologic Studies Depression Scale[38], body mass index (BMI; kg/m2), alcohol use (number of drinks per week), injection drug use (yes/no), cholesterol medication use (yes/no), hepatitis-C virus (HCV) serostatus, and the component parts of the CVD risk scores (total cholesterol, HDL, systolic blood pressure (SBP), hypertension medication use, diabetes, cigarette smoking).
Statistical Analysis
We defined the baseline visit for each participant as the first ascertainment of the CVD risk scores. Given the different age ranges applicable for the outcome measures, baseline assessments for FRS-H and AHA/ACC-PCE could occur at different visits. Therefore, we examined the distribution of study variables according to abuse history type within strata of HIV separately for the baseline FRS-H and AHA/ACC-PCE visit. Also, we tested the baseline cross-sectional associations between abuse history, CVD risk, and other study variables within HIV strata using χ2-square and ANOVA tests for categorical and continuous variables respectively. Next, longitudinal associations between abuse history with CVD risk were estimated using repeated measures linear regression models with generalized estimating equations (GEE), with exchangeable covariance structures. GEE is an extension of generalized linear models applicable to longitudinal data as it can accommodate missing observations and account for correlated observations within participants, and models the average (marginal) response over time[39]. Unadjusted and adjusted GEE models were fit sequentially. Potential pathways were assessed in two ways. First, pathway variables that were not part of the risk scores were added to the adjusted models. We interpreted attenuated coefficients with the addition of pathway variables to models by approximately 10% and change in significance thresholds to suggest possible mediation[40,41]. Second, we examined longitudinal associations between abuse history with component parts of the CVD risk scores and other possible pathways factors as separate outcomes with GEE linear and logistic regressions. All models were stratified by HIV serostatus. Adjusted models included both time-varying (e.g., CD4 count) and time-invariant (e.g., study site) covariates. The regression parameter estimates for CVD risk were back-transformed from the natural log scale and can be interpreted as percent change in 10-year CVD risk according to abuse exposure. Statistical significance was defined as two-tailed α<0.05. Analyses were conducted using SAS 9.4 (Cary, NC).
RESULTS
Sexual Abuse History and CVD Risk at Baseline
Table 1 lists participant characteristics according to SA history and HIV serostatus. There were no differences in baseline CVD risk according to SA among WLWH and WLWOH. Among WLWH, the prevalence of childhood and adulthood SA was 18% and 20% respectively, and the average baseline FRS-H was 1.61% (SD=2.08) and ACC/AHA-PCE was 5.21% (SD=8.05). Among WLWOH, the prevalence of childhood and adulthood SA was 19% and 20% respectively, and the average initial FRS-H was 1.95% (SD=3.15) and ACC/AHA-PCE was 3.64% (SD=7.22). There was a small degree of overlap for sexual and physical abuse history: 4.69% of WLWH reported both SA and PA in childhood, and 14.06% reported both forms of abuse in adulthood. For WLWOH, 4.15% reported both SA and PA occurring in childhood and 14.92% reported both in adulthood. Some significant differences according to SA history were noted at baseline, including race, education, depressive symptoms, smoking, HCV, injection drug use and antiretroviral use among WLWH. A high proportion of WLWH had detectable viral load across SA groups at the baseline visit (67–70%). Those with child SA history had the lowest antiretroviral use rates (p=0.04). Among WLWOH, significant baseline differences in age at first FRS-H assessment, race, depressive symptoms, injection drug use, and HCV were also observed. At the last study visit for WLWH, the prevalence of detectable viral load was 35.8%.
Table 1.
Participant characteristics at baseline according to sexual abuse history and HIV serostatus among WIHS women
| WLWOH (n=771) | WLWH (n=1963) | |||||||
|---|---|---|---|---|---|---|---|---|
|
| ||||||||
| Characteristic | Child sexual abuse (n=147) | Adult sexual abuse (n=155) | No sexual abuse (n=469) | p^ | Child sexual abuse (n=361) | Adult sexual abuse (n=401) | No sexual abuse (n=1201) | p^ |
|
| ||||||||
| CVD risk scores, Mean (SD) or n (%) | ||||||||
| FRS-H, 10-year risk % | 1.79 (2.72) | 2.13 (3.55) | 1.94 (3.14) | 0.3 | 1.59 (2.67) | 1.80 (3.07) | 1.55 (2.75) | 0.35 |
| ACC/AHA-PCE, 10-year risk % | 4.83 (5.84) | 5.61 (8.17) | 5.94 (7.22) | 0.59 | 4.72 (7.93) | 5.15 (7.60) | 5.41 (8.27) | 0.19 |
| CVD components, covariates, M (SD) or n (%) | ||||||||
| Age, first FRS-H assessment, years | 42.33 (7.86) | 44.76 (8.54) | 42.78 (8.03) | 0.03 | 43.11 (7.54) | 43.26 (7.34) | 42.62 (8.01) | 0.31 |
| Age, first ACC/AHA-PCE assessment, years | 47.67 (5.24) | 49.09 (6.43) | 48.08 (5.67) | 0.24 | 47.01 (5.57) | 47.19 (5.43) | 47.78 (5.60) | 0.12 |
| Total cholesterol, mg/dL | 181.12 (40.24) | 181.24 (37.23) | 182.72 (38.17) | 0.89 | 180.09 (40.21) | 183.18 (43.44) | 181.17 (42.47) | 0.6 |
| HDL cholesterol, mg/dL | 54.15 (15.80) | 56.04 (17.90) | 55.60 (17.81) | 0.65 | 49.87 (19.08) | 48.86 (18.39) | 49.67 (18.81) | 0.72 |
| Systolic blood pressure, mmHg | 120.02 (15.40) | 124.16 (22.69) | 124.34 (21.11) | 0.12 | 120.35 (17.75) | 118.76 (15.69) | 121.35 (18.36) | 0.05 |
| Body mass index, kg/m2 | 32.59 (8.52) | 32.71 (8.66) | 31.53 (8.33) | 0.28 | 29.86 (8.79) | 29.56 (8.26) | 30.27 (8.41) | 0.37 |
| Depressive symptoms, CES-D score | 17.43 (14.31) | 19.23 (13.54) | 13.41 (12.39) | 0.0003 | 16.43 (12.76) | 16.73 (12.62) | 13.71 (11.75) | 0.0002 |
| Alcohol use, drinks per week | 7.30 (18.84) | 5.02 (14.56) | 6.47 (18.35) | 0.57 | 5.03 (27.09) | 2.42 (8.34) | 3.23 (10.67) | 0.05 |
| Race, black | 74 (61.16) | 70 (51.47) | 248 (70.25) | 0.0004 | 188 (56.63) | 220 (56.99) | 720 (67.23) | <0.0001 |
| Education, high school or less | 72 (59.50) | 80 (58.82) | 228 (64.89) | 0.39 | 211 (63.55) | 220 (56.99) | 752 (70.28) | <0.0001 |
| Cholesterol medication use | 10 (8.26) | 11 (8.09) | 25 (7.08) | 0.88 | 22 (6.63) | 30 (7.77) | 88 (8.22) | 0.64 |
| Blood pressure medication use | 28 (23.14) | 46 (33.82) | 91 (25.78) | 0.11 | 88 (26.51) | 112 (29.02) | 297 (27.73) | 0.75 |
| Current smoker | 79 (65.29) | 80 (58.82) | 201 (56.94) | 0.27 | 185 (55.72) | 195 (50.52) | 473 (44.16) | 0.0005 |
| Diabetes | 8 (6.61) | 19 (13.97) | 28 (10.76) | 0.16 | 28 (8.43) | 33 (8.55) | 93 (8.68) | 0.99 |
| Hepatitis C virus infection | 19 (15.70) | 34 (25.19) | 54 (15.34) | 0.03 | 94 (28.40) | 122 (31.77) | 202 (18.91) | <0.0001 |
| Injection drug use | 32 (26.45) | 42 (30.88) | 53 (15.01) | 0.0001 | 121 (36.45) | 132 (34.20) | 193 (18.04) | <0.0001 |
| Covariates for WLWH, n (%) or mean (SD) | ||||||||
| HIV viral load, detectable | 220 (67.90) | 255 (67.28) | 725 (70.39) | 0.45 | ||||
| Antiretroviral therapy use | 204 (61.45) | 242 (62.69) | 726 (67.79) | 0.04 | ||||
| CD4 count, cells/uL | 461.79 (314.28) | 452.31 (288.12) | 426.91 (291.90) | 0.26 | ||||
p-value for ANOVA or χ2 tests for continuous and categorical variables respectively. Cell entries reflect characteristics at the first FRS-H assessment unless otherwise specified.
HIV=Human immunodeficiency virus; WIHS=Women’s Interagency HIV Study: WLWOH=Women living without HIV; WLWH=Women living with HIV; CVD=Cardiovascular disease; SD=Standard deviation; FRS-H=Framingham risk score for hard events; ACC/AHA-PCE=American college of cardiology/American heart association-pooled cohort equations; HDL=High-density lipoprotein; CES-D=Centers for epidemiologic studies depression scale; CD4=Cluster of differentiation 4.
Physical Abuse History and CVD Risk at Baseline
Table 2 lists participant characteristics according to PA history and HIV serostatus. Among WLWH, the prevalence of childhood and adulthood PA was 10.55% and 41.77% respectively. There were no differences in baseline CVD risk according to PA history for WLWH. Among WLWOH, the prevalence of childhood and adulthood PA was 10.12% and 45.01% respectively. There were significant baseline differences in FRS-H for WLWH, with the childhood PA group having the lowest (Mean=1.51, SD=2.76) and the adulthood group having the highest FRS-H scores (Mean=1.71, SD=2.74). Some differences according to PA history at baseline were also noted for WLWH, including age at first AHA/ACC-PCE assessment, race, depressive symptoms, smoking, injection drug use, HCV, and HIV viral load below the LLQ. A high proportion of WLWH had detectable viral load across PA groups at the baseline visit (65.17%–70.14%). Those with adult PA history had the lowest antiretroviral use rates (p=0.02). Among WLWOH, baseline differences were evident for age at first FRS-H assessment, total cholesterol, blood pressure, BMI, depressive symptoms, injection drug use, and HCV.
Table 2.
Participant characteristics at baseline according to physical abuse history and HIV serostatus among WIHS women
| WLWOH (n=771) | WLWH (n=1963) | |||||||
|---|---|---|---|---|---|---|---|---|
|
|
||||||||
| Characteristic | Child physical abuse (n=78) | Adult physical abuse (n=347) | No physical abuse (n=346) | p^ | Child physical abuse (n=207) | Adult physical abuse (n=820) | No physical abuse (n=936) | p^ |
|
| ||||||||
| CVD risk scores, Mean (SD) or n (%) | ||||||||
| FRS-H, 10-year risk % | 1.46 (1.79) | 2.24 (3.57) | 1.72 (2.82) | 0.08 | 1.51 (2.76) | 1.71 (2.74) | 1.54 (2.78) | 0.03 |
| ACC/AHA-PCE, 10-year risk % | 5.18 (4.64) | 5.92 (7.82) | 5.35 (6.84) | 0.73 | 4.99 (8.57) | 4.79 (7.01) | 5.73 (8.98) | 0.2 |
| CVD components, covariates, M (SD) or n (%) | ||||||||
| Age, first FRS-H assessment, years | 41.44 (7.55) | 44.12 (8.13) | 42.27 (8.18) | 0.01 | 40.85 (7.28) | 43.77 (7.39) | 42.42 (8.20) | <0.0001 |
| Age, first ACC/AHA-PCE assessment, years | 47.16 (5.24) | 48.59 (5.97) | 48.01 (5.68) | 0.35 | 46.28 (5.51) | 47.52 (5.37) | 47.70 (5.76) | 0.07 |
| Total cholesterol, mg/dL | 181.19 (38.89) | 181.85 (37.58) | 182.55 (39.02) | 0.04 | 178.89 (40.44) | 179.97 (40.68) | 183.32 (44.05) | 0.19 |
| HDL cholesterol, mg/dL | 49.88 (13.90) | 56.03 (18.32) | 55.96 (16.89) | 0.05 | 49.44 (16.78) | 49.78 (19.47) | 49.34 (18.52) | 0.9 |
| Systolic blood pressure, mmHg | 115.83 (16.42) | 124.88 (20.61) | 123.49 (20.97) | 0.01 | 120.00 (18.18) | 119.70 (16.81) | 120.60 (18.43) | 0.09 |
| Body mass index, kg/m2 | 32.98 (9.57) | 32.94 (8.47) | 30.62 (7.97) | 0.004 | 30.19 (8.59) | 29.65 (8.19) | 30.37 (8.64) | 0.26 |
| Depressive symptoms, CES-D score | 17.09 (14.48) | 17.89 (13.78) | 12.17 (12.53) | <0.0001 | 18.65 (13.49) | 16.68 (12.78) | 12.42 (10.86) | <0.0001 |
| Alcohol use, drinks per week | 2.79 (6.20) | 7.60 (20.97) | 5.47 (14.72) | 0.09 | 4.27 (16.23) | 3.79 (18.39) | 2.81 (9.88) | 0.29 |
| Race, black | 39 (66.10) | 190 (62.50) | 163 (65.99) | 0.66 | 94 (51.37) | 464 (59.26) | 570 (69.26) | <0.0001 |
| Education, high school or less | 33 (55.93) | 191 (62.83) | 156 (63.16) | 0.57 | 129 (70.49) | 499 (63.81) | 555 (67.44) | 0.13 |
| Cholesterol medication use | 5 (8.47) | 29 (9.54) | 12 (4.86) | 0.11 | 9 (4.92) | 55 (7.02) | 76 (9.23) | 0.08 |
| Blood pressure medication use | 12 (20.34) | 93 (30.59) | 60 (24.29) | 0.12 | 46 (25.14) | 208 (26.56) | 243 (29.93) | 0.29 |
| Current smoker | 36 (61.02) | 186 (61.18) | 138 (55.87) | 0.43 | 107 (58.47) | 416 (53.13) | 330 (40.10) | <0.0001 |
| Diabetes | 6 (10.17) | 31 (10.20) | 28 (11.34) | 0.9 | 13 (7.10) | 66 (8.43) | 75 (9.11) | 0.66 |
| Hepatitis C virus infection | 9 (15.25) | 66 (21.71) | 32 (29.91) | 0.03 | 52 (28.42) | 235 (30.09) | 131 (16.00) | <0.0001 |
| Injection drug use | 9 (15.25) | 81 (26.64) | 37 (14.98) | 0.002 | 63 (34.43) | 260 (33.25) | 123 (14.95) | <0.0001 |
| Covariates for WLWH, n (%) or mean (SD) | ||||||||
| HIV viral load, detectable | 116 (65.17) | 525 (69.26) | 559 (70.14) | 0.43 | ||||
| Antiretroviral therapy use | 120 (65.57) | 486 (60.27) | 566 (68.77) | 0.02 | ||||
| CD4 count cells/uL | 459.82 (308.18) | 449.56 (294.39) | 424.20 (293.61) | 0.14 | ||||
p-value for ANOVA or χ2 tests for continuous and categorical variables respectively. Cell entries reflect characteristics at the first FRS-H assessment unless otherwise specified.
HIV=Human immunodeficiency virus; WIHS=Women’s Interagency HIV Study: WLWOH=Women living without HIV; WLWH=Women living with HIV; CVD=Cardiovascular disease; SD=Standard deviation; FRS-H=Framingham risk score for hard events; ACC/AHA-PCE=American college of cardiology/American heart association-pooled cohort equations; HDL=High-density lipoprotein; CES-D=Centers for epidemiologic studies depression scale; CD4=Cluster of differentiation 4.
Supplemental Tables 1–2 list participant characteristics according to abuse history and HIV status at the baseline AHA/ACC-PCC visit. Results were similar to the characteristics described for FRS-H.
Associations Between Life Course Sexual and Physical Abuse History and CVD risk
Table 3 shows the longitudinal associations between SA and PA and CVD risk. Among WLWH, childhood SA was significantly associated with higher FRS-H (β=1.25, SE=1.08, p=0.005) and ACC/AHA-PCE (β=1.14, SE=1.07, p=0.04), adjusted for covariates. Associations were attenuated for FRS-H (β=1.16, SE=1.08, p=0.06; percent change=7.2%) and ACC/AHA-PCE (β=1.08, SE=1.06, p=0.21; percent change=5.3%) when adding pathway variables to the models, suggesting that behavioral and clinical factors may help explain the associations. Also among WLWH, adulthood SA was positively associated with FRS-H (β=1.39, SE=1.08, p<0.0001), but not ACC/AHA-PCE (β=1.11, SE=1.07, p=0.10) in covariate adjusted models. Among WLWOH, there was no significant association between childhood SA and either measure of CVD risk. Adulthood SA was positively associated with FRS-H (β=1.58, SE=1.14, p=0.0006) but not ACC/AHA-PCE (β=1.21, SE=1.12, p=0.10) in adjusted analyses for WLWOH.
Table 3.
Repeated measures generalized estimating equations for the associations between cardiovascular risk scores with sexual and physical abuse history among WIHS womena
| WLWOH | WLWH | ||||||
|---|---|---|---|---|---|---|---|
| Unadjusted | Covariates | Pathways | Unadjusted | Covariates | Pathways | ||
|
| |||||||
| FRS-H | |||||||
| Child sexual abuse | 1.21 (1.15) | 1.25 (1.14) | 1.22 (1.14) | 1.26 (1.08) | 1.25 (1.08) | 1.16 (1.08) | |
| 0.15 | 0.1 | 0.14 | 0.003 | 0.005 | 0.06 | ||
| Adult sexual abuse | 1.51 (1.14) | 1.58 (1.14) | 1.42 (1.14) | 1.34 (1.08) | 1.39 (1.08) | 1.30 (1.08) | |
| 0.002 | 0.0006 | 0.008 | <0.0001 | <0.0001 | 0.0004 | ||
| No sexual abuse (Reference) | -- | -- | -- | -- | -- | -- | |
| Child physical abuse | 1.09 (1.20) | 1.12 (1.19) | 1.03 (1.19) | 1.12 (1.11) | 1.12 (1.11) | 1.01 (1.11) | |
| 0.62 | 0.52 | 0.85 | 0.26 | 0.27 | 0.92 | ||
| Adult physical abuse | 1.67 (1.12) | 1.68 (1.12) | 1.42 (1.11) | 1.41 (1.07) | 1.44 (1.07) | 1.29 (1.07) | |
| <0.0001 | <0.0001 | 0.0004 | <0.0001 | <0.0001 | <0.0001 | ||
| No physical abuse (Reference) | -- | -- | -- | -- | -- | -- | |
| ACC/AHA-PCE | |||||||
| Child sexual abuse | 0.90 (1.13) | 0.90 (1.13) | 0.85 (1.12) | 1.12 (1.07) | 1.14 (1.07) | 1.08 (1.06) | |
| 0.4 | 0.38 | 0.18 | 0.09 | 0.04 | 0.21 | ||
| Adult sexual abuse | 1.21 (1.13) | 1.21 (1.12) | 1.16 (1.12) | 1.07 (1.07) | 1.11 (1.07) | 1.05 (1.07) | |
| 0.11 | 0.1 | 0.19 | 0.29 | 0.1 | 0.42 | ||
| No sexual abuse (Reference) | -- | -- | -- | -- | -- | -- | |
| Child physical abuse | 0.88 (1.19) | 0.88 (1.18) | 0.81 (1.17) | 0.95 (1.09) | 1.01 (1.10) | 0.96 (1.10) | |
| 0.44 | 0.44 | 0.18 | 0.7 | 0.94 | 0.64 | ||
| Adult physical abuse | 1.26 (1.11) | 1.24 (1.11) | 1.10 (1.10) | 1.14 (1.06) | 1.18 (1.06) | 1.12 (1.05) | |
| 0.02 | 0.03 | 0.33 | 0.01 | 0.002 | 0.03 | ||
| No physical abuse (Reference) | -- | -- | -- | -- | -- | -- | |
Top cell entries are β (Standard Error), bottom cell entries are p-values. Models among WLWH included 1963 individuals with 25,597 and 20,303 observations for FRS-H and ACC/AHA-PCE respectively. Models among WLWOH included 771 individuals with 10,535 and 7699 observations for FRS-H and ACC/AHA-PCE respectively. Models for sexual and physical abuse were fit separately. Covariate models include enrollment wave, study site, education, and for WLWH antiretroviral treatment, CD4 count, and detectable viral load. Pathways models additionally include for cholesterol medication use, body mass index, depressive symptoms, hepatitis-C infection, alcohol use, and injection drug use. FRS-H models also adjust for race and diabetes. WIHS=Women’s Interagency HIV Study: WLWOH=Women living without HIV; WLWH=Women living with HIV; FRS-H=Framingham risk score for hard events; ACC/AHA-PCE=American college of cardiology/American heart association-pooled cohort equations.
There were no significant associations between childhood PA and CVD risk among WLWH or WLWOH (Table 3). For both WLWH and WLWOH, there was a positive association between adulthood PA and FRS-H (WLWH: β=1.44, SE=1.07, p<0.0001; WLWOH: β=1.68, SE=1.12, p<0.0001) and ACC/AHA-PCE (WLWH β=1.18, SE=1.06, p=0.002; WLWOH: β=1.24, SE=1.11, p=0.03) in adjusted models. Parameter estimates were attenuated in pathways models but remained significant for both groups except for ACC/AHA-PCE among WLWOH. Among WLWOH, percent change in parameter estimates from covariate adjusted to pathways models was 15% and 11% for adulthood PA with FRS-H and ACC/AHA-PCE assessed CVD risk respectively.
Pathway Factors by Sexual Abuse History
Table 4 shows longitudinal associations between SA and possible pathway variables. Among WLWH, childhood SA was associated with more depressive symptoms (β=3.54, SE=0.59, p<0.0001), cholesterol medication use (OR=1.74, 95%CI:1.18, 2.57), smoking (OR=1.71, 95%CI:1.35, 2.16), and HCV infection (OR=1.54, 95%CI:1.16, 2.05). Also, among WLWH, adulthood SA was associated with depressive symptoms (β=3.57, SE=0.54, p<0.0001), smoking (OR=1.64, 95%CI:1.31, 2.05), and HCV (OR=2.03, 95%CI:1.55, 2.66) compared to those with no abuse history. Among WLWOH, childhood (β =4.28, SE=0.92, p<0.0001) and adulthood (β =5.78, SE=0.84, p<0.0001) SA were each positively associated with higher depression scores compared to those with no abuse. Childhood SA was associated with systolic blood pressure (β =−2.66, SE=1.30, p=0.04) and alcohol use (β=1.72, SE=0.85, p=0.04). Adulthood SA was also associated with HCV (OR=2.19, 95%CI:1.36, 3.53).
Table 4.
Repeated measures linear and logistic GEE models for the association between components of the CVD risk scores and pathway variables with sexual abuse history among WIHS womena
| WLWOH | WLWH | |||
|---|---|---|---|---|
| β (SE) or OR | p or 95% CI | β (SE) or OR | p or 95% CI | |
|
| ||||
| Total cholesterol | ||||
| Child sexual abuse | 2.41 (3.30) | 0.47 | −2.21 (1.96) | 0.26 |
| Adult sexual abuse | −2.75 (2.84) | 0.33 | 1.80 (1.92) | 0.35 |
| Reference | -- | -- | -- | -- |
| HDL | ||||
| Child sexual abuse | 0.05 (1.40) | 0.97 | −0.55 (0.94) | 0.55 |
| Adult sexual abuse | 0.77 (1.50) | 0.61 | 0.80 (0.87) | 0.93 |
| Reference | -- | -- | -- | -- |
| Systolic blood pressure | ||||
| Child sexual abuse | −2.66 (1.30) | 0.04 | −0.84 (0.80) | 0.29 |
| Adult sexual abuse | −1.33 (1.41) | 0.34 | −0.87 (0.76) | 0.25 |
| Reference | -- | -- | -- | -- |
| Body mass index | ||||
| Child sexual abuse | 1.05 (0.82) | 0.2 | 1.02 (0.54) | 0.06 |
| Adult sexual abuse | −0.26 (0.80) | 0.74 | −0.08 (0.46) | 0.87 |
| Reference | -- | -- | -- | -- |
| Depressive symptoms | ||||
| Child sexual abuse | 4.28 (0.92) | <0.0001 | 3.54 (0.59) | <0.0001 |
| Adult sexual abuse | 5.78 (0.84) | <0.0001 | 3.57 (0.54) | <0.0001 |
| Reference | -- | -- | -- | -- |
| Alcohol use | ||||
| Child sexual abuse | 1.72 (0.85) | 0.04 | 0.74 (0.39) | 0.06 |
| Adult sexual abuse | 1.40 (0.97) | 0.15 | 0.44 (0.38) | 0.25 |
| Reference | -- | -- | -- | -- |
| Blood pressure medication | ||||
| Child sexual abuse | 0.86 | 0.53, 1.39 | 1.01 | 0.74, 1.37 |
| Adult sexual abuse | 1.43 | 0.96, 2.13 | 1 | 0.75, 1.33 |
| Reference | -- | -- | -- | -- |
| Cholesterol medication | ||||
| Child sexual abuse | 1.52 | 0.86, 2.72 | 1.74 | 1.18, 2.57 |
| Adult sexual abuse | 0.96 | 0.54, 1.70 | 1.05 | 0.68, 1.62 |
| Reference | -- | -- | -- | -- |
| Diabetes | ||||
| Child sexual abuse | 1.05 | 0.48, 2.31 | 1.5 | 0.93, 2.42 |
| Adult sexual abuse | 1.62 | 0.82, 3.22 | 0.84 | 0.50, 1.44 |
| Reference | -- | -- | -- | -- |
| Smoker | ||||
| Child sexual abuse | 1.38 | 0.94, 2.05 | 1.71 | 1.35, 2.16 |
| Adult sexual abuse | 1.38 | 0.94, 2.02 | 1.64 | 1.31, 2.05 |
| Reference | -- | -- | -- | -- |
| HCV | ||||
| Child sexual abuse | 1.06 | 0.55, 1.88 | 1.54 | 1.16, 2.05 |
| Adult sexual abuse | 2.19 | 1.36, 3.53 | 2.03 | 1.55, 2.66 |
| Reference | -- | -- | -- | -- |
| IDU | ||||
| Child sexual abuse | 2.38 | 0.74, 7.63 | 1.37 | 0.69, 2.70 |
| Adult sexual abuse | 1.2 | 0.37, 3.93 | 1.88 | 0.94, 3.75 |
| Reference | -- | -- | -- | -- |
All models adjusted for age, race, education, wave of enrollment, and study site. Models for WLWH are additional adjusted for antiretroviral therapy, CD4 count, and detectable viral load.
GEE=Generalized estimating equations; CVD=Cardiovascular disease; WIHS=Women’s Interagency HIV Study; WLWOH=Women living without HIV; WLWH=Women living with HIV; SE=Standard error; OR=Odds ratio; p=p-value; CI=Confidence interval; HDL=high-density lipoprotein; HCV=Hepatitis-C infection; IDU=injection drug use.
Pathway Factors by Physical Abuse History
Table 5 shows the associations between PA and possible pathway variables. Among WLWH, childhood PA was associated with depressive symptoms (β=4.64, SE=0.74, p<0.0001), alcohol use (β=1.11, SE=0.46, p=0.02), smoking (OR=2.04, 95%CI:1.51, 2.74) and HCV (OR=1.92, 95%CI:1.32, 2.78) compared to WLWH with no abuse history. Also, among WLWH, adulthood PA was associated with systolic blood pressure (β=−1.40, SE=0.62, p=0.02), depressive symptoms (β=3.84, SE=0.44, p<0.0001), alcohol use (β=1.33, SE=0.31, p<0.0001), smoking (OR=1.91, 95%CI:1.58, 2.31), HCV (OR=2.14, 95%CI:1.68, 2.72), and IDU (OR=2.00, 95%CI: 1.06, 3.77). Among WLWOH, child PA was associated with BMI (β=2.59, SE=1.18, p=0.03), and depressive symptoms (β=5.85, SE=1.18, p<0.0001) compared to those with no abuse history. Also, among WLWOH, adulthood PA was positively associated with BMI (β=1.86, SE=0.63, p=0.003), depressive symptoms (β=5.19, SE=0.66, p<0.0001), alcohol use (β=1.49, SE=0.73, p=0.04), smoking (OR=1.80, 95%CI:1.32, 2.45), HCV (OR=2.20, 95%CI:1.41, 3.44), and IDU (OR=3.58, 95%CI:1.30, 9.82).
Table 5.
Repeated measures linear and logistic GEE models for the association between components of the CVD risk scores and pathway variables with physical abuse history among WIHS womena
| WLWOH | WLWH | |||
|---|---|---|---|---|
| β (SE) or OR | p or 95% CI | β (SE) or OR | p or 95% CI | |
|
| ||||
| Total cholesterol | ||||
| Child physical abuse | 0.41 (4.17) | 0.92 | −4.85 (2.50) | 0.05 |
| Adult physical abuse | −0.47 (2.48) | 0.85 | −2.13 (1.62) | 0.19 |
| Reference | -- | -- | -- | -- |
| HDL | ||||
| Child physical abuse | −3.06 (1.64) | 0.06 | 0.17 (1.08) | 0.87 |
| Adult physical abuse | −1.75 (1.17) | 0.14 | −0.003 (0.75) | 0.99 |
| Reference | -- | -- | -- | -- |
| Systolic blood pressure | ||||
| Child physical abuse | −3.45 (1.78) | 0.05 | 1.22 (1.13) | 0.28 |
| Adult physical abuse | −0.67 (1.14) | 0.55 | −1.40 (0.62) | 0.02 |
| Reference | -- | -- | -- | -- |
| Body mass index | ||||
| Child physical abuse | 2.59 (1.18) | 0.03 | 1.15 (0.65) | 0.08 |
| Adult physical abuse | 1.86 (0.63) | 0.003 | −0.03 (0.40) | 0.96 |
| Reference | -- | -- | -- | -- |
| Depressive symptoms | ||||
| Child physical abuse | 5.85 (1.18) | <0.0001 | 4.64 (0.74) | <0.0001 |
| Adult physical abuse | 5.19 (0.66) | <0.0001 | 3.84 (0.44) | <0.0001 |
| Reference | -- | -- | -- | -- |
| Alcohol use | ||||
| Child physical abuse | −0.09 (1.08) | 0.93 | 1.11 (0.46) | 0.02 |
| Adult physical abuse | 1.49 (0.73) | 0.04 | 1.33 (0.31) | <0.0001 |
| Reference | -- | -- | -- | -- |
| Blood pressure medication | ||||
| Child physical abuse | 1.04 | 0.56, 1.91 | 0.89 | 0.58, 1.35 |
| Adult physical abuse | 1.19 | 0.84, 1.70 | 0.91 | 0.72, 1.15 |
| Reference | -- | -- | -- | -- |
| Cholesterol medication | ||||
| Child physical abuse | 1.64 | 0.78, 3.46 | 0.73 | 0.41, 1.29 |
| Adult physical abuse | 0.92 | 0.56, 1.49 | 0.98 | 0.71, 1.35 |
| Reference | -- | -- | -- | -- |
| Diabetes | ||||
| Child physical abuse | 1.32 | 0.53, 3.29 | 1.23 | 0.64, 2.37 |
| Adult physical abuse | 0.77 | 0.42, 1.41 | 1.08 | 0.71, 1.65 |
| Reference | -- | -- | -- | -- |
| Smoker | ||||
| Child physical abuse | 1.32 | 0.79, 2.18 | 2.04 | 1.51, 2.74 |
| Adult physical abuse | 1.8 | 1.32, 2.45 | 1.91 | 1.58, 2.31 |
| Reference | -- | -- | -- | -- |
| HCV | ||||
| Child physical abuse | 1.32 | 0.62, 2.95 | 1.92 | 1.32, 2.78 |
| Adult physical abuse | 2.2 | 1.41, 3.44 | 2.14 | 1.68, 2.72 |
| Reference | -- | -- | -- | -- |
| IDU | ||||
| Child physical abuse | 1.9 | 0.45, 7.97 | 1.52 | 0.62, 3.71 |
| Adult physical abuse | 3.58 | 1.30, 9.82 | 2 | 1.06, 3.77 |
| Reference | -- | -- | -- | -- |
All models adjusted for age, race, education, wave of enrollment, and study site. Models for WLWH are additional adjusted for antiretroviral therapy, CD4 count, and detectable viral load.
GEE=Generalized estimating equations; CVD=Cardiovascular disease; WIHS=Women’s Interagency HIV Study; WLWOH=Women living without HIV; WLWH=Women living with HIV; SE=Standard error; OR=Odds ratio; p=p-value; CI=Confidence interval; HDL=high-density lipoprotein; HCV=Hepatitis-C infection; IDU=injection drug use.
DISCUSSION
Using 25 years of prospective data, we found that experiencing childhood SA, one of the most severe ACEs, was associated with higher 10-year risk of hard CVD events among WLWH. Several behavioral and clinical factors helped explain the association. Experiences of SA and PA in adulthood were associated with increased CVD risk among WLWH and women at high risk of acquiring HIV. These findings are noteworthy as they indicate likelihood of hard CVD events, including myocardial infarction, stroke, and cardiovascular death. WLWH and women at risk of acquiring HIV experience many abuses over the life course. This study shows the cardiovascular impact of such abuses and addresses a key omission articulated by the AHA and NIH/NHLBI[6,10] by examining life course adversity and CVD risk in this vulnerable population.
Childhood SA was positively associated with CVD risk among WLWH. This is consistent with the general population patterns where childhood SA predicts later life CVD, often working through behavioral, physiologic, and psychological mechanisms[42]. Several pathway factors that were associated with child SA among WLWH, including depression, smoking, and HCV. While depression and smoking as pathways is consistent with the general population evidence[10,42], our study shows HCV infection could be a novel pathway linking ACEs and CVD risk for WLWH. HCV is prevalent among PLWH[43], and can affect cholesterol metabolism among PLWH[44] and those without HIV[45]; HCV also predicts CVD events among PLWH[27,43]. From a life course perspective[46,47], these findings suggest that childhood SA may lead to depression and smoking, as well as the uptake of risky sexual behavior and injection drug use, and consequent HIV and HCV infection, which may cumulatively influence CVD risk.
For WLWH and WLWOH, experiences of adulthood SA and PA were strongly associated with higher FRS-H CVD risk. Our study considered the age at which the abuse first occurred. Therefore, those with adulthood abuse did not also experience childhood abuse. Adulthood abuse, perhaps attributable in part to intimate partner violence which is prevalent among WLWH and those at risk for HIV infection[18,19], might be relevant. For pathways, we observed significant associations between adult abuse with depression, smoking, and HCV, as well as alcohol use and IDU for adult PA abuse. However, the association between adulthood abuse with CVD risk was largely maintained when controlling for pathways factors, suggesting behaviors and comorbidities explain some of the variation in CVD risk but not all.
It was surprising that childhood SA was not associated with CVD risk among seronegative women at risk of acquiring HIV, and that null findings were observed for childhood PA among all participants. Low prevalence of childhood PA (10%) and small sample size for childhood SA among WLWOH (n=148) could have constrained power to detect associations. These groups did, however, exhibit higher levels of behavioral and clinical risks over time (e.g., smoking, HCV), suggesting that it may be possible to detect risk in larger samples. Also, while the pattern of associations between SA history and ACC/AHA-PCE CVD risk were similar to FRS-H, the associations were less robust. This could be due to the differences between the CVD risk algorithms. Compared to FRS-H, ACC/AHA-PCE predicts more and different CVD endpoints, incorporates race and diabetes as component parts, and applies to an older population.
This study has some limitations. We used retrospective self-report of abuse which due to stigma and recall might contribute to misclassification of exposure and underestimated associations. Also, the CVD scores were developed for the general population and may underestimate risk among WLWH[48,49]. We examined these scores as they are common in clinical practice and allowed for comparisons between WLWH and WLWOH. While some recent work finds ACC/AHA-PCE to be a valid tool to predict CVD risk for PLWH[24,50], we acknowledge that risk scores derived for PLWH might yield different associations with abuse history. Also, we did not conduct formal mediation tests and acknowledge that an alternative explanation for the parameter estimate changes could reflect confounding control. This study has strengths, including capitalizing on a large well-characterized longitudinal sample of women and the use of standardized CVD risk prediction tools commonly employed in clinical practice.
There are several future directions for research. First, CVD risk scores do not confirm diagnoses or describe disease progression. Future work should consider abuse history in association with major adverse cardiovascular events (MACEs) and carotid artery plaque progression among WLWH. Additionally, the pathways analysis provides a first look at possible mechanisms for abuse history and CVD among WLWH and WLWOH. Future work should conduct formal mediation analyses such as structural equation modelling[51] or effect decomposition in causal mediation[52] including with longitudinal data[53], and also consider potential reverse causation (e.g., depression, IDU, and HCV can give rise to abuse and HIV infection). Mediation analyses could also explore education attainment as mechanism. Finally, future work should study the accumulation and duration of childhood and adulthood abuse in association with CVD risk for WLWH, and associations between abuse history and CVD association among HIV-seropositive men.
Assessing life course abuse experiences in clinical encounters may help contextualize cardiovascular risk among this vulnerable population and inform intervention. The prevention and management of CVD among PLWH tends to focus on proximal risks, including lifestyle factors and medication therapy[27]. While critically important, this study additionally shows that life course abuse history, often originating decades prior to any manifestation of disease or lifestyle factor, can modulate CVD risk among WLWH. Moreover, this study also suggests that managing comorbidities like depression and HCV with a trauma-informed care lens for WLWH and women at high risk of acquiring HIV may help prevent and reduce CVD. As lifespans continue to improve for PLWH, clinicians and researchers alike should consider abuse history in clinical encounters and in the study of disease etiology and progression.
Supplementary Material
ACKNOWLEDGMENTS:
The authors gratefully acknowledge the contributions of the study participants and dedication of the staff at the MWCCS sites.
SOURCES OF FUNDING:
Data in this manuscript were collected by the Women’s Interagency HIV Study (WIHS), now the MACS/WIHS Combined Cohort Study (MWCCS). The contents of this publication are solely the responsibility of the authors and do not represent the official views of the National Institutes of Health (NIH). MWCCS (Principal Investigators): Atlanta CRS (Ighovwerha Ofotokun, Anandi Sheth, and Gina Wingood), U01-HL146241; Baltimore CRS (Todd Brown and Joseph Margolick), U01-HL146201; Bronx CRS (Kathryn Anastos, David Hanna, and Anjali Sharma), U01-HL146204; Brooklyn CRS (Deborah Gustafson and Tracey Wilson), U01-HL146202; Data Analysis and Coordination Center (Gypsyamber D’Souza, Stephen Gange and Elizabeth Topper), U01-HL146193; Chicago-Cook County CRS (Mardge Cohen and Audrey French), U01-HL146245; Chicago-Northwestern CRS (Steven Wolinsky, Frank Palella, Valentina Stosor), U01-HL146240; Northern California CRS (Bradley Aouizerat, Jennifer Price, and Phyllis Tien), U01-HL146242; Los Angeles CRS (Roger Detels and Matthew Mimiaga), U01-HL146333; Metropolitan Washington CRS (Seble Kassaye and Daniel Merenstein), U01-HL146205; Miami CRS (Maria Alcaide, Margaret Fischl, and Deborah Jones), U01-HL146203; Pittsburgh CRS (Jeremy Martinson and Charles Rinaldo), U01-HL146208; UAB-MS CRS (Mirjam-Colette Kempf, Jodie Dionne-Odom, and Deborah Konkle-Parker), U01-HL146192; UNC CRS (Adaora Adimora and Michelle Floris-Moore), U01-HL146194. The MWCCS is funded primarily by the National Heart, Lung, and Blood Institute (NHLBI), with additional co-funding from the Eunice Kennedy Shriver National Institute Of Child Health & Human Development (NICHD), National Institute On Aging (NIA), National Institute Of Dental & Craniofacial Research (NIDCR), National Institute Of Allergy And Infectious Diseases (NIAID), National Institute Of Neurological Disorders And Stroke (NINDS), National Institute Of Mental Health (NIMH), National Institute On Drug Abuse (NIDA), National Institute Of Nursing Research (NINR), National Cancer Institute (NCI), National Institute on Alcohol Abuse and Alcoholism (NIAAA), National Institute on Deafness and Other Communication Disorders (NIDCD), National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), National Institute on Minority Health and Health Disparities (NIMHD), and in coordination and alignment with the research priorities of the National Institutes of Health, Office of AIDS Research (OAR). MWCCS data collection is also supported by UL1-TR000004 (UCSF CTSA), UL1-TR003098 (JHU ICTR), UL1-TR001881 (UCLA CTSI), P30-AI-050409 (Atlanta CFAR), P30-AI-073961 (Miami CFAR), P30-AI-050410 (UNC CFAR), P30-AI-027767 (UAB CFAR), P30-MH-116867 (Miami CHARM), UL1-TR001409 (DC CTSA), KL2-TR001432 (DC CTSA), and TL1-TR001431 (DC CTSA). Additional sources of funding include K01HL137557 (DH), K23HL152903 (CM) and K12HL143958 (JW).
Non-standard Abbreviations and Acronyms:
- ACEs
Adverse Childhood Experiences
- WLWH
Women Living with HIV
- WLWOH
Women Living without HIV
- FRS-H
Framingham Risk Score for Hard Events
- ACC/AHA-PCE
American College of Cardiology/American Heart Association-Pooled Cohort Equation
Footnotes
DISCLOSURES: All authors have no disclosures and declare no conflicts.
DATA AVAILABILITY STATEMENT
Access to individual-level data from the MACS/WIHS Combined Cohort Study Data (MWCCS) may be obtained upon review and approval of a MWCCS concept sheet. Links and instructions for online concept sheet submission are on the study website: https://statepi.jhsph.edu/mwccs/work-with-us/
REFERENCES
- 1.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. American Journal of Preventive Medicine 1998; 14:245–258. [DOI] [PubMed] [Google Scholar]
- 2.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; 6:228–235. [DOI] [PubMed] [Google Scholar]
- 3.Merrick MT, Ford DC, Ports KA, Guinn AS. Prevalence of Adverse Childhood Experiences From the 2011–2014 Behavioral Risk Factor Surveillance System in 23 States. JAMA Pediatr 2018; 172:1038–1044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Merrick MT. 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. doi: 10.15585/mmwr.mm6844e1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Shonkoff JP, Boyce WT, McEwen BS. Neuroscience, molecular biology, and the childhood roots of health disparities. Journal of the American Medical Association 2009; 301:2252–2259. [DOI] [PubMed] [Google Scholar]
- 6.Suglia SF, Campo RA, Brown AGM, Stoney C, Boyce CA, Appleton AA, et al. Social Determinants of Cardiovascular Health: Early Life Adversity as a Contributor to Disparities in Cardiovascular Diseases. J Pediatr 2020; 219:267–273. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Loveday S, Hall T, Constable L, Paton K, Sanci L, Goldfeld S, et al. Screening for Adverse Childhood Experiences in Children: A Systematic Review. Pediatrics 2022; 149:e2021051884. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Ford K, Hughes K, Hardcastle K, Di Lemma LCG, Davies AR, Edwards S, et al. The evidence base for routine enquiry into adverse childhood experiences: A scoping review. Child Abuse & Neglect 2019; 91:131–146. [DOI] [PubMed] [Google Scholar]
- 9.Shimkhada R, Miller J, Magnan E, Miller M, Coffman J, Corbett G. Policy Considerations for Routine Screening for Adverse Childhood Events (ACEs). J Am Board Fam Med 2022; 35:862–866. [DOI] [PubMed] [Google Scholar]
- 10.Suglia SF, Koenen KC, Boynton-Jarrett R, Chan PS, Clark CJ, Danese A, et al. Childhood and Adolescent Adversity and Cardiometabolic Outcomes: A Scientific Statement From the American Heart Association. Circulation 2018; 137:e15–e28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Jakubowski KP, Cundiff JM, Matthews KA. Cumulative childhood adversity and adult cardiometabolic disease: A meta-analysis. Health Psychology 2018; 37:701–715. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Appleton AA, Holdsworth E, Ryan M, Tracy M. Measuring childhood adversity in life course cardiovascular research: A systematic review. Psychosomatic medicine 2017; 79:434–440. [DOI] [PubMed] [Google Scholar]
- 13.Bertolino DF, Sanchez TH, Zlotorzynska M, Sullivan PS. Adverse childhood experiences and sexual health outcomes and risk behaviors among a nationwide sample of men who have sex with men. Child Abuse Negl 2020; 107:104627. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Andersen JP, Blosnich J. Disparities in Adverse Childhood Experiences among Sexual Minority and Heterosexual Adults: Results from a Multi-State Probability-Based Sample. PLOS ONE 2013; 8:e54691. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.McCabe SE, Hughes TL, West BT, Evans-Polce RJ, Veliz PT, Dickinson K, et al. Sexual Orientation, Adverse Childhood Experiences, and Comorbid DSM-5 Substance Use and Mental Health Disorders. J Clin Psychiatry 2020; 81. doi: 10.4088/JCP.20m13291 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.LoSchiavo C, Halkitis PN, Kapadia F. Sexual Orientation and Gender Identity Victimization Among Young Adults in the New York City Metropolitan Area: The P18 Cohort Study. Psychol Sex Orientat Gend Divers 2019; 6:399–407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Zhu J, Exner-Cortens D, Dobson K, Wells L, Noel M, Madigan S. Adverse childhood experiences and intimate partner violence: A meta-analysis. Dev Psychopathol 2023;:1–15. [DOI] [PubMed] [Google Scholar]
- 18.Campbell JC, Baty ML, Ghandour RM, Stockman JK, Francisco L, Wagman J. The intersection of intimate partner violence against women and HIV/AIDS: a review. International Journal of Injury Control and Safety Promotion 2008; 15:221–231. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Cheng LJ, Cheng JY, Yen KY, Lau ST, Lau Y. Global Prevalence and Factors Related to Intimate Partner Violence Amongst People Living with Human Immunodeficiency Virus/Acquired Immunodeficiency Syndrome: A Systematic Review, Meta-Analysis, and Meta-Regression. Trauma, Violence, & Abuse 2022;:15248380221097436. [DOI] [PubMed] [Google Scholar]
- 20.Wada N, Jacobson LP, Cohen M, French A, Phair J, Muñoz A. Cause-Specific Life Expectancies After 35 Years of Age for Human Immunodeficiency Syndrome-Infected and Human Immunodeficiency Syndrome-Negative Individuals Followed Simultaneously in Long-term Cohort Studies, 1984–2008. American Journal of Epidemiology 2013; 177:116–125. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ballocca F, D’Ascenzo F, Gili S, Grosso Marra W, Gaita F. Cardiovascular disease in patients with HIV. Trends Cardiovasc Med 2017; 27:558–563. [DOI] [PubMed] [Google Scholar]
- 22.Smith CJ, Ryom L, Weber R, Morlat P, Pradier C, Reiss P, et al. Trends in underlying causes of death in people with HIV from 1999 to 2011 (D:A:D): a multicohort ollaboration. The Lancet 2014; 384:241–248. [DOI] [PubMed] [Google Scholar]
- 23.So-Armah K, Freiberg MS. HIV and Cardiovascular Disease: Update on Clinical Events, Special Populations, and Novel Biomarkers. Curr HIV/AIDS Rep 2018; 15:233–244. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Delabays B, Cavassini M, Damas J, Beuret H, Calmy A, Hasse B, et al. Cardiovascular risk assessment in people living with HIV compared to the general population. Eur J Prev Cardiol 2022; 29:689–699. [DOI] [PubMed] [Google Scholar]
- 25.Kaplan RC, Hanna DB, Kizer JR. Recent Insights into Cardiovascular Disease (CVD) Risk Among HIV-Infected Adults. Curr HIV/AIDS Rep 2016; 13:44–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Rodger AJ, Lodwick R, Schechter M, Deeks S, Amin J, Gilson R, et al. Mortality in well controlled HIV in the continuous antiretroviral therapy arms of the SMART and ESPRIT trials compared with the general population. AIDS 2013; 27:973–979. [DOI] [PubMed] [Google Scholar]
- 27.Feinstein MJ, Hsue PY, Benjamin LA, Bloomfield GS, Currier JS, Freiberg MS, et al. Characteristics, Prevention, and Management of Cardiovascular Disease in People Living With HIV: A Scientific Statement From the American Heart Association. Circulation 2019; 140:e98–e124. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Kovacs L, Kress TC, Belin de Chantemèle EJ. HIV, Combination Antiretroviral Therapy, and Vascular Diseases in Men and Women. JACC Basic Transl Sci 2022; 7:410–421. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Currier JS, Lundgren JD, Carr A, Klein D, Sabin CA, Sax PE, et al. Epidemiological Evidence for Cardiovascular Disease in HIV-Infected Patients and Relationship to Highly Active Antiretroviral Therapy. Circulation 2008; 118:e29–e35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Mdodo R, Frazier EL, Dube SR, Mattson CL, Sutton MY, Brooks JT, et al. Cigarette smoking prevalence among adults with HIV compared with the general adult population in the United States: cross-sectional surveys. Ann Intern Med 2015; 162:335–344. [DOI] [PubMed] [Google Scholar]
- 31.Frederick T, Burian P, Terrault N, Cohen M, Augenbraun M, Young M, et al. Factors associated with prevalent hepatitis C infection among HIV-infected women with no reported history of injection drug use: the Women’s Interagency HIV Study (WIHS). AIDS Patient Care STDS 2009; 23:915–923. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Barkan SE, Melnick SL, Preston-Martin S, Weber K, Kalish LA, Miotti P, et al. The Women’s Interagency HIV Study. WIHS Collaborative Study Group. Epidemiology 1998; 9:117–125. [PubMed] [Google Scholar]
- 33.D’Souza G, Bhondoekhan F, Benning L, Margolick JB, Adedimeji AA, Adimora AA, et al. Characteristics of the MACS/WIHS Combined Cohort Study: Opportunities for Research on Aging With HIV in the Longest US Observational Study of HIV. Am J Epidemiol 2021; 190:1457–1475. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Expert Panel on Detection Evaluation and Treatment of High Blood Cholesterol in Adults. Executive Summary of The Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III). Journal of the American Medical Association 2001; 285:2486–2497. [DOI] [PubMed] [Google Scholar]
- 35.Wilson PWF, D’Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of coronary heart disease using risk factor categories. Circulation 1998; 97:1837–1847. [DOI] [PubMed] [Google Scholar]
- 36.NIH/NHLBI. Assessing Cardiovascular Risk: Systematic Evidence Review from the Risk Assessment Work Group.; 2013. https://www.nhlbi.nih.gov/health-topics/assessing-cardiovascular-risk (accessed 11 Aug 2022).
- 37.Arnett DK, Blumenthal RS, Albert MA, Buroker AB, Goldberger ZD, Hahn EJ, et al. 2019 ACC/AHA Guideline on the Primary Prevention of Cardiovascular Disease: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Circulation 2019; 140:e596–e646. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Radloff LS. CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement 1977; 1:385–401. [Google Scholar]
- 39.LIANG K-Y, ZEGER SL. Longitudinal data analysis using generalized linear models. Biometrika 1986; 73:13–22. [Google Scholar]
- 40.Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology 1986; 51:1173–1182. [DOI] [PubMed] [Google Scholar]
- 41.MacKinnon DP, Fairchild AJ, Fritz MS. Mediation analysis. Annual Review of Psychology 2007; 58:593–614. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Soares AG, Howe LD, Heron J, Hammerton G, Rich-Edwards J, Magnus MC, et al. How does childhood maltreatment influence cardiovascular disease? A sequential causal mediation analysis. Int J Epidemiol 2022; 51:555–566. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Osibogun O, Ogunmoroti O, Michos ED, Spatz ES, Olubajo B, Nasir K, et al. HIV/HCV coinfection and the risk of cardiovascular disease: A meta-analysis. J Viral Hepat 2017; 24:998–1004. [DOI] [PubMed] [Google Scholar]
- 44.Kuniholm MH, Liang H, Anastos K, Gustafson D, Kassaye S, Nowicki M, et al. Association of a 3’ untranslated region polymorphism in proprotein convertase subtilisin/kexin type 9 with HIV viral load and CD4+ levels in HIV/hepatitis C virus coinfected women. AIDS 2017; 31:2483–2492. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Butt AA, Yan P, Simon TG, Chung RT, Abou-Samra A-B, ERCHIVES study team. Changes in circulating lipids level over time after acquiring HCV infection: results from ERCHIVES. BMC Infect Dis 2015; 15:510. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Ben-Shlomo Y, Kuh D. A life course approach to chronic disease epidemiology: conceptual models, empirical challenges and interdisciplinary perspectives. International Journal of Epidemiology 2002; 31:285–293. [PubMed] [Google Scholar]
- 47.Kuh D, Ben-Shlomo Y, Lynch J, Hallqvist J, Power C. Life course epidemiology. Journal of epidemiology and community health 2003; 57:778–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Achhra AC, Lyass A, Borowsky L, Bogorodskaya M, Plutzky J, Massaro JM, et al. Assessing Cardiovascular Risk in People Living with HIV: Current Tools and Limitations. Curr HIV/AIDS Rep 2021; 18:271–279. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Krikke M, Hoogeveen R, Hoepelman A, Visseren F, Arends J. Cardiovascular risk prediction in HIV-infected patients: comparing the Framingham, atherosclerotic cardiovascular disease risk score (ASCVD), Systematic Coronary Risk Evaluation for the Netherlands (SCORE-NL) and Data Collection on Adverse Events of Anti-HIV Drugs (D:A:D) risk prediction models. HIV Medicine 2016; 17:289–297. [DOI] [PubMed] [Google Scholar]
- 50.Krikke M, Hoogeveen R, Hoepelman A, Visseren F, Arends J. Cardiovascular risk prediction in HIV-infected patients: comparing the Framingham, atherosclerotic cardiovascular disease risk score (ASCVD), Systematic Coronary Risk Evaluation for the Netherlands (SCORE-NL) and Data Collection on Adverse Events of Anti-HIV Drugs (D:A:D) risk prediction models. HIV Medicine 2016; 17:289–297. [DOI] [PubMed] [Google Scholar]
- 51.Beran TN, Violato C. Structural equation modeling in medical research: a primer. BMC Research Notes 2010; 3:267. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.VanderWeele TJ. A three-way decomposition of a total effect into direct, indirect, and interactive effects. Epidemiology (Cambridge, Mass) 2013; 24:224–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Bind M- AC, Vanderweele TJ, Coull BA, Schwartz JD. Causal mediation analysis for longitudinal data with exogenous exposure. Biostatistics 2016; 17:122–134. [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
Access to individual-level data from the MACS/WIHS Combined Cohort Study Data (MWCCS) may be obtained upon review and approval of a MWCCS concept sheet. Links and instructions for online concept sheet submission are on the study website: https://statepi.jhsph.edu/mwccs/work-with-us/