The questions in the 2002 Texas BRFSS provided a unique opportunity to obtain and report estimates for 7 categories of ACEs in a population-based sample. Our findings demonstrate that abuse and household dysfunction are common, and the findings are similar to findings from the CDC–Kaiser Permanente ACE Study (12
) among adult health maintenance organization members, in which 56% reported experiencing at least 1 ACE.
Using data obtained from the Texas BRFSS, we were also able to examine the prevalence of ACEs by sociodemographic characteristics. We observed a higher prevalence of ACEs among people with lower levels of education and a gradient in the proportion of ACEs by education level. Similarly, we observed a higher burden of ACEs for people with lower annual household income compared with those in the higher income brackets and for people unable to work compared with people employed for wages. These findings demonstrate that childhood adversity may be disproportionately represented in segments of the population where health disparities are often observed, and they suggest that adverse experiences in childhood may be associated with lower socioeconomic status later in life.
Among adults who experienced abuse or were exposed to serious dysfunction, a higher percentage reported being current smokers compared with people reporting no ACEs, similar to findings in a previous study (19
). In another study in 4 separate birth cohorts dating back to 1900 (20
), childhood stressors were associated with a lifetime history of smoking, which demonstrates that the childhood family milieu may be a salient factor to consider, despite any influences of social and secular trends to change the behavior during the past century.
Among adults who reported any form of childhood abuse, there was a higher proportion of people with a BMI ≥30 kg/m2
compared with people who did not report a history of abuse or household dysfunction in childhood, even after adjusting for sociodemographic factors. In a weight loss program conducted by Kaiser Permanente, clinicians and researchers discovered that sexual abuse in childhood was common among adult program dropouts and that abuse predated obesity (13
). Severity of abuse was also associated with obesity in the Kaiser cohort (21
). Our observations help us understand the potential contributing role of ACEs to overweight and obesity in adulthood.
Self-reported poor or fair general health among Texas adults was more prevalent among people reporting household dysfunction, childhood abuse, or both compared with people reporting no ACEs, and the association persisted even after controlling for sociodemographic factors. This is not surprising, given that people who report ACEs may have a wide variety of physical and mental health problems. Prior studies support this possibility: people who experienced childhood adversity had lower scores across all 8 domains of health-related quality of life, as measured by the Standard Form-36 (22
). There is evidence that exposure to ACEs is related to difficulties in emotional self-regulation, which may lead to reports of fair or poor general health (23
Epidemiologic studies documenting the associations between childhood adversity and negative health outcomes in adulthood are converging with studies in the neurosciences that have documented physiological and anatomical changes in the brains of people who experienced childhood abuse. These studies may provide biological plausibility for our findings (25
). For example, a study that used electroencephalograms to measure limbic irritability (7
) found a higher percentage of clinically significant brain-wave abnormalities among people who had a history of early trauma than among those who did not experience early trauma (7
). Magnetic resonance imaging has revealed reductions in hippocampal volumes among severely sexually abused women, and reductions in the intracranial and cerebral volumes among maltreated children compared with those who were not maltreated (5
). Additionally, the limbic system, which is responsible for emotional response, is adversely affected (5
Anatomic and functional neurologic changes may occur among people who experienced 1 or more forms of abuse compared with nonabused people (5
) through repeated or chronic activation of the stress response. The relationships we observed between specific ACE groupings with smoking, obesity, and self-rated health may indicate the inherent human stress response (27
); effects of the adrenal release of catecholamines and corticosteroids on developing neurons and neural networks is a biologic phenomenon that cannot be ruled out as a mechanism for the associations observed.
Several limitations should be considered when interpreting these results. First, the BRFSS is a telephone survey by landline and not all people in the United States have landline telephones. This could limit the generalizability of the results of this survey. Although the Texas BRFSS data are representative of the Texas population, they differ from that of the US population and, therefore, generalizability to the US population is limited. For example, Texas has a higher percentage of foreign-born people, people who speak a language other than English at home, and families living below poverty level compared with the US population. Texas also has a lower percentage of adults who graduate high school, people aged 18 years or older, adults aged 65 years or older, and people who self-identify as Hispanic or Latino compared with the overall US population. Second, the responses to BRFSS questions are self-reports, and independent verification of reported exposures is not possible. However, longitudinal follow-up of adults whose childhood abuse was documented has shown that their retrospective reports of their experiences are likely to underestimate actual occurrence (28
). Moreover, the test-retest reliability of retrospective reports of ACEs from adults was in the good-to-excellent range (30
). In addition, because self-reported general health is subjective, it is possible that it is providing a measure of psychological well-being.
The method of questionnaire administration may have led to underreporting of exposures. Disclosure of sensitive topics (such as childhood abuse) to a stranger conducting a telephone survey may result in a downward bias in estimating prevalence compared with surveys in which such disclosure was obtained in a more private manner. We did find some differences between the prevalence estimates from the CDC–Kaiser Permanente ACE Study (13
) and the BRFSS results. For instance, the prevalence of growing up with substance abuse in the home was somewhat lower than in other national surveys. Studies have reported that 1 in 4 of the adult population report growing up in homes with alcoholism (31
); our study found that 1 in 5 reported this exposure. Similarly, the prevalence of sexual abuse (8%) in our analysis is substantially lower than most prevalence estimates, which have ranged from 15% to 25% (32
). Finally, some of the estimates had large confidence intervals due to small sample sizes, indicating that caution must be used when interpreting the findings.
Despite these limitations, the findings suggest that growing up with abuse and serious forms of family dysfunction among adults in Texas is common. The findings also highlight the health effects associated with ACEs among adults in Texas. Continued public health attention is needed to prevent child abuse and concomitant stressful family exposures and to address ACEs in association with health problems. Such efforts will lead to improved well-being in the nation as a whole.