Vital Signs

Estimated Proportion of Adult Health Problems Attributable to Adverse Childhood Experiences and Implications for Prevention — 25 States, 2015–2017

Melissa T. Merrick, PhD; Derek C. Ford, PhD; Katie A. Ports, PhD; Angie S. Guinn, MPH; Jieru Chen, PhD; Joanne Klevens, MD, PhD; Marilyn Metzler, MPH; Christopher M. Jones, PharmD, DrPH; Thomas R. Simon, PhD; Valerie M. Daniel, MPH; Phyllis Ottley, PhD; James A. Mercy, PhD


Morbidity and Mortality Weekly Report. 2019;68(44):999-1005. 

In This Article


The Behavioral Risk Factor Surveillance System (BRFSS)* is a state-based telephone survey of noninstitutionalized adults administered annually within each state, the District of Columbia, and U.S. territories. Participants report on a range of health conditions and risk behaviors. During the 2015–2017 data collection years, 27 states included state-added adverse childhood experience questions, in addition to the standardized set of BRFSS questions. These 11 state-added questions assess exposure to eight types of adverse childhood experiences: three types of abuse (physical, emotional, and sexual) and five types of household challenges (household member substance misuse, incarceration, mental illness, parental divorce, or witnessing intimate partner violence) before age 18 years. The adverse childhood experience items administered on the California and New Hampshire BRFSS surveys were inconsistent with those administered by the other states and were excluded, leaving 25 states in these analyses. Data were collected from 144,017 respondents who answered all adverse childhood experience questions and provided responses for age, race/ethnicity, and sex. Each respondent was classified into one of the following adverse childhood experience exposure categories based on the number of adverse childhood experience types reported: zero, one, two or three, and four or more types of adverse childhood experience exposure. The content and scoring of BRFSS adverse childhood experience items have been previously described.[17]

Associations between outcomes and adverse childhood experience exposure were assessed. Coronary heart disease, stroke, asthma, chronic obstruction pulmonary disease (COPD), cancer (excluding skin cancer), kidney disease, diabetes, and depression were measured by asking respondents whether they had ever been told by a health care professional that they had the condition. Body mass index (BMI), calculated from self-reported height and weight, was used to determine each participant's overweight or obesity status (overweight defined as BMI of ≥25 kg/m2; obesity defined as BMI of ≥30 kg/m2). Current smoking was defined as lifetime smoking of at least 100 cigarettes and currently smoking on at least some days. Heavy drinking was defined as adult men consuming at least 15 alcoholic beverages per week or adult women consuming at least eight alcoholic beverages per week in the past 30 days. Socioeconomic challenges included current lack of health insurance, current unemployment status, and attainment of less than a high school diploma or equivalent education.

As a preliminary step, the frequency distributions, including weighted percentages and corresponding 95% confidence intervals (CIs) of adverse childhood experience exposure by sociodemographic characteristics, were estimated. The overall bivariate associations between adverse childhood experience score and each sociodemographic variable were subsequently tested using chi-squared tests of independence. Logistic regression models were used to quantify the associations between adverse childhood experience exposure and each of the health outcomes, health risk behaviors, and socioeconomic challenges. All models were adjusted for race/ethnicity (non-Hispanic white [white], non-Hispanic black [black], non-Hispanic American Indian/Alaska Native [AI/AN], non-Hispanic Asian [Asian], Hispanic, and non-Hispanic other [Other]§); sex (male or female); and age group (18–24, 25–34, 35–44, 45–54, 55–64, and ≥65 years). Population attributable fractions, adjusted for age, race/ethnicity, and sex, were estimated using the predicted probabilities from the models to ascertain the percentage reduction in the number of observed cases of each outcome that would be expected if adverse childhood experience exposure were incrementally reduced or eliminated in the study population.[18] R (version 3.6.0; R Core Team) was used for all analyses and accounted for the complex survey design. Response rates for the states analyzed ranged from 30.6% to 59.0%.

In 2015, data were analyzed from Alaska, Kansas, Kentucky, Maryland, Ohio, South Carolina, and Texas; in 2016, from Arizona, Arkansas, Georgia, Louisiana, Michigan, New York, Oklahoma, Pennsylvania, and Utah; and in 2017, from Connecticut, Illinois, Iowa, Nevada, Oregon, South Dakota, Tennessee, Virginia, and Wisconsin.
§The Other race/ethnicity category consisted of participants self-reporting as non-Hispanic Native Hawaiian or Other Pacific Islander, non-Hispanic multiracial, or non-Hispanic other.