Advertisement

Relationship of Childhood Abuse and Household Dysfunction to Many of the Leading Causes of Death in Adults

The Adverse Childhood Experiences (ACE) Study

      Abstract

      Background: The relationship of health risk behavior and disease in adulthood to the breadth of exposure to childhood emotional, physical, or sexual abuse, and household dysfunction during childhood has not previously been described.
      Methods: A questionnaire about adverse childhood experiences was mailed to 13,494 adults who had completed a standardized medical evaluation at a large HMO; 9,508 (70.5%) responded. Seven categories of adverse childhood experiences were studied: psychological, physical, or sexual abuse; violence against mother; or living with household members who were substance abusers, mentally ill or suicidal, or ever imprisoned. The number of categories of these adverse childhood experiences was then compared to measures of adult risk behavior, health status, and disease. Logistic regression was used to adjust for effects of demographic factors on the association between the cumulative number of categories of childhood exposures (range: 0–7) and risk factors for the leading causes of death in adult life.
      Results: More than half of respondents reported at least one, and one-fourth reported ≥2 categories of childhood exposures. We found a graded relationship between the number of categories of childhood exposure and each of the adult health risk behaviors and diseases that were studied (P < .001). Persons who had experienced four or more categories of childhood exposure, compared to those who had experienced none, had 4- to 12-fold increased health risks for alcoholism, drug abuse, depression, and suicide attempt; a 2- to 4-fold increase in smoking, poor self-rated health, ≥50 sexual intercourse partners, and sexually transmitted disease; and a 1.4- to 1.6-fold increase in physical inactivity and severe obesity. The number of categories of adverse childhood exposures showed a graded relationship to the presence of adult diseases including ischemic heart disease, cancer, chronic lung disease, skeletal fractures, and liver disease. The seven categories of adverse childhood experiences were strongly interrelated and persons with multiple categories of childhood exposure were likely to have multiple health risk factors later in life.
      Conclusions: We found a strong graded relationship between the breadth of exposure to abuse or household dysfunction during childhood and multiple risk factors for several of the leading causes of death in adults.

      Keywords

      Only recently have medical investigators in primary care settings begun to examine associations between childhood abuse and adult health risk behaviors and disease.[
      • Springs F
      • Friedrich WN
      Health risk behaviors and medical sequelae of childhood sexual abuse.
      ,
      • Felitti VJ
      Long-term medical consequences of incest, rape, and molestation.
      ,
      • Felitti VJ
      Childhood sexual abuse, depression and family dysfunction in adult obese patients a case control study.
      ,
      • Gould DA
      • Stevens NG
      • Ward NG
      • Carlin AS
      • Sowell HE
      • Gustafson B
      Self-reported childhood abuse in an adult population in a primary care setting.
      ,
      • McCauley J
      • Kern DE
      • Kolodner K
      • Schroeder AF
      • et al.
      Clinical characteristics of women with a history of childhood abuse.
      ]These associations are important because it is now clear that the leading causes of morbidity and mortality in the United States[
      Mortality patterns: United States, 1993
      ]are related to health behaviors and lifestyle factors; these factors have been called the “actual” causes of death.[
      • McGinnis JM
      • Foege WH
      Actual causes of death in the United States.
      ]Insofar as abuse and other potentially damaging childhood experiences contribute to the development of these risk factors, then these childhood exposures should be recognized as the basic causes of morbidity and mortality in adult life.
      Although sociologists and psychologists have published numerous articles about the frequency[
      • Landis J
      Experiences of 500 children with adult sexual deviation.
      ,
      • Straus MA
      • Gelles RJ
      Societal change and change in family violence from 1975 to 1985 as revealed by two national surveys.
      ,
      • Wyatt GE
      • Peters SD
      Methodological considerations in research on the prevalence of child sexual abuse.
      ,
      • Berger AM
      • Knutson JF
      • Mehm JG
      • Perkins KA
      The self-report of punitive childhood experiences of young adults and adolescents.
      ,
      • Finkelhor D
      • Hotaling G
      • Lewis IA
      • Smith C
      Sexual abuse in a national survey of adult men and women prevalence, characteristics, and risk factors.
      ]and long-term consequences[
      • Egelend B
      • Sroufe LA
      • Erickson M
      The developmental consequence of different patterns of maltreatment.
      ,
      • Finkelhor D
      • Browne A
      The traumatic impact of child sexual abuse.
      ,
      • Beitchman JH
      • Zucker KJ
      • Hood JE
      • DaCosta GA
      • Akman D
      • Cassavia E
      A review of the long-term effects of sexual abuse.
      ]of childhood abuse, understanding their relevance to adult medical problems is rudimentary. Furthermore, medical research in this field has limited relevance to most primary care physicians because it is focused on adolescent health,[
      • Hibbard RA
      • Ingersoll GM
      • Orr DP
      Behavioral risk, emotional risk, and child abuse among adolescents in a nonclinical setting.
      ,
      • Nagy S
      • Adcock AG
      • Nagy MC
      A comparison of risky health behaviors of sexually active, sexually abused, and abstaining adolescents.
      ,
      • Cunningham RM
      • Stiffman AR
      • Dore P
      The association of physical and sexual abuse with HIV risk behaviors in adolescence and young adulthood implications for public health.
      ,
      Council on Scientific Affairs
      Adolescents as victims of family violence.
      ,
      • Nelson DE
      • Higginson GK
      • Grant-Worley JA
      Physical abuse among high school students. Prevalence and correlation with other health behaviors.
      ]mental health in adults,[
      • Nelson DE
      • Higginson GK
      • Grant-Worley JA
      Physical abuse among high school students. Prevalence and correlation with other health behaviors.
      ]or on symptoms among patients in specialty clinics.[
      • Drossman DA
      • Leserman J
      • Nachman G
      • Li Z
      • et al.
      Sexual and physical abuse in women with functional or organic gastrointestinal disorders.
      ,
      • Harrop-Griffiths J
      • Katon W
      • Walker E
      • Holm L
      • Russo J
      • Hickok L
      The association between chronic pelvic pain, psychiatric diagnoses, and childhood sexual abuse.
      ]Studies of the long-term effects of childhood abuse have usually examined single types of abuse, particularly sexual abuse, and few have assessed the impact of more than one type of abuse.[
      • McCauley J
      • Kern DE
      • Kolodner K
      • Schroeder AF
      • et al.
      Clinical characteristics of women with a history of childhood abuse.
      ,
      • Briere J
      • Runtz M
      Multivariate correlates of childhood psychological and physical maltreatment among university women.
      ,
      • Briere J
      • Runtz M
      Differential adult symptomatology associated with three types of child abuse histories.
      ,
      • Claussen AH
      • Crittenden PM
      Physical and psychological maltreatment relations among types of maltreatment.
      ,
      • Moeller TP
      • Bachman GA
      • Moeller JR
      The combined effects of physical, sexual, and emotional abuse during childhood long-term health consequences for women.
      ,
      • Bryant SL
      • Range LM
      Suicidality in college women who were sexually and physically punished by parents.
      ]Conditions such as drug abuse, spousal violence, and criminal activity in the household may co-occur with specific forms of abuse that involve children. Without measuring these household factors as well, long-term influence might be wrongly attributed solely to single types of abuse and the cumulative influence of multiple categories of adverse childhood experiences would not be assessed. To our knowledge, the relationship of adult health risk behaviors, health status, and disease states to childhood abuse and household dysfunction[
      • Zeitlen H
      Children with alcohol misusing parents.
      ,
      • Dore MM
      • Doris JM
      • Wright P
      Identifying substance abuse in maltreating families a child welfare challenge.
      ,
      • Ethier LS
      • Lacharite C
      • Couture G
      Childhood adversity, parental stress, and depression of negligent mothers.
      ,
      • Spaccarelli S
      • Coatsworth JD
      • Bowden BS
      Exposure to family violence among incarcerated boys; its association with violent offending and potential mediating variables.
      ,
      • McCloskey LA
      • Figueredo AJ
      • Koss MP
      The effects of systemic family violence on children’s mental health.
      ,
      • Brent DA
      • Perper JA
      • Moritz G
      • Schweers J
      • Balach L
      • Roth C
      Familial risk factors for adolescent suicide a case-control study.
      ,
      • Shaw DS
      • Vondra JI
      • Hommerding KD
      • Keenan K
      • Dunn M
      Chronic family adversity and early child behavior problems a longitudinal study of low income families.
      ]has not been described.
      We undertook the Adverse Childhood Experiences (ACE) Study in a primary care setting to describe the long-term relationship of childhood experiences to important medical and public health problems. The ACE Study is assessing, retrospectively and prospectively, the long-term impact of abuse and household dysfunction during childhood on the following outcomes in adults: disease risk factors and incidence, quality of life, health care utilization, and mortality. In this initial paper we use baseline data from the study to provide an overview of the prevalence and interrelation of exposures to childhood abuse and household dysfunction. We then describe the relationship between the number of categories of these deleterious childhood exposures and risk factors and those diseases that underlie many of the leading causes of death in adults.[
      Mortality patterns: United States, 1993
      ,
      • McGinnis JM
      • Foege WH
      Actual causes of death in the United States.
      ,

      U.S. Department of Health and Human Services. Physical activity and health: A report of the Surgeon General. Atlanta, Georgia. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion; 1996.

      ,
      • Rivara FP
      • Mueller BA
      • Somes G
      • Mendoza CT
      • Rushforth NB
      • Kellerman AL
      Alcohol and illicit drug abuse and the risk of violent death in the home.
      ]

      1. Methods

      1.1 Study Setting

      The ACE Study is based at Kaiser Permanente’s San Diego Health Appraisal Clinic. More than 45,000 adults undergo standardized examinations there each year, making this clinic one of the nation’s largest free-standing medical evaluation centers. All enrollees in the Kaiser Health Plan in San Diego are advised through sales literature about the services (free for members) at the clinic; after enrollment, members are advised again of its availability through new-member literature. Most members obtain appointments by self-referral; 20% are referred by their health care provider. A recent review of membership and utilization records among Kaiser members in San Diego continuously enrolled between 1992 and 1995 showed that 81% of those 25 years and older had been evaluated in the Health Appraisal Clinic.
      Health appraisals include completion of a standardized medical questionnaire that requests demographic and biopsychosocial information, review of organ systems, previous medical diagnoses, and family medical history. A health care provider completes the medical history, performs a physical examination, and reviews the results of laboratory tests with the patient.

      1.2 Survey Methods

      The ACE Study protocol was approved by the Institutional Review Boards of the Southern California Permanente Medical Group (Kaiser Permanente), the Emory University School of Medicine, and the Office of Protection from Research Risks, National Institutes of Health. All 13,494 Kaiser Health Plan members who completed standardized medical evaluations at the Health Appraisal Clinic between August–November of 1995 and January–March of 1996 were eligible to participate in the ACE Study. Those seen at the clinic during December were not included because survey response rates are known to be lower during the holiday period.[
      • Dillman DA
      ]
      In the week after visiting the clinic, and hence having their standardized medical history already completed, members were mailed the ACE Study questionnaire that included questions about childhood abuse and exposure to forms of household dysfunction while growing up. After second mailings of the questionnaire to persons who did not respond to the first mailing, the response rate for the survey was 70.5% (9,508/13,494).
      A second survey wave of approximately the same number of patients as the first wave was conducted between June and October of 1997. The data for the second survey wave is currently being compiled for analysis. The methods for the second mail survey wave were identical to the first survey wave as described above. The second wave was done to enhance the precision of future detailed analyses on special topics and to reduce the time necessary to obtain precise statistics on follow-up health events. An overview of the total ACE Study design is provided in Fig. 1.
      Figure thumbnail gr1
      Fig. 1ACE Study design. ∗After exclusions, 59.7% of the original wave I sample (8,056/13,494) were included in this analysis.

      1.3 Comparison of Respondents and Nonrespondents

      We abstracted the completed medical evaluation for every person eligible for the study; this included their medical history, laboratory results, and physical findings. Respondent (n = 9,508) and nonrespondent (n = 3,986) groups were similar in their percentages of women (53.7% and 51.0%, respectively) and in their mean years of education (14.0 years and 13.6 years, respectively). Respondents were older than nonrespondents (means 56.1 years and 49.3 years) and more likely to be white (83.9% vs. 75.3%) although the actual magnitude of the differences was small.
      Respondents and nonrespondents did not differ with regard to their self-rated health, smoking, other substance abuse, or the presence of common medical conditions such as a history of heart attack or stroke, chronic obstructive lung disease, hypertension, or diabetes, or with regard to marital status or current family, marital, or job-related problems (data not shown). The health appraisal questionnaire used in the clinic contains a single question about childhood sexual abuse that reads “As a child were you ever raped or sexually molested?” Respondents were slightly more likely to answer affirmatively than nonrespondents (6.1% vs. 5.4%, respectively).

      1.4 Questionnaire Design

      We used questions from published surveys to construct the ACE Study questionnaire. Questions from the Conflicts Tactics Scale[
      • Straus M
      • Gelles RJ
      ]were used to define psychological and physical abuse during childhood and to define violence against the respondent’s mother. We adapted four questions from Wyatt[
      • Wyatt GE
      The sexual abuse of Afro-American and White-American women in childhood.
      ]to define contact sexual abuse during childhood. Questions about exposure to alcohol or drug abuse during childhood were adapted from the 1988 National Health Interview Survey.[

      National Center for Health Statistics. Exposure to alcoholism in the family: United States, 1988. Advance Data, No. 205. U.S. Department of Health and Human Services, Washington, DC; September 30, 1991.

      ]All of the questions we used in this study to determine childhood experiences were introduced with the phrase “While you were growing up during your first 18 years of life …”
      Questions about health-related behaviors and health problems were taken from health surveys such as the Behavioral Risk Factor Surveys[

      Siegel PZ, Frazier EL, Mariolis P, et al. Behaviorial risk factor surveillance, 1991; Monitoring progress toward the Nation’s Year 2000 Health Objectives. Morb Mortal Wkly Rep 1992;42(SS-4)1–15.

      ]and the Third National Health and Nutrition Examination Survey,[
      • Crespo CJ
      • Keteyian SJ
      • Heath GW
      • Sempos CT
      Leisure-time physical activity among US adults Results from the Third National Health and Nutrition Examination Survey.
      ]both of which are directed by the Centers for Disease Control and Prevention. Questions about depression came from the Diagnostic Interview Schedule of the National Institute of Mental Health (NIMH).[
      • Robins LN
      • Helzer JE
      • Groughan J
      • Ratliff K
      National Institute of Mental Health Diagnostic Interview Schedule its history, characteristics, and validity.
      ]Other information for this analysis such as disease history was obtained from the standardized questionnaire used in the Health Appraisal Clinic. (A copy of the questionnaires used in this study may be found at www.elsevier.com/locate/amepre.)

      1.5 Defining Childhood Exposures

      We used three categories of childhood abuse: psychological abuse (2 questions), physical abuse (2 questions), or contact sexual abuse (4 questions). There were four categories of exposure to household dysfunction during childhood: exposure to substance abuse (defined by 2 questions), mental illness (2 questions), violent treatment of mother or stepmother (4 questions), and criminal behavior (1 question) in the household. Respondents were defined as exposed to a category if they responded “yes” to 1 or more of the questions in that category. The prevalence of positive responses to the individual questions and the category prevalences are shown in Table 1.
      Table 1Prevalence of Childhood Exposure to Abuse and Household Dysfunction
      Category of childhood exposure
      An exposure to one or more items listed under the set of questions for each category.
      Prevalence (%)Prevalence(%)
      Abuse by category
      Psychological11.1
      (Did a parent or other adult in the household …)
      Often or very often swear at, insult, or put you down?10.0
      Often or very often act in a way that made you afraid that you would be physically hurt?4.8
      Physical10.8
      (Did a parent or other adult in the household …)
      Often or very often push, grab, shove, or slap you?4.9
      Often or very often hit you so hard that you had marks or were injured?9.6
      Sexual22.0
      (Did an adult or person at least 5 years older ever …)
      Touch or fondle you in a sexual way?19.3
      Have you touch their body in a sexual way?8.7
      Attempt oral, anal, or vaginal intercourse with you?8.9
      Actually have oral, anal, or vaginal intercourse with you?6.9
      Household dysfunction by category
      Substance abuse25.6
      Live with anyone who was a problem drinker or alcoholic?23.5
      Live with anyone who used street drugs?4.9
      Mental illness18.8
      Was a household member depressed or mentally ill?17.5
      Did a household member attempt suicide?4.0
      Mother treated violently12.5
      Was your mother (or stepmother)
      Sometimes, often, or very often pushed, grabbed, slapped, or had something thrown at her?11.9
      Sometimes, often, or very often kicked, bitten, hit with a fist, or hit with something hard?6.3
      Ever repeatedly hit over at least a few minutes?6.6
      Ever threatened with, or hurt by, a knife or gun?3.0
      Criminal behavior in household
      Did a household member go to prison?3.43.4
      Any category reported52.1%
      a An exposure to one or more items listed under the set of questions for each category.
      We used these 7 categories of childhood exposures to abuse and household dysfunction for our analysis. The measure of childhood exposure that we used was simply the sum of the categories with an exposure; thus the possible number of exposures ranged from 0 (unexposed) to 7 (exposed to all categories).

      1.6 Risk Factors and Disease Conditions Assessed

      Using information from both the study questionnaire and the Health Appraisal Clinic’s questionnaire, we chose 10 risk factors that contribute to the leading causes of morbidity and mortality in the United States.[
      Mortality patterns: United States, 1993
      ,
      • McGinnis JM
      • Foege WH
      Actual causes of death in the United States.
      ,

      U.S. Department of Health and Human Services. Physical activity and health: A report of the Surgeon General. Atlanta, Georgia. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion; 1996.

      ,
      • Rivara FP
      • Mueller BA
      • Somes G
      • Mendoza CT
      • Rushforth NB
      • Kellerman AL
      Alcohol and illicit drug abuse and the risk of violent death in the home.
      ]The risk factors included smoking, severe obesity, physical inactivity, depressed mood, suicide attempts, alcoholism, any drug abuse, parenteral drug abuse, a high lifetime number of sexual partners (≥50), and a history of having a sexually transmitted disease.
      We also assessed the relationship between childhood exposures and disease conditions that are among the leading causes of mortality in the United States.[
      Mortality patterns: United States, 1993
      ]The presence of these disease conditions was based upon medical histories that patients provided in response to the clinic questionnaire. We included a history of ischemic heart disease (including heart attack or use of nitroglycerin for exertional chest pain), any cancer, stroke, chronic bronchitis, or emphysema (COPD), diabetes, hepatitis or jaundice, and any skeletal fractures (as a proxy for risk of unintentional injuries). We also included responses to the following question about self-rated health: “Do you consider your physical health to be excellent, very good, good, fair, or poor?” because it is strongly predictive of mortality.[
      • Idler E
      • Angel RJ
      Self-rated health and mortality in the NHANES I Epidemiologic Follow-up Study.
      ]

      1.7 Definition of Risk Factors

      We defined severe obesity as a body mass index (kg/meter2) ≥35 based on measured height and weight; physical inactivity as no participation in recreational physical activity in the past month; and alcoholism as a “Yes” response to the question “Have you ever considered yourself to be an alcoholic?” The other risk factors that we assessed are self-explanatory.

      1.8 Exclusions from Analysis

      Of the 9,508 survey respondents, we excluded 51 (0.5%) whose race was unstated and 34 (0.4%) whose educational attainment was not reported. We also excluded persons who did not respond to certain questions about adverse childhood experiences. This involved the following exclusions: 125 (1.3%) for household substance abuse, 181 (1.9%) for mental illness in the home, 148 (1.6%) for violence against mother, 7 (0.1%) for imprisonment of a household member, 109 (1.1%) for childhood psychological abuse, 44 (0.5%) for childhood physical abuse, and 753 (7.9%) for childhood sexual abuse. After these exclusions, 8,056 of the original 9,508 survey respondents (59.7% of the original sample of 13,494) remained and were included in the analysis. Procedures for insuring that the findings based on complete data were generalizable to the entire sample are described below.
      The mean age of the 8,506 persons included in this analysis was 56.1 years (range: 19–92 years); 52.1% were women; 79.4% were white. Forty-three percent had graduated from college; only 6.0% had not graduated from high school.

      1.9 Statistical Analysis

      We used the Statistical Analysis System (SAS)[

      SAS Procedures Guide. SAS Institute Inc. Version 6, 3rd edition, Cary, NC: SAS Institute; 1990.

      ]for our analyses. We used the direct method to age-adjust the prevalence estimates. Logistic regression analysis was employed to adjust for the potential confounding effects of age, sex, race, and educational attainment on the relationship between the number of childhood exposures and health problems.
      To test for a dose-response relationship to health problems, we entered the number of childhood exposures as a single ordinal variable (0, 1, 2, 3, 4, 5, 6, 7) into a separate logistic regression model for each risk factor or disease condition.

      1.10 Assessing the Possible Influence of Exclusions

      To determine whether our results were influenced by excluding persons with incomplete information on any of the categories of childhood exposure, we performed a separate sensitivity analysis in which we included all persons with complete demographic information but assumed that persons with missing information for a category of childhood exposure did not have an exposure in that category.

      2. Results

      2.1 Adverse Childhood Exposures

      The level of positive responses for the 17 questions included in the seven categories of childhood exposure ranged from 3.0% for a respondent’s mother (or stepmother) having been threatened with or hurt by a gun or knife to 23.5% for having lived with a problem drinker or alcoholic (Table 1). The most prevalent of the 7 categories of childhood exposure was substance abuse in the household (25.6%); the least prevalent exposure category was evidence of criminal behavior in the household (3.4%). More than half of respondents (52%) experienced ≥1 category of adverse childhood exposure; 6.2% reported ≥4 exposures.

      2.2 Relationships between Categories of Childhood Exposure

      The probability that persons who were exposed to any single category of exposure were also exposed to another category is shown in Table 2. The relationship between single categories of exposure was significant for all comparisons (P < .001; chi-square). For persons reporting any single category of exposure, the probability of exposure to any additional category ranged from 65%–93% (median: 80%); similarly, the probability of ≥2 additional exposures ranged from 40%–74% (median: 54.5%).
      Table 2Relationships between Categories of Adverse Childhood Exposure
      Percent (%) Exposed to Another Category
      First Category of Childhood ExposureSample Size
      Number exposed to first category. For example, among persons who were psychologically abused, 52% were also physically abused. More persons were a second category than would be expected by chance (P < .001; chi-square).
      Psychological AbusePhysical AbuseSexual AbuseSubstance AbuseMental IllnessTreated ViolentlyImprisoned MemberAny One Additional CategoryAny Two Additional Categories
      Childhood Abuse:
      Psychological89852
      Number exposed to first category. For example, among persons who were psychologically abused, 52% were also physically abused. More persons were a second category than would be expected by chance (P < .001; chi-square).
      4751503999374
      Physical abuse874544445383598664
      Sexual abuse1770242239312366541
      Household dysfunction:
      Substance abuse2064221934342986940
      Mental illness1512302237462677447
      Mother treated violently10103431415938108662
      Member imprisoned2712929406242378664
      median29.525.440.548.538328.58054.5
      range(22–54)(19–52)(34–47)(39–62)(31–50)(23–39)(6–10)(65–93)(40–74)
      a Number exposed to first category. For example, among persons who were psychologically abused, 52% were also physically abused. More persons were a second category than would be expected by chance (P < .001; chi-square).
      The number of categories of childhood exposures by demographic characteristics is shown in Table 3. Statistically, significantly fewer categories of exposure were found among older persons, white or Asian persons, and college graduates (P < .001). Because age is associated with both the childhood exposures as well as many of the health risk factors and disease outcomes, all prevalence estimates in the tables are adjusted for age.
      Table 3Prevalence of Categories of Adverse Childhood Exposures by Demographic Characteristics
      CharacteristicSample size (N)Number of categories (%)
      The number of categories of exposure was simply the sum of each of the seven individual categories that were assessed (see Table 1).
      01234
      Age group (years)
      19–3480735.425.417.211.010.9
      35–492,06339.325.115.69.110.9
      50–642,57746.525.213.97.96.6
      ≥652,61060.024.58.94.22.4
      Gender
      Prevalence estimates adjusted for age.
      Women4,19745.424.013.48.78.5
      Men3,85953.725.811.65.03.9
      Race
      Prevalence estimates adjusted for age.
      White6,43249.725.312.46.76.0
      Black38538.825.716.312.37.0
      Hispanic43142.924.913.77.411.2
      Asian50866.019.09.93.41.7
      Other30041.023.513.99.512.1
      Education
      Prevalence estimates adjusted for age.
      No HS diploma48056.521.58.46.57.2
      HS graduate1,53651.624.511.37.45.2
      Any college2,54144.125.514.87.87.8
      College graduate3,49951.425.112.16.15.3
      All participants8,05649.524.912.56.96.2
      a The number of categories of exposure was simply the sum of each of the seven individual categories that were assessed (see Table 1).
      b Prevalence estimates adjusted for age.

      2.3 Relationship between Childhood Exposures and Health Risk Factors

      Both the prevalence and risk (adjusted odds ratio) increased for smoking, severe obesity, physical inactivity, depressed mood, and suicide attempts as the number of childhood exposures increased (Table 4). When persons with 4 categories of exposure were compared to those with none, the odds ratios ranged from 1.3 for physical inactivity to 12.2 for suicide attempts (Table 4).
      Table 4Number of Categories of Adverse Childhood Exposure and the Adjusted Odds of Risk Factors Including Current Smoking, Severe Obesity, Physical Inactivity, Depressed Mood, and Suicide Attempt
      Health problemNumber of categoriesSample size (N)
      Sample sizes will vary due to incomplete or missing information about health problems.
      Prevalence (%)
      Prevalence estimates are adjusted for age.
      Adjusted odds ratio
      Odds ratios adjusted for age, gender, race, and educational attainment.
      95% confidence interval
      Current smoker
      Indicates information recorded in the patient’s chart before the study questionnaire was mailed.
      03,8366.81.0Referent
      12,0057.91.1(0.9–1.4)
      21,04610.31.5(1.1–1.8)
      358713.92.0(1.5–2.6)
      4 or more54416.52.2(1.7–2.9)
      Total8,0188.6
      Severe obesity
      Indicates information recorded in the patient’s chart before the study questionnaire was mailed.
      (BMI ≥ 35)
      03,8505.41.0Referent
      12,0047.01.1(0.9–1.4)
      21,0419.51.4(1.1–1.9)
      359010.31.4(1.0–1.9)
      4 or more54312.01.6(1.2–2.1)
      Total8,0287.1
      No leisure-time physical activity03,63418.41.0Referent
      11,91722.81.2(1.1–1.4)
      21,00622.01.2(1.0–1.4)
      355926.61.4(1.1–1.7)
      4 or more52326.61.3(1.1–1.6)
      Total7,63921.0
      Two or more weeks of depressed mood in the past year03,79914.21.0Referent
      11,98421.41.5(1.3–1.7)
      21,03631.52.4(2.0–2.8)
      358436.22.6(2.1–3.2)
      4 or more54250.74.6(3.8–5.6)
      Total7,94522.0
      Ever attempted suicide03,8521.21.0Referent
      11,9972.41.8(1.2–2.6)
      21,0484.33.0(2.0–4.6)
      35879.56.6(4.5–9.8)
      4 or more54418.312.2(8.5–17.5)
      Total8,0283.5
      a Sample sizes will vary due to incomplete or missing information about health problems.
      b Prevalence estimates are adjusted for age.
      c Odds ratios adjusted for age, gender, race, and educational attainment.
      d Indicates information recorded in the patient’s chart before the study questionnaire was mailed.
      Similarly, the prevalence and risk (adjusted odds ratio) of alcoholism, use of illicit drugs, injection of illicit drugs, ≥50 intercourse partners, and history of a sexually transmitted disease increased as the number of childhood exposures increased (Table 5). In comparing persons with ≥4 childhood exposures to those with none, odds ratios ranged from 2.5 for sexually transmitted diseases to 7.4 for alcoholism and 10.3 for injected drug use.
      Table 5Number of Categories of Adverse Childhood Exposure and the Prevalence and Risk (Adjusted Odds Ratio) of Health Risk Factors Including Alcohol or Drug Abuse, High Lifetime Number of Sexual Partners, or History of Sexually Transmitted Disease
      Health problemNumber of categoriesSample size (N)
      Sample sizes will vary due to incomplete or missing information about health problems.
      Prevalence (%)
      Prevalence estimates are adjusted for age.
      Adjusted odds ratio
      Odds ratios adjusted for age, gender, race, and educational attainment.
      95% confidence interval
      Considers self an alcoholic03,8412.91.0Referent
      11,9935.72.0(1.6–2.7)
      21,04210.34.0(3.0–5.3)
      358611.34.9(3.5–6.8)
      4 or more54016.17.4(5.4–10.2)
      Total8,0025.9
      Ever used illicit drugs03,8566.41.0Referent
      11,99811.41.7(1.4–2.0)
      21,04519.22.9(2.4–3.6)
      358921.53.6(2.8–4.6)
      4 or more54128.44.7(3.7–6.0)
      Total8,02911.6
      Ever injected drugs03,8550.31.0Referent
      11,9960.51.3(0.6–3.1)
      21,0441.43.8(1.8–8.2)
      35872.37.1(3.3–15.5)
      4 or more5403.410.3(4.9–21.4)
      Total8,0220.8
      Had 50 or more intercourse partners03,4003.01.0Referent
      11,8125.11.7(1.3–2.3)
      29266.12.3(1.6–3.2)
      35266.33.1(2.0–4.7)
      4 or more4746.83.2(2.1–5.1)
      Total7,1384.4
      Ever had a sexually transmitted disease
      Indicates information recorded in the patient’s chart before the study questionnaire was mailed.
      03,8485.61.0Referent
      12,0018.61.4(1.1–1.7)
      21,04410.41.5(1.2–1.9)
      358813.11.9(1.4–2.5)
      4 or more54216.72.5(1.9–3.2)
      Total80238.2
      a Sample sizes will vary due to incomplete or missing information about health problems.
      b Prevalence estimates are adjusted for age.
      c Odds ratios adjusted for age, gender, race, and educational attainment.
      d Indicates information recorded in the patient’s chart before the study questionnaire was mailed.

      2.4 Childhood Exposures and Clustering of Health Risk Factors

      We found a strong relationship between the number of childhood exposures and the number of health risk factors for leading causes of death in adults (Table 6). For example, among persons with no childhood exposures, 56% had none of the 10 risk factors whereas only 14% of persons with ≥4 categories of childhood exposure had no risk factors. By contrast, only 1% of persons with no childhood exposures had four or more risk factors, whereas 7% of persons with ≥4 childhood exposures had four or more risk factors (Table 6).
      Table 6Relationship Between Number of Categories of Childhood Exposure and Number of Risk Factors for the Leading Causes of Death
      Risk factors include: smoking, severe obesity, physical inactivity, depressed mood, suicide attempt, alcoholism, any drug use, injected drug use, ≥50 lifetime sexual partners, and history of a sexually transmitted disease.
      Number of categoriesSample size% with number of risk factors
      01234
      03,86156291041
      12,00942331662
      21,051313320104
      3590243320137
      ≥4545142628177
      Total8,05644311573
      a Risk factors include: smoking, severe obesity, physical inactivity, depressed mood, suicide attempt, alcoholism, any drug use, injected drug use, ≥50 lifetime sexual partners, and history of a sexually transmitted disease.

      2.5 Relationship between Childhood Exposures and Disease Conditions

      When persons with 4 or more categories of childhood exposure were compared to those with none, the odds ratios for the presence of studied disease conditions ranged from 1.6 for diabetes to 3.9 for chronic bronchitis or emphysema (Table 7). Similarly, the odds ratios for skeletal fractures, hepatitis or jaundice, and poor self-rated health were 1.6, 2.3, and 2.2, respectively (Table 8).
      Table 7Number of Categories of Adverse Childhood Exposure and the Prevalence and Risk (Adjusted Odds Ratio) of Heart Attack, Cancer, Stroke, COPD, and Diabetes
      Disease condition
      Indicates information recorded in the patient’s chart before the study questionnaire was mailed.
      Number of categoriesSample size (N)
      Sample sizes will vary due to incomplete or missing information about health problems.
      Prevalence (%)
      Prevalence estimates are adjusted for age.
      Adjusted odds ratio
      Odds ratios adjusted for age, gender, race, and educational attainment.
      95% confidence interval
      Ischemic heart disease03,8593.71.0Referent
      12,0093.50.9(0.7–1.3)
      21,0503.40.9(0.6–1.4)
      35904.61.4(0.8–2.4)
      4 or more5455.62.2(1.3–3.7)
      Total8,0223.8
      Any cancer03,8421.91.0Referent
      11,9951.91.2(1.0–1.5)
      21,0431.91.2(1.0–1.5)
      35881.91.0(0.7–1.5)
      4 or more5431.91.9(1.3–2.7)
      Total8,0111.9
      Stroke03,8322.61.0Referent
      11,9932.40.9(0.7–1.3)
      21,0422.00.7(0.4–1.3)
      35882.91.3(0.7–2.4)
      4 or more5434.12.4(1.3–4.3)
      Total7,9982.6
      Chronic bronchitis or emphysema03,7582.81.0Referent
      11,9394.41.6(1.2–2.1)
      21,0094.41.6(1.1–2.3)
      35655.72.2(1.4–3.3)
      4 or more5128.73.9(2.6–5.8)
      Total7,7834.0
      Diabetes03,8504.31.0Referent
      12,0024.11.0(0.7–1.3)
      21,0463.90.9(0.6–1.3)
      35875.01.2(0.8–1.9)
      4 or more5425.81.6(1.0–2.5)
      Total8,0274.3
      a Sample sizes will vary due to incomplete or missing information about health problems.
      b Prevalence estimates are adjusted for age.
      c Odds ratios adjusted for age, gender, race, and educational attainment.
      d Indicates information recorded in the patient’s chart before the study questionnaire was mailed.
      Table 8Number of Categories of Adverse Childhood Exposure and the Prevalence and Risk (Adjusted Odds Ratio) of Skeletal Fracture, Hepatitis or Jaundice, and Poor Self-rated Health
      Disease conditionNumber of categoriesSample size (N)
      Sample sizes will vary due to incomplete or missing information about health problems.
      Prevalence (%)
      Prevalence estimates are adjusted for age and gender.
      Adjusted odds ratio
      Odds ratios adjusted for age, gender, race, and educational attainment.
      95% confidence interval
      Ever had a skeletal fracture03,8433.61.0Referent
      11,9984.01.1(1.0–1.2)
      21,0484.51.4(1.2–1.6)
      35874.01.2(1.0–1.4)
      4 or more5444.81.6(1.3–2.0)
      Total8,0203.9
      Ever had hepatitis or jaundice03,8465.31.0Referent
      12,0065.51.1(0.9–1.4)
      21,0457.71.8(1.4–2.3)
      359010.21.6(1.2–2.3)
      4 or more54310.72.4(1.8–3.3)
      Total8,0306.5
      Fair or poor self-rated health03,76216.31.0Referent
      11,95717.81.2(1.0–1.4)
      21,02919.91.4(1.2–1.7)
      358420.31.4(1.1–1.7)
      4 or more52728.72.2(1.8–2.7)
      Total7,85918.2
      Indicates information recorded in the patient’s chart before the study questionnaire was mailed.
      a Sample sizes will vary due to incomplete or missing information about health problems.
      b Prevalence estimates are adjusted for age and gender.
      c Odds ratios adjusted for age, gender, race, and educational attainment.

      2.6 Significance of Dose-Response Relationships

      In logistic regression models (which included age, gender, race, and educational attainment as covariates) we found a strong, dose-response relationship between the number of childhood exposures and each of the 10 risk factors for the leading causes of death that we studied (P < .001). We also found a significant (P < .05) dose-response relationship between the number of childhood exposures and the following disease conditions: ischemic heart disease, cancer, chronic bronchitis or emphysema, history of hepatitis or jaundice, skeletal fractures, and poor self-rated health. There was no statistically significant dose-response relationship for a history of stroke or diabetes.

      2.7 Assessment of the Influence of Exclusions

      In the sensitivity analysis where missing information for a category of childhood exposure was considered as no exposure, the direction and strength of the associations between the number of childhood exposures and the risk factors and disease conditions were nearly identical (data not shown). Thus, the results we present appear to be unaffected by our decision to exclude persons for whom information on any category of childhood exposure was incomplete.

      3. Discussion

      We found a strong dose response relationship between the breadth of exposure to abuse or household dysfunction during childhood and multiple risk factors for several of the leading causes of death in adults. Disease conditions including ischemic heart disease, cancer, chronic lung disease, skeletal fractures, and liver disease, as well as poor self-rated health also showed a graded relationship to the breadth of childhood exposures. The findings suggest that the impact of these adverse childhood experiences on adult health status is strong and cumulative.
      The clear majority of patients in our study who were exposed to one category of childhood abuse or household dysfunction were also exposed to at least one other. Therefore, researchers trying to understand the long-term health implications of childhood abuse may benefit from considering a wide range of related adverse childhood exposures. Certain adult health outcomes may be more strongly related to unique combinations or the intensity of adverse childhood exposures than to the total breadth of exposure that we used for our analysis. However, the analysis we present illustrates the need for an overview of the net effects of a group of complex interactions on a wide range of health risk behaviors and diseases.
      Several potential limitations need to be considered when interpreting the results of this study. The data about adverse childhood experiences are based on self-report, retrospective, and can only demonstrate associations between childhood exposures and health risk behaviors, health status, and diseases in adulthood. Second, some persons with health risk behaviors or diseases may have been either more, or less, likely to report adverse childhood experiences. Each of these issues potentially limits inferences about causality. Furthermore, disease conditions could be either over- or under-reported by patients when they complete the medical questionnaire. In addition, there may be mediators of the relationship between childhood experiences and adult health status other than the risk factors we examined. For example, adverse childhood experiences may affect attitudes and behaviors toward health and health care, sensitivity to internal sensations, or physiologic functioning in brain centers and neurotransmitter systems. A more complete understanding of these issues is likely to lead to more effective ways to address the long-term health problems associated with childhood abuse and household dysfunction.
      However, our estimates of the prevalence of childhood exposures are similar to estimates from nationally representative surveys, indicating that the experiences of our study participants are comparable to the larger population of U.S. adults. In our study, 23.5% of participants reported having grown up with an alcohol abuser; the 1988 National Health Interview Survey estimated that 18.1% of adults had lived with an alcohol abuser during childhood.[

      National Center for Health Statistics. Exposure to alcoholism in the family: United States, 1988. Advance Data, No. 205. U.S. Department of Health and Human Services, Washington, DC; September 30, 1991.

      ]Contact sexual abuse was reported by 22% of respondents (28% of women and 16% of men) in our study. A national telephone survey of adults in 1990 using similar criteria for sexual abuse estimated that 27% of women and 16% of men had been sexually abused.[
      • Finkelhor D
      • Hotaling G
      • Lewis IA
      • Smith C
      Sexual abuse in a national survey of adult men and women prevalence, characteristics, and risk factors.
      ]
      There are several reasons to believe that our estimates of the long-term relationship between adverse childhood experiences and adult health are conservative. Longitudinal follow-up of adults whose childhood abuse was well documented has shown that their retrospective reports of childhood abuse are likely to underestimate actual occurrence.[
      • Femina DD
      • Yeager CA
      • Lewis DO
      Child abuse adolescent records vs. adult recall.
      ,
      • Williams LM
      Recovered memories of abuse in women with documented child sexual victimization histories.
      ]Underestimates of childhood exposures would result in downwardly biased estimates of the relationships between childhood exposures and adult health risk behaviors and diseases. Another potential source of underestimation of the strength of these relationships is the lower number of childhood exposures reported by older persons in our study. This may be an artifact caused by premature mortality in persons with multiple adverse childhood exposures; the clustering of multiple risk factors among persons with multiple childhood exposures is consistent with this hypothesis. Thus, the true relationships between adverse childhood exposures and adult health risk behaviors, health status, and diseases may be even stronger than those we report.
      An essential question posed by our observations is, “Exactly how are adverse childhood experiences linked to health risk behaviors and adult diseases?” The linking mechanisms appear to center on behaviors such as smoking, alcohol or drug abuse, overeating, or sexual behaviors that may be consciously or unconsciously used because they have immediate pharmacological or psychological benefit as coping devices in the face of the stress of abuse, domestic violence, or other forms of family and household dysfunction. High levels of exposure to adverse childhood experiences would expectedly produce anxiety, anger, and depression in children. To the degree that behaviors such as smoking, alcohol, or drug use are found to be effective as coping devices, they would tend to be used chronically. For example, nicotine is recognized as having beneficial psychoactive effects in terms of regulating affect[
      • Carmody TP
      Affect regulation, nicotine addiction, and smoking cessation.
      ]and persons who are depressed are more likely to smoke.[
      • Anda RF
      • Williamson DF
      • Escobedo LG
      • Mast EE
      • Giovino GA
      • Remington PL
      Depression and the dynamics of smoking. A national perspective.
      ,
      • Glassman AH
      • Helzer JE
      • Covey LS
      • Cottler LB
      • Stetner F
      • Tipp JE
      • Johnson J
      Smoking, smoking cessation, and major depression.
      ]Thus, persons exposed to adverse childhood experiences may benefit from using drugs such as nicotine to regulate their mood.[
      • Carmody TP
      Affect regulation, nicotine addiction, and smoking cessation.
      ,
      • Hughes JR
      Clonidine, depression, and smoking cessation.
      ]
      Consideration of the positive neuroregulatory effects of health-risk behaviors such as smoking may provide biobehavioral explanations[
      • Pomerlau OF
      • Pomerlau CS
      Neuroregulators and the reinforcement of smoking towards a biobehavioral explanation.
      ]for the link between adverse childhood experiences and health risk behaviors and diseases in adults. In fact, we found that exposure to higher numbers of categories of adverse childhood experiences increased the likelihood of smoking by the age of 14, chronic smoking as adults, and the presence of smoking-related diseases. Thus, smoking, which is medically and socially viewed as a “problem” may, from the perspective of the user, represent an effective immediate solution that leads to chronic use. Decades later, when this “solution” manifests as emphysema, cardiovascular disease, or malignancy, time and the tendency to ignore psychological issues in the management of organic disease make improbable any full understanding of the original causes of adult disease (Fig. 2). Thus, incomplete understanding of the possible benefits of health risk behaviors leads them to be viewed as irrational and having solely negative consequences.
      Figure thumbnail gr2
      Fig. 2Potential influences throughout the lifespan of adverse childhood experiences.
      Because adverse childhood experiences are common and they have strong long-term associations with adult health risk behaviors, health status, and diseases, increased attention to primary, secondary, and tertiary prevention strategies is needed. These strategies include prevention of the occurrence of adverse childhood experiences, preventing the adoption of health risk behaviors as responses to adverse experiences during childhood and adolescence, and, finally, helping change the health risk behaviors and ameliorating the disease burden among adults whose health problems may represent a long-term consequence of adverse childhood experiences.
      Primary prevention of adverse childhood experiences has proven difficult[
      • Hardy JB
      • Street R
      Family support and parenting education in the home an effective extension of clinic-based preventive health care services for poor children.
      ,
      • Olds DL
      • Henderson CR
      • Chamberlin R
      • Tatelbaum R
      Preventing child abuse and neglect a randomized trial of nurse home visitation.
      ]and will ultimately require societal changes that improve the quality of family and household environments during childhood. Recent research on the long-term benefit of early home visitation on reducing the prevalence of adverse childhood experiences is promising.[
      • Olds DL
      • Eckenrode J
      • Henderson CR
      • Kitzman H
      • et al.
      Long-term effects of home visitation on maternal life course and child abuse and neglect Fifteen-year follow-up of a randomized trial.
      ]In fact, preliminary data from the ACE Study provided the impetus for the Kaiser Health Plan to provide funding to participate at 4 locations (including San Diego County, California) in the Commonwealth Fund’s “Healthy Steps” program. This program extends the traditional practice of pediatrics by adding one or more specialists in the developmental and psychosocial dimensions of both childhood and parenthood. Through a series of office visits, home visits, and a telephone advice line for parents, these specialists develop close relationships between children and their families from birth to 3 years of age. This approach is consistent with the recommendation of the U.S. Advisory Board on Child Abuse and Neglect that a universal home visitation program for new parents be developed[

      U.S. Advisory Board on Child Abuse and Neglect. Child abuse and neglect: critical first steps in response to a national emergency. Washington, DC: U.S. Government Printing Office; August 1990; publication no. 017-092-00104-5.

      ,

      U.S. Advisory Board on Child Abuse and Neglect. Creating caring communities: blueprint for an effective federal policy on child abuse and neglect. Washington, DC: U.S. Government Printing Office; September 1991.

      ]and provides an example of a family-based primary prevention effort that is being explored in a managed care setting. If these types of approaches can be replicated and implemented on a large scale, the long-term benefits may include, somewhat unexpectedly, substantial improvements in overall adult health.
      Secondary prevention of the effects of adverse childhood experiences will first require increased recognition of their occurrence and second, an effective understanding of the behavioral coping devices that commonly are adopted to reduce the emotional impact of these experiences. The improbability of giving up an immediate “solution” in return for a nebulous long-term health benefit has thwarted many well-intended preventive efforts. Although articles in the general medical literature are alerting the medical community to the fact that childhood abuse is common,[
      • MacMillan HL
      • Fleming JE
      • Trocme N
      • Boyle MH
      • et al.
      Prevalence of child physical and sexual abuse in the community. Results from the Ontario Health Supplement.
      ]adolescent health care is often inadequate in terms of psychosocial assessment and anticipatory guidance.[
      • Rixey S
      Family violence and the adolescent.
      ]Clearly, comprehensive strategies are needed to identify and intervene with children and families who are at risk for these adverse experiences and their related outcomes.[
      • Chamberlin RW
      Preventing low birth weight, child abuse, and school failure the need for comprehensive, community-wide approaches.
      ]Such strategies should include increased communication between and among those involved in family practice, internal medicine, nursing, social work, pediatrics, emergency medicine, and preventive medicine and public health. Improved understanding is also needed of the effects of childhood exposure to domestic violence.[
      Council on Scientific Affairs
      Adolescents as victims of family violence.
      ,
      • Kashani JH
      • Daniel AE
      • Dandoy AC
      • Holcomb WR
      Family violence impact on children.
      ]Additionally, increased physician training[
      • Dubowitz H
      Child abuse programs and pediatric residency training.
      ]is needed to recognize and coordinate the management of all persons affected by child abuse, domestic violence, and other forms of family adversity such as alcohol abuse or mental illness.
      In the meantime, tertiary care of adults whose health problems are related to experiences such as childhood abuse[
      • McCauley J
      • Kern DE
      • Kolodner K
      • Schroeder AF
      • et al.
      Clinical characteristics of women with a history of childhood abuse.
      ]will continue to be a difficult challenge. The relationship between childhood experiences and adult health status is likely to be overlooked in medical practice because the time delay between exposure during childhood and recognition of health problems in adult medical practice is lengthy. Moreover, these childhood exposures include emotionally sensitive topics such as family alcoholism[
      • Zeitlen H
      Children with alcohol misusing parents.
      ,
      • Dore MM
      • Doris JM
      • Wright P
      Identifying substance abuse in maltreating families a child welfare challenge.
      ]and sexual abuse.[
      • Tabachnick J
      • Henry F
      • Denny L
      Perceptions of child sexual abuse as a public health problem. Vermont, September 1995.
      ]Many physicians may fear that discussions of sexual violence and other sensitive issues are too personal even for the doctor-patient relationship.[
      • Sugg NK
      • Inui T
      Primary care physicians’ response to domestic violence. Opening Pandora’s box.
      ]For example, the American Medical Association recommends screening of women for exposure to violence at every entrance to the health system;[
      Council on Scientific Affairs
      American Medical Association Diagnostic and Treatment Guidelines on Domestic Violence.
      ]however, such screening appears to be rare.[
      • Hamberger LK
      • Saunders DG
      • Hovey M
      Prevalence of domestic violence in community practice and rate of physician inquiry.
      ]By contrast, women who are asked about exposure to sexual violence say they consider such questions to be welcome and germane to routine medical care,[
      • Friedman LS
      • Samet JH
      • Roberts MS
      • Hans P
      Inquiry about victimization experiences. A survey of patient preferences and physician practices.
      ]which suggests that physicians’ fears about patient reactions are largely unfounded.
      Clearly, further research and training are needed to help medical and public health practitioners understand how social, emotional, and medical problems are linked throughout the lifespan (Fig. 2). Such research and training would provide physicians with the confidence and skills to inquire and respond to patients who acknowledge these types of childhood exposures. Increased awareness of the frequency and long-term consequences of adverse childhood experiences may also lead to improvements in health promotion and disease prevention programs. The magnitude of the difficulty of introducing the requisite changes into medical and public health research, education, and practice can be offset only by the magnitude of the implications that these changes have for improving the health of the nation.[
      • Mullen PE
      • Roman-Clarkson SE
      • Walton VA
      • Herbison GP
      Impact of sexual and physical abuse on women’s mental health.
      ]

      Acknowledgements

      We thank Naomi Howard for her dedication to the ACE Study. This research is supported by the Centers for Disease Control and Prevention via cooperative agreement TS-44-10/12 with the Association of Teachers of Preventive Medicine.

      References

        • Springs F
        • Friedrich WN
        Health risk behaviors and medical sequelae of childhood sexual abuse.
        Mayo Clin Proc. 1992; 67: 527-532
        • Felitti VJ
        Long-term medical consequences of incest, rape, and molestation.
        South Med J. 1991; 84: 328-331
        • Felitti VJ
        Childhood sexual abuse, depression and family dysfunction in adult obese patients.
        South Med J. 1993; 86: 732-736
        • Gould DA
        • Stevens NG
        • Ward NG
        • Carlin AS
        • Sowell HE
        • Gustafson B
        Self-reported childhood abuse in an adult population in a primary care setting.
        Arch Fam Med. 1994; 3: 252-256
        • McCauley J
        • Kern DE
        • Kolodner K
        • Schroeder AF
        • et al.
        Clinical characteristics of women with a history of childhood abuse.
        JAMA. 1997; 277: 1362-1368
        • Mortality patterns: United States, 1993
        Morb Mortal Wkly Rep. 1996; 45: 161-164
        • McGinnis JM
        • Foege WH
        Actual causes of death in the United States.
        JAMA. 1993; 270: 2207-2212
        • Landis J
        Experiences of 500 children with adult sexual deviation.
        Psychiatr Q. 1956; 30: 91-109
        • Straus MA
        • Gelles RJ
        Societal change and change in family violence from 1975 to 1985 as revealed by two national surveys.
        J Marriage Family. 1986; 48: 465-479
        • Wyatt GE
        • Peters SD
        Methodological considerations in research on the prevalence of child sexual abuse.
        Child Abuse Negl. 1986; 10: 241-251
        • Berger AM
        • Knutson JF
        • Mehm JG
        • Perkins KA
        The self-report of punitive childhood experiences of young adults and adolescents.
        Child Abuse Negl. 1988; 12: 251-262
        • Finkelhor D
        • Hotaling G
        • Lewis IA
        • Smith C
        Sexual abuse in a national survey of adult men and women.
        Child Abuse Negl. 1990; 14: 19-28
        • Egelend B
        • Sroufe LA
        • Erickson M
        The developmental consequence of different patterns of maltreatment.
        Child Abuse Negl. 1983; 7: 459-469
        • Finkelhor D
        • Browne A
        The traumatic impact of child sexual abuse.
        Am J Orthopsychiatry. 1985; 55: 530-541
        • Beitchman JH
        • Zucker KJ
        • Hood JE
        • DaCosta GA
        • Akman D
        • Cassavia E
        A review of the long-term effects of sexual abuse.
        Child Abuse Negl. 1992; 16: 101-118
        • Hibbard RA
        • Ingersoll GM
        • Orr DP
        Behavioral risk, emotional risk, and child abuse among adolescents in a nonclinical setting.
        Pediatrics. 1990; 86: 896-901
        • Nagy S
        • Adcock AG
        • Nagy MC
        A comparison of risky health behaviors of sexually active, sexually abused, and abstaining adolescents.
        Pediatrics. 1994; 93: 570-575
        • Cunningham RM
        • Stiffman AR
        • Dore P
        The association of physical and sexual abuse with HIV risk behaviors in adolescence and young adulthood.
        Child Abuse Negl. 1994; 18: 233-245
        • Council on Scientific Affairs
        Adolescents as victims of family violence.
        JAMA. 1993; 270: 1850-1856
        • Nelson DE
        • Higginson GK
        • Grant-Worley JA
        Physical abuse among high school students. Prevalence and correlation with other health behaviors.
        Arch Pediatr Adolesc Med. 1995; 149: 1254-1258
        • Mullen PE
        • Roman-Clarkson SE
        • Walton VA
        • Herbison GP
        Impact of sexual and physical abuse on women’s mental health.
        Lancet. 1988; 1: 841-845
        • Drossman DA
        • Leserman J
        • Nachman G
        • Li Z
        • et al.
        Sexual and physical abuse in women with functional or organic gastrointestinal disorders.
        Ann Intern Med. 1990; 113: 828-833
        • Harrop-Griffiths J
        • Katon W
        • Walker E
        • Holm L
        • Russo J
        • Hickok L
        The association between chronic pelvic pain, psychiatric diagnoses, and childhood sexual abuse.
        Obstet Gynecol. 1988; 71: 589-594
        • Briere J
        • Runtz M
        Multivariate correlates of childhood psychological and physical maltreatment among university women.
        Child Abuse Negl. 1988; 12: 331-341
        • Briere J
        • Runtz M
        Differential adult symptomatology associated with three types of child abuse histories.
        Child Abuse Negl. 1990; 14: 357-364
        • Claussen AH
        • Crittenden PM
        Physical and psychological maltreatment.
        Child Abuse Negl. 1991; 15: 5-18
        • Moeller TP
        • Bachman GA
        • Moeller JR
        The combined effects of physical, sexual, and emotional abuse during childhood.
        Child Abuse Negl. 1993; 17: 623-640
        • Bryant SL
        • Range LM
        Suicidality in college women who were sexually and physically punished by parents.
        Violence Vict. 1995; 10: 195-201
        • Zeitlen H
        Children with alcohol misusing parents.
        Br Med Bull. 1994; 50: 139-151
        • Dore MM
        • Doris JM
        • Wright P
        Identifying substance abuse in maltreating families.
        Child Abuse Negl. 1995; 19: 531-543
        • Ethier LS
        • Lacharite C
        • Couture G
        Childhood adversity, parental stress, and depression of negligent mothers.
        Child Abuse Negl. 1995; 19: 619-632
        • Spaccarelli S
        • Coatsworth JD
        • Bowden BS
        Exposure to family violence among incarcerated boys; its association with violent offending and potential mediating variables.
        Violence Vict. 1995; 10: 163-182
        • McCloskey LA
        • Figueredo AJ
        • Koss MP
        The effects of systemic family violence on children’s mental health.
        Child Dev. 1995; 66: 1239-1261
        • Brent DA
        • Perper JA
        • Moritz G
        • Schweers J
        • Balach L
        • Roth C
        Familial risk factors for adolescent suicide.
        Acta Psychiatr Scand. 1994; 89: 52-58
        • Shaw DS
        • Vondra JI
        • Hommerding KD
        • Keenan K
        • Dunn M
        Chronic family adversity and early child behavior problems.
        J Child Psychol Psychiatry. 1994; 35: 1109-1122
      1. U.S. Department of Health and Human Services. Physical activity and health: A report of the Surgeon General. Atlanta, Georgia. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion; 1996.

        • Rivara FP
        • Mueller BA
        • Somes G
        • Mendoza CT
        • Rushforth NB
        • Kellerman AL
        Alcohol and illicit drug abuse and the risk of violent death in the home.
        JAMA. 1997; 278: 569-575
        • Dillman DA
        Mail and telephone surveys. John Wiley & Sons, New York1978
        • Straus M
        • Gelles RJ
        Physical violence in American families. Transaction Press, New Brunswick1990
        • Wyatt GE
        The sexual abuse of Afro-American and White-American women in childhood.
        Child Abuse Negl. 1985; 9: 507-519
      2. National Center for Health Statistics. Exposure to alcoholism in the family: United States, 1988. Advance Data, No. 205. U.S. Department of Health and Human Services, Washington, DC; September 30, 1991.

      3. Siegel PZ, Frazier EL, Mariolis P, et al. Behaviorial risk factor surveillance, 1991; Monitoring progress toward the Nation’s Year 2000 Health Objectives. Morb Mortal Wkly Rep 1992;42(SS-4)1–15.

        • Crespo CJ
        • Keteyian SJ
        • Heath GW
        • Sempos CT
        Leisure-time physical activity among US adults.
        Arch Intern Med. 1996; 156: 93-98
        • Robins LN
        • Helzer JE
        • Groughan J
        • Ratliff K
        National Institute of Mental Health Diagnostic Interview Schedule.
        Arch Gen Psychiatry. 1981; 38: 381-389
        • Idler E
        • Angel RJ
        Self-rated health and mortality in the NHANES I Epidemiologic Follow-up Study.
        Am J Pub Health. 1990; 80: 446-452
      4. SAS Procedures Guide. SAS Institute Inc. Version 6, 3rd edition, Cary, NC: SAS Institute; 1990.

        • Femina DD
        • Yeager CA
        • Lewis DO
        Child abuse.
        Child Abuse Negl. 1990; 14: 227-231
        • Williams LM
        Recovered memories of abuse in women with documented child sexual victimization histories.
        J Traumatic Stress. 1995; 8: 649-673
        • Carmody TP
        Affect regulation, nicotine addiction, and smoking cessation.
        J Psychoactive Drugs. 1989; 21: 331-342
        • Anda RF
        • Williamson DF
        • Escobedo LG
        • Mast EE
        • Giovino GA
        • Remington PL
        Depression and the dynamics of smoking. A national perspective.
        JAMA. 1990; 264: 1541-1545
        • Glassman AH
        • Helzer JE
        • Covey LS
        • Cottler LB
        • Stetner F
        • Tipp JE
        • Johnson J
        Smoking, smoking cessation, and major depression.
        JAMA. 1990; 264: 1546-1549
        • Hughes JR
        Clonidine, depression, and smoking cessation.
        JAMA. 1988; 259: 2901-2902
        • Pomerlau OF
        • Pomerlau CS
        Neuroregulators and the reinforcement of smoking.
        Neurosci Biobehav Rev. 1984; 8: 503-513
        • Hardy JB
        • Street R
        Family support and parenting education in the home.
        J Pediatr. 1989; 115: 927-931
        • Olds DL
        • Henderson CR
        • Chamberlin R
        • Tatelbaum R
        Preventing child abuse and neglect.
        Pediatrics. 1986; 78: 65-78
        • Olds DL
        • Eckenrode J
        • Henderson CR
        • Kitzman H
        • et al.
        Long-term effects of home visitation on maternal life course and child abuse and neglect.
        JAMA. 1997; 278: 637-643
      5. U.S. Advisory Board on Child Abuse and Neglect. Child abuse and neglect: critical first steps in response to a national emergency. Washington, DC: U.S. Government Printing Office; August 1990; publication no. 017-092-00104-5.

      6. U.S. Advisory Board on Child Abuse and Neglect. Creating caring communities: blueprint for an effective federal policy on child abuse and neglect. Washington, DC: U.S. Government Printing Office; September 1991.

        • MacMillan HL
        • Fleming JE
        • Trocme N
        • Boyle MH
        • et al.
        Prevalence of child physical and sexual abuse in the community. Results from the Ontario Health Supplement.
        JAMA. 1997; 278: 131-135
        • Rixey S
        Family violence and the adolescent.
        Maryland Med J. 1994; 43: 351-353
        • Chamberlin RW
        Preventing low birth weight, child abuse, and school failure.
        Pediatr Rev. 1992; 13: 64-71
        • Kashani JH
        • Daniel AE
        • Dandoy AC
        • Holcomb WR
        Family violence.
        J Am Acad Child Adolesc Psychiatry. 1992; 31: 181-189
        • Dubowitz H
        Child abuse programs and pediatric residency training.
        Pediatrics. 1988; 82: 477-480
        • Tabachnick J
        • Henry F
        • Denny L
        Perceptions of child sexual abuse as a public health problem. Vermont, September 1995.
        Morb Mortal Wkly Rep. 1997; 46: 801-803
        • Sugg NK
        • Inui T
        Primary care physicians’ response to domestic violence. Opening Pandora’s box.
        JAMA. 1992; 267: 3157-3160
        • Council on Scientific Affairs
        American Medical Association Diagnostic and Treatment Guidelines on Domestic Violence.
        Arch Fam Med. 1992; 1: 38-47
        • Hamberger LK
        • Saunders DG
        • Hovey M
        Prevalence of domestic violence in community practice and rate of physician inquiry.
        Fam Med. 1992; 24: 283-287
        • Friedman LS
        • Samet JH
        • Roberts MS
        • Hans P
        Inquiry about victimization experiences. A survey of patient preferences and physician practices.
        Arch Int Med. 1992; 152: 1186-1190