ANNOTATED

www.elsevier.com/locate/jad

Journal of Affective Disorders 82 (2004) 217–225

Research report

Adverse childhood experiences and the risk of depressive

disorders in adulthood

Daniel P. Chapman a,*, Charles L. Whitfield

b , Vincent J. Felitti

c , Shanta R. Dube

a ,

Valerie J. Edwards a , Robert F. Anda

a

a Division of Adult and Community Health, National Center for Chronic Disease Prevention and Health Promotion,

Centers for Disease Control and Prevention, 4770 Buford Highway NE, Mailstop K-67, Atlanta, GA 30341, USA b Private Practice of Addiction Medicine and Trauma Psychology, Atlanta, GA, USA

c Department of Preventive Medicine, Southern California Kaiser Permanente Medical Group (Kaiser Permanente), San Diego, CA, USA

Received 24 July 2003; accepted 3 December 2003

Abstract

Background: Research examining the association between childhood abuse and depressive disorders has frequently assessed

abuse categorically, thus not permitting discernment of the cumulative impact of multiple types of abuse. As previous research

has documented that adverse childhood experiences (ACEs) are highly interrelated, we examined the association between the

number of such experiences (ACE score) and the risk of depressive disorders. Methods: Retrospective cohort study of 9460

adult health maintenance organization members in a primary care clinic in San Diego, CA who completed a survey addressing a

variety of health-related concerns, which included standardized assessments of lifetime and recent depressive disorders,

childhood abuse and household dysfunction. Results: Lifetime prevalence of depressive disorders was 23%. Childhood

emotional abuse increased risk for lifetime depressive disorders, with adjusted odds ratios (ORs) of 2.7 [95% confidence

interval (CI), 2.3–3.2] in women and 2.5 (95% CI, 1.9–3.2) in men. We found a strong, dose–response relationship between

the ACE score and the probability of lifetime and recent depressive disorders (P < 0.0001). This relationship was attenuated

slightly when a history of growing up with a mentally ill household member was included in the model, but remained

significant (P < 0.001). Conclusions: The number of ACEs has a graded relationship to both lifetime and recent depressive

disorders. These results suggest that exposure to ACEs is associated with increased risk of depressive disorders up to decades

after their occurrence. Early recognition of childhood abuse and appropriate intervention may thus play an important role in the

prevention of depressive disorders throughout the life span.

Published by Elsevier B.V.

Keywords: Child abuse; Depressive disorders

Research has documented an increased prevalence

of psychiatric disorders among individuals experienc-

0165-0327/$ - see front matter. Published by Elsevier B.V.

doi:10.1016/j.jad.2003.12.013

* Corresponding author. Tel.: +1-770-488-5463; fax: +1-770-

488-5965.

E-mail address: [email protected] (D.P. Chapman).

ing childhood abuse and trauma relative to their age

peers (Briere et al., 1997; Silverman et al., 1996). In

particular, childhood abuse has been associated with

subsequent development of posttraumatic stress dis-

order (PTSD) (Widom, 1999), borderline personality

disorder (Herman et al., 1989), dissociative symptoms

D.P. Chapman et al. / Journal of Affective Disorders 82 (2004) 217–225218

(Chu et al., 1999), and depression (Sansone et al.,

2001).

Childhood physical abuse (Goldberg, 1994; Kauf-

man, 1991), sexual abuse, (Goldberg, 1994; Kinard,

1995) and psychological or emotional abuse (Fergu-

son and Dacey, 1997; Kaufman, 1991) have each been

associated with an increased prevalence of depressive

disorders. However, the effects of each of these types

of abuse have generally been examined categorically,

and in ways not permitting assessment of the cumu-

lative impact of multiple types of abuse or assessment

of a dose–response relationship between the number

of types of abuse and the prevalence of depressive

disorders.

Because abuse, domestic violence, and other forms

of household dysfunction, which we term adverse

childhood experiences (ACEs), are interrelated (Anda

et al., 1999; Felitti et al., 1998) and have repeatedly

demonstrated a strong graded relationship to a variety

of health problems (Anda et al., 2001; Dube et al.,

2001), we assessed the relationship of each of these

ACEs to the lifetime risk of depressive disorders.

Using a cumulative stressor model, we then examined

the relationship between the total number of adverse

childhood experiences (ACEs) and the prevalence of

depressive disorders.

1. Methods

The data were collected as part of the ACE Study, a

collaboration between Kaiser Permanente (San Diego,

CA) and the Centers for Disease Control and Preven-

tion (CDC, Atlanta). The study was approved by the

institutional review boards of Kaiser Permanente and

the Office of Protection from Research Risks at the

National Institutes of Health. Potential participants

received letters that accompanied the ACE study

questionnaire informing them that their participation

was voluntary, their answers would be held in strictest

confidence, and would never become part of their

medical records.

1.1. Study population and data collection

The study population consisted of adult members

of the Kaiser Health Plan who received a standard-

ized medical and biopsychosocial examination at

Kaiser’s Health Appraisal Center in San Diego,

CA. In any 4-year period, 81% of all adult members

received the examination, and more than 50,000

members are examined annually. The primary pur-

pose of the examination is to conduct a complete

health assessment rather than provide symptom- or

illness-based care.

The ACE Study consisted of two survey waves.

Wave I was conducted among 13,494 consecutive

members visiting Kaiser’s Health Appraisal Center

between August 1995 and March 1996, with a

response rate of 70% (n = 9508). Because the full

questionnaire used to screen for depressive disor-

ders was contained in Wave I only, this analysis is

restricted to Wave I data. The details of the study

have been published elsewhere (Felitti et al.,

1998).

The ACE Study questionnaire was mailed to each

member 2 weeks after their examination and collected

information on ACEs, including abuse (emotional,

physical, or sexual), or household dysfunction (paren-

tal separation or divorce, having a battered mother or

a substance abusing, criminal or mentally ill house-

hold member), as well as health-related behaviors

from adolescence to adulthood.

1.2. Assessment of representatives and response or

reporting bias

Standardized health examination data were ab-

stracted for both respondents and nonrespondents to

the ACE Study questionnaire enabling a detailed

assessment of possible bias in terms of demograph-

ic characteristics and health-related issues. Although

nonrespondents tended to be younger, less educated,

and more likely to be members of racial and ethnic

minority groups, the prevalence of both psychoso-

cial and health problems was remarkably similar

between respondents and nonrespondents after con-

trolling for demographic differences (Edwards et al.,

2001).

1.3. Exclusions from the study cohort

We excluded 34 respondents with missing infor-

mation about race and 14 with missing educational

attainment. Thus, the final study cohort included 98%

of the respondents (9460/9508).

Affect

1.3.1. Definitions of adverse childhood experiences

All questions about ACEs pertained to the respon-

dent’s first 18 years of life. Questions used to define

emotional and physical abuse, and growing up with a

battered mother were adapted from the Conflict Tac-

tics Scale (CTS) (Straus, 1979) with the response

categories of never, once or twice, sometimes, often,

or very often.

1.3.2. Emotional abuse

Participants were defined as being emotionally

abused during childhood if they responded often or

very often to either of the following two questions:

‘‘How often did a parent, stepparent, or adult living in

your home swear at you, insult you, or put you

down?’’ and ‘‘How often did a parent, stepparent, or

adult living in your home act in a way that made you

afraid that you might be physically hurt?’’

1.3.3. Physical abuse

Two items were adapted from the Conflict Tac-

tics Scale (CTS) (Straus, 1979). Respondents who

indicated that they had been pushed, grabbed,

shoved, slapped, or had something thrown at them

‘‘often’’ or ‘‘very often,’’ or who indicated they had

been hit so hard that they had marks or were

injured ‘‘once’’ or more were considered victims

of physical abuse.

1.3.4. Sexual abuse

Assessed by four categorical questions adapted

from Wyatt (1985), that covered fondling, attempted

intercourse, and intercourse. An affirmative answer to

any of the four items resulted in classification as

sexually abused.

1.3.5. Battered mother

Four items adapted from CTS were used to

determine whether respondents were exposed to

family violence. Answers indicating that the respon-

dent had witnessed their mother ‘‘sometimes’’ or

more being pushed, grabbed, slapped or seen some-

thing thrown at her, or who witnessed more serious

violence (kicking, biting, hit with a fist or something

hard, repeatedly hit over at least a few minutes,

threatened with a knife or gun, or used a knife or

gun to hurt her) once or more were considered

exposed to a battered mother.

D.P. Chapman et al. / Journal of

1.3.6. Household substance abuse

Two questions were used to determine whether

respondents during their childhood, lived with a

problem drinker or alcoholic (Schoenborn, 1995) or

anyone who used street drugs.

1.3.7. Parental separation or divorce

This adverse experience was defined as an affir-

mative response to the question ‘‘Were your parents

ever separated or divorced?’’

1.3.8. Criminal household member

The respondent was defined as having childhood

exposure to a criminal household member if anyone in

the household had gone to prison during the respon-

dent’s childhood.

1.4. Mental illness in household

A respondent was defined as being exposed to

mental illness if anyone in the household was

depressed or mentally ill or had attempted suicide

during the respondent’s childhood. Because expo-

sure to mental illness during childhood can exert an

effect on the risk of depressive disorders during

adulthood through both experiential and potential

genetic influences, we treated this ACE as a sepa-

rate type of exposure which was not included in the

ACE score. Rather, we controlled for the effect of

this ACE separately in our analysis of the associa-

tion between the ACE score and the risk of depres-

sive disorders.

1.5. The ACE score

The total number of ACEs (excluding mental

illness in the household) experienced by respondents

became their ACE score, which was used to assess the

cumulative effect of multiple ACEs.

1.6. Personal history of depressive disorders

We used a screening instrument for depressive

disorders (major depression and dysthymia) devel-

oped for the Medical Outcomes Study (Burnam et

al., 1988). This instrument used data from primary

care and mental health subsamples of the Los Angeles

Epidemiological Catchment Area Study (Burnam et

ive Disorders 82 (2004) 217–225 219

D.P. Chapman et al. / Journal of Affective Disorders 82 (2004) 217–225220

al., 1987) and the Psychiatric Screening Question-

naires for Primary Care Patients (Hough et al., 1983).

The screening instrument (Burnam et al., 1988)

includes the following two questions from the Diag-

nostic Interview Schedule (DIS) of the National Insti-

tute of Mental Health (Robins et al., 1981): (1) ‘‘In the

past year, have you had 2 weeks or more during which

you felt sad, blue, or depressed, or lost pleasure in

things that you usually cared about or enjoyed?’’ and

(2) ‘‘Have you had 2 years or more in your life when

you felt depressed or sad most days, even if you felt

okay sometimes? (If yes) Have you felt depressed or

sad much of the time in the past year?’’

The instrument also included six questions from

the Center for Epidemiologic Studies Depression

Scale (CES-D) (Roberts and Vernon, 1983). The

CES-D items ask how often in the past week the

respondent had experienced the following: (1) ‘‘I felt

depressed.’’ (2) ‘‘My sleep was restless.’’ (3) ‘‘I

enjoyed life.’’ (4) ‘‘I had crying spells.’’ (5) ‘‘I felt

sad.’’ (6) ‘‘I felt that people dislike me.’’ The response

scale for these questions was less than 1, 1–2, 3–4,

and 5–7 days. Using the prediction equation devel-

oped for this screener, we used the cutoff point of

0.009 to define a lifetime history of depression disor-

der and the cutoff point of 0.06 for recent depressive

disorders occurring during the past year (see Burnam

et al., 1988, for further details).

1.7. Statistical analysis

We used the Statistical Analysis System (SAS) for

all analyses. Persons with incomplete information

about an ACE were considered not to have had that

experience. This would likely result in conservative

estimates of the relationship between ACEs and health

outcomes because persons who had potentially been

exposed to an experience would be misclassified as

unexposed. This type of misclassification would bias

our results towards the null. However, to assess this

potential effect, we repeated our analyses after ex-

cluding all respondents with missing information on

any of the ACEs.

Adjusted odds ratios (ORs) and 95% confidence

intervals (CIs) from logistic regression models were

used to assess the associations between each category

of ACE and the risk of depressive disorders. The

number of ACEs was summed for each respondent

(range: 0–7); analyses were repeated with five di-

chotomous variables (yes/no) with 0 ACEs as the

referent. To test for observed trends in the ORs from

the models using five dichotomous variables, the

summed score was entered as an ordinal variable (0,

1, 2, 3, 4, or z 5). We also modeled the relationship of the ACE score to depressive disorders with and

without controlling for exposure to mental illness in

the household.

Attributable risk fractions (ARFs) were calculated

using adjusted ORs from logistic regression models

based upon z 1 ACE with 0 ACEs as the referent. This analysis was done because a substantial increase

in the risk of depressive disorders was seen for

persons reporting at least 1 ACE. We used Levin’s

formula for these calculations, ARF = P1(RR-1)/

1 + P1(RR-1), where P1 is the prevalence of an ACE

score z 1 and RR = OR of depressive disorders for an ACE score z 1. The ARF is an estimate of the proportion of the health problem (e.g., depressive

disorders) that would not have occurred if no persons

had been exposed to the risk factor being assessed

(ACEs) (Haddix et al., 1996).

2. Results

Of the 9460 respondents, 54% were women. The

mean age of respondents was 56.6 years. Seventy-five

percent were white; 42% were college graduates and

only 7% had not graduated from high school. The

prevalence of a lifetime history of depressive disor-

ders was greater among women than men (28.9% vs.

19.4%), as was the prevalence of recent depressive

disorders (15.7% vs. 8.4%). Approximately one in

five women reported the presence of a mentally ill

household member while they were growing up, with

a slightly lower prevalence reported among men

(15.0%). Among both men and women, the preva-

lence of each of the ACEs ranged from over 3%

reporting a criminal household member to about 30%

indicating they had been physically abused while

growing up. Notably, 20.8% of women and 14.0%

of men reported they had experienced three or more

ACEs while growing up, excluding mental illness in

the household.

Table 1 summarizes the associations between each

of the seven ACEs and the lifetime history and recent

Table 1

Adverse childhood experiences and the prevalence and risk (adjusted odds ratio) of a lifetime history of depressive disorders or of recent

depressive disorders

Adverse childhood Women reporting Men reporting Lifetime prevalence Recent prevalence

experience (ACE) ACE (n) ACE (n) Women adjusted

odds ratio

Men adjusted

odds ratio

Women adjusted

odds ratio

Men adjusted

odds ratio

Emotional abuse 709 319 2.7 (2.3–3.2) 2.5 (1.9–3.2) 3.1 (2.6–3.8) 3.3 (2.4–4.4)

Physical abuse 1456 1394 2.1 (1.8–2.4) 1.6 (1.4–1.9) 2.3 (2.0–2.7) 1.8 (1.4–2.2)

Sexual abuse 1246 650 1.8 (1.5–2.0) 1.6 (1.3–2.0) 2.0 (1.7–2.3) 1.6 (1.2–2.1)

Battered mother 676 478 2.1 (1.8–2.5) 1.5 (1.2–1.9) 2.2 (1.8–2.7) 1.5 (1.1–2.1)

Household substance abuse 1413 965 1.7 (1.5–2.0) 1.3 (1.1–1.5) 1.8 (1.5–2.1) 1.3 (1.1–1.6)

Parental separation or divorce 1174 929 1.4 (1.2–1.6) 1.1 (1.0–1.4) 1.4 (1.2–2.7) 1.0 (0.9–1.3)

Criminal household member 177 141 1.7 (1.2–2.3) 1.5 (1.1–2.1) 1.6 (1.1–2.2) 0.9 (0.5–1.6)

D.P. Chapman et al. / Journal of Affective Disorders 82 (2004) 217–225 221

prevalence of depressive disorders for women and

men. Among women, adjusted odds ratios reveal

significant associations between each ACE and both

a lifetime history of depressive disorders and recent

depressive disorders. Notably, women reporting child-

hood emotional abuse were 2.7 and 3.1 times as likely

as those not reporting emotional abuse to have a

lifetime history of depressive disorders or recent

depressive disorders, respectively. Most of the indi-

vidual ACEs were also significantly associated with a

recent and lifetime history of depressive disorders

among men, with the exception of growing up with

a criminal household member and parental separation

or divorce. As was observed among women, child-

Table 2

Relationship of the ACE score and a history of growing up with a mentall

and recent depressive disorders among women

ACE score (N) Lifetime history of depressive di

% Separate models a

adjusted odds ratio

Sin

adj

0 (1984) 18.5 1.0 (referent) 1.0

1 (1289) 25.8 1.4 (1.2–1.6) 1.3

2 (742) 32.7 1.8 (1.5–2.2) 1.6

3 (503) 44.7 3.0 (2.4–3.7) 2.5

4 (322) 47.5 3.0 (2.3–3.9) 2.4

z 5 (236) 61.0 5.0 (3.7–6.7) 3.7

Mentally ill No (4018) 24.3 1.0 (referent) 1.0

household member Yes (1058) 46.1 2.5 (2.1–2.8) 1.8

a Separate logistic models were run for ACE score and a history of g

included both the ACE score and a history of mental illness in the househ

model with the ACE score vs. the single model for a lifetime history of dep

the X 2 for the difference in the log likelihood ratios for recent depressive

hood emotional abuse posed the greatest risk of any of

the ACEs for both a lifetime history of depressive

disorders and recent depressive disorders (adjusted

odds ratios: 2.5 and 3.3, respectively) among men.

The cumulative effects of ACEs on the probability

of a lifetime history of depressive disorders and

current depressive disorders are summarized for wom-

en and for men in Tables 2 and 3, respectively. As the

presence of a mentally ill household member could

potentially be confounded with genetic influences on

the etiology of depressive disorders, we analyzed the

effect of this ACE in separate logistic models, as well

as in a single model in which both the ACE score and

a childhood history of having a mentally ill household

y ill household member to a lifetime history of depressive disorders

sorders Recent depressive disorders

gle model a

usted odds ratio

% Separate models a

adjusted odds ratio

Single model a

adjusted odds ratio

(referent) 8.3 1.0 (referent) 1.0 (referent)

(1.1–1.5) 13.5 1.6 (1.2–2.0) 1.4 (1.1–1.8)

(1.3–2.0) 18.7 2.1 (1.7–2.7) 1.8 (1.4–2.4)

(2.0–3.1) 24.1 2.9 (2.2–3.8) 2.3 (1.7–3.0)

(1.9–3.2) 29.8 3.6 (2.7–4.8) 2.7 (2.0–3.7)

(2.7–5.0) 44.1 6.4 (4.7–8.7) 4.4 (3.2–6.1)

(referent) 12.0 1.0 (referent) 1.0 (referent)

(1.5–2.1) 29.9 2.9 (2.4–3.4) 2.0 (1.7–2.5)

rowing up with a mentally ill household member; the single model

old. X 2 for the difference in the log likelihood ratios for the single

ressive disorders was 65 (1 degree of freedom; P<0.0001). Similarly,

disorders was 48 (1 degree of freedom; P<0.0001).

D.P. Chapman et al. / Journal of Affective Disorders 82 (2004) 217–225222

member were entered simultaneously. This approach

is conservative as it likely overcontrolled for genetic

influences while simultaneously disregarding experi-

ential influences; thus, the single model likely under-

estimates the strength of the associations.

Compared to women reporting no ACEs, those

who reported five or more ACEs had a fivefold

increased risk for a lifetime history of depressive

disorders and a greater than sixfold increased risk

for recent depressive disorders in the separate logistic

models. This relationship was attenuated somewhat in

the single logistic models in which the ACE score and

childhood history of having a mentally ill household

member were entered simultaneously, but remained

statistically significant for both women and men

(Tables 2 and 3) ( P < 0.0001). Despite these attenu-

ations, in this model, women reporting five or more

ACEs had adjusted odd ratios of 3.7 and 4.4 for

lifetime and recent depressive disorders, respectively.

Cumulative exposure to ACEs generally assumed a

stronger dose–response relationship with depressive

disorders among women than men. However, the

presence of ACEs was also associated with strong

and significantly increased risks of both lifetime and

current depressive disorders in men. Among men,

ACEs assumed cumulative effects on lifetime history

of depressive disorders in separate models (adjusted

odds ratios = 1.4 for one ACE, 1.8 for two ACEs, and

Table 3

Relationship of the ACE score and a history of growing up with a mentall

and recent depressive disorders among men

ACE score (N) Lifetime history of depressive di

% Separate models a

adjusted odds ratio

Sin

adj

0 (1823) 13.8 1.0 (referent) 1.0

1 (1256) 19.0 1.4 (1.1–1.7) 1.3

2 (693) 24.4 1.8 (1.4–2.2) 1.6

3 (340) 26.5 1.9 (1.4–2.5) 1.5

4 (176) 36.9 2.9 (2.0–4.0) 2.3

z 5 (96) 35.4 2.4 (1.5–3.7) 1.7

Mentally ill No (3726) 16.7 1.0 (referent) 1.0

household member Yes (658) 34.6 2.4 (2.0–2.9) 2.1

a Separate logistic models were run for ACE score and a history of g

included both the ACE score and a history of mental illness in the househ

model with the ACE score vs. the single model for a lifetime history of dep

the X 2 for the difference in the log likelihood ratios for recent depressive

2.4 for five or more ACEs) (Table 3). Our estimate of

the attributable risk fraction (ARF) for lifetime de-

pressive disorders was 35% (women, 38%; men, 29%)

and 40% for recent depressive disorders (women,

46%; men, 29%).

3. Discussion

This investigation represents a departure from

previous studies characteristically restricted to exam-

ination of the association between single forms of

abuse and depressive disorders. Our results indicate

that the majority of respondents reported at least one

ACE and approximately one-third of adults experi-

enced at least two ACEs during their childhood. This

finding, and prior publications from the ACE Study

(Anda et al., 1999; Dube et al., 2002; Felitti et al.,

1998), suggest that detection of one ACE should alert

the clinician to assess the patient for a history of

exposure to other forms of abuse or household dys-

function. Moreover, a strong graded relationship was

generally evident between the number of ACEs and

recent and lifetime depressive disorders among men

and women. These findings suggest that experiencing

multiple forms of abuse or household dysfunction

during childhood may pose particularly deleterious

consequences on adult mental health.

y ill household member to a lifetime history of depressive disorders

sorders Recent depressive disorders

gle model a

usted odds ratio

% Separate models a

adjusted odds ratio

Single model a

adjusted odds ratio

(referent) 5.5 1.0 (referent) 1.0 (referent)

(1.1–1.6) 8.2 1.4 (1.1–1.9) 1.4 (1.1–1.8)

(1.3–2.0) 10.8 1.8 (1.3–2.5) 1.6 (1.2–2.2)

(1.1–2.1) 11.2 1.8 (1.2–2.7) 1.4 (1.0–2.2)

(1.6–3.3) 18.7 3.2 (2.0–4.9) 2.5 (1.6–3.9)

(1.1–2.8) 17.7 2.6 (1.5–4.6) 1.8 (1.1–3.3)

(referent) 6.8 1.0 (referent) 1.0 (referent)

(1.7–2.5) 17.0 2.5 (1.9–3.1) 2.1 (1.6–2.8)

rowing up with a mentally ill household member; the single model

old. X 2 for the difference in the log likelihood ratios for the single

ressive disorders was 46 (1 degree of freedom; P<0.0001). Similarly,

disorders was 33 (1 degree of freedom; P<0.0001).

D.P. Chapman et al. / Journal of Affective Disorders 82 (2004) 217–225 223

The lifetime prevalence of depressive disorders in

this study (women, 28.9%; men, 19.4%) is similar to

that obtained in other studies conducted in clinical

populations (Nagel et al., 1998; Zung et al., 1993).

However, it must be acknowledged that respondents

reporting depressive symptoms cannot be presumed to

be suffering from major depression. Nevertheless,

previous research has documented the negative effect

of depressive symptoms on quality of life (Pyne et al.,

1997) and activities of daily living (Alexopoulos et

al., 1996).

Consistent with previous research (Weissman et al.,

1991; Zung et al., 1993), we found a greater preva-

lence of depressive disorders among women than

men. These findings are paralleled by a greater

reported prevalence of ACEs—with the exception of

physical abuse—among women than men. While

potentially mediated by a number of variables unex-

amined here, our data suggest the increased preva-

lence of depressive disorders among women may be

at least partially attributable to the higher prevalence

of ACEs and their stronger association with depres-

sive disorders in women.

A slight reduction in the effect of ACEs on

depressive symptoms was observed among respond-

ents who reported the presence of a mentally ill

household member, relative to those who did not.

Yet, even after controlling for this factor, ACEs

remained strongly associated with the development

of depressive disorders across the life span. However,

the mentally ill household members inquired about

were not necessarily suffering from depression and

may not have been blood relatives. Thus, conclusions

about the role of genetics in the development of

depression throughout the life span cannot be drawn

from these data.

While permitting examination of a wide variety of

childhood experiences and subsequent behavioral and

health outcomes, the ACE Study has several potential

limitations. The duration of each ACE was not

assessed. Moreover, these data cannot definitively

establish the temporal relationship between ACEs

and lifetime depressive disorders, as it is conceivable

that the onset of a depressive disorder may have

preceded exposure to ACEs. However, the finding

of strong, positive associations between ACEs and

depressive disorders occurring during the past year

would appear to obviate this concern, although the

possibility of a mood state bias altering recall of

childhood events must be acknowledged. Nonethe-

less, as there is only a modest attenuation of the

cumulative effect of ACEs on recent onset relative

to lifetime depressive disorders, these data strongly

suggest that ACEs exert deleterious consequences

threatening mental health throughout adulthood.

As the median age of respondents was 57 years,

the positive association of the ACE score with recent

depressive symptoms suggests the consequences of

ACEs persist for several decades after their occur-

rence. This observation thus extends previous find-

ings of increased depressive symptomatology among

abused children (Stern et al., 1995) and suggests that

the depressogenic potential of child abuse extends far

into adulthood. Further research identifying character-

istics distinguishing adults who experienced child-

hood abuse and do not manifest increased adulthood

depressive symptomatology from those who do is

needed.

Of all the individual ACEs, emotional abuse

exhibited the strongest relationship to both measures

of depressive symptoms among both men and women.

The increased risk evident for emotional abuse was

statistically significant in comparison with most of the

other ACEs. These findings corroborate previous

investigations documenting the deleterious conse-

quences of emotional (Roy, 1999) or psychological

(Ferguson and Dacey, 1997) abuse on mental health.

Our results suggest that emotional abuse is character-

istically combined with other forms of abuse, thereby

potentiating its impact. Succinctly stated, ‘‘names do

hurt’’ and assessment for childhood emotional abuse

may provide an important benchmark for other forms

of abuse and a heightened risk for depressive symp-

toms in adulthood.

Adverse childhood experiences have a strong,

graded relationship to the risk of lifetime and

current depressive disorders that extends into adult-

hood. Because ACEs are interrelated (Anda et al.,

1999; Felitti et al., 1998), it is important to con-

sider abuse and household dysfunction as a set of

experiences that affect the risk of depressive dis-

orders. Moreover, ACEs are common and account

for a considerable proportion of depressive disor-

ders—as evidenced by the estimates of the popula-

tion attributable risk. Prevention of ACEs and early

treatment of persons affected by them will likely

D.P. Chapman et al. / Journal of Affective Disorders 82 (2004) 217–225224

substantially decrease the serious burden of depres-

sive disorders.

Acknowledgements

The Adverse Childhood Experiences Study was

supported by a grant from the Garfield Memorial

Fund.

References

Alexopoulos, G.S., Vrontou, C., Kakuma, T., Meyers, B.S., Young,

R.C., Klausner, E., Clarkin, J., 1996. Disability in geriatric de-

pression. Am. J. Psychiatry 153, 877–885.

Anda, R.F., Croft, J.S., Felitti, V.J., Nordenberg, D., Giles, W.H.,

Williamson, D.F., 1999. Adverse childhood experiences and

smoking during adolescence and adulthood. JAMA 282,

1652–1658.

Anda, R.F., Felitti, V.J., Chapman, D.P., Croft, J.S., Williamson,

D.F., Santelli, J., Dietz, P.M., Marks, J.S., 2001. Abused boys,

battered mothers, and male involvement in teen pregnancy.

Pediatrics 107, e19.

Briere, J., Woo, R., McRae, B., Foltz, J., Sitzman, R., 1997. Life-

time victimization history, demographics, and clinical status in

female psychiatric emergency room patients. J. Nerv. Ment. Dis.

185, 95–101.

Burnam, M.A., Hough, R.L., Escobar, J.I., Karno, M., Timbers,

D.M., Telles, C.A., Locke, B.Z., 1987. Six-month prevalence of

specific psychiatric disorders among Mexican–Americans and

non-Hispanic Whites in Los Angeles. Arch. Gen. Psychiatry

44, 687–694.

Burnam, M.A., Wells, K.B., Leake, B., Landsverk, J., 1988. Devel-

opment of a brief screening instrument for detecting depressive

disorders. Med. Care 26, 775–789.

Chu, J.A., Frey, L.M., Ganzel, B.L., Matthews, J.A., 1999. Mem-

ories of childhood abuse: dissociation, amnesia, and corrobora-

tion. Am. J. Psychiatry 156, 749–755.

Dube, S.R., Anda, R.F., Felitti, V.J., Chapman, D.P., Williamson,

D.F., Giles, W.H., 2001. Childhood abuse, household dysfunc-

tion, and the risk of attempted suicide throughout the life span:

findings from the adverse childhood experiences study. JAMA

286, 3089–3096.

Dube, S.R., Anda, R.F., Felitti, V.J., Edwards, V.J., Williamson,

D.F., 2002. Exposure to abuse, neglect, and household dys-

function. Violence Vict. 17, 3–17.

Edwards, V.J., Anda, R.F., Nordenberg, D.F., Felitti, V.J., William-

son, D.F., Wright, J.A., 2001. Bias assessment for child abuse

survey: factors affecting probability of response to a survey

about child abuse. Child Abuse Negl. 25, 307–312.

Felitti, V.J., Anda, R.F., Nordenberg, D., Williamson, D.F., Spitz,

A.M., Edwards, V., Koss, M.P., Marks, J.S., 1998. Relation-

ship of childhood abuse and household dysfunction to many

of the leading causes of death in adults: the Adverse Child-

hood Experiences (ACE) study. Am. J. Prev. Med. 14,

245–258.

Ferguson, K.S., Dacey, C.M., 1997. Anxiety, depression, and

dissociation in women health care providers reporting a his-

tory of childhood psychological abuse. Child Abuse Negl. 21,

941–952.

Goldberg, R.T., 1994. Childhood abuse, depression, and chronic

pain. Clin. J. Pain 10, 277–281.

Haddix, A.C., Teutsch, S.M., Shaeffer, P.A., Dunet, P.O., 1996.

Prevention Effectiveness: a Guide to Decision Analysis and

Economic Evaluation. Oxford Univ. Press, New York, NY.

Herman, J.L., Perry, J.C., van der Kolk, B.A., 1989. Childhood

trauma in borderline personality disorder. Am. J. Psychiatry

146, 490–495.

Hough, R.L, Landsverk, J.A., Stone, J.D., et al., 1983. Psychiatric

screening scale project: final report. Contract #DB-81-0036.

National Institutes of Mental Health, Bethesda, MD.

Kaufman, J., 1991. Depressive disorders in maltreated children.

J. Am. Acad. Child Adolesc. Psych. 30, 257–265.

Kinard, E.M., 1995. Mother and teacher assessments of behavior

problems in abused children. J. Am. Acad. Child Adolesc.

Psych. 34, 1043–1053.

Nagel, R., Lynch, D., Tamburrino, M., 1998. Validity of the

medical outcomes study depression screener in family practice

training centers and community settings. J. Fam. Med. 30,

362–365.

Pyne, J.M., Patterson, T.L., Kaplan, R.M., Gillin, J.C., Koch, W.L.,

Grant, I., 1997. Assessment of the quality of life of patients with

major depression. Psychiatr. Serv. 48, 224–230.

Roberts, R.E., Vernon, S.W., 1983. The center for epidemiologic

studies depression scale: its use in a community sample. Am. J.

Psychiatry 140, 41–46.

Robins, L.N., Helzer, J.E., Groughan, J., Ratliff, K.S., 1981. Na-

tional Institute of Mental Health diagnostic interview schedule:

its history, characteristics, and validity. Arch. Gen. Psychiatry

38, 381–389.

Roy, A., 1999. Childhood trauma and depression in alcoholics:

relationship to hostility. J. Affect. Disord. 56, 215–218.

Sansone, R.A., Wiederman, M.W., Sansone, L.A., 2001. Adult so-

matic preoccupation and its relationship to childhood trauma.

Violence Vict. 16, 39–47.

Schoenborn, C.A., 1995. Exposure to alcoholism in the family:

United States 1988. Advance Data from Vital and Health Sta-

tistics, vol. 205. National Center for Health Statistics, Hyatts-

ville MD. Publication PHS 95-1880.

Silverman, A.B., Reinherz, H.Z., Gianconia, R.M., 1996. The long-

term sequelae of child and adolescent abuse: a longitudinal

community study. Child Abuse Negl. 20, 709–723.

Stern, A.E., Lynch, D.L., Oates, R.K., O’Toole, B.I., Cooney, G.,

1995. Self esteem, depression, behaviour and family functioning

in sexually abused children. J. Child Psychol. Psychiatry 36,

1077–1089.

Straus, M.A., 1979. Measuring intrafamily conflict and violence:

the Conflict Tactics (CT) scales. J. Marriage Fam. 41, 75–88.

Weissman, M.M., Bruce, M.L., Leaf, P.J., Florio, L.P., Holzer, C.,

1991. Affective disorders. In: Robins, L.N., Regier, D.A. (Eds.),

D.P. Chapman et al. / Journal of Affective Disorders 82 (2004) 217–225 225

Psychiatric Disorders in America. Free Press, New York, NY,

pp. 53–80.

Widom, C.S., 1999. Posttraumatic stress disorder in abused

and neglected children grown up. Am. J. Psychiatry 156,

1223–1229.

All in-text references underlined in blue are linked to publications

Wyatt, G.E., 1985. The sexual abuse of Afro-American and white

American women in childhood. Child Abuse Negl. 9, 507–519.

Zung, W.W., Broadhead, W.E., Roth, M.E., 1993. Prevalence of

depressive symptoms in primary care. J. Fam. Pract. 37,

337–344.

on ResearchGate, letting you access and read them immediately.