Adverse Childhood Experiences and Chronic

Obstructive Pulmonary Disease in Adults

Robert F. Anda, MD, MS, David W. Brown, MSPH, MS, Shanta R. Dube, PhD, MPH, J. Douglas Bremner, MD,
Vincent J. Felitti, MD, Wayne H. Giles, MD, MS

Background: Chronic obstructive pulmonary disease (COPD) is an important cause of morbidity and
mortality in the U.S. However, little is known about the influence of childhood stressors on
its occurrence.
Methods: Data were from 15,472 adult HMO members enrolled in the Adverse Childhood Experi-
ences (ACE) Study from 1995 to 1997 and eligible for the prospective phase. Eight ACEs
were assessed: abuse (emotional, physical, sexual); witnessing domestic violence; growing
up with substance-abusing, mentally ill, or criminal household members; and parental
separation or divorce. The number of ACEs (ACE Score) was used to examine the
relationship of childhood stressors to the risk of COPD. Three methods of case ascertain-
ment were used to define COPD: baseline reports of prevalent COPD, incident hospital-
izations with COPD as a discharge diagnosis, and rates of prescription medications to treat
COPD during follow-up. Follow-up data were available through 2004.
Results: The ACE Score had a graded relationship to each of three measures of the occurrence of
COPD. Compared to people with an ACE Score of 0, those with an ACE Score of ⱖ5 had
2.6 times the risk of prevalent COPD, 2.0 times the risk of incident hospitalizations, and 1.6
times the rates of prescriptions (p⬍0.01 for all comparisons). These associations were only
modestly reduced by adjustment for smoking. The mean age at hospitalization decreased
as the ACE Score increased (p⬍0.01).
Conclusions: Decades after they occur, adverse childhood experiences increase the risk of COPD.
Because this increased risk is only partially mediated by cigarette smoking, other mecha-
nisms by which ACEs may contribute to the occurrence of COPD merit consideration.
(Am J Prev Med 2008;34(5):396 – 403) © 2008 American Journal of Preventive Medicine

Introduction involved in the pathogenesis of COPD include reduced

lung growth during childhood through young adult-

C
hronic obstructive pulmonary disease (COPD)
is a heterogeneous group of disorders classified
into three subtypes: asthmatic, bronchitic, and
emphysematous. In 2000, an estimated 10 million U.S.
adults reported physician-diagnosed COPD, of whom
approximately 726,000 were hospitalized.1 While smok-
ing is the primary risk factor for COPD,2 multiple
factors other than smoking play a role in COPD devel-
opment and progression,3– 6 including nutrition7 and
childhood exposures to respiratory infection.8 Pathways
From the ACE Study Group, National Center for Chronic Disease
Prevention and Health Promotion, CDC (Anda, Brown, Dube, Giles),
Atlanta, Georgia; the Departments of Psychiatry and Radiology and
Emory Center for Positron Emission Tomography, Emory University
School of Medicine (Bremner), Atlanta, Georgia; Atlanta VA Medical
Center (Bremner), Decatur, Georgia; and the Department of Preven-
tive Medicine, Southern California Permanente Medical Group (Fe-
litti), San Diego, California
Address correspondence and reprint requests to: Robert F. Anda,
MD, MS, CDC, National Center for Chronic Disease Prevention and
Health Promotion, Division of Adult and Community Health, 4770
Buford Highway NE, MS K-67, Atlanta GA 30341-3717. E-mail:
rfa1@cdc.gov.
hood, a premature decline in lung function when it
should be stable in young adulthood, and accelerated
decline in lung function after age 35.9 Improved un-
derstanding of childhood influences on the natural
history of lung function may lead to interventions to
prevent or slow the irreversible loss of lung function
during adulthood.10,11
Asthma was originally called asthma nervosa,12 yet
evidence remains scant for a causal link between trau-
matic stress during childhood and lung disease in
adults.13 Using retrospective cohort data from the first
half of the Adverse Childhood Experiences (ACE)
Study, this paper reported graded relationships be-
tween the number of categories of ACEs (ACE Score)
and early smoking initiation (by age 14),14 the preva-
lence of smoking in adults,14 and the prevalence of
self-reported chronic bronchitis or emphysema.15 The
relationship of the ACE Score to health-related out-
comes theoretically parallels the total exposure of the
developing central nervous system and other organ
systems to the activated stress response16; biologic plau-

396 Am J Prev Med 2008;34(5) 0749-3797/08/$–see front matter

© 2008 American Journal of Preventive Medicine • Published by Elsevier Inc. doi:10.1016/j.amepre.2008.02.002
sibility for this thesis is reinforced by data demonstrat-
ing a relationship of childhood abuse to differences in
brain structure and function, hypothalamic–pituitary–
adrenal (HPA) axis physiology, and autonomic nervous
system function.17,18
This paper assesses the relationship of the ACE Score
to the occurrence of COPD using prevalence and
prospective data from the ACE Study cohort.19 –35
Three methods of case ascertainment were used:
(1) prevalent COPD based on patient histories at
baseline; (2) incident hospitalizations from ICD-9 –
coded hospital discharge records that listed chronic
bronchitis, emphysema, or asthma as a discharge diag-
nosis; and (3) use of prescription medications for the
treatment of COPD during follow-up.
Methods
Study Population
14,29
The ACE Study has been described in detail elsewhere.
Members of the Kaiser Foundation Health Plan in San Diego
CA who attended its Health Appraisal Clinic (HAC) were
invited to participate. At the HAC they completed a standard-
ized evaluation that included an assessment of health history
and health-related behaviors, a clinical review of systems, and
psychosocial evaluations.14,15,29 The ACE Study was approved
by the IRB of Kaiser Permanente.
Each member who attended the HAC from August 1995 to
October 1997 was mailed an ACE Study questionnaire during
two separate survey waves that contained questions about
childhood exposure to abuse, neglect, domestic violence, and
forms of serious and interrelated household dysfunc-
tion.14,15,29 The second survey wave contained additional
questions. A total of 17,421 members (68%) responded; 84 of
them had incomplete information on race and educational
attainment, leaving an analytic sample of 17,337 persons.
Eligibility for the Prospective (Follow-Up) Phase of the
Study
Of the 17,337 participants included in prior analyses of the
baseline data, 708 (4.1%) were excluded from the prospective
phase of the study; either their HMO membership had lapsed
prior to their evaluation at the HAC or their member record
number was not considered valid. Prospective data included
that available through December 31, 2004.
Of the people who disenrolled and re-enrolled at least once
(median/mean: 1 time; range: 1– 6 times) during the fol-
low-up period, there were 1157 (6.7%) whose ratio of time
disenrolled/total possible time enrolled during follow-up
exceeded 20%; these people were also excluded, as they were
considered to have inadequate continuity of follow-up to
merit consideration for inclusion in the prospective analysis.
From the baseline sample, 15,472 people (89.2%) were
included in the follow-up analysis.
Relationship of the ACE Score to Exclusion from
Follow-Up
The potential contribution of ACEs to the exclusion from
follow-up due to lack of continuity in follow-up was assessed as
May 2008
a source of bias. A logistic model was used to include the
following: the ACE Score (0, 1, 2, 3, ⱖ4); age; gender; race;
and education, with exclusion from follow-up as the depen-
dent variable. The risks (ORs) of exclusion from follow-up for
people with 1, 2, 3, or ⱖ4 ACEs were 1.0 (0.9 –1.1); 1.0
(0.9 –1.2); 1.0 (0.9 –1.3); and 1.2 (1.0 –1.4), respectively. Thus,
ACEs had no substantial relationship to exclusion from
follow-up.
Adverse Childhood Experiences
Details of the ACE Study definitions, prevalence, and the
interrelatedness of ACEs have been published elsewhere.19
Briefly, all questions about ACEs referred to a respondent’s
first 18 years of life.
Questions used to define emotional and physical abuse and
growing up with domestic violence were adapted from the
conflict tactics scale (CTS)36 with the response categories of
never, once or twice, sometimes, often, or very often.
Emotional abuse. Two questions were used: 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? A respon-
dent was defined as being emotionally abused during child-
hood if the response was either often or very often to the first
question or sometimes, often, or very often to the second.
Physical abuse. Two questions were used: How often did a
parent, stepparent, or adult living in your home (1) push,
grab, slap, or throw something at you, or (2) hit you so hard
that you had marks or were injured? A respondent was
defined as being physically abused during childhood if the
response was either sometimes, often, or very often to the first
question, or if there was any response other than never to the
second question.
Sexual abuse. Questions used to assess contact sexual abuse
were adapted from Wyatt.37 Each respondent was asked
whether an adult, relative, family friend, or stranger who
was at least 5 years older than the respondent had ever
(1) touched or fondled the respondent’s body in a sexual way;
(2) had the respondent touch his or her body in a sexual
way; (3) attempted to have any type of sexual intercourse
(oral, anal, or vaginal) with the respondent; or (4) actually
had any type of sexual intercourse (oral, anal, or vaginal) with
the respondent. Respondents were classified as sexually
abused during childhood if they responded affirmatively to
any of the four questions.
Domestic violence. Four questions from the CTS36 were used
to consider childhood exposure to domestic violence, all of
them preceded by the following statement: Sometimes phys-
ical blows occur between parents. While you were growing up
in your first 18 years of life, how often did your father (or
stepfather) or mother’s boyfriend do any of these things to
your mother (or stepmother): (1) push, grab, slap, or throw
something at her; (2) kick, bite, hit her with a fist, or hit her
with something hard; (3) repeatedly hit her for at least a few
minutes; or (4) threaten her with a knife or gun, or use a
knife or gun to hurt her? A positive indication for witnessed
domestic violence was a response of sometimes, often, or very
Am J Prev Med 2008;34(5) 397
often to either one of the first two questions, or any response
other than never to either the third or fourth question.
Household substance abuse. Two questions were used to
determine whether respondents lived with a problem drinker,
an alcoholic,38 or a street-drug user.
Mental illness in household. A respondent was defined as
being exposed to mental illness if anyone in the household
was depressed, mentally ill, or had attempted suicide during
the respondent’s childhood.
Parental separation or divorce. This adverse experience was
defined as an affirmative response to the question Were your
parents ever separated or divorced during your first 18 years?
Criminal household member. If anyone in the household
had gone to prison during the respondent’s childhood, the
respondent was defined as being exposed to a criminal
household member.
The ACE Score
The ACE Score ranges from 0 to 8, but because of smaller
sample sizes within high scores the highest category was
designated ⱖ5. The analyses were conducted with the
summed score as five dichotomous variables with 0 ACEs as
the referent. The statistical validity of the ACE Score has been
published elsewhere.19
Current smokers were defined as those who had smoked
ⱖ100 cigarettes during their lifetime and who were currently
smokers, while former smokers were those who had smoked
at least 100 cigarettes but who were not currently smoking.
People with a BMI (kg/m2) ⱖ35 were considered obese.
Pulmonary lung function data obtained by spirometry were
available for people from the second survey wave. Criteria
from the Global Initiative for Chronic Obstructive Lung
Disease (GOLD)39 were used to categorize people as having
normal, restricted, or severely restricted lung function.
Case Ascertainment of COPD
Three methods of case ascertainment were used to define
COPD: (1) prevalent COPD based on self-reports, (2) inci-
dent hospitalizations from ICD-9 – coded hospital-discharge
records during follow-up diagnosis, and (3) use of prescrip-
tion medications for the treatment of COPD during
follow-up.
Prevalent COPD at baseline. Prevalent COPD was defined by
an affirmative response to either of the following baseline survey
questions: Have you ever been told that you have chronic
bronchitis or emphysema? or Do you currently have asthma?
Because information on current asthma was asked only during
survey Wave 2, analyses of prevalent COPD are limited to Wave-2
participants who were eligible for follow-up (N⫽7801).
Hospitalizations for COPD. People who were hospitalized
during follow-up and identified with ICD-9-CM40 codes 491
(chronic bronchitis), 492 (emphysema), 493 (asthma), and
496 (chronic airway obstruction, not elsewhere classified) in
the discharge record were considered to have incident hos-
pitalizations for COPD.
Prescriptions for bronchodilators and other medications
for COPD. Prescriptions for the bronchodilator medications
used for treatment of COPD were identified. The following
classes of medications were included: inhaled corticosteroids
398 American Journal of Preventive Medicine, Volume 34, Number 5
(e.g., flunisolide); adrendergic bronchodilators (e.g., ter-
butaline); theophylline bronchodilators (theophyline); anti-
cholinergic bronchodilators (e.g., ipratropium); nonsteroidal
anti-allergics (e.g., cromolyn); and leukotriene modifiers
(e.g., montelukast). Prescription medication data first be-
came available January 1, 1997; prescription rates were calcu-
lated from that date through December 31, 2004.
Data Analysis
All analyses were completed using SAS, version 8, along with
the 2000 U.S. Standard Population for direct age-standardiza-
tion of prevalences and risks. Logistic regression was used to
obtain the relative odds of prevalent COPD across ACE Score
groups. Cox proportional hazard models were used to esti-
mate the relative risk of hospitalization for COPD during
follow-up. Finally, negative binomial regression was used to
estimate relative rates of prescription-bronchodilator use dur-
ing follow-up. The natural logarithm of follow-up time be-
tween January 1, 1997, and December 31, 2004, was used as
the offset.
Variables included in the multivariate models included age
at baseline; gender; race (white, nonwhite); education
(⬍high school, high school graduate, some college, college
graduate); smoking status (never, former, current); obesity
(yes, no); and diabetes (yes, no). All statistical inferences were
based on a significance level of ␣ (2-sided)⫽0.05.
Results
Characteristics of Study Population
The study population included 8355 women (54%) and
7117 men (46%). The mean age (SD) was 56 (15)
years. Seventy-six percent of participants were white,
12% Hispanic, 4% black, 7% Asian, ⬍1% Native Amer-
ican, and 2% other. Forty percent were college gradu-
ates, 36% had some college education, and 17% were
high school graduates; only 7% had not graduated
from high school. Half (51%) of the participants were
never smokers, 8% were current smokers, and 41%
were former smokers. The prevalence of current
asthma at baseline was 4.6% (n⫽359), while that of
chronic bronchitis/emphysema was 4.9% (n⫽378).
The prevalence of each of the eight individual ACEs
were as follows: emotional abuse, 10%; physical abuse,
28%; sexual abuse, 21%; witnessed domestic violence,
13%; household substance abuse, 27%; mental illness
in the home, 19%; parental separation or divorce, 23%;
and criminal household member, 5%. The prevalence
of ACE Scores of 0, 1, 2, 3, 4, and ⱖ5 were 36.4%,
26.2%, 15.9%, 9.3%, 6.1%, and 6.1%, respectively.
The prevalence of current asthma at baseline was
inversely related to age (⬍40: 5.4%, OR⫽1.0[ref];
40 –54: 3.9%, OR⫽0.7 [0.5–1.0]; 55– 64: 3.7%, OR⫽0.7
[0.5–1.0]; 65–74: 2.5%, OR⫽0.5 [0.3– 0.7]; ⱖ75: 2.4%,
OR⫽0.5 [0.3– 0.9]), while chronic bronchitis/emphy-
sema at baseline increased with increasing age (⬍40:
2.9%, OR⫽1.0 [ref]; 40 –54: 3.0%, OR⫽1.1 [0.7–1.6];
55– 64: 3.8%, OR⫽1.3 [0.9 –2.0]; 65–74: 4.5%, OR⫽1.6
www.ajpm-online.net
Table 1. Proportion of adults identified with COPD using two different case ascertainment methods by selected participant
characteristics: Adverse Childhood Experiences (ACE) Study
Method of case ascertainment
COPD identified during follow-up by
COPD identified by self-report hospital discharge records
n (%) ORa (95% CI) n (%) Hazard ratioa (95% CI)
Age (years)
⬍40 104 (8.8) 1.0 (ref) 35 (1.5) 1.0 (ref)
40–54 183 (7.3) 0.8 (0.6–1.1) 91 (1.9) 1.0 (0.7–1.5)
55–64 143 (8.5) 1.0 (0.4–1.3) 153 (4.5) 2.3 (1.6–3.3)
65–74 132 (8.2) 0.9 (0.7–1.2) 283 (8.4) 4.2 (2.9–5.9)
ⱖ75 101 (12.2) 1.4 (1.1–1.9) 179 (10.6) 6.0 (4.2–8.6)
Smoking status
Never 297 (7.5) 1.0 (ref) 231 (2.9) 1.0 (ref)
Former 296 (9.2) 1.3 (1.1–1.5) 385 (6.1) 1.8 (1.5–2.1)
Current 70 (11.3) 1.7 (1.3–2.2) 125 (9.7) 4.9 (3.9–6.1)
Pulmonary functionb
At risk 443 (6.6) 1.0 (ref) 166 (2.5) 1.0 (ref)
Mild 92 (15.3) 2.6 (2.0–3.3) 58 (9.6) 4.0 (2.9–5.4)
Moderate or severe 112 (37.3) 8.4 (6.5–10.9) 94 (31.3) 11.4 (8.8–14.7)
Self-reported diabetes mellitus (ever)
No 624 (8.4) 1.0 (ref) 682 (4.6) 1.0 (ref)
Yes 39 (11.0) 1.3 (0.9–1.8) 59 (8.4) 1.5 (1.2–2.0)
Obese (BMI>35 kg/m2)
No 583 (8.1) 1.0 (ref) 650 (4.5) 1.0 (ref)
Yes 80 (12.7) 1.7 (1.3–2.2) 91 (7.9) 2.3 (1.9–2.9)
a
Estimated ORs, hazard ratios, and 95% CIs adjusted for age
b
Analyses completed among participants from Wave 2 only of the ACE Study (N⫽7801)
ACE, adverse childhood experiences; COPD, chronic obstructive pulmonary disease.

[1.1–2.4]; ⱖ75: 7.3%, OR⫽2.6 [1.7– 4.0]). As a result,
when these two forms of case ascertainment were
combined, there was no consistent association of prev-
alent COPD to the age of the respondents (Table 1).
Hospitalizations were strongly associated with older
age at baseline (Table 1). Both self-reported prevalence
and incident hospitalizations for COPD were associated
with smoking status and with pulmonary function based
on the GOLD criteria39 measured by spirometry at
baseline (Table 1). The prevalence and risk of hospi-
talizations for COPD were slightly greater among peo-
ple with self-reported diabetes mellitus and substan-
tially increased among people who were severely obese
(BMIⱖ35kg/m2) (Table 1).
Prevalent COPD at Baseline and ACE Score
The age-adjusted prevalence and risk (adjusted OR) of
COPD at baseline increased in a graded fashion as the
ACE Score increased (Table 2). Adjustment for demo-
graphic factors (Table 2, Model A) and demographic
factors, smoking status, diabetes, and obesity (Table 2,
Model B) had little effect on the strength of the
relationship between the ACE Score and the risk of

Table 2. Association between the ACE score and COPD ascertained by self-report at baseline for 7801 adults by ACE score:
Adverse Childhood Experiences (ACE) Study
Multivariable-adjusted
Age-adjusted Age-adjusted Model Aa Model Bb
ACE score n % (SE) OR (95% CI) OR (95% CI) OR (95% CI)
0 2770 5.4 (0.6) 1.0 (ref) 1.0 (ref) 1.0 (ref)
1 2050 8.3 (0.8) 1.4 (1.1–1.8) 1.4 (1.1–1.8) 1.4 (1.1–1.8)
2 1250 9.7 (1.0) 1.6 (1.3–2.1) 1.6 (1.3–2.1) 1.5 (1.2–2.0)
3 727 9.9 (1.3) 1.8 (1.3–2.4) 1.7 (1.3–2.3) 1.7 (1.2–2.2)
4 485 13.4 (2.1) 2.0 (1.5–2.8) 1.9 (1.4–2.7) 1.8 (1.3–2.5)
5 or more 519 13.8 (1.7) 2.6 (1.9–3.5) 2.3 (1.7–3.2) 2.1 (1.6–2.9)
a
Model A adjusts for age, gender, race/ethnicity, and education.
b
Model B adjusts for Model A variables as well as diabetes, obesity, and smoking.
ACE, adverse childhood experiences; COPD, chronic obstructive pulmonary disease.

May 2008 Am J Prev Med 2008;34(5) 399

Table 3. Association between the ACE score and COPD identified during follow-up by hospital discharge records between
baseline and December 31, 2004, among 15,472 adults: ACE Study
Multivariable-adjusted
Person-time Number of Age-adjusted HR Model Aa Model Bb
ACE Score (years) hospitalizations (95% CI) HR (95% CI) HR (95% CI)
0 38,248 274 1.0 (ref) 1.0 (ref) 1.0 (ref)
1 27,212 175 1.0 (0.8–1.2) 1.0 (0.8–1.2) 1.0 (0.8–1.2)
2 16,337 123 1.3 (1.1–1.7) 1.3 (1.1–1.6) 1.2 (1.0–1.5)
3 9,205 82 1.8 (1.4–2.3) 1.7 (1.4–2.2) 1.5 (1.2–2.0)
4 6,046 42 1.5 (1.1–2.1) 1.5 (1.1–2.1) 1.2 (0.9–1.7)
5 or more 5,891 45 2.0 (1.4–2.7) 1.8 (1.3–2.5) 1.4 (1.0–2.0)
a
Model A adjusts for age, gender, race/ethnicity, and education.
b
Model B adjusts for Model A variables as well as diabetes, obesity, and smoking.
ACE, adverse childhood experiences; COPD, chronic obstructive pulmonary disease; HR, hazard ratio; RR, relative rate.

COPD. In each model, the risk of COPD for people
with an ACE Score of ⱖ5 was increased more than
two-fold compared to people with an ACE Score of 0.
Risk of Hospitalizations for COPD
and ACE Score
The age-adjusted relative risk (hazard ratio) of hospital-
ization for COPD also increased in a graded fashion as the
ACE Score increased (Table 3, Model A). Adjustment for
demographics alone (Table 3, Model A) had little effect
on the hazard ratios; adjustment for demographic factors,
smoking status, diabetes, and obesity (Table 3, Model B)
moderately decreased the strength of the association
between the ACE Score and the risk hospitalizations for
COPD. Similar estimates were observed following further
adjustment for a history of parental smoking during the
respondent’s childhood (a crude proxy for exposure to
secondhand smoke; data not shown). Relationships be-
tween the ACE Score and COPD were also observed both
for never smokers and current smokers (data not shown).
Assessment of Asthma and Chronic Bronchitis/
Emphysema as Separate Entities
To ensure that the results were not due primarily to a
strong relationship of the ACE Score to either asthma
or chronic bronchitis/emphysema, those two were as-
sessed separately. After adjustment for demographic
factors and smoking status, the ACE Score showed a
graded relationship to both prevalent asthma and
prevalent chronic bronchitis/emphysema when as-
sessed as separate entities (data not shown). Similarly,
the ACE Score had a graded relationship to the relative
risk (hazard ratio) of hospitalizations for asthma (ICD-9
code 493) and chronic bronchitis/emphysema (ICD-9
codes 491, 492, and 496; data not shown).
Relative Rates of Prescriptions for
Bronchodilators and ACE Score
Rates of prescriptions for medications used to prevent
and treat COPD increased substantially as the ACE
Score increased (Table 4). Adjustment for demo-
graphic factors (Table 4, Model A) and for demo-
graphic factors, smoking status, diabetes, and obesity
(Table 4, Model B) resulted in modest decreases in the
strength of this association.
Assessment of Confounding and Interaction by
Smoking Exposure
For each of the three case definitions, similar estimates
were observed following further adjustment for a his-
tory of parental smoking during the respondent’s child-
hood (a crude proxy for exposure to secondhand

Table 4. Association between the ACE Score and the rate of prescriptions for the treatment of COPD between January 1,
1997, and December 31, 2004: ACE Study
Multivariable-adjusted
Person-time Number of Age-adjusted RR Model Aa Model Bb
ACE Score (years) prescriptions (95% CI) RR (95% CI) RR (95% CI)
0 35,618 11,542 1.0 (ref) 1.0 (ref) 1.0 (ref)
1 25,295 9,414 1.3 (1.1–1.5) 1.3 (1.1–1.6) 1.3 (1.1–1.6)
2 15,206 5,379 1.3 (1.0–1.5) 1.2 (1.0–1.5) 1.2 (1.0–1.4)
3 8,624 3,204 1.4 (1.1–1.8) 1.3 (1.0–1.7) 1.3 (1.0–1.6)
4 5,648 2,541 1.7 (1.3–2.3) 1.7 (1.2–2.2) 1.5 (1.1–1.9)
5 or more 5,530 1,711 1.6 (1.2–2.1) 1.4 (1.1–2.0) 1.3 (1.0–1.8)
a
Model A adjusts for age, gender, race/ethnicity, and education.
b
Model B adjusts for Model A variables as well as diabetes, obesity, and smoking.
ACE, adverse childhood experiences; COPD, chronic obstructive pulmonary disease; RR, relative rate.

400 American Journal of Preventive Medicine, Volume 34, Number 5 www.ajpm-online.net


Figure 1. Age at hospitalization for obstructive lung disease
between baseline and December 31, 2004, by ACE Score:
Adverse Childhood Experiences (ACE) Study.
smoke; data not shown). In addition, the ACE Score
had a graded relationship (p⬍0.05) to each of the three
definitions of COPD for both never smokers and cur-
rent smokers (data not shown).
Age at Hospitalization for Asthma and Chronic
Bronchitis/Emphysema and ACEs
As the ACE Score increased, the mean age at the time
of hospitalization for COPD decreased (for trend
p⬍0.001; Figure 1); this pattern was also seen when
asthma and chronic bronchitis/emphysema were as-
sessed as separate diagnostic categories (data not
shown).
Discussion
This study is the first to exploring the relationship of
the ACE Score to a disease outcome (e.g., COPD) using
data from the prospective phase of the ACE Study. The
risk of self-reported prevalent disease, and, in the prospec-
tive data, of incident hospitalizations and the rates of
prescriptions to treat COPD increased in a consistently
graded fashion as the ACE Score increased. Notably,
the strength of the relationships between the ACE
Score and each of these measures of the occurrence of
COPD were only modestly reduced by adjustment for
smoking histories.
An important thesis of the ACE Study was that
childhood stressors would be associated with known
risk factors for disease, which, in turn, would be strong
mediators of any ACE– disease relationships uncovered.
The data presented herein suggest that this thesis is
incomplete. In fact, when adjustment is made for
smoking exposures in the multivariate models, it ap-
pears that smoking is not the primary mediator of the
ACE–COPD relationships. Thus the role of childhood
stressors in the pathophysiology of COPD merits seri-
ous consideration.
May 2008
Because of the secular decrease in the prevalence of
smoking that has occurred over the past several de-
cades,41 in the future the proportion of cases of COPD
that are attributable to smoking will also necessarily
decline. In fact, a 1993 study estimated that 17% of
COPD mortality in the U.S. population occurs in never
smokers.42 National data indicate that 20% of the
estimated 11.5 million adults with low lung-function
had never smoked.43 It will become increasingly impor-
tant to understand mechanisms other than smoking
that lead to COPD.
There are several mechanisms by which stress could
mediate an increase in the diagnosis of asthma and
related conditions. As noted above, ACEs (stressors)
are associated with an increase in smoking and alcohol
use that can independently increase asthma risk. ACEs
are also associated with risk factors for chronic disease
conditions such as obesity,44 diabetes,15 ischemic heart
disease,45 and liver disease30 that may result in an
increased risk of exacerbating underlying lung diseases
or negatively affect general health, leading to disease
progression, risk of hospitalization, or the need for
prescription medications.
The ACE Score has a graded relationship to both
depression46 and panic reactions.16 Mental disorders
may make it more likely that unidentified COPD may
become severe enough to require treatment, or mental
disorders and asthma may interact to exacerbate each
other.
Occurrences of asthma are more numerous among the
poor and in minority populations. This may be due to the
stresses of racism, increased exposure to violence, or
limited access to medical care. Indeed, a relationship
between exposure to violence and asthma has been found
in inner-city, impoverished minorities.47
Exposure to traumatic stress during childhood can
induce lasting changes in the central nervous system,
including increased activity of the HPA axis,48 which
may affect both a person’s lung development during
childhood and adolescence and the functioning of the
cardiorespiratory system throughout his or her lifespan.
Acute stress is associated with an increase in cortisol
that results in a suppression of the inflammatory re-
sponse.49 With chronic stressors (including childhood
abuse), however, there is dysregulation of the HPA
axis,48 which may lead to a decrease in cortisol and
increased inflammatory markers. Indeed, stressed
women with asthma were found to have lower cortisol
levels.50 Asthma has been hypothesized to be triggered
by inflammation51; increased inflammation with stress
may therefore lead to asthma. Although studies have
shown an increase in stress-induced asthma response,
they have not been able to directly link stress, inflam-
mation, and pulmonary dysfunction.
Stress is associated with an acute suppression of the
inflammatory response, which could increase suscepti-
bility to infection. Stress has been associated with an
Am J Prev Med 2008;34(5) 401
increased risk of colds and other upper respiratory
infections in a dose-dependent manner.51,52 Such in-
fections in childhood may cause tissue damage and
impair lung capacity, leading to an increased risk for
COPD later in life.
Maternal exposure to a hostile environment with its
resultant stress response may be transmitted to the
fetus, resulting in altered development of the fetal HPA
axis and immune system that might increase the risk for
COPD.52–55 Maternal cortisol affects the developing
fetal immune system by increasing placental-corticotro-
phin–releasing factor, which stimulates the production
of both maternal and fetal cortisol. Increased maternal
cortisol with stress causes a decrease in the TH1/ TH2
T-helper cell ratio that increases the risk for asthma and
atopy.56 Women with a high ACE Score are at increased
risk of having a miscarriage or stillbirth33— evidence
that the in utero environment is altered by ACEs.
Patients with a history of exposure to traumatic stress
and the diagnosis of post-traumatic stress disorder
(PTSD) have chronic elevations of the sympathetic
system.57– 60 Although stress, which results in sympa-
thetic activation, would paradoxically appear to result
in bronchial dilation, after the acute sympathetic acti-
vation it is possible that a parasympathetic rebound can
occur, causing bronchial constriction.61 Also, chronic
elevations in the sympathetic system could cause a
down-regulation of the beta-sympathetic system (which
occurs in chronic users of beta-agonists for the treat-
ment of asthma), decreasing the sensitivity of the
beta-adrenergic system and increasing vulnerability to
asthma.
The prevalence of childhood exposures that this
study reported is nearly identical to those reported in
surveys of the general population. It was found that
16% of the men and 25% of the women met the case
definition for childhood sexual abuse, similar to find-
ings by Finkelhor et al.62 that 16% of men and 27% of
women had been sexually abused. Twenty-eight percent
of the men in this study reported experiencing physical
abuse as boys, which closely parallels the percentage
(31%) found in a recent population-based study of
Ontario men that used questions from the same
scales.63 The similarity of the estimates from the ACE
Study to those of population-based studies suggests that
this study’s findings are likely to be applicable in other
settings.
For each of three methods of case ascertainment a
graded relationship was found between the ACE Score
and COPD. Prevalent COPD was based on patients’
self-reports of diagnosed disease. ICD-9 – coded hospi-
tal-discharge diagnoses necessarily require a physician
to record the diagnosis in the patient’s chart. Rates of
prescriptions for medications treating COPD were
based on the requirement that a physician write them
and a pharmacist fill them. The first method of case
ascertainment may be limited by validity of self-
402 American Journal of Preventive Medicine, Volume 34, Number 5
reported data, but the latter two methods do not suffer
from this potential limitation. Additionally, ICD-
9 – coded hospitalizations and prescriptions for the
treatments of COPD were assessed prospectively and
represent different measures of disease occurrence.
The ACE Score had a graded relationship to the
presence of COPD for prevalent disease at baseline, risk
of hospitalizations, and use of prescription medications
for treating COPD. The relationships of the ACE Score
to these three measures of the occurrence and burden
of COPD are only partially mediated by cigarette smok-
ing, suggesting other potential mechanisms by which
ACEs influence the development of COPD.
The Adverse Childhood Experiences Study was supported
under a cooperative agreement, #TS-44-10/11, from the CDC
through the Association of Teachers of Preventive Medicine
and a grant from the Garfield Memorial Fund.
JDB has provided expert testimony and received consulting
fees for PTSD cases. No other financial disclosures were
reported.
References
1. Mannino DM, Homa DM, Akinabi LJ, et al. Chronic obstructive pulmonary
disease surveillance—U.S., 1971–2000. MMWR CDC Surveill Summ
2002;51:1–16.
2. Silverman EK, Speizer FE. Risk factors for the development of chronic
obstructive pulmonary disease. Med Clin N Am 1996;80:501–22.
3. Sparrow D, O’Connor G, Weiss ST. The relation of airways responsiveness
and atopy to the development of chronic obstructive pulmonary disease.
Epidemiol Rev 1998;10:29 – 47.
4. Abbey DE, Burchette RJ, Knutson SF, et al. Long-term particulate and other
air pollutants and lung function in nonsmokers. Am J Respir Crit Care Med
1998;158:289 –98.
5. Chen Y, Daler R, Krewski D, et al. Increased effects of smoking and obesity
on asthma among female Canadians: the national population health
survey, 1994 –1995. Am J Epidemiol 1999;150:255– 62.
6. Bakke PS, Hanoa R, Gulsvik A. Educational level and obstructive lung
disease given smoking habits and occupational airborne exposure: a
Norwegian community study. Am J Epidemiol 1995;141:1080 – 8.
7. Hu G, Cassano PA. Antioxidant nutrients and pulmonary function: the
Third National Health and Nutrition Examination Survey (NHANES III).
Am J Epidemiol 2000;151:975– 81.
8. Britton J, Martinez FD. The relationship of childhood respiratory infection
to growth and decline in lung function. Am J Respir Crit Care Med
1996;154(6 Pt 2):S240 –S245.
9. O’Byrne PM, Postma DS. The many faces of airway inflammation: asthma
and chronic obstructive pulmonary disease: Asthma Research Group. Am J
Respir Crit Care Med 1999;159(5 Pt 2):S41–S63.
10. Edwards CA, Osman LM, Godden DJ, Douglas JG. Wheezy bronchitis in
childhood: a distinct clinical entity with lifelong significance? Chest
2003;124:18 –24.
11. von Mutius E. Childhood experiences take away your breath as a young
adult. Am J Respir Crit Care Med 2002;165:1467– 8.
12. Wright RJ, Rodriguez M, Cohen S. Review of psychosocial stress and
asthma: an integrated biopsychosocial approach. Thorax 1998;53:1066 –74.
13. Goodwin RD, Wamboldt MZ, Pine DS. Lung disease and internalizing
disorders. Is childhood abuse a shared etiologic factor? J Psychosom Res
2003;55:215–9.
14. Anda RF, Croft JB, Felitti VJ, et al. Adverse childhood experiences and
smoking during adolescence and adulthood. JAMA 1999;282:1652– 8.
15. Felitti VJ, Anda RF, Nordenberg D, et al. The relationship of childhood
abuse and household dysfunction to many of the leading causes of death in
adults. Am J Prev Med 1998;14:245–58.
16. Anda RF, Felitti VJ, Bremner JD, et al. The enduring effects of abuse and
related adverse experiences in childhood: a convergence of evidence from
www.ajpm-online.net
 neurobiology and epidemiology. Eur Arch Psychiatry Clin Neurosci
2006;256:174 – 86.
17. Perry BD, Pollard R. Homeostasis, stress, trauma, and adaptation. A
neurodevelopmental view of childhood trauma. Child Adolesc Psychiatr
Clin N Am 1998;7:33–51,viii.
18. Teicher MH, Andersen SL, Polcari A, et al. Developmental neurobiology of
childhood stress and trauma. Psychiatr Clin North Am 2002;25:397–
426,vii–viii.
19. Dong M, Anda RF, Felitti VJ, et al. The interrelatedness of multiple forms
of childhood abuse, neglect, and household dysfunction. Child Abuse Negl
2004;28:771– 84.
20. Dube SR, Anda RF, Felitti VJ, Chapman D, Williamson DF, Giles WH.
Childhood abuse, household dysfunction, and the risk of attempted suicide
throughout the life span: findings from the adverse childhood experiences
study. JAMA 2001;286:3089 –96.
21. Dietz PM, Spitz AM, Anda RF, et al. Unintended pregnancy among adult
women exposed to abuse or household dysfunction during their child-
hood. JAMA 1999;282:1359 – 64.
22. Hillis SD, Anda RF, Felitti VJ, Nordenberg D, Marchbanks PA. Adverse
childhood experiences and sexually transmitted diseases in men and
women: a retrospective study. Pediatrics 2000;106:E11.
23. Hillis SD, Anda RF, Felitti VJ, Marchbanks PA. Adverse childhood experi-
ences and sexual risk behaviors in women: a retrospective cohort study.
Fam Plann Perspect 2001;33:206 –11.
24. Anda RF, Felitti VJ, Chapman DP, et al. Abused boys, battered mothers, and
male involvement in teen pregnancy. Pediatrics 2001;107:e19.
25. Dube SR, Anda RF, Felitti VJ, Edwards VJ, Croft JB. Adverse childhood
experiences and personal alcohol abuse as an adult. Addict Behav 2002;
27:713–25.
26. Anda RF, Chapman DP, Felitti VJ, et al. Adverse childhood experiences and
risk of paternity in teen pregnancy. Obstet Gynecol 2002;100:37– 45.
27. Anda RF, Whitfield CL, Felitti VJ, et al. Adverse childhood experiences,
alcoholic parents and later risk of alcoholism and depression. Psychiatr
Serv 2002;53:1001–9.
28. Dube SR, Felitti VJ, Chapman DP, Giles WH, Anda RF. Childhood abuse,
neglect, and household dysfunction and the risk of illicit drug use: the
adverse childhood experience study. Pediatrics 2003;111:564 –72.
29. Dube SR, Felitti VJ, Dong M, Giles WH, Anda RF. The impact of adverse
childhood experiences on health problems: evidence from four birth
cohorts dating back to 1900. Prev Med 2003;37:268 –77.
30. Dong M, Dube SR, Felitti VJ, Giles WH, Anda RF. Adverse childhood
experiences and self-reported liver disease: new insights into a causal
pathway. Arch Intern Med 2003;163:1949 –56.
31. Edwards VJ, Holden GW, Anda RF, Felitti VJ. Relationship between
multiple forms of childhood maltreatment and adult mental health in
community respondents: results from the adverse childhood experiences
(ACE) study. Am J Psychiatry 2003;160:53– 60.
32. Edwards VJ, Anda RF, Felitti VJ, Dube SR. Adverse childhood experiences
and health-related quality of life as an adult. In: Kendall-Tackett, Ed.
Victimization and health. Washington DC: American Psychological Associ-
ation, 2003.
33. Hillis SD, Anda RF, Dube SR, Felitti VJ, Marchbanks PA, Marks JS. The
association between adolescent pregnancy, long-term psychosocial out-
comes, and fetal death. Pediatrics 2004;113:320 –7.
34. Chapman DP, Whitfield CL, Felitti VJ, et al. Adverse childhood experiences
and the risk of depressive disorders in adulthood. J Affect Disord 2004;82:
217–25.
35. Whitfield CL, Anda RF, Dube SR, Felitti VJ. Violent childhood experiences
and the risk of intimate partner violence in adults: assessment in a large
health maintenance organization. J Interpers Violence 2003;18:166 – 85.
36. Straus M, Gelles RJ. Physical violence in American families: risk factors and
adaptations to violence in 8,145 families. New Brunswick NJ: Transaction
Press, 1990.
37. Wyatt GA. The sexual abuse of Afro-American and white American women
in childhood. Child Abuse Negl 1985;9:507–19.
38. Schoenborn CA. Exposure to alcoholism in the family: U.S., 1988. Adv Data
1991;205:1–13.
39. Pauwels RA, Buist AS, Calverley PM, Jenkins CR, Hurd SS. Global strategy
for the diagnosis, management, and prevention of chronic obstructive
pulmonary disease. NHLBI/WHO global initiative for chronic obstructive
May 2008
lung disease (GOLD) workshop summary. Am J Respir Crit Care Med
2001;163:1256 –76.
40. ICD-9-CM: International classification of diseases, 9th revision; clinical
modification. U.S. DHHS, Public Health Service, Health Care Financing
Administration. DHHS Publication No. (PHS) 80-1260.
41. U.S. DHHS. Reducing tobacco use: a report of the surgeon general. Atlanta
GA: U.S. DHHS, CDC, National Center for Chronic Disease Prevention and
Health Promotion, Office of Smoking and Health, 2000.
42. Meyer PA, Mannino DM, Redd SC, et al. Characteristics of adults dying with
chronic obstructive pulmonary disease. Chest 2002;122:2003– 8.
43. Mannino DM, Gagnon RC, Petty TL, et al. Obstructive lung disease and low
lung function in adults in the U.S.: data from the National Health and
Nutrition Examination Survey. Arch Intern Med 2000;160:1683–9.
44. Williamson DF, Thompson TJ, Anda RF, Dietz WH, Felitti VJ. Body weight,
obesity, and self-reported abuse in childhood. Int J Obes 2002;26:1075– 82.
45. Dong M, Giles WH, Felitti VJ, et al. Insights into causal pathways for
ischemic heart disease: the adverse childhood experiences study. Circula-
tion 2004;110:1761– 6.
46. Chapman DP, Anda RF, Felitti VJ, Dube SR, Edwards VJ, Whitfield CL.
Epidemiology of adverse childhood experiences and depressive disorders
in a large health maintenance organization population. J Affect Disord
2004;82:217–25.
47. Wright RJ, Hanrahan JP, Tager I, Speizer FE. Effect of the exposure to
violence on the occurrence and severity of childhood asthma in an
inner-city population. Am J Respir Crit Care Med 1997;155:A972.
48. Bremner JD, Vermetten E. Stress and development: behavioral and biolog-
ical consequences. Developmental Psychopathology 2001;13:473– 89.
49. Forsythe P, Ebeling C, Gordon JR, Befus AD, Vliagoftis H. Opposing effects
of short- and long-term stress on airway inflammation. Am J Respir Crit
Care Med 2004;169:220 – 6.
50. Laube BL, Curbow BA, Costello RW, Fitzgerald ST. A pilot study examining
the relationship between stress and serum cortisol concentrations in
women with asthma. Respir Med 2002; 96:823– 8.
51. Cohen S. Psychological stress, immunity, and upper respiratory infections.
Curr Dir Psychol Sci 1996;5:86 –90.
52. Cohen S, Tyrrell DA, Smith AP. Psychological stress and susceptibility to the
common cold. N Engl J Med 1991;325:606 –12.
53. Owen D, Andrews MH, Matthews SG. Maternal adversity, glucocorticoids
and programming of neuroendocrine function and behaviour. Neurosci
Biobehav Rev 2005;29:202–26.
54. Wadhwa PD, Dunkel-Schetter C, Chicz-DeMet A, Porto M, Sandman CA.
Prenatal psychosocial factors and the neuroendocrine axis in human
pregnancy. Psychosom Med 1996;58:432– 46.
55. Avishai-Eliner S, Brunson KL, Sandman CA, Baram TZ. Stressed-out, or in
(utero)? Trends Neurosci 2002;25:518 –24.
56. von Hertzen LC. Maternal stress and T-cell differentiation of the develop-
ing immune system: possible implications for the development of asthma
and atopy. J Allergy Clin Immunol 2002;109:923– 8.
57. Bremner JD, Innis RB, Ng CK, et al. PET measurement of cerebral
metabolic correlates of yohimbine administration in posttraumatic stress
disorder. Arch Gen Psychiatry 1997;54:246 –56.
58. Bremner JD, Krystal JH, Southwick SM, Charney DS. Noradrenergic
mechanisms in stress and anxiety: II. Clinical studies. Synapse 1996;
23:39 –51.
59. Pitman RK, Orr SP, Forgus DF, de Jong JB, Claiborn JM. Psychophysiologic
assessment of posttraumatic stress disorder imagery in Vietnam combat
veterans. Arch Gen Psychiatry 1987;44:970 –5.
60. Southwick SM, Krystal JH, Morgan, CA, et al. Abnormal noradrenergic
function in posttraumatic stress disorder. Arch Gen Psychiatry 1993;
50:295–305.
61. Lehrer P. Emotionally triggered asthma: a review of research literature and
some hypotheses for self-regulating therapies. Appl Psychophysiol Biofeed-
back 1998;23:13– 41.
62. Finkelhor D, Hotaling G, Lewis IA, Smith C. Sexual abuse in a national
survey of adult men and women: prevalence, characteristics, and risk
factors. Child Abuse Negl 1990;14:19 –28.
63. MacMillan HL, Fleming JE, Trocme N, et al. Prevalence of child physical
and sexual abuse in the community: results from the Ontario Health
Supplement. JAMA 1997;278:131–5.
Am J Prev Med 2008;34(5) 403