Professional Documents
Culture Documents
ACE y EPOC
ACE y EPOC
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
C
hronic obstructive pulmonary disease (COPD)
hood, a premature decline in lung function when it
is a heterogeneous group of disorders classified
should be stable in young adulthood, and accelerated
into three subtypes: asthmatic, bronchitic, and
decline in lung function after age 35.9 Improved un-
emphysematous. In 2000, an estimated 10 million U.S.
derstanding of childhood influences on the natural
adults reported physician-diagnosed COPD, of whom
history of lung function may lead to interventions to
approximately 726,000 were hospitalized.1 While smok-
prevent or slow the irreversible loss of lung function
ing is the primary risk factor for COPD,2 multiple
during adulthood.10,11
factors other than smoking play a role in COPD devel-
Asthma was originally called asthma nervosa,12 yet
opment and progression,3– 6 including nutrition7 and
evidence remains scant for a causal link between trau-
childhood exposures to respiratory infection.8 Pathways
matic stress during childhood and lung disease in
adults.13 Using retrospective cohort data from the first
From the ACE Study Group, National Center for Chronic Disease half of the Adverse Childhood Experiences (ACE)
Prevention and Health Promotion, CDC (Anda, Brown, Dube, Giles), Study, this paper reported graded relationships be-
Atlanta, Georgia; the Departments of Psychiatry and Radiology and
Emory Center for Positron Emission Tomography, Emory University tween the number of categories of ACEs (ACE Score)
School of Medicine (Bremner), Atlanta, Georgia; Atlanta VA Medical and early smoking initiation (by age 14),14 the preva-
Center (Bremner), Decatur, Georgia; and the Department of Preven- lence of smoking in adults,14 and the prevalence of
tive Medicine, Southern California Permanente Medical Group (Fe-
litti), San Diego, California self-reported chronic bronchitis or emphysema.15 The
Address correspondence and reprint requests to: Robert F. Anda, relationship of the ACE Score to health-related out-
MD, MS, CDC, National Center for Chronic Disease Prevention and comes theoretically parallels the total exposure of the
Health Promotion, Division of Adult and Community Health, 4770
Buford Highway NE, MS K-67, Atlanta GA 30341-3717. E-mail: developing central nervous system and other organ
rfa1@cdc.gov. systems to the activated stress response16; biologic plau-
Study Population Emotional abuse. Two questions were used: How often did a
14,29 parent, stepparent, or adult living in your home swear at you,
The ACE Study has been described in detail elsewhere.
insult you, or put you down? and How often did a parent,
Members of the Kaiser Foundation Health Plan in San Diego
stepparent, or adult living in your home act in a way that
CA who attended its Health Appraisal Clinic (HAC) were
made you afraid that you might be physically hurt? A respon-
invited to participate. At the HAC they completed a standard-
dent was defined as being emotionally abused during child-
ized evaluation that included an assessment of health history
hood if the response was either often or very often to the first
and health-related behaviors, a clinical review of systems, and
question or sometimes, often, or very often to the second.
psychosocial evaluations.14,15,29 The ACE Study was approved
by the IRB of Kaiser Permanente. Physical abuse. Two questions were used: How often did a
Each member who attended the HAC from August 1995 to parent, stepparent, or adult living in your home (1) push,
October 1997 was mailed an ACE Study questionnaire during grab, slap, or throw something at you, or (2) hit you so hard
two separate survey waves that contained questions about that you had marks or were injured? A respondent was
childhood exposure to abuse, neglect, domestic violence, and defined as being physically abused during childhood if the
forms of serious and interrelated household dysfunc- response was either sometimes, often, or very often to the first
tion.14,15,29 The second survey wave contained additional question, or if there was any response other than never to the
questions. A total of 17,421 members (68%) responded; 84 of second question.
them had incomplete information on race and educational
attainment, leaving an analytic sample of 17,337 persons. Sexual abuse. Questions used to assess contact sexual abuse
were adapted from Wyatt.37 Each respondent was asked
Eligibility for the Prospective (Follow-Up) Phase of the whether an adult, relative, family friend, or stranger who
Study was at least 5 years older than the respondent had ever
(1) touched or fondled the respondent’s body in a sexual way;
Of the 17,337 participants included in prior analyses of the
(2) had the respondent touch his or her body in a sexual
baseline data, 708 (4.1%) were excluded from the prospective
way; (3) attempted to have any type of sexual intercourse
phase of the study; either their HMO membership had lapsed
(oral, anal, or vaginal) with the respondent; or (4) actually
prior to their evaluation at the HAC or their member record
had any type of sexual intercourse (oral, anal, or vaginal) with
number was not considered valid. Prospective data included
that available through December 31, 2004. the respondent. Respondents were classified as sexually
Of the people who disenrolled and re-enrolled at least once abused during childhood if they responded affirmatively to
(median/mean: 1 time; range: 1– 6 times) during the fol- any of the four questions.
low-up period, there were 1157 (6.7%) whose ratio of time Domestic violence. Four questions from the CTS36 were used
disenrolled/total possible time enrolled during follow-up to consider childhood exposure to domestic violence, all of
exceeded 20%; these people were also excluded, as they were them preceded by the following statement: Sometimes phys-
considered to have inadequate continuity of follow-up to ical blows occur between parents. While you were growing up
merit consideration for inclusion in the prospective analysis. in your first 18 years of life, how often did your father (or
From the baseline sample, 15,472 people (89.2%) were stepfather) or mother’s boyfriend do any of these things to
included in the follow-up analysis. your mother (or stepmother): (1) push, grab, slap, or throw
something at her; (2) kick, bite, hit her with a fist, or hit her
Relationship of the ACE Score to Exclusion from
with something hard; (3) repeatedly hit her for at least a few
Follow-Up
minutes; or (4) threaten her with a knife or gun, or use a
The potential contribution of ACEs to the exclusion from knife or gun to hurt her? A positive indication for witnessed
follow-up due to lack of continuity in follow-up was assessed as domestic violence was a response of sometimes, often, or very
[1.1–2.4]; ⱖ75: 7.3%, OR⫽2.6 [1.7– 4.0]). As a result, tially increased among people who were severely obese
when these two forms of case ascertainment were (BMIⱖ35kg/m2) (Table 1).
combined, there was no consistent association of prev-
alent COPD to the age of the respondents (Table 1).
Prevalent COPD at Baseline and ACE Score
Hospitalizations were strongly associated with older
age at baseline (Table 1). Both self-reported prevalence The age-adjusted prevalence and risk (adjusted OR) of
and incident hospitalizations for COPD were associated COPD at baseline increased in a graded fashion as the
with smoking status and with pulmonary function based ACE Score increased (Table 2). Adjustment for demo-
on the GOLD criteria39 measured by spirometry at graphic factors (Table 2, Model A) and demographic
baseline (Table 1). The prevalence and risk of hospi- factors, smoking status, diabetes, and obesity (Table 2,
talizations for COPD were slightly greater among peo- Model B) had little effect on the strength of the
ple with self-reported diabetes mellitus and substan- 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.
COPD. In each model, the risk of COPD for people sessed separately. After adjustment for demographic
with an ACE Score of ⱖ5 was increased more than factors and smoking status, the ACE Score showed a
two-fold compared to people with an ACE Score of 0. graded relationship to both prevalent asthma and
prevalent chronic bronchitis/emphysema when as-
Risk of Hospitalizations for COPD sessed as separate entities (data not shown). Similarly,
and ACE Score the ACE Score had a graded relationship to the relative
risk (hazard ratio) of hospitalizations for asthma (ICD-9
The age-adjusted relative risk (hazard ratio) of hospital- code 493) and chronic bronchitis/emphysema (ICD-9
ization for COPD also increased in a graded fashion as the codes 491, 492, and 496; data not shown).
ACE Score increased (Table 3, Model A). Adjustment for
demographics alone (Table 3, Model A) had little effect
Relative Rates of Prescriptions for
on the hazard ratios; adjustment for demographic factors,
Bronchodilators and ACE Score
smoking status, diabetes, and obesity (Table 3, Model B)
moderately decreased the strength of the association Rates of prescriptions for medications used to prevent
between the ACE Score and the risk hospitalizations for and treat COPD increased substantially as the ACE
COPD. Similar estimates were observed following further Score increased (Table 4). Adjustment for demo-
adjustment for a history of parental smoking during the graphic factors (Table 4, Model A) and for demo-
respondent’s childhood (a crude proxy for exposure to graphic factors, smoking status, diabetes, and obesity
secondhand smoke; data not shown). Relationships be- (Table 4, Model B) resulted in modest decreases in the
tween the ACE Score and COPD were also observed both strength of this association.
for never smokers and current smokers (data not shown).
Assessment of Confounding and Interaction by
Assessment of Asthma and Chronic Bronchitis/ Smoking Exposure
Emphysema as Separate Entities
For each of the three case definitions, similar estimates
To ensure that the results were not due primarily to a were observed following further adjustment for a his-
strong relationship of the ACE Score to either asthma tory of parental smoking during the respondent’s child-
or chronic bronchitis/emphysema, those two were as- 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.