Adverse Childhood Experiences and Chronic Obstructive Pulmonary Disease in Adults

Abstract

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 Experiences (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 ascertainment were used to define COPD: baseline reports of prevalent COPD, incident hospitalizations 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 ≥5 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 mechanisms by which ACEs may contribute to the occurrence of COPD merit consideration.

Introduction

Chronic 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 smoking is the primary risk factor for COPD,² multiple factors other than smoking play a role in COPD development and progression,^{3– 6} including nutrition⁷ and childhood exposures to respiratory infection.⁸ Pathways involved in the pathogenesis of COPD include reduced lung growth during childhood through young adulthood, 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 understanding 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 traumatic 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 between the number of categories of ACEs (ACE Score) and early smoking initiation (by age 14),¹⁴ the prevalence 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 outcomes theoretically parallels the total exposure of the developing central nervous system and other organ systems to the activated stress response¹⁶; biologic plausibility for this thesis is reinforced by data demonstrating 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 diagnosis; and (3) use of prescription medications for the treatment of COPD during follow-up.

Methods

Study Population

The ACE Study has been described in detail elsewhere.^14,29 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 standardized 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 dysfunction.^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 follow-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 a source of bias. A logistic model was used to include the following: the ACE Score (0, 1, 2, 3, ≥4); age; gender; race; and education, with exclusion from follow-up as the dependent variable. The risks (ORs) of exclusion from follow-up for people with 1, 2, 3, or ≥4 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 respondent was defined as being emotionally abused during childhood 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 physical 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 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 ≥5. 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 ≥100 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) ≥35 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) incident hospitalizations from ICD-9–coded hospital-discharge records during follow-up diagnosis, and (3) use of prescription 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 hospitalizations 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 (e.g., flunisolide); adrendergic bronchodilators (e.g., terbutaline); theophylline bronchodilators (theophyline); anticholinergic bronchodilators (e.g., ipratropium); nonsteroidal anti-allergics (e.g., cromolyn); and leukotriene modifiers (e.g., montelukast). Prescription medication data first became available January 1, 1997; prescription rates were calculated 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-standardization 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 estimate the relative risk of hospitalization for COPD during follow-up. Finally, negative binomial regression was used to estimate relative rates of prescription-bronchodilator use during follow-up. The natural logarithm of follow-up time between 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 α (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 American, and 2% other. Forty percent were college graduates, 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 ≥5 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]; ≥75: 2.4%, OR=0.5 [0.3–0.9]), while chronic bronchitis/emphysema 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 [1.1–2.4]; ≥75: 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 prevalent COPD to the age of the respondents (Table 1).

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 by self-report		COPD identified during follow-up by 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)
≥75	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)

^aEstimated ORs, hazard ratios, and 95% CIs adjusted for age

^bAnalyses completed among participants from Wave 2 only of the ACE Study (N=7801)

ACE, adverse childhood experiences; COPD, chronic obstructive pulmonary disease.

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 hospitalizations for COPD were slightly greater among people with self-reported diabetes mellitus and substantially increased among people who were severely obese (BMI≥35kg/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 demographic 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 COPD. In each model, the risk of COPD for people with an ACE Score of ≥5 was increased more than two-fold compared to people with an ACE Score of 0.

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
ACE score	n	Age-adjusted % (SE)	Age-adjusted OR (95% CI)	Model Aa OR (95% CI)	Model Bb 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)

^aModel A adjusts for age, gender, race/ethnicity, and education.

^bModel B adjusts for Model A variables as well as diabetes, obesity, and smoking.

ACE, adverse childhood experiences; COPD, chronic obstructive pulmonary disease.

Risk of Hospitalizations for COPD and ACE Score

The age-adjusted relative risk (hazard ratio) of hospitalization 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 between the ACE Score and COPD were also observed both for never smokers and current smokers (data not shown).

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
ACE Score	Person-time (years)	Number of hospitalizations	Age-adjusted HR (95% CI)	Model Aa HR (95% CI)	Model Bb 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)

^aModel A adjusts for age, gender, race/ethnicity, and education.

^bModel 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.

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 assessed 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 assessed 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 demographic factors (Table 4, Model A) and for demographic factors, smoking status, diabetes, and obesity (Table 4, Model B) resulted in modest decreases in the strength of this association.

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
ACE Score	Person-time (years)	Number of prescriptions	Age-adjusted RR (95% CI)	Model Aa RR (95% CI)	Model Bb 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)

^aModel A adjusts for age, gender, race/ethnicity, and education.

^bModel 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.

Assessment of Confounding and Interaction by Smoking Exposure

For each of the three case definitions, 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). 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 current 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 assessed as separate diagnostic categories (data not shown).

Figure 1.

Age at hospitalization for obstructive lung disease between baseline and December 31, 2004, by ACE Score: Adverse Childhood Experiences (ACE) Study.

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 prospective 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 appears that smoking is not the primary mediator of the ACE–COPD relationships. Thus the role of childhood stressors in the pathophysiology of COPD merits serious consideration.

Because of the secular decrease in the prevalence of smoking that has occurred over the past several decades,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 important 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 response.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, inflammation, and pulmonary dysfunction.

Stress is associated with an acute suppression of the inflammatory response, which could increase susceptibility to infection. Stress has been associated with an increased risk of colds and other upper respiratory infections in a dose-dependent manner.^51,52 Such infections 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-corticotrophin–releasing factor, which stimulates the production of both maternal and fetal cortisol. Increased maternal cortisol with stress causes a decrease in the T_H1/ T_H2 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 sympathetic activation, would paradoxically appear to result in bronchial dilation, after the acute sympathetic activation 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 treatment 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 findings 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 hospital-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-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 smoking, suggesting other potential mechanisms by which ACEs influence the development of COPD.