Mild and Moderate-to-Severe COPD in Nonsmokers* Distinct Demographic Profiles Carolyn E. Behrendt, PhD Study objective: To investigate the risk of COPD among nonsmokers. Design: Case-control study, logistic regression analysis. Setting: Third National Health and Nutrition Examination Survey, from 1988 to 1994. Participants: Community residents 18 to 80 years of age, of white, black, or Mexican-American ethnicity. Nonsmokers included never-smokers and former smokers with a < 5 pack-year smoking history who had never smoked cigars or pipes. Measurements: COPD (FEV 1 /FVC < 70%) was classified as mild (FEV 1 > 80% predicted) or moderate to severe (FEV 1 23 to 79% predicted). Results: Among 13,995 examinees, 51.3 0.4% were female, mean age was 42.2 0.4 years, 48.7 0.9% were nonsmokers, 8.8 0.3% had mild COPD, and 4.1 0.3% had moderate-tosevere COPD [ SE]. One fourth of mild and moderate-to-severe cases were nonsmokers. Among 7,526 nonsmokers, 4.7 0.3% had mild COPD (n 403; age, 60.9 1.3 years) and were mostly female (82.5%), while 1.9 0.3% had moderate-to-severe COPD (n 92, age 39.3 1.3) and were mostly male (88.1%). Few nonsmokers with COPD (12.1 2.4%) had a previous diagnosis of chronic bronchitis or emphysema. Among nonsmokers, physician-diagnosed asthma increased the risk of mild and especially of moderate-to-severe COPD. Independently of asthma, risk of mild COPD in nonsmokers increased with age (doubling every 12 years), before age 60 was lower among men than women, and was inversely associated with current exposure to tobacco smoke at home and at work. In contrast, the risk of moderate-to-severe COPD in nonsmokers was markedly associated with male gender, peaked in middle age, and was inversely associated with nonwhite ethnicity. COPD risks did not vary by minimal smoking history, longest-held occupation, urban residence, income, allergies, thyroid disease, or Helicobacter pylori antibody. Conclusions: Among nonsmokers, mild and moderate-to-severe COPD are associated with asthma but otherwise have distinct demographic profiles, suggesting that moderate-to-severe disease is not a mere progression of mild COPD. (CHEST 2005; 128:1239 1244) Key words: asthma; COPD; epidemiology; nonsmoker Abbreviations: CI confidence interval; OR odds ratio COPD most commonly refers to chronic bronchitis, emphysema, and the subset of asthma characterized by irreversible or partly reversible airflow obstruction. 1 Although the majority of COPD occurs in current or former smokers, the disease also occurs in persons who have never smoked. According to the *From Epidemiology, Pfizer Global Research and Development. This work was performed at Pfizer La Jolla Laboratories, San Diego, CA. Financial support was provided by Pfizer, Inc. Manuscript received October 7, 2004; revision accepted January 29, 2005. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml). Correspondence to: Carolyn Behrendt, PhD, 3413 Paseo del Campo, Palos Verdes Estates, CA 90274; e-mail: carolynbehrendt@ yahoo.com previous epidemiologic study 2 of COPD in nonsmokers, prevalence is greater among women than men until the age of 60 years, when prevalence ceases to differ by sex. Neither urban residence nor occupational category is associated with COPD in nonsmokers. 2 In the general population, COPD is independently associated with smoking, age, and asthma but not with atopy alone. 3,4 In addition, genetic predisposition, environmental tobacco smoke, air pollution, Helicobacter pylori infection, and autoimmune thyroid disease have been proposed as risk factors for COPD. 5 9 The possibility that risk factors for COPD differ according to the severity of disease has not been investigated to date. Using data from a national health examination, the current study identifies and www.chestjournal.org CHEST / 128 / 3/ SEPTEMBER, 2005 1239
compares risk factors for mild and moderate-tosevere COPD among nonsmokers. Spirometry Materials and Methods In the Third National Health and Nutrition Examination Survey (from 1988 to 1994), a representative sample of the civilian, noninstitutionalized population of the United States consented to an interview (assessing demographic characteristics, medical history, medication use, risk behaviors) and a medical examination (including spirometry and a battery of laboratory tests). 10 Spirometry was conducted by a trained technician either in the mobile examination center or in the home of examinees aged 60 years who were unwilling or unable to come to the center. 10 Excluded from spirometry were examinees who had undergone chest or abdominal surgery within 3 weeks or had been hospitalized for myocardial infarction, chest pain, or congestive heart failure within 6 weeks. Study Population Current subjects were white, black, or Mexican-American examinees aged 18 to 80 years whose spirometry findings met the reproducibility and reliability criteria of the American Thoracic Society 11,12 and who provided data on smoking history and height. The upper age limit for the study was chosen to accord with formulas used to derive predicted values of FEV 1 and FVC. 13 Most of the current analysis was restricted to nonsmokers, a group that included both lifelong nonsmokers of cigarettes, cigars, or pipe tobacco and former cigarette smokers with a 5 pack-year smoking history who had never smoked cigars or pipes. 9 Pack-years of smoking were calculated taking into account not only the most recent level of daily smoking but also periods of higher usage or abstinence. COPD. In addition, long-term occupational exposure to airway irritants was considered likely for persons in the following categories of longest held occupation: extractive or precision production, farm or nursery work, cleaning or building services, laborer, cook, and waiter. Income was analyzed using the poverty index, a ratio of family income to the Census Bureau poverty threshold value for the calendar year. 10 Statistical Analysis Sampling weights were taken into account in all analyses (SUDAAN; Research Triangle Institute; Research Triangle Park, NC). Means and percentages were reported with their SEs. Logistic regression analysis was used to identify independent risk factors. Variables found significant (p 0.05) in univariate analysis were tested in multivariate analysis and retained if they improved the fit of the model. As suggested by published data 2 and Figure 1, an interaction term for the effect of sex 60 years and 60 years was tested in the model of mild COPD. Study Sample Results Of the 16,238 examinees, aged 18 to 80 years and identified as white, black, or Mexican American, 95.4 0.3% underwent spirometry; of these, 95.2 0.3% produced reproducible and reliable results. After excluding 6 persons with acceptable spirometry who lacked data on smoking history or height, eligibility criteria were met by 13,995 examinees, who represented 155.82 million adults nationwide (female gender, 51.3 0.4%; mean age, 42.2 0.4 years). Among this general sample, one half (48.7 0.9%) were nonsmokers (n 7,526), who represented 75.95 million adults (female gender, 63.2 0.8%; mean age, 40.1 0.4 years). A Definitions Using a modification of the Global Initiative for Chronic Obstructive Lung Disease criteria, COPD was defined as a FEV 1 /FVC 70% and categorized as mild (FEV 1 80% of predicted) or moderate to severe (FEV 1 80% predicted). 14,15 Respiratory symptoms (cough or phlegm on most days for 3 months per year, dyspnea on exertion, or wheezing during the past year) and physician diagnosis of asthma, chronic bronchitis, and/or emphysema were recorded; however, none of these was a current criterion for COPD. Allergies included history of allergic reaction (to insect sting, food, allergy shot, skin test), allergy symptoms (nose, eyes) during the past year, or physician diagnosis of hay fever. Thyroid disease was defined as thyroid stimulating hormone level of 0.1 or 4.5 miu/l, a physician diagnosis of goiter or other thyroid disease, or current use of thyroid medication. 16 H pylori antibody was measured by enzyme-linked immunoassay during phase 1 (from 1988 to 1991) of the survey. 17 Current exposure to environmental tobacco smoke was reported separately for the home and the workplace. Exposure at home was present if a household member smoked cigarettes in the home. Exposure at work was present if examinees smelled tobacco smoke at the workplace for 4 h/d; lesser daily exposure at work was currently found not to associate significantly with Figure 1. Prevalence of mild and moderate-to-severe COPD in nonsmoking adults, by age and sex. 1240 Clinical Investigations
minority of nonsmokers (12.2 0.6%) had smoked 5 pack-years before quitting at age 27.2 0.4 years. Among the general sample (smokers and nonsmokers), mild COPD was present in 8.8 0.3% and moderate-to-severe COPD was present in 4.1 0.3%. Nonsmokers comprised approximately one fourth of COPD cases: 26.1 1.7% of mild COPD cases, and 22.3 3.0% of moderate-tosevere cases. Among nonsmokers, mild COPD was present in 4.7 0.3% (n 403) and moderate-tosevere COPD was present in 1.9 0.3% (n 92, of which 31 cases were severe, with FEV 1 23 to 49% predicted). Nonsmokers with and without COPD are described in Table 1. Nonsmokers With COPD The prevalence of mild and moderate-to-severe COPD varied by age and sex (Fig 1). Most nonsmokers (76.1 3.7%) with mild COPD were older (age 50 years), while most nonsmokers (84.3 5.0%) with moderate-to-severe COPD were younger (age 50). Women comprised the majority of nonsmokers with mild COPD, but a small minority of those with moderate-to-severe disease (Table 1). The ratio Table 1 Characteristics, Respiratory Symptoms, and Physician Diagnoses Among Nonsmokers Aged 18 to 80 Years, by Level of COPD* Characteristics No COPD Mild COPD of mild to moderate-to-severe COPD was 1:2 among nonsmoking men but 17.2:1 among nonsmoking women. The percentage of nonsmokers who had received a physician diagnosis of asthma increased with the degree of COPD (Table 1). Nonsmokers median age at diagnosis of asthma was 10 years (interquartile range, 5 to 28 years). Among nonsmokers with asthma (n 470), nearly one fourth had some degree of COPD at the time of the survey, 11.3 1.8% having mild COPD, and 12.3 2.9% having moderate-to-severe disease. Few nonsmokers (63 of 495 nonsmokers [12.1 2.4%]) with COPD had ever received a physician diagnosis of chronic bronchitis or emphysema (Table 1). According to multivariate analysis, such a diagnosis among nonsmokers with COPD was more likely in women age 50 years (odds ratio [OR], 7.4; 95% confidence interval [CI], 2.6 to 20.7) and persons reporting allergies (OR, 15.2; 95% CI, 3.4 to 67.9), chronic cough (OR, 6.1; 95% CI, 2.7 to 14.0), wheezing (OR, 6.0; 95% CI, 2.3 to 15.6), or dyspnea on exertion (OR, 3.3; 95% CI, 1.3 to 8.2). Likelihood of diagnosis did not vary by severity of COPD, diagnosis of asthma, minimal smoking his- Moderate-to- Severe COPD Subjects, No. 7,031 403 92 Population represented, millions 70.923 3.587 1.437 Mean age, yr 39.1 0.4 60.9 1.3 39.3 1.3 Female gender 63.2 0.9 82.5 2.3 11.9 4.2 Ever smoked 12.1 0.6 15.3 2.8 9.1 4.4 Now exposed to smoking at home 15.2 1.0 6.2 1.4 16.9 5.5 Now exposed to smoking at work 9.7 0.7 1.3 0.7 23.1 8.3 Thyroid disease 10.1 0.7 22.6 3.3 12.8 8.3 H pylori antibody positive 27.1 1.9 41.9 4.6 35.9 15.8 Education 12 yr 16.7 1.0 30.3 2.4 10.6 3.2 White race 79.5 1.2 87.1 1.6 88.8 2.1 Respiratory symptoms in past year Chronic cough 3.9 0.3 12.3 2.7 8.3 3.5 Chronic phlegm 4.1 0.4 7.0 1.2 7.3 2.6 Wheezing 10.0 0.6 21.6 3.1 43.0 10.2 Dyspnea on exertion 15.0 0.9 26.9 2.8 23.4 6.2 Any of the above symptoms 25.5 1.0 41.1 2.8 52.0 9.7 Physician diagnosis, ever Asthma 5.7 0.4 16.6 2.6 45.1 8.3 Chronic bronchitis 3.2 0.3 12.7 2.6 6.8 3.3 Emphysema 0.2 0.1 2.7 1.2 1.6 1.1 Any of the above diagnoses 8.1 0.5 24.3 3.2 45.8 8.3 *Data are presented as % ( SE) unless otherwise indicated. p 0.001 vs nonsmokers without COPD. p 0.01 vs nonsmokers without COPD. p 0.05 vs nonsmokers without COPD. www.chestjournal.org CHEST / 128 / 3/ SEPTEMBER, 2005 1241
tory, current exposure to smoking in the home or workplace, urban residence, income, occupation, or education. Nonsmoker Risk of Mild COPD Univariate associations with thyroid disease, H pylori antibody, education, and ethnicity (Table 1) became nonsignificant after adjustment for age and sex. Instead, multivariate analysis (Table 2) indicated that risk of mild COPD increased with asthma and with age (doubling every 12 years). Risk was reduced among men before age 60 years; thereafter, risk did not differ significantly by sex. Compared to persons without COPD, mild cases were half as likely to be currently exposed to smoking in the home and one fourth as likely to be currently exposed to smoking in the workplace. Nonsmoker Risk of Moderate-to-Severe COPD Asthma was even more strongly associated with moderate-to-severe COPD than with mild COPD (Table 2). Unlike mild COPD, the risk of moderateto-severe COPD was markedly increased among males, peaked in middle age, and was inversely associated with nonwhite ethnicity. Also unlike mild COPD, moderate-to-severe COPD was not associated with current exposure to smoking at home or at work. Nonrisks for COPD Neither mild nor moderate-to-severe COPD among nonsmokers was associated with ever smoking up to 5 pack-years (Table 1), occupation involving exposure to airway irritants, urban residence, income, or allergies (data not shown). Associations between COPD and asthma (described above) did not vary by age at asthma diagnosis. Discussion According to the current study, nonsmokers account for one fourth (24.9 1.4%) of COPD cases in the United States. Similar proportions of nonsmokers have been reported among COPD cases in the United Kingdom and Spain (22.9% and 23.4%, respectively). 9,18 Few nonsmokers with COPD in the current study had had a previous diagnosis of chronic bronchitis or emphysema. Mild and moderate-tosevere COPD were currently associated with distinct demographic profiles among nonsmokers, suggesting that moderate-to-severe disease is not a mere progression of mild COPD. The previous epidemiologic study of COPD among nonsmokers employed data from earlier National Health and Nutrition Examination Surveys (from 1971 to 1975, from 1976 to 1980, and from 1982 to 1984). 2 The current study differed from the previous study by defining COPD cases through spirometry rather than reported diagnosis of chronic bronchitis or emphysema, by considering mild and moderate-to-severe COPD separately, by considering asthma and current exposure to tobacco smoke as risk factors, and by reporting the prevalence of respiratory symptoms among COPD cases. In addition, while the previous study defined nonsmokers as persons who had not smoked 100 cigarettes in their lifetime, the current study demonstrated the appropriateness of including among nonsmokers those former smokers with minimal smoking history. 9 Table 2 Characteristics Independently Associated with Mild and Moderate-to-Severe COPD Among Nonsmokers Aged 18 to 80 Years Characteristics Mild COPD (397 Cases, 6,922 Controls*), OR (95% CI) Moderate-to-Severe COPD (92 cases, 7,030 Controls*), OR (95% CI) Asthma diagnosis 4.27 (2.59 7.04) 18.8 (8.6 40.8) Per year of age 1.06 (1.04 1.08) 1.04 (1.02 1.07) Per year of age 60 0.85 (0.74 0.97) Male gender 18.3 (7.7 43.4) Gender by age category, yr Male, age 60 0.12 (0.05 0.28) Female, age 60 0.72 (0.39 1.35) Male, age 60 1.29 (0.80 2.10) Female, age 60 1.00 Nonwhite ethnicity 0.58 (0.37 0.89) Now exposed to smoking at home 0.56 (0.35 0.90) Now exposed to smoking at work 0.26 (0.08 0.82) *Controls are nonsmokers without COPD. Six cases and 109 controls were excluded from the model of mild COPD due to missing data, chiefly on current exposure to smoking at work. The R 2 values for the fit of the two models are 0.09 and 0.05, respectively. 1242 Clinical Investigations
As the previous study reported for COPD in general, mild COPD was currently found more prevalent among women than men before age 60 years; in contrast, moderate-to-severe COPD was currently found to affect predominantly men throughout adulthood. The current study confirmed the finding of the previous study 2 of no association between COPD in nonsmokers and occupational category or urban residence. Among a prior case series of nonsmokers with COPD (n 22), none was currently exposed to smoking in the home or workplace; however, half had 20- to 30-year histories of such exposure. 9 In the current study, patients with mild COPD tended to avoid current exposure to smoking in the home and at work, but patients with moderate-tosevere COPD did not. Information on exposure history was unavailable. A previous survival analysis 19 detected no increased mortality hazard from moderate-to-severe COPD among nonsmokers. However, current inverse associations with age 60 years and nonwhite ethnicity suggest that moderate-to-severe COPD may indeed compromise survival among specific groups of nonsmokers. Longitudinal studies of COPD in nonsmokers are needed. The current study confirmed findings from earlier COPD studies not restricted to nonsmokers, specifically independent associations with increasing age and asthma but not with allergy alone. 3,4 The magnitude of the association between asthma and COPD was currently shown to increase with the severity of COPD. Limitations of the current study include its cross-sectional nature, limited numbers of nonsmokers with moderate and severe COPD (precluding separate analyses), and nonsampling of nursing home residents and homeless persons. 20 Also, as noted in the footnote to Table 2, current models explained a small fraction of the risk of mild and moderate-to-severe COPD among nonsmokers, indicating that important risk factors remain to be identified. Data on several proposed risks were not available for the current study: gestational age, birth weight, lower respiratory tract infections before age 7 years, parental smoking during gestation and childhood, exposure to environmental tobacco smoke and other airway irritants throughout adulthood, genetic markers, and family history of COPD. 5,6,21 23 Future spirometric studies, especially if lifetime medical and exposure histories are obtained and subjects are followed up prospectively, will provide a means of testing and improving on current models of COPD risk among nonsmokers. References 1 Mannino DM. Chronic obstructive pulmonary disease: definition and epidemiology. Respir Care 2003; 48:1185 1191 2 Whittemore AS, Perlin SA, DiCiccio Y. Chronic obstructive pulmonary disease in lifelong nonsmokers: results from NHANES. Am J Public Health 1995; 85:702 706 3 Silva GE, Sherrill DL, Guerra S, et al. Asthma as a risk factor for COPD in a longitudinal study. Chest 2004; 126:59 65 4 Hospers JJ, Schouten JP, Weiss ST, et al. Asthma attacks with eosinophilia predict mortality from chronic obstructive pulmonary disease in a general population sample. Am J Respir Crit Care Med 1999; 160:1869 1874 5 Molfino NA. Genetics of COPD. Chest 2004; 125:1929 1940 6 Jaakkola MS, Jaakkola JJ. Effects of environmental tobacco smoke on the respiratory health of adults. Scand J Work Environ Health 2002; 28(Suppl):52 70 7 Sunyer J. Urban air pollution and chronic obstructive pulmonary disease: a review. Eur Respir J 2001; 17:1024 1033 8 Shiotani A. Linking Helicobacter pylori and chronic bronchitis: fact or fancy? J Gastroenterol 2002; 37:402 403 9 Birring SS, Brightling CE, Bradding P, et al. Clinical, radiologic, and induced sputum features of chronic obstructive pulmonary disease in nonsmokers: a descriptive study. Am J Respir Crit Care Med 2002; 166:1078 1083 10 U.S. Department of Health and Human Services. National Center for Health Statistics. Plan and Operation of the Third National Health and Nutrition Examination Survey, 1988 1994. Hyattsville, MD: Centers for Disease Control and Prevention, 1994. Available at: http://www.cdc.gov/nchs/data/ series/sr_01/sr01_032.pdf. Accessed December, 23, 2004 11 Standardization of spirometry, 1987 update: statement of the American Thoracic Society. Am Rev Respir Dis 1987; 136: 1285 1298 12 Third National Health and Nutrition Examination Survey, spirometry procedure manual. Rockville, MD: Westat, 1988. Available at: http://www.cdc.gov/nchs/data/nhanes/nhanes3/ cdrom/nchs/manuals/spiro.pdf. Accessed December 23, 2004 13 Hankinson JL, Odencrantz JR, Fedan KB. Spirometric reference values from a sample of the general U.S. population. Am J Respir Crit Care Med 1999; 159:179 187 14 Pauwels RA, Buist AS, Calverley PM, et al. Global strategy for the diagnosis, management and prevention of chronic obstructive pulmonary disease: NHLBI/WHO Global Initiative for Chronic Obstructive Lung Disease (GOLD) workshop summary. Am J Respir Crit Care Med 2001; 163:1256 1276 15 Mannino DM, Ford ES, Redd SC. Obstructive and restrictive lung disease and markers of inflammation: data from the Third National Health and Nutrition Examination. Am J Med 2003; 114:758 762 16 Hollowell JG, Staehling NW, Flanders WD, et al. Serum TSH, T 4, and thyroid antibodies in the United States population (1988 to 1994): National Health and Nutrition Examination Survey (NHANES III). J Clin Endocrinol Metab 2002; 87:489 499 17 U.S. Department of Health and Human Services. National Center for Health Statistics. Third National Health and Nutrition Examination Survey, 1988 1994, NHANES III second laboratory data file documentation (Series 11, No. 2A). Hyattsville, MD: Centers for Disease Control and Prevention, 1998. Available at: http://www.cdc.gov/nchs/data/ nhanes/nhanes3/lab2-acc.pdf. Accessed December 23, 2004 18 Pena VS, Miravitlles M, Gabriel R, et al. Geographic variations in prevalence and underdiagnosis of COPD: results of the IBERPOC multicentre epidemiological study. Chest 2000; 118:981 989 19 Mannino DM, Buist AS, Petty TL, et al. Lung function and www.chestjournal.org CHEST / 128 / 3/ SEPTEMBER, 2005 1243
mortality in the United States: data from the First National Health and Nutrition Examination Survey follow up study. Thorax 2003; 58:388 393 20 Snyder LD, Eisner MD. Obstructive lung disease among the urban homeless. Chest 2004; 125:1719 1725 21 Bakke PS. Factors affecting growth of FEV 1. Monaldi Arch Chest Dis 2003; 59:103 107 22 Sandford AJ, Silverman EK. Chronic obstructive pulmonary disease: 1. Susceptibility factors for COPD, the genotypeenvironment interaction. Thorax 2002; 57:736 741 23 Montnemery P, Lanke J, Lindholm LH, et al. Familial related risk-factors in the development of chronic bronchitis/emphysema as compared to asthma assessed in a postal survey. Eur J Epidemiol 2000; 16:1003 1007 1244 Clinical Investigations