Prevalence of Respiratory Symptoms, Bronchial Hyperreactivity, and ...

Prevalence of Respiratory Symptoms, Bronchial Hyperreactivity, and Asthma in a Megacity Results of the European Community Respiratory Health Survey in Mumbai (Bombay) ROHINI V. CHOWGULE, VASANT M. SHETYE, JITENDRA R. PARMAR, AJAY M. BHOSALE, MANGALAR R. KHANDAGALE, SUHAS V. PHALNITKAR, and PRAKASH C. GUPTA Department of Chest Medicine, Bombay Hospital Institute of Medical Sciences and Tata Institute of Fundamental Research, Mumbai, India

To estimate adult asthma prevalence in the world’s most rapidly growing mega-city, we applied epidemiologic surveillance tools, as a cooperating center of the European Community Respiratory Health Survey, to a randomly selected sample of Mumbai (Bombay) residents in 1992 through 1995. From a metropolitan population of over 10 million, we took a one-in-ten random sample from electoral rolls in a socially diverse residential district, and examined asthma symptoms in adults age 20 to 44 yr. In Phase I, we interviewed 2,313 adults about symptoms, asthma diagnosis, and medications in the previous 12 mo. In Phase II, family and smoking history, socioeconomic data, housing characteristics, serum IgE, allergy skin tests, spirometry, and methacholine challenge tests were obtained in a subset of 20% of those who had completed Phase I. House dust mite was the most common positive skin test (18% prevalence) and the only one of the nine applied that was significantly associated with asthma symptoms and physician-diagnosed asthma. Asthma prevalence was 3.5% by physician diagnosis, and 17% using a very broad definition including those with asymptomatic bronchial hyperreactivity. Asthma prevalence was strongly associated with positive house dust mite skin test, family history of asthma, and total IgE. Chowgule RV, Shetye VM, Parmar JR, Bhosale AM, Khandagale MR, Phalnitkar SV, Gupta PC. Prevalence of respiratory symptoms, bronchial hyperreactivity, and asthma in a megacity: results of the European Community Respiratory Health Survey in AM J RESPIR CRIT CARE MED 1998;158:547–554. Mumbai (Bombay).

Both prevalence and mortality from asthma appear to have increased in many parts of the world during a time when better asthma medications have been available to more patients suffering from asthma (1). Striking geographic variations in asthma prevalence are evidence of important variations in asthma susceptibility and equally important—and often dynamic—environmental determinants of asthma. Surveillance of asthma in North America, Western Europe, and Australia has demonstrated recent increases in asthma prevalence and mortality. Application of a standardized asthma surveillance methodology in 48 centers worldwide, the European Community Respiratory Health Survey (ECRHS), allows the most valid available comparisons of asthma prevalence and of determinants associated with asthma, including patterns of asthma therapy. India is projected to become the world’s most populous nation by the year 2050. Industrialization and urban growth are

(Received in original form August 13, 1997 and in revised form April 7, 1998) Supported by the Bombay Hospital Trust and the Indian Institute of Environmental Medicine, Mumbai, India. Correspondence and requests for reprints should be addressed to R. V. Chowgule, M.D., Bombay Hospital, MR & C Building, First Floor, Room 12-5, New Marine Lines, Mumbai 400020, India. Am J Respir Crit Care Med Vol 158. pp 547–554, 1998 Internet address: www.atsjournals.org

now occurring at an unprecedented rate in this previously predominantly agrarian society. Mumbai (formerly Bombay), the largest ECRHS city, is the nation’s financial and commercial center and a major industrial port. It is the world’s most rapidly growing mega-city, with a projected population of 15 million in the year 2000. The city gains over 250,000 rural-to-urban emigrants annually. Like another international commercial center, New York City, Mumbai lies on a narrow peninsula running north to south, connected to the mainland at the northern tip, constricting the geographic spread of its growing population. Situated at approximately 198 north latitude, the coastal city has the warm moist climate of a tropical savanna. Sulfur reductions in fossil fuels in recent years have helped reduce ambient sulfur dioxide levels, but the increasing density of motor vehicles contributes to rising concentrations of total suspended particulate. The ECRHS was designed to assess prevalence of asthma and bronchial hyperresponsiveness in urban centers, to estimate the variation in exposure to risk factors and assess their relationship to the variations of asthma prevalence, and to estimate variation in the treatment practice for asthma. We present here a survey of these factors from a large residential district in Mumbai. Because of its size and random sampling methodology, this study represents our best view to date of asthma prevalence in India, and will serve as the baseline for future studies of changing asthma prevalence.

548

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE

METHODS Mumbai was one of the participating centers of the ECRHS, which included centers on five continents. Details of the uniform protocols and an initial comparison of the prevalence of respiratory symptoms among 48 centers in 20 countries have been published elsewhere (2). To summarize this uniform design, each center selected for study all residents of a preexisting urban administrative district with a total population of approximately 150,000 or more, and included adults age 20 to 44 yr. Questions were taken from the bronchial symptoms questions of the International Union Against Tuberculosis and Lung Disease (IUATLD) questionnaire (3), and questions about the home environment were based on the Children’s Health Study (Harvard School of Public Health and Canadian Health and Welfare), a survey of 24 communities in the U.S. and Canada. Additional questions were asked about occupational and social status, smoking, and medication use (4). Translation of the English-language questionnaire into standard Marathi and Hindi language questionnaires was performed by members of the research team. Mumbai was one of only three participating centers to administer these questionnaires in person during home visits, rather than through postal surveys. For the larger, initial Phase I of the survey, 1,750 female and 1,750 male subjects were selected from the electoral rolls of a mainly residential district, one of 20 political wards in Mumbai (Figure 1). The

VOL 158

1998

sample size was chosen for the ECRHS multicenter study to detect a twofold difference in the prevalence of asthma between any two participating centers, assuming a baseline asthma prevalence estimate of 5%. Questionnaires were administered by trained, trilingual social workers, in the language favored by the respondent. Social workers made 4 to 5 visits to interview subjects before applying exclusion criteria. In Phase II, a randomly selected 20% subset of 240 males and 276 females were selected for an additional detailed, interviewer-administered, 13-page questionnaire concerning symptoms, allergies, family history, smoking, home environment, socioeconomic status, education, and medication use. Self-selection occurred when randomly selected Phase I subjects declined to participate in Phase II, and additional Phase I subjects were selected randomly until the total of 516 was reached. The Phase II group received skin tests with nine allergens and a positive control, serum total IgE level, spirometry, and methacholine challenge testing, using the standard instruments, reagents, and protocols of the ECRHS. Spirometry was with a Spirotech dry rolling seal spirometer (Spirotech, Smyrna, GA) using American Thoracic Society technical criteria for acceptability and reproducibility (5) and using five to nine expiratory maneuvers. Prediction equations for mean lung function values based on height, gender, and age are those suggested by the ECRHS, and are shown in APPENDIX B. Methacholine challenge (Provocholine; Hoffman La Roche, Basel, Switzerland) was performed with the Mefar MB3 dosimeter (Mefar srl, Bovezzi, Italy). Serial methacholine dilutions (0.195 mg/ml to 12.5 mg/ml) were made from a 25 mg/ml stock solution. Subjects in Mumbai who showed more than 20% fall at any dose (2.0 mg cumulative dose or less, with maximal concentration of 12.5 mg/ml) were considered hyperresponsive. Within-center variation in nebulizer percentage aerosol output in a sample of 161 nebulizers from 21 participating centers was less than 10% (6). Allergy testing used standardized antigen-coated skin-prick lancets (Phazets; Pharmacia Diagnostics AB, Uppsala, Sweden) including an uncoated negative control and a histamine-coated positive control. Skin pricks were applied with uniform pressure and read 15 min later. Wheal diameters in two orientations 90 degrees apart were recorded to the nearest millimeter, and mean diameter > 4 mm was considered a positive skin test. Total serum IgE and eosinophilic cationic protein were determined by radioimmunoassay from serum transferred in dry ice to the United States National Institute for Occupational Safety and Health Laboratories, Morgantown, West Virginia. Descriptive statistics with stratification were performed to examine characteristics of this representative random sample of the general population of Mumbai in relationship to symptoms, allergy, asthma, environment, demographics, and medication usage. To correlate asthma symptoms with potential independent asthma determinants, logistic regression analysis was performed.

RESULTS

Figure 1. Mumbai (formerly Bombay) with a current population of approximately 12 million, lies on a narrow peninsula in the Arabian Sea on the west coast of India. Population density is high in part because of limited area for expansion. The study was performed in one of 20 urban political districts (D-Ward), a primarily residential area with diverse socioeconomic characteristics.

The crude response rate for the Mumbai center (those randomly selected for Phase I of the study who were interviewed) was 63.2%. After removing candidates from the sample who were selected but ineligible (i.e., in the wrong age category, resided outside the administrative district, had moved away, died, or refused), the adjusted response rate was 81.0%. This was higher than the median adjusted response rate of 77.8% for all 48 participating centers. By comparison, the adjusted response rate for Portland, OR was 68.8%, for Paris 74.3%, and for Melbourne 72.7%. The prevalence in Phase I of airway symptoms, medication use, diagnosis, and nasal allergy during the previous 12 mo, broken down by 5-yr age cohorts, is shown in Table 1. The same symptom data, shown overall and by gender, are presented in Table 2. The prevalence of airway symptoms in the previous 12 mo (see APPENDIX A for text of Phase I questions) ranged from 4 to 11% of these adults, with a prevalence of doctor-diagnosed asthma (positive response to questions 5 or 6) of 3.5%. Comparison with selected ECRHS cities is shown in Table 7. Positive responses to these questions were at or be-

549

Chowgule, Shetye, Parmar, et al.: Respiratory Symptoms and Asthma in Mumbai TABLE 1 PREVALENCE OF RESPIRATORY SYMPTOMS IN LAST 12 mo, BY AGE GROUP, PHASE I Age Range

20–24

25–29

30–34

35–39

40–45

Total

Number (%)

57 (2.6%)

460 (20.8%)

580 (26.2%)

506 (22.9%)

610 (27.6%)

2,213 (100%)

1. Wheeze* 1.1. Wheeze and breathlessness 1.2. Wheeze and no cold 2. Chest tightness 3. Breathlessness at night 4. Cough at night 5. Asthma attack 6. Asthma medication 5. or 6. Diagnosed asthma 7. Nasal allergy

4 (7%) 4 (7%) 1 (1.8%) 5 (8.8%) 7 (12.3%) 9 (15.8%) 2 (3.5%) 3 (5.3%) 3 (5.3%) 9 (15.8%)

17 (3.6%) 11 (2.4%) 9 (2%) 33 (7.2%) 36 (7.8%) 58 (12.6%) 11 (2.4%) 11 (2.4%) 16 (3.5%) 45 (9.8%)

19 (3.3%) 14 (2.4%) 10 (1.7%) 44 (7.6%) 37 (6.6%) 63 (10.9%) 12 (2.1%) 11 (1.9%) 16 (2.8%) 59 (10.2%)

23 (4.5%) 15 (3%) 15 (3%) 42 (8.3%) 29 (5.7%) 58 (11.5%) 14 (2.8%) 11 (2.2%) 16 (3.2%) 33 (6.5%)

34 (5.6%) 21 (3.4%) 13 (2.1%) 43 (7%) 39 (6.4%) 61 (10%) 20 (3.3%) 23 (3.8%) 26 (4.3%) 67 (11%)

96 (4.4%) 65 (2.9%) 48 (2.1%) 167 (7.5%) 148 (6.7%) 249 (11.2%) 59 (2.7%) 59 (2.7%) 77 (3.5%) 213 (9.6%)

* Complete questions for each response in this column are listed in APPENDIX A.

low the median, and for many questions in the lowest quartile, of ECRHS participating centers (2). The completion rate for subjects in Phase II was 73% (53% female, 94% Hindu religion, 42% with education to grade 10 or less, 58% with undergraduate or graduate education). Sixteen percent of Phase II subjects reported smoking; all were males. All respondents in Phase II were divided into three symptom groups: Group 1 asymptomatic with respect to wheeze, chest tightness, and shortness of breath at rest during the last 12 mo, (n 5 385, 75%); Group 2 with positive response to one or more of these three questions without a physician diagnosis of asthma (n 5 99, 19%); and Group 3 symptomatic with previously diagnosed asthma (n 5 32, 6%). Table 3 contrasts the demographic and clinical characteristics of the Phase II subjects in these three groups, which stratify all Phase II subjects by a combination of symptom severity and physician diagnosis of asthma. With greater severity of symptoms, the groups show increasing prevalence of low percentage of predicted FEV1, of family history of asthma and atopy, skin prick test positivity, and serum IgE. The prevalence of positive skin tests was highest for house dust mite (18% overall, 14% in Group 1, 21% in Group 2, and 60% in Group 3), and was significantly higher for Group 3 (asthma diagnosis) subjects than for Group I (asymptomatic) subjects. The prevalence of any positive test was 4% or less for each of the other eight skin test antigens (results shown in Table 3), and none of these other skin tests was significantly associated with symptom groups. Use of asthma medications was also

TABLE 3

TABLE 2 PREVALENCE OF AIRWAY RESPIRATORY SYMPTOMS DURING LAST 12 mo IN 2,213 SUBJECTS, PHASE I

1. Wheeze* 1.1. Wheeze and breathlessness 1.2. Wheeze and no cold 2. Chest tightness 3. Breathlessness at night 4. Cough at night 5. Asthma attack 6. Asthma medication 5. or 6. Diagnosed asthma 7. Nasal allergy

correlated with symptom classifications (not shown). In Table 4, the same three symptom groups are contrasted by percentage of predicted FEV1. Overall, there is an association of higher symptom groups with lower percentage of predicted FEV1. The distribution of methacholine challenge test positives (20% fall in FEV1 at or below the indicated methacholine cumulative dose) is shown in Table 5. Overall, 87 of 516 (14%) of the challenges were “positive” at < 2.0 mg methacholine cumulative dose, and the prevalence of positive responses in those with asthma medication or an attack of asthma in the previous year was 80%. Asthma may be underdiagnosed, and the prevalence of “doctor-diagnosed” asthma may underestimate the true prevalence. In order to estimate the prevalence of asthma in this randomly selected adult population, we applied two different case definitions to our population. As an additional, more inclusive definition using Phase II subjects, we added to those with a previous physician diagnosis of asthma all additional subjects who responded to < 1 mg/ml methacholine with a 20% or greater fall from diluent FEV1 response (n 5 56) , and all subjects from symptom groups 1 and 2 with < 70% predicted FEV1 in the absence of smoking and a history of chronic bronchitis (n 5 34). Using this definition, the prevalence of asthma in Phase II was 23.6% (n 5 122), six times the prevalence of doctor-diagnosed asthma. However, self-selection occurred in Phase II and selection may have been biased toward including asthmatics.

Male (n 5 1,151)

Female (n 5 1,062)

Total (n 5 2,213)

46 (4%) 30 (2.6%) 26 (2.3%) 86 (7.5%) 73 (6.3%) 128 (11.1%) 35 (3%) 38 (3.3%) 44 (3.8%) 107 (9.3%)

50 (4.7%) 35 (3.3%) 22 (2.1%) 81 (7.6%) 75 (7.1%) 121 (11.4%) 24 (2.3%) 21 (2%) 33 (3.1%) 106 (10%)

96 (4.4%) 65 (2.9%) 48 (2.1%) 167 (7.5%) 148 (6.7%) 249 (11.2%) 59 (2.7%) 59 (2.7%) 77 (3.5%) 213 (9.6%)

* Numbers refer to question numbers from the European Community Respiratory Health Survey, as shown in the APPENDIX.

CHARACTERISTICS BY SYMPTOM GROUPS,* PHASE II

Mean age Proportion male (%) Mean level of FEV1 Subjects FEV1 , 70% predicted Family history of asthma Family history of atopy Skin prick test positive, > 4 mm IgE, geometric mean kU/L‡ Eosinophilic cationic protein, mean 6 SD, µg/L‡

Group 1 (n 5 385)

Group 2 (n 5 99)

Group 3 (n 5 32)

36 6 6 188 (49) 2.46 6 0.64 63 (16%) 81 (21%) 57 (15%) 66 (17%) 229.7 (212)

37 6 5 35 (35) 2.10 6 0.56 28 (28%) 25 (25%) NS 23 (23%) NS 26 (26%) NS 363.1 (59)

38 6 5 17 (53) 1.94 6 0.88 18 (56%) 17 (53%)† 11 (34%) NS 21 (66%)† 489.8 (21)

36 6 33 (76)

38 6 35 (25)

22 6 14 (9)

Definition of abbreviation: NS 5 nonsignificant. * For explanation of groups, see Table 1. † p , 0.05. ‡ Numbers in parentheses refer to the number of subjects tested in each category.

550


VOL 158

TABLE 4

TABLE 5

LEVEL OF LUNG FUNCTION BY SYMPTOM GROUPS,* PHASE III

FREQUENCY OF AIRWAYS RESPONSIVENESS BY SYMPTOM GROUPS,* PHASE II

% of Predicted FEV1

Group 1 (n 5 385)

Group 2 (n 5 99)

Group 3 (n 5 32)

Less than 60% 61–69% 70–79%† More than 80%† Poor reproducibility

14 (4%) 49 (13%) 110 (29%) 196 (51%) 16 (4%)

13 (13%) 15 (15%) 30 (30%) 39 (40%) 2 (2%)

12 (38%) 6 (19%) 8 (25%) 4 (13%) 2 (6%)

* For explanation of groups, see Table 1. † Subjects with more than 70% of predicted FEV1 were given the methacholine challenge test.

Using a third definition including only those with physician-diagnosed asthma and/or high airway reactivity (provocative concentration of methacholine producing a 20% fall in FEV1 [PC20] , 2.0 mg) the prevalence of asthma in Phase II is 17%. This definition is also subject to selection bias, and includes subjects with asymptomatic airway hyperresponsiveness. Oral asthma medication was used by 34 asthmatic subjects and inhaled medication by 16 in Phase II. Beta agonists and steroids were the most frequently used.

PC20 Methacholine response positive methacholine, cumulative dose, mg 0.0156 0.0625 0.125 0.25 0.50 1.00 2.00

Group 1 (n 5 261)

Group 2 (n 5 55)

Group 3 (n 5 10)

65 (25%) 0 2 2 13 13 16 19

14 (25%) 1 2 1 2 2 3 3

8 (80%) 2 1 1 3 1 1 0

* Group 1: no wheeze, chest tightness, shortness of breath at rest in last 12 mo; Group 2: wheeze, chest tightness or shortness of breath in last 12 mo but without physician-diagnosed asthma; Group 3: currently symptomatic with physician-diagnosed asthma.

Table 6 summarizes logistic regression analyses using three slightly different definitions of asthma in Phase II subjects and correlating diagnosis with potential asthma determinants of sex, education, type of home, occupation, smoking status, family history of asthma or allergy, and skin test status to house

TABLE 6 FACTORS ASSOCIATED WITH POSITIVE RESPONSES FOR ASTHMA-RELATED QUESTIONS IN PHASE II SUBJECTS, BY LOGISTIC REGRESSION ANALYSIS Variable “Ever had asthma” Sex 0 - Male 1 - Female Education 0 - No education/primary 1 - Secondary 2 - Undergraduate 3 - Graduate/postgraduate Home 0 - Hutment/single room 1 - Apartment/chawl Occupation 0 - Housewife/student/unskilled 1 - Professional 2 - Intermediate 3 - Skilled (nonmanual) 4 - Skilled (manual) Ever smoked 0 - No 1 - Yes Family history of asthma 0 - No 1 - Yes Family history of allergy 0 - No 1 - Yes House dust mite 0 - Negative 1 - Positive Physician-diagnosed asthma Sex 0 - Male 1 - Female Education 0 - No education/primary 1 - Secondary 2 - Undergraduate 3 - Graduate/postgraduate

1998

n (%)

Relative Odds

240 (47%) 276 (53%)

1.03

0.57–1.87

61 (12%) 158 (31%) 151 (29%) 146 (28%)

4.03 2.79 3.13

0.91–17.86 0.61–12.70 0.69–14.21

155 (30%) 361 (70%)

1.96

0.93–4.16

267 (52%) 23 (5%) 44 (8%) 99 (19%) 83 (16%)

1.01 2.01 1.06 0.98

0.22–4.58 0.80–4.99 0.47–2.38 0.43–2.23

433 (84%) 83 (16%)

0.73

0.29–1.79

393 (76%) 123 (24%)

2.17

1.11–4.22

425 (82%) 91 (18%)

1.20

0.56–2.55

422 (82%) 94 (18%)

5.69

3.06–10.57

240 (47%) 276 (53%)

0.75

0.37–1.54

61 (12%) 158 (31%) 151 (29%) 146 (28%)

1.99 2.09 2.17

0.42–9.37 0.44–9.84 0.46–10.20

95% CI

(Continued )

551

Chowgule, Shetye, Parmar, et al.: Respiratory Symptoms and Asthma in Mumbai

dust mite (the most frequently positive skin test). Family history of asthma was associated with asthma diagnosis using all definitions. Positive house dust mite skin test was associated with asthma based on positive responses to the question, “Have you ever had asthma?” and with physician-diagnosed asthma. In addition, living in an apartment (or chawl) was associated with asthma defined as a positive response to the question, “Have you ever had asthma?”

DISCUSSION As Mumbai is India’s largest and fastest growing city and a major destination for rural Indians from all parts of the country, this study provides the best available estimate of adult

asthma, bronchial reactivity, allergic symptoms, atopy, and specific allergy in India, whose population is projected to exceed one billion by the turn of the century. While Mumbai’s ethnic diversity reflects that of the nation, India is geographically diverse and Mumbai’s moist coastal climate is only one of several, ranging from tropical to alpine, throughout the country. The ECRHS is to date the most widely applied, physiologically based asthma and airways disease surveillance methodology, and permits the most valid comparisons of asthma prevalence, bronchial responsiveness, and asthma-related factors between widely differing urban centers. Mumbai is the largest of the 48 participating ECRHS centers by total city population. The study of these urban centers summarizes the effects

TABLE 6 CONTINUED Variable Home 0 - Hutment/single room 1 - Apartment/chawl Occupation 0 - Housewife/student/unskilled 1 - Professional 2 - Intermediate 3 - Skilled (nonmanual) 4 - Skilled (manual) Ever smoked 0 - No 1 - Yes Family history of asthma 0 - No 1 - Yes Family history of allergy 0 - No 1 - Yes House dust mite 0 - Negative 1 - Positive “Wheeze and breathlessness” and/or “wheeze when no cold” Sex 0* - Male 1 - Female Education 0 - No education/primary 1 - Secondary 2 - Undergraduate 3 - Graduate/postgraduate Home 0 - Hutment/single room 1 - Apartment/chawl Occupation 0 - Housewife/student/unskilled 1 - Professional 2 - Intermediate 3 - Skilled (nonmanual) 4 - Skilled (manual) Ever smoked 0 - No 1 - Yes Family history of asthma 0 - No 1 - Yes Family history of allergy 0 - No 1 - Yes House dust mite 0 - Negative 1 - Positive * 0 5 reference variable.

n (%)

Relative Odds

155 (30%) 361 (70%)

3.11

1.09–9.20

267 (52%) 23 (5%) 44 (8%) 99 (19%) 83 (16%)

1.72 1.81 1.17 1.41

0.37–8.08 0.57–5.58 0.44–3.12 0.52–3.79

433 (84%) 83 (16%)

0.73

0.25–2.15

393 (76%) 123 (24%)

3.46

1.58–7.58

425 (82%) 91 (18%)

1.60

0.69–3.70

422 (82%) 94 (18%)

7.97

3.77–16.82

240 (47%) 276 (53%)

0.76

0.27–2.12 NS

61 (12%) 158 (31%) 151 (29%) 146 (28%)

0.49 0.16 0.17

0.14–1.66 NS 0.03–0.88 NS 0.03–0.88 NS

155 (30%) 361 (70%)

1.12

0.77–3.86 NS

267 (52%) 23 (5%) 44 (8%) 99 (19%) 83 (16%)

1.17 0.001 0.26 0.96

0.14–9.61 NS 0.00–8.42 NS 0.03–2.08 0.26–3.59

433 (84%) 83 (16%)

1.89

0.59–6.11

393 (76%) 123 (24%)

4.55

1.46–14.22

425 (82%) 91 (18%)

1.91

0.60–6.06

422 (82%) 94 (18%)

0.87

0.19–3.95

95% CI

552


VOL 158

1998

TABLE 7 PERCENT PREVALENCE OF POSITIVE RESPONSES TO PHASE I ASTHMA QUESTIONNAIRES AT DIFFERENT CENTERS* Center Mumbai, India Göteborg, Sweden Hamburg, Germany Cambridge, UK Paris, France Barcelona, Spain Athens, Greece Wellington, New Zealand Melbourne, Australia Portland, USA Algiers, Algeria

Wheeze

Wheeze and Breathlessness

Wheeze, No Cold

Chest Tightness

Breathlessness at Night

Cough at Night

Asthma Attack

Asthma Medicine

Nasal Allergy

Diagnosed Asthma

4.1 23.2 21.1 25.2 14.2 19.2 16.0 27.3 28.8 25.7 4.2

3.0 12.3 8.0 13.9 9.3 5.6 9.4 16.0 16.1 10.5 3.3

2.0 13.5 13.5 17.7 9.0 11.0 9.8 18.0 20.7 14.9 2.8

7.0 14.7 9.6 17.4 16.8 7.0 11.7 18.1 20.5 16.6 6.4

6.8 7.1 5.0 8.4 4.7 4.6 5.7 10.4 11.4 7.7 4.4

11.2 28.2 25.8 27.4 26.0 28.2 17.8 31.2 28.5 32.5 6.0

2.6 3.1 3.0 5.7 4.3 2.1 2.4 8.6 9.7 5.8 2.4

2.8 4.8 3.4 6.8 3.2 2.2 2.2 9.8 9.3 4.8 2.5

10.1 22.2 23.0 29.2 30.3 13.1 18.4 36.6 40.9 39.4 9.5

3.5 5.8 4.4 8.4 5.1 3.1 2.9 11.3 11.9 7.1 3.0

* Adapted from Reference 2.

both of markedly different genetic risks and different environmental risks for asthma, both of which play a major part in varying asthma prevalence rates. Asthma is frequently underdiagnosed, precluding adequate treatment (7). Due to lack of consistency in the definition of asthma, estimates of the prevalence of asthma in the general population in different locations have been difficult to compare without a standardized methodology (8, 9). Comparison of our results with the prevalence of asthma symptoms in all 48 participating ECRHS centers (2) shows that the prevalence of some asthma symptoms, and of nasal allergy/hay fever, is significantly lower in Mumbai than the median for all centers. Comparisons with selected other ECRHS centers are shown in Table 7. Doctor-diagnosed asthma and use of asthma medications were also lower in Mumbai, although differences with ECRHS medians were not statistically significant using 95% confidence intervals. When stratifying adults into 5-yr age cohorts, the youngest group (age 20 to 24) had the highest prevalence of doctordiagnosed asthma and medication use. However, this age stratum included only seven subjects. When the 20 to 24 yr and 25 to 29 yr strata were collapsed into one group, there were no age-related patterns of asthma prevalence or symptoms. The contrast between numbers of physician-diagnosed cases of asthma and the prevalence of asthma symptoms leads us to believe that asthma is relatively underdiagnosed in Mumbai. Our additional examination using a very broad definition of asthma adds to those with a previous physician diagnosis subjects with marked bronchial hyperreactivity (PC20 , 1 mg cumulative dose) who may however be asymptomatic, and airflow obstruction in the absence of smoking or bronchitis markedly increases the population estimate of asthma, but includes individuals who are minimally symptomatic or asymptomatic. Of demographic and environmental factors tested for association with asthma diagnosis, positive skin test response to house dust mite (Dermatophagoides pteronyssinus) was the factor most strongly associated with asthma diagnosis. Comparing skin test responses (a sensitive bioassay for specific IgE) to measured specific IgE from serum of random selected subjects from 37 centers in 16 other ECRHS countries, the 18% prevalence of specific house dust mite sensitization in Mumbai is comparable to that in many other centers (the median for the 37 centers being 20% positive) while the prevalence of specific IgE for pollens and grass in Mumbai was well below the median for ECRHS centers, generally in the lowest quartile (2% Cladosporium fungus, 3% Alternaria fungus, 2% timothy grass, 2% olive pollen, 1% birch pollen, and 1% pari-

etaria pollen). The low rate of skin test positivity for the common grass and Cladosporium spp. fungal antigen is consistent with the relatively low positive response to the question of nasal allergy, in Phase I, compared with most other centers (Table 6). D. pteronyssinus was the most frequently positive specific IgE among the four common allergens tested at all 35 centers participating in this comparison study (10). The strong association of house dust mite, but not other skin test antigens, with asthma diagnosis (Table 6) suggests an important role for this indoor antigen in asthma causation or asthma severity in Mumbai. It is possible that the indoor antigen D. pteronyssinus is specific to asthma risk while other aeroallergens (fungus and pollen) are more associated with nasal allergies. This has been suggested in other geographic areas, at low altitude with high humidity (including one in North America), where sensitization to dust mites is a dominant risk factor for asthma (11). Strong correlation of asthma diagnosis with family history of asthma is consistent with most epidemiological studies of asthma determinants in adults, including an analysis of 30 participating ECRHS centers in which asthma prevalence ranged from 2.1% to 16.2%. Applying gene segregation models to the data from these centers, a major gene which could also be involved in allergy exists, but asthma was not fully described by a single-gene model (12). The association of a history of “ever had asthma” with residence in a chawl (apartment) compared with residence in a single room hutment is of uncertain significance. Because this factor was analyzed in the model with house dust mite skin test positivity, residence in a chawl does not confound the association of asthma with house dust mite allergy. In general, dwelling in a chawl is associated with higher socioeconomic status. Thus the significance of this association is not yet known but suggests an important asthma risk related to type of dwelling but independent of educational level. Using our broad definition of asthma and comparing it with the prevalence of doctor-diagnosed asthma, there is a ratio of 6.7 untreated to each treated asthmatic in this survey. Smoking prevalence in Mumbai is low compared with many Asian and European countries. The reported cigarette smoking prevalence of 16% among Phase II male participants (age 20 to 44) is consistent with a previous survey of smoking practices in 100,000 Bombay residents (by one of our research group, P.C.G.) in which 23% of men age 35 and older reported cigarette smoking, and fewer than 1% of women reported smoking. Environmental tobacco smoke is thus also much less prevalent than in some of the other ECRHS centers. This study will serve as the benchmark for asthma epidemi-

Chowgule, Shetye, Parmar, et al.: Respiratory Symptoms and Asthma in Mumbai

ology in India both for within-nation comparisons and to measure secular trends in asthma and symptom prevalence as Mumbai continues to grow. The symptom prevalence of the Phase I sample also permits comparison with urban populations in many other parts of the world. The ECRHS spirometric prediction equations used by all participating centers are based on surveys of Europeans. As a cooperating ECRHS center, we have applied these norms to our spirometry measurements to facilitate comparisons among ECRHS centers. Normal spirometric values for our subjects may differ, and this may have led to an increase in the percentage of subjects with FEV1 less than 70% predicted. Patterns of Asthma in India

The asthma prevalence in the Mumbai sample (3.5% based on physician diagnosis in Phase I, but 9 to 12% when including symptomatic subjects without diagnosis) exceeds the 2.78% prevalence of asthma in 30 to 49 yr-olds seen in a 5% population sample questionnaire survey, utilizing questions similar to those of Phase I, of the Indian interior city Patna reported 30 yr ago, the only other large survey of asthma in India we have found (13). In comparing the two studies, constellations of symptoms strongly suggestive of asthma were used in both, whereas physician diagnosis was used only in the present study. Thus symptom prevalence constitutes the most valid basis for comparison, and in comparing studies by symptoms, there is a markedly higher prevalence in Mumbai. One possible explanation for the difference in asthma prevalence is related to climate. Higher humidity in Mumbai than in Patna may accelerate dust mite reproduction, causing higher levels of house dust mites in homes. Although the relationship of associations between sensitization to environmental allergens and clinical asthma in adults needs further clarification, there are striking geographic variations in the nature of allergen/asthma association. Older adult male asthmatics in Boston, for example, are more likely to be sensitive to cat allergen than to dust mite (14). It has been suggested that urbanization and the attendant changes in multiple social, lifestyle, and dietary factors may play an important role in increasing asthma prevalence in certain groups. This might also help explain the difference in asthma prevalence seen between Patna in the 1960s and Mumbai in the 1990s. One comparison of asthma prevalence (without a uniform sampling methodology) of children living in Patna with children born in India who emigrated to England, and children born in England of Indian parents, showed increasing prevalence of asthma with migration (15). Such migration effects—having often to do more with lifestyle rather than purely geographic changes—may have occurred within India during the past 30 yr. Even allowing for large potential differences in sampling methodology, the much higher prevalence of asthma as estimated by symptoms alone in Mumbai (9 to 12%) suggests either a marked geographical disparity, or possibly the effect of factors associated with a rapidly changing lifestyle in urban India. Whether generally higher asthma prevalence and morbidity are features of multiple factors associated with changing urban lifestyle, as suggested by PlattsMills and Carter (16), can now be tested by reapplication of this methodology over geographic and temporal distance. Because of the extremely rapid in-migration to Mumbai, it may be possible in the future to perform studies of asthma prevalence in relation to rural-to-urban migration patterns. It should also be noted that treatment of common helminthic infections in low socioeconomic status residents of a rural tropical region has recently been shown to improve asthma symptoms (17).

553

Conclusions

Hypotheses raised in this study by the prevalence and determinants of asthma among adults in Mumbai can be further examined in the future by comparisons among participating ECRHS centers. House dust mite skin test positivity is moderately high in Mumbai, and comparable to other moist coastal or low-lying cities in North America, Australia, and Europe. In some of these other areas, house dust mite skin test positivity has been associated with the population prevalence of asthma. Correlation of dust mite exposure and skin test with asthma is strong (18, 19), raising the possibility that sensitization to these antigens may be an important contributor to asthma in Mumbai, as it has been found in some other urban centers and particularly for children living in moist areas where house dust mite growth and antigen levels are higher than in dry climates (20). A committee of experts setting international priorities in Environmental Epidemiology recently identified asthma incidence as a high priority outcome measure for research in air contaminants (21). Other environmental factors not measured in this study which might also contribute to asthma symptom severity include ambient total suspended particulate (TSP), which in 1992–1993 has exceeded recommended World Health Organization annual TSP upper guidelines of 90 mg per cubic meter annual average at an air pollution monitoring station close to the administrative district studied here (22). The asthma prevalence data presented here may now serve as a representative baseline for future studies of air contaminants in relation to respiratory symptoms and asthma. Acknowledgment : Mumbai was a participating center in the European Community Respiratory Health Survey and received methacholine chloride and skin testing materials from the survey center (Project Leader: Professor Peter Burney) but did not receive European Community member state or Cooperation in Science and Technical Research funds. IgE concentrations were measured by Dr. Val Vallyathan, the U.S. National Institute for Occupational Safety and Health (NIOSH), Morgantown, W. Va. We are grateful for the critical review of William S. Beckett, M.D., MPH, Department of Environmental Medicine, University of Rochester School of Medicine and Dentistry.

References 1. Alderson, M. 1987. Trends in morbidity and mortality from asthma. Population Trends 49:18–23. 2. Anonymous. 1996. Variations in the prevalence of respiratory symptoms, self-reported asthma attacks, and use of asthma medication in the European Community Respiratory Health Survey (ECRHS). Eur. Respir. J. 9:687–695. 3. Burney, P. G. J., C. Luczynska, S. Chinn, and D. Jarvis. 1994. The European Community Respiratory Health Survey. Eur. Respir. J. 7:954960. 4. Burney, P. G. J., and S. Chinn. 1987. Developing a new questionnaire for measuring the prevalence and distribution of asthma. Chest 91:79S– 83S. 5. American Thoracic Society. 1979. Standardization of spirometry. Am. Rev. Respir. Dis. 119:831–838. 6. Chinn, S., W. A. Arossa, D. L. Jarvis, C. M. Luczynska, and P. G. J. Burney. 1997. Variations in nebulizer aerosol output and weight from the Mefar dosimeter—implications for multicentre studies. Eur. Respir. J. 10:452–456. 7. Gregg, I. 1986. Epidemiological research in asthma: the need for a broad perspective. Clin. Allergy 16:17–23. 8. Toelle, B. G., J. K. Peat, C. M. Salome, and A. J. Woolcock. 1992. Towards a definition of asthma for epidemiology. Am. Rev. Respir. Dis. 146:633–637. 9. Samet, J. M. 1987. Epidemiological approaches for the identification of asthma. Chest 91:74S–89S. 10. Burney, P., E. Malmberg, S. Chinn, D. Jarvis, C. Luczynska, E. Lai, on behalf of the European Community Respiratory Health Survey. 1997. The distribution of total and specific serum IgE in the European Community Respiratory Health Survey. J. Allergy Clin. Immunol. 99:414–322.

554


11. Squillace, S. P., R. B. Sporik, G. Rakes, N. Couture, A. Lawrence, S. Merriman, J. Zhang, A. E. Platts-Mills. 1997. Sensitization to dust mites as a dominant risk factor for asthma among adolescents living in central Virginia. Am. J. Respir. Crit. Care Med. 156:1760–1764. 12. European Community Respiratory Health Survey Group. 1997. Genes for asthma? An analysis of the European Community Respiratory Health Survey. Am. J. Respir. Crit. Care Med. 156:1773–1780. 13. Viswanathan, R. M., M. Prasad. A. K. Thakur, S. P. Sinha, N. Prakash, R. K. Mody, T. R. B. P. N. Singh, and S. N. Prasad. 1966. Epidemiology of asthma in an urban population: a random morbidity survey. J. Med. Assoc. India 46:480–483. 14. Litunja, A. A., D. Sparrow, S. T. Weiss, G. T. O’Connor, A. A. Long, and I. L. Ohman. 1997. Sensitization to cat allergen is associated with asthma in older men and predicts new-onset allergy and hypersensitivity. Am. J. Respir. Crit. Care Med. 156:23–27. 15. Spears, J. 1975. The prevalence of allergic disease in young British-born schoolchildren of different ethnic origin. J. Roy. Coll. Gen. Pract. 25: 282–285. 16. Platts-Mills, T. A. E., and M. C. Carter. 1997. Asthma and indoor exposures to allergens [Editorial]. N. Engl. J. Med. 336:1382–1384. 17. Lynch, N. R., M. Palenque, I. Hagel, and M. C. DiPrisco. 1997. Clinical improvement of asthma after antihelminthic treatment in a tropical situation. Am. J. Respir. Crit. Care Med. 156:50–54. 18. Platts-Mills, T. A., W. R. Thomas, R. C. Aalberse, D. Vervloet, and M. D. Chapman. 1992. Dust mite allergens and asthma: report of a second international workshop. J. Allergy Clin. Immunol. 89:1046–1060. 19. Sporik, R., M. D. Chapman, and T. A. Platts-Mills. 1992. House dust mite exposure as a cause of asthma. Clin. Exp. Allergy 22:897–906. 20. Peat, J. K., E. Tovey, B. G. Toelle, M. M. Haby, W. Xuan, and A. J. Woolcock. 1996. House dust mite allergens: a major risk factor for childhood asthma in Australia. Am. J. Respir. Crit. Care Med. 153: 141–146. 21. World Health Organization Regional Office for Europe. 1994. International workshop: setting priorities in environmental epidemiology. Arch. Environ. Health 49:239–245. 22. Larssen, S., F. Gram, L. O. Hagen, J. Huib, and O. Xander. 1994. Urbair. Bombay City Specific Report (draft 13.5.1994), Norwegian Institute for Air Research and Free University, Amsterdam. 23. Burney, P. G. J., C. Luczynska, S. Chinn, and D. Jarvis. 1994. For the Eu-

VOL 158

1998

ropean Community Respiratory Health Survey. Eur. Respir. J. 7:954– 960.

APPENDIX A Asthma and Allergy Screening Questionnaire Used in Phase I (3) 1. Have you had wheezing or whistling in your chest at any time in the last 12 months?

IF “NO” GO TO QUESTION 2 IF “YES”: 1.1. Have you been at all breathless when the wheezing noise was present? 1.2. Have you had this wheezing or whistling when you did not have a cold? 2. Have you woken up with a feeling of tightness in your chest at any time in the last 12 months? 3. Have you been woken by an attack of shortness of breath at any time in the last 12 months? 4. Have you been woken by an attack of coughing at any time in the last 12 months? 5. Have you had an attack of asthma in the last 12 months? 6. Are you currently taking any medicine (including inhalers, aerosols, or tablets) for asthma? 7. Do you have any nasal allergies including hay fever? 8. What is your date of birth 9. What is today’s date? 10. Are you male or female?

APPENDIX B Prediction Equations for Mean Lung Function Values Prediction equations for expected FEV1 values as suggested in the ECRHS manual (23): Males: FEV1 5 4.30 H 2 0.029 A 2 2.49 Females: FEV1 5 3.95 H 2 0.025 A 2 2.60 where H 5 height in meters, A 5 age in years (range, 25 to 44).