Effect of environmental interventions to reduce exposure to asthma ...

Journal of Exposure Analysis and Environmental Epidemiology (2004) 14, S133–S143 r 2004 Nature Publishing Group All rights reserved 1053-4245/04/$25.00

www.nature.com/jea

Effect of environmental interventions to reduce exposure to asthma triggers in homes of low-income children in Seattle TIM K. TAKARO,a,b JAMES W. KRIEGERa,b,c,d AND LIN SONGc a

University of Washington School of Public Health and Community Medicine, USA University of Washington School of Medicine, USA c Public Health – Seattle and King County, USA d Seattle Partners for Healthy Communities, USA b

The effectiveness of community health workers (CHWs) assisting families in reducing exposure to indoor asthma triggers has not been studied. In all, 274 low-income asthmatic children were randomly assigned to high- or low-intensity groups. CHWs visited all homes to assess exposures, develop action plans and provide bedding encasements. The higher-intensity group also received cleaning equipment and five to nine visits over a year focusing on asthma trigger reduction. The asthma trigger composite score decreased from 1.56 to 1.19 (D ¼ 0.37, 95% CI 0.13, 0.61) in the higher-intensity group and from 1.63 to 1.43 in the low-intensity group (D ¼ 0.20, 95% CI 0.004, 0.4). The difference in this measure due to the intervention was significant at the P ¼ 0.096 level. The higher-intensity group also showed improvement during the intervention year in measurements of condensation, roaches, moisture, cleaning behavior, dust weight, dust mite antigen, and total antigens above a cut point, effects not demonstrated in the low-intensity group. CHWs are effective in reducing asthma trigger exposure in low-income children. Further research is needed to determine the effectiveness of specific interventions and structural improvements on asthma trigger exposure and health. Journal of Exposure Analysis and Environmental Epidemiology (2004) 14, S133–S143. doi:10.1038/sj.jea.7500367

Keywords: asthma, indoor environment, antigen exposure, interventions, community health workers, Healthy Homes, inner city.

Introduction Asthma affects 15 million Americans (7% of the population), a third of them under the age of 18 years (Mannino et al., 2002). It is the most common chronic disease in children, the leading noninjury cause of hospitalization for children aged 0–15 years and the most common medical cause of missed school days (Graves and Kozak, 1998; Akinbami and Schoendorf, 2002; Mannino et al., 2002). Asthma prevalence, health service utilization, and mortality have increased among children and young adults in the US since 1980. The self-reported prevalence of childhood asthma increased by 75% between 1980 and 1994. From 1975 to 1993–1995, the estimated annual number of pediatric office visits for asthma more than doubled, from 4.6 million to 10.4 million, and the hospitalization rate also increased by 1.4% per year on average. The mortality of childhood asthma increased by 118% between 1978 and 1995 (Gergen, 1992; Mannino et al., 1998; Akinbami and Schoendorf, 2002).

1. Address all correspondence to: Dr. Tim K. Takaro, Occupational and Environmental Medicine, University of Washington, 4225 Roosevelt Way NE, Suite 100, Seattle, WA 98105, USA. Tel.: þ 1-206-616-7458. Fax: þ 1-206-616-4875. E-mail: [email protected]

Asthma is an immunologic disease triggered by specific allergens as well as respiratory irritants. These triggers induce airway inflammation and accompanying bronchial hyperresponsiveness. Exposure to indoor asthma triggers plays an important role in the development and exacerbation of childhood asthma (On allergens and asthma, 2001). Sensitized or atopic individuals are at greater risk of developing disease and are more likely to have severe disease (Sears et al., 1993; Nelson et al., 1999; Dharmage et al., 2001). Although we cannot yet quantify the precise role of the indoor environment in the increase in asthma, a variety of exposures concentrated in the indoor environment have been associated with asthma. The most reported exposures that trigger asthma are house dust mites (De Blay et al., 1992; Van der Heide et al., 1994; Carswell et al., 1996; Arlian and PlattsMills, 2001), environmental tobacco smoke (Burchfield et al., 1986; Weitzman et al., 1990; Young et al., 1991; Chilmonczyk et al., 1993), dampness and mold (Brunekreef et al., 1989; Verhoeff et al., 1995; Andriessen et al., 1998; Dharmage et al., 1999, 2001; Bush and Portnoy, 2001), household pets (Dales, 1991; De Blay et al., 1991; InfanteRivard, 1993; Bierman, 1996; Institute of Medicine, 2000), and cockroaches (Rosenstreich et al., 1997; Institute of Medicine, 2000; Eggleston and Arruda, 2001). Viral infections, endotoxins, and residues from combustion also play a role in childhood asthma (Johnston et al., 1995;

Environmental interventions for asthma triggers

Takaro et al.

Institute of Medicine, 2000; Wooton and Ashley, 2000). While rodents appear to be a significant asthma trigger in laboratory workers (Hollander et al., 1996; Nieuwenhuijsen et al., 2003), and the US National Cooperative Inner City Asthma Study found 19% of children allergic to rats (Kattan et al., 1997), the role of these pests in asthma is not well defined (Institute of Medicine, 2000). Reported exposure assessments vary widely, with many studies using both questionnaire data and quantitative environmental measures to characterize exposures. Dharmage et al. (1999) suggest that interview and visual inspection can provide valid measures of home environmental conditions when compared with the researcher’s assessment for cat antigen, relative humidity, and ergosterol (a surrogate for mold). Antigen assessment in house dust has been correlated with increases in asthma activity (Rosenstreich et al., 1997; Shapiro et al., 1999; Institute of Medicine, 2000; Platts-Mills et al., 2000; Carter et al., 2001). A few studies have demonstrated that home environmental interventions can reduce symptoms of asthma and bronchial hyper-responsiveness, through reduction of exposure to single triggers such as dust mite antigen (Shapiro et al., 1999; Platts-Mills et al., 2000; Carter et al., 2001; Maestrelli et al., 2001) and tobacco smoke (Greenberg et al., 1994), but none have assessed the benefit of a global reduction in indoor asthma triggers (Institute of Medicine, 2000). Most patients with asthma are sensitive and exposed to multiple allergens. Therefore, a global approach to reducing asthma triggers in the home environment is likely to be the most effective and efficient approach. Despite the lack of adequate evidence supporting such a comprehensive approach and the need for additional randomized controlled trials to test its efficacy, the American Academy of Asthma, Allergy and Immunology has taken a precautionary approach and recommended that physicians include indoor allergen avoidance measures in their therapeutic plan for patients with chronic allergic asthma (Eggleston and Bush, 2001). Their recommendations along with others (Institute of Medicine) target multiple exposures (e.g. moisture reduction impacting both mold and dust mites, and vacuuming deep dust from the carpet influencing an allergen reservoir). Deep dust tends to collect in carpets as they age even when they receive regular cleaning (Roberts et al., 1999). The surface dust is removed by normal cleaning, but the deep dust builds up in most carpets, and there is no practical way to monitor deep dust. On a carpet that has had normal vacuuming, the deep dust becomes surface and airborne dust after activity (Leese et al., 1997). The introduction of vacuum cleaners with dirt finders in the early 1990s made it possible to measure and control deep dust. Removing shoes at the door or using a commercial grade doormat retards track-in and also reduces the build up of deep dust. An approach that includes removing deep dust from carpets can also S134

reduce exposure to other hazards such as pesticides and carcinogens (Leese et al., 1997; Ott and Roberts, 1998; Roberts et al., 1999). The study described here was designed to test the effectiveness of using community health workers (CHWs) to promote such a global reduction in asthma trigger exposure in the homes of low-income children in order to decrease asthma morbidity. The exposure changes in homes receiving 1-year higher-intensity and 1-year lowerintensity CHW interventions are compared along with behavioral changes and knowledge associated with exposure reduction.

Methods The Seattle-King County Healthy Homes Project (SKCHH) is a randomized, controlled trial of an outreach/education intervention to improve asthma-related health status by reducing exposure to allergens, irritants, and toxicants in the home described in more detail by Krieger et al. (2002). We randomly assigned 274 low-income children with asthma aged 4–12 years to either a higher- or lower-intensity intervention group. In the higher-intensity group, CHWs conducted initial home environmental assessments, provided individualized action plans and made additional visits (average of 7) to each home over a 12-month period. The CHWs provided a protocol-defined package of materials to reduce exposures including bedding covers, vacuum cleaners with dirt finders (Hoover, Model V5270-930), and doublelayer, reduced emissions vacuum bags, commercial quality door mats, cleaning kits, mops, buckets, rubber gloves, food storage containers; assistance with roach and rodent eradication; and support for behavior changes. The dirt finder on the vacuum cleaner consisted of a red light and green light. The red light illuminated when the dust coming from the carpet created a sound on an impact plate at the top of the dirt tube. A microphone attached to the impact plate picked up this sound. The green light illuminated when the sound stopped. These lights provided real-time feedback that told the resident when they had thoroughly cleaned the deep dust from the carpet and where dust still remained in the carpet. Members of the lower-intensity group received the initial assessment, home action plan, limited education during the assessment visit, and bedding covers. At 1 year after joining, low-intensity group participants received the full package of resources and additional advice to address remaining indoor environmental quality concerns. Details of the CHWs training and intervention have been described previously (Krieger et al., 2002). Exposures were assessed by measuring antigen levels in dust samples, walk-through visual inspection, and a standardized interviewer-administered questionnaire. The questionnaire portions of the data were collected using the Journal of Exposure Analysis and Environmental Epidemiology (2004) 14(S1)


Healthy Homes Baseline Questionnaire and the Home Environmental Assessment List F II (HEAL-II), both of which are available on the SKCHH website (Public Health F Seattle and King County http://www.metrokc.gov/ health/asthma/healthyhomes/accessed15Apr, 2003). The latter was adapted from the HEAL developed by the Master Home Environmentalist Program of the American Lung Association of Washington (Dickey, 1998). Exposure questions included: tobacco smoke, exposure to allergen sources (mites, cockroaches, rodents, dust, pets), dust control behaviors (track-in, vacuuming/cleaning, use of allergencontrol bedding covers), mold and moisture problems, and contributing structural factors (condensation, water infiltration and damage, sources of leaks), ventilation (windows, fans, appliances, weatherization, heating, insulation, vapor barriers), structural conditions (carpeting, building age, condition of paint, structural deficits, recent remodeling), food debris and storage, trash, clutter, heating system filters and ducts, heating and cooking sources, location of garage, use and storage of hazardous and toxic products. The walk-through inspection corroborated parts of the questionnaire and documented visible mold, moldy odor, and water damage. We defined condensation presence as ‘‘windows often fog up in heating season’’ or ‘‘bathroom windows/mirror remain fogged for 415 min after shower’’; we defined a roach presence as visible evidence of roaches or self-report of roach problems in the past 3 months; rats as visible evidence of rats or self-report of rat problems in the past 3 months; mold as visible evidence of mold in the home; and a working bathroom fan as one able to hold 1-ply toilet paper to the grill from 901. Water damage was defined as visible water damage, moisture, or leaks in child’s bedroom, playing area, kitchen, or bathroom; and tobacco smoke as daily exposure to any smoker in the home. Prior to analyzing the data, summary scores were constructed for asthma triggers and moisture. The trigger score assigns one point for daily exposure to tobacco smoke, mold, roaches, pets, and poor ventilation in the home. The moisture score consists of condensation, water damage, or visible mold in any room, each given one point for a maximum of three. In addition to the exposure measures, intermediate outcomes included knowledge of asthma triggers assessed by counting the number of correct answers to 24 agree/ disagree questions and behaviors. We asked participants about many behaviors related to trigger exposure and control, and constructed a summary measure that assigns one point for each positive behavior noted. Primary healthrelated outcomes were: (1) urgent health-care utilization defined as any self-reported urgent health service utilization during the past 2 months (emergency department, hospital, or unscheduled clinic visit); (2) symptom days or nights defined as self-reported number of 24-h periods during the 2 weeks prior to interview with any daytime asthma symptoms (wheeze, tightness in chest, cough, shortness of breath, Journal of Exposure Analysis and Environmental Epidemiology (2004) 14(S1)

Takaro et al.

slowing down activities because of asthma) or nighttime symptoms (waking up because of wheeze, chest tightness, or cough); and (3) Pediatric Asthma Caregiver Quality of Life Scale score (which ranges from 1 to 7, with higher scores indicating better quality of life, and 0.5 considered to be clinically significant; Juniper et al., 1996). These results are reported in detail elsewhere (Krieger et al., manuscript under review). Dust samples to ascertain surface dust antigen concentrations were collected from the child’s bedroom or primary play area using the High Volume Small Surface Sampler (HVS3); a 1 m2 or larger area was sampled (ASTM Method D5438-00) (Roberts et al., 1999). The sample was sieved with 100 mesh to remove particles larger than 150 mm, then weighed and stored at 41C (ASTM, 1994). The stored dust samples were shipped on ice to Dr. Thomas Platts-Mills’ laboratory and analyzed for dust mite (Der p1 and Der f1), roach (Bla g2), cat (Fel d1), and dog (Can f1) antigen concentrations using validated ELISA techniques described in detail elsewhere (Chapman et al., 1988; Luczynska et al., 1989; de Groot et al., 1991). Owing to cost constraints, 60 homes were randomly selected for antigen analysis. For analysis, a threshold value of 5 mg/m2 was selected based upon a bimodal distribution in the data for the most prevalent antigen Der p1. This value was used as an indicator of potentially significant exposure for an asthmatic child. We assessed knowledge of asthma triggers by counting the number of correct answers to 24 agree/disagree questions. We assessed behaviors related to trigger exposure and control. These are described separately elsewhere (Krieger et al., manuscript under review), but a summary measure that assigns one point for each positive behavior present is included here. No participants crossed over from one arm to the other, so analysis was based on original allocation. w2 analysis, t-test, or Wilcoxon’s rank-sum test was used to compare group differences and the association between asthma triggers or cleaning behavior and the antigen concentration at baseline. We then used paired t, signed-rank, and McNemar tests to examine within-group baseline-to-exit changes among participants for whom both baseline and exit data were available. The effect of intervention on categorical behavior change and exposure variables was examined using logistic regression with exit behavior as the dependent variable, intervention group as the independent variable, and baseline behavior as covariate. We used generalized estimating equation (GEE) models with the robust option (using the Huber/White/ sandwich estimator of variance) and the independent withingroup working correlation structure (Hardin and Hilbe, 2003) to assess the intervention effect on continuous variables. Models included the outcome as the dependent variable, and group assignment (coded 0 for lower-intensity group and 1 for higher), time (baseline or exit), and the interaction of these two variables as independent variables. S135


Takaro et al.

Results

A significant intervention effect was present if the coefficient of the interaction term generated a P-value of o0.05. We tested for potential confounding by other baseline variables (child’s age, gender, and asthma severity; household income; caregiver’s race/ethnicity, employment status and education; and single caregiver status) by assessing whether inclusion of the variable changed the coefficient of the interaction term by more than 10%. No confounding was present, so no other variables were included in the models. A P-value of 0.05 for the interaction term was used as the threshold for statistical significance of across-group differences. We also used the GEE models to determine the presence of significant (P o0.05) interactions between the groups by time interaction term and each of the other variables. All data analyses were performed using Stata V7.0 (Stata Corporation, 1997). We obtained informed consent from caregivers, and assent from child participants aged 12 years and older. The Children’s Hospital and Regional Medical Center Institutional Review Board approved research protocols and participant contact. The project was sponsored by Seattle Partners for Healthy Communities, a CDC-funded Urban Research Center that promotes adherence to communitybased participatory research principles (Kone et al., 2000; Sullivan et al., 2001, 2002).

Of the 274 households randomized for study, 214 (78%) completed the exit measurements. As shown in Table 1, the mean age of the study sample at baseline was 7.3 years. Boys accounted for 59% of the sample. Of the primary caregivers, 84% were non-White, 83% were renters, 52% were employed, 39% had less than high school education, and 31% were single parent households. Asthma triggers such as roaches, pets, mold, moisture problems, and environmental tobacco smoke were noted in three-quarters of the homes at baseline. The prevalence of these triggers individually ranged from 18% to 44%. In all, 36% had two or more triggers (Figure 1). Participants were distributed across the spectrum of asthma severity. Threequarters had persistent asthma and nearly one-third of children had severe asthma. In total, 51 children began the study with persistent asthma, but upon exit only met the criteria for intermittent asthma. Caregivers reported a major impact on the quality of their lives from their children’s asthma. Randomization produced balanced lower- and higherintensity intervention groups on most baseline characteristics. Compared to the lower-intensity group, the higher-intensity

Table 1. Baseline demographic characteristics of study participants who completed the study compared with those randomized at entry All randomized

Completed exit

Noncomplete Antigen analysis

High (n ¼ 138) Low (n ¼ 136) High (n ¼ 110) Low (n ¼ 104) Total (n ¼ 214) (n ¼ 60)

Pre/post pairs (n ¼ 60)

Child’s age (mean years) Child’s gender (% male) Caregiver’s ethnicity (%) White African-American Vietnamese Other Asian Hispanic Other

7.4 55.8

7.3 61.8

7.4 55.5

7.3 62.5

7.3 58.9

7.2 58.3

7.9 63.3

11.6 31.9 25.4 8.7 17.4 5.1

20.6 27.2 22.1 5.2 16.9 8.1

10.0 31.8 27.3 10.9 14.6 5.5

21.2 26.9 26.0 5.8 15.4 4.8

15.4 29.4 26.6 8.4 15.0 5.1

18.3 30.0 13.3 1.7 25.0 11.7

16.7 30.0 23.3 11.7 15.0 3.3

Household income (%) o100% poverty 100–149% poverty 150–200% poverty

51.9 33.3 14.8

60.9 24.1 15.0

53.3 31.8 15.0

60.8 22.6 16.7

56.9 27.3 15.8

54.2 33.9 11.9

52.5 25.4 22.0

Caregiver’s education (%) Less than high school 40.9 High school graduate/GED 25.8 Some college 26.5 College graduate 6.8

37.6 27.8 25.6 9.0

41.9 27.6 23.8 6.7

36.6 29.7 23.8 9.9

39.3 28.6 23.8 8.3

39.0 20.3 33.9 6.8

28.8 39.0 25.4 6.8

Single caregiver HH (%)* Caregiver employed (%) Caregiver’s mean age (years) Caregiver rents home (%)

23.5 57.4 34.6 83.0

33.6 46.4 36.3 80.9

21.2 59.6 34.2 80.6

27.6 52.8 35.3 80.8

36.7 51.7 35.3 88.3

23.3 52.6 35.2 78.0

35.5 47.8 35.9 81.9

*Po0.05 for between-group comparisons among all participants randomized and those completing the study. No other statistically significant differences were observed.

S136

Journal of Exposure Analysis and Environmental Epidemiology (2004) 14(S1)


Takaro et al.

Figure 1. Baseline environmental characteristics of the higher- and lower-intensity groups.

Table 2. Baseline asthma severity and trigger exposure prevalence among study participants who completed the study compared with those randomized at entry All randomized

Smoker in home (%) Pet in home (%) Mold (%) Water damage/moisture/leak (%) Roaches (observed or reported) (%) Rodents (observed or reported) (%)#

Completed exit

High (n ¼ 138)

Low (n ¼ 136)

High (n ¼ 110)

Low (n ¼ 104)

39.9 19.9 41.1 17.8 19.6 0.0

41.9 26.9 46.2 23.9 15.6 3.5

36.4 19.3 39.8 19.4 20.9 0.0

42.3 22.1 44.8 24.0 14.4 4.2

#

Po0.05 for between-group comparison among all participants randomized and those completing the study.

group had significantly more single parent households (Pvalue ¼ 0.03) and lower caregiver quality of life (Pvalue ¼ 0.027). The differences between the two randomized groups on the other demographic, asthma severity, or exposure factors were not statistically significant. Those who completed the study differed in ethnicity from those who did not (more likely to be Asian and less likely to be Hispanic). Exposures were similar for those who completed the study compared to those who did not except for the rare presence of rodents, which were slightly more prevalent in the exit population (3.5% vs. 4.2%, Table 2). The completers were otherwise similar to the noncompleters with respect to demographic characteristics, asthma severity, and exposure to triggers at baseline. Table 3 shows changes in behaviors that impact exposure. Both groups showed significant improvement in dust control measures (three for higher, one for lower), use of allergy control covers and ventilation. Significant improvements Journal of Exposure Analysis and Environmental Epidemiology (2004) 14(S1)

were found for vacuuming frequency, vacuuming of upholstery, use of mattress and pillow covers for dust mites, use of doormats, and bathroom fan presence and use, but not for pets in home, stuffed toys in bedroom, bedding washing frequency, smoking behavior, or reduction of food clutter. As with the quantitative environmental laboratory measures, the low-intensity group also achieved behavioral improvements, although in other domains, that is, allergy control cover use, kitchen clutter, and stuffed toys. The subjects’ knowledge about exposures to asthma triggers also increased over the study in both groups, but was greater in the higherintensity group as summarized in the trigger knowledge score. The higher-intensity group showed significant improvements during the intervention year in mean scores for condensation, roaches, moisture score, dust weight, and asthma trigger composite score by sign-rank test, while the lower-intensity groups had no significant improvements (Table 4). The asthma trigger composite score decreased S137


Takaro et al.

Table 3. Behavior change across the 1-year intervention in higher- and lower-intensity groups High group (n ¼ 110)

Behaviors Vacuum child’s bedroom at least twice/2 weeksd Dust child’s bedroom at least twice/2 weeksd Vacuum cloth-covered furniture at least twice/2 weeks or remove itd Use doormat and/or remove shoesd Use allergy control covers on mattress and pillowsd Wash sheets weekly and use hot wash or rinsed No pet in the homed No pet in child’s bedroom Caregiver does not smoke Not allow smoking in the homed Wash stuffed animals or no stuffed animal Clutter in child bedroom Clutter in the kitchen Working bath exhaust fan present and usedd Working kitchen exhaust fan present and usedd

Low group (n ¼ 104)

a

Base

Exit

D (95% CI)

P

63.9

78.7

14.8 (4.5, 25.1)

63.9

70.4

31.0

b

Across-group effect

a

P

OR (95% CI)

Pc

0.706

2.2 (1.2, 4.2)

0.014

2.8 (15.0, 9.2)

0.639

1.5 (0.8, 3.1)

0.239

11.2 (1.0, 21.5)

0.028

2.2 (1.1, 4.4)

0.023

D (95% CI)

b

Base

Exit

0.004

62.5

64.4

1.9 (8.1, 11.9)

6.5 (4.9, 17.9)

0.262

69.2

66.3

63.0

32.0 (17.6, 46.4)

0.000

22.4

33.7

67.3

88.1

20.8 (8.9, 32.7)

0.000

70.6

77.5

6.9 (–3.9, 17.6)

0.209

2.9 (1.2, 6.7)

0.016

5.7

85.9

80.2 (57.3, 100)

0.000

7.8

71.8

64.1 (43.9, 84.2)

0.000

3.8 (1.2, 11.8)

0.020

47.3

41.8

5.5 (17.8, 6.9)

0.387

42.3

42.3

1.00

0.8 (0.4, 1.5)

0.447

81.5 93.5 78.9 80.0 39.8

81.5 93.5 83.5 77.3 47.2

0.0 0.0 4.6 2.7 7.4

(5.7, 5.7) (4.4, 4.4) (1.4, 10.6) (13.0, 7.5) (5.8, 20.6)

1.00 1.00 0.132 0.602 0.267

77.9 91.4 74.0 76.0 37.3

75.0 87.5 72.1 79.8 51.0

2.9 3.8 1.9 3.8 13.7

23.2 22.1 55.5

29.3 17.9 73.3

6.1 (4.5, 16.6) 4.2 (15.9, 7.5) 17.8 (5.1, 30.6)

0.257 0.480 0.004

36.6 19.8 68.1

26.9 9.9 68.1

9.7 (20.0, 0.6) 9.9 (20.0, 0.2) 0.0 (13.3, 13.3)

0.061 0.0495 1.00

1.9 (0.9, 3.8) 2.0 (0.7, 5.6) 2.0 (0.9, 4.2)

0.071 0.203 0.081

70.9

67.3

3.6 (13.6, 6.1)

0.465

54.8

70.2

15.4 (3.4, 27.4)

0.009

0.4 (0.2, 0.9)

0.016

0.0 (11.9, 11.9) (9.7,3.9) (10.4, 2.7) (8.5, 4.6) (5.0, 12.7) (1.6, 25.9)

0.405 0.248 0.564 0.394 0.023

Summary scores Behavior summary score Trigger knowledge score

1.0 1.3 1.5 0.7 0.57

(0.6, 1.8) (0.5, 3.3) (0.9, 2.6) (0.3, 1.5) (0.48, 0.67)

0.962 0.657 1.44 0.333 0.503

Coeff. (95% CI) 6.4 16.0

7.9 17.7

1.5 (1.1, 2.0) 1.6 (0.9, 2.4)

0.000 0.000

6.3 16.3

7.3 17.1

1.0 (0.6, 1.4) 0.9 (0.1, 1.6)

0.000 0.005

0.39 (0.21, 1.0) 0.203 0.61 (0.4, 1.63) 0.235

Change (exit F baseline value). Within-group comparison of baseline and exit values using the Wilcoxon’s matched-pairs test or the McNemar test. c OR refers to odds ratio for the high-intensity group relative to the low-intensity group calculated from the coefficient of the group time interaction term in the GEE model, adjusted for baseline differences in the dependent variable. d Items included in the behavior summary score. *Indicates P-value o0.05. **Indicates P-value o0.01. All values are percentage of group unless otherwise indicated. a

b

from 1.56 to 1.19 (D ¼ 0.37, 95% CI 0.13, 0.61) in the higher-intensity group and from 1.63 to 1.43 in the lowintensity group (D ¼ 0.20, 95% CI 0.004, 0.4). The difference in this measure due to the intervention was significant at the P ¼ 0.096 level. Antigen levels measured as micrograms of antigen per one square meter appear reduced in both higher- and lowerintensity intervention groups, except for cat in the higherintensity group and mite and dog in the lower-intensity group (Table 5). However, the variance is generally large. Der p1 (mite) and Can f1 (dog) in the higher-intensity group and Fel d1 (cat) in the lower-intensity group reached statistical significance. At baseline both groups had the same number of antigens above the theorized threshold effect level S138

as measured by surface dust loading of antigens (g/m2). A significant reduction in this measure was obtained in the higher-intensity group over the year-long intervention, corresponding to the reduction in total surface dust loading noted due to the intervention. When measuring antigens as concentration in total sieved dust (mg/g), no significant effects are seen within or across the intervention groups.

Discussion The relationship between indoor environmental exposures and asthma-related health disparities has sparked widespread enthusiasm for interventions to improve the environmental Journal of Exposure Analysis and Environmental Epidemiology (2004) 14(S1)


Takaro et al.

Table 4. Comparison of the 1-year intervention effect on exposures High (n ¼ 110)

Low (n ¼ 104)

Intervention effect

High (n ¼ 102), low (n ¼ 93)

Base

Exit

P

Base

Exit

P

P*

Condensation Mold Water damage Moisture score (the above 3) Roaches Rodents Pets inside the home Exposed to tobacco smoke daily Dust weight (g/m2 after 100 mesh)

61.3 39.2 17.7 1.2 20.9 7.3 18.1 14.2 2.64

34.0 35.3 20.6 0.9 8.2 9.1 18.1 17.0 0.96

0.000 0.564 0.590 0.033 0.003 0.617 1.00 0.549 0.008

59.2 46.2 24.7 1.3 14.4 7.7 22.1 21.4 2.67

49.5 45.2 21.5 1.1 7.7 7.7 25.0 22.3 1.74

0.077 0.853 0.513 0.116 0.071 1.00 0.405 0.808 0.172

0.013 0.214 0.933 0.150 0.831 0.694 0.328 0.585 0.042

1.56

1.19

0.005

1.63

1.43

0.069

0.096

Trigger score (mean)a

*Intervention effect P-value is based on the group time interaction term in the GEE model, adjusted for baseline differences in the dependent variable. a Trigger score includes condensation, mold, roaches, pets, and daily exposure to tobacco smoke, each contributing one point to the score.

Table 5. Antigen results, percent of homes above cutoff at baseline (base), and exit F all cases Higher (n ¼ 31)

Lower (n ¼ 31)

Base

Mean

SD

Exit

Mean

SD

P*

Base

Mean

SD

Exit

Mean

SD

P*

Surface loading (mg/m2) Fel d1 (cat) 4 ¼ 5 Can f1 (dog) 4 ¼ 5 Der f1 (mite1) 4 ¼ 5 Der p1 (mite2) 4 ¼ 5 Bla g2 (roach) 4 ¼ 5 Fraction w/antigens 4cutoff point

21.4 26.7 3.2 38.7 6.5 0.97

11.8 61.9 1.5 7.9 1.1

38.3 302 7.7 11.5 4.0

21.4 6.7 0 22.6 0 0.48

11.6 4.4 0.1 4.8 0.1

28.8 14.8 0.30 10.4 0.39

1.0 0.034 0.317 0.025 0.157 0.006

46.2 20.7 6.5 10.7 6.5 0.97

426 5.5 2.7 10.2 2.6

2132 13.5 12.9 45.9 12.6

19.2 27.6 6.5 18.9 0 0.65

141 13.1 1.0 8.3 0.1

573 33.7 3.2 28.9 0.37

0.020 0.414 1.0 0.414 0.157 0.185

Antigen concentration (mg/g) Fel d1 (cat) 4 ¼ 8 Can f1 (dog) 4 ¼ 8 Der f1 (mite1) 4 ¼ 2 Der p1 (mite2) 4 ¼ 2 Bla g2 (roach) 4 ¼ 2

12.9 9.7 3.2 35.5 6.5

16.1 9.7 0 32.3 3.2

12.7 9.6 0.07 4.7 0.15

41.1 36.6 0.19 9.1 0.75

0.655 1.0 0.317 0.655 0.317

22.6 16.1 6.5 22.6 6.5

112 4.1 0.59 3.4 0.25

575 9.6 2.5 10.5 0.7

19.4 12.9 3.2 29.0 3.2

106.8 7.9 0.44 3.5 0.34

539 23.4 1.5 5.5 1.9

0.655 0.655 0.317 0.480 0.317

Fraction w/antigens 4cutoff point

0.68

6.5 5.9 0.13 5.6 0.79

21.5 20.6 0.46 9.3 3.1

0.61

0.732

0.74

0.68

0.758

*Within-group McNemar’s test.

quality of homes, particularly those of low-income people and people of color. Exposure to asthma triggers was common in the diverse, low-income population reported here. Prevalence of roaches, pets, mold, moisture problems, and smokers ranged from 18% to 44% in this cohort. Threequarters of the homes had at least one trigger present, while 36% had two or more. This 1-year intervention by CHWs to reduce asthma triggers in the homes of low-income children demonstrated significant improvements in exposure-related behaviors. Vacuuming and other dust control measures, use of allergy control covers, and use of ventilation all improved. The intervention also demonstrated some reductions in indicators of exposure, with visible or reported condensation and Journal of Exposure Analysis and Environmental Epidemiology (2004) 14(S1)

surface dust loading statistically significant. The higherintensity group had significant reductions in dust mite and dog antigens, number of antigens with surface loading Z5 mg/m2 (cat, dog, dust mite, and roach), moisture, roach sightings, dust surface-loading, and composite asthma trigger score. One measure of exposure that was reduced in the lower-intensity group was surface loading of cat antigen. We measured clinical outcomes (asthma symptom days, asthma-related quality of lie, and asthma-related urgent health services utilization) and, as will be reported in a forthcoming publication, noted improvements in these measures (Krieger et al., manuscript under review). As demonstrated in the exposure behavior measures, some improvements were also seen in the lower-intensity intervenS139

Takaro et al.

tion group. The effectiveness of the intervention was shown in a population of low-income children with asthma recruited from across urban King County, including Seattle. Study participants were similar to those who did not enroll (Krieger et al., manuscript under review) and the intervention itself was performed by community members in a field setting similar to what would be feasible in a clinical setting given adequate resources for such outreach activity (Krieger et al., 2002). The approach may therefore have generalizability to similar low-income urban populations. Vacuuming behaviors were significantly improved in the higher-intensity group but not in the lower-intensity group, except for vacuuming of cloth furniture for which the lower group also marginally improved. Since one-quarter of the homes had no working vacuum cleaners at baseline, cleaning behaviors are likely to have been influenced heavily by the gift of the vacuum cleaner with dirt finder in the higherintensity group. Some of the trigger reduction effects were likely due to changes in cleaning behaviors. This is suggested by the correlations found between vacuuming at least twice a week in the child’s bedroom and the surface-loading dust (mg/m2), dog, cat, or at least one of the five antigens in collected dust. The correlations found between the presence of dogs, cats, and roaches assessed by observation and the detection of these antigens in dust samples supports the assumption that interventions directed at the sources of these exposures, as well as vacuuming itself, can reduce the surface loading of antigens. In the higher-intensity intervention group, surface loading for antigens appears reduced for all but cat antigen (four of five antigens), for number of antigens above the 5 mg/m2 cutoff. In the lower-intensity group, antigen loading appears decreased only for cat antigen. The reduction was statistically significant for two antigens in the higher-intensity group (dog and mite) and one in the lower-intensity group (cat) when adjusted for baseline levels. The antigen analysis was hampered by high variance in the measure and few homes with detectable antigens, for all but dust mite Der p1. The variance in pet antigens was especially high and may account for the inconsistencies in the higher and lower groups. Sampling from a larger number of homes or collection of more dust might have improved the consistency of this data. In a post hoc power analysis based upon reports of Der p1 reductions from published reports given an ability to detect a 0.45 mg/m2 difference across the intervention, 150 homes would be required to provide a power of 0.80 (alpha 0.05, beta 0.20). Dust mite antigens in this study were Der p1 and Der f1. The latter was rarely detected suggesting that Dematophagoides farinae is not a common species in Seattle. Surface loading for Der p1, the most commonly detected antigen, did not correlate strongly with vacuuming, nor did it correlate with water damage and visible mold (a food source for dust mites) as was expected. Humidity is required for dust mite growth, and water damage and visible mold may not be S140


reasonable surrogates for this measure. Dust mite antigen is a significant component of deep dust. Pet antigens adhere more to surfaces (O’Meara and Tovey, 2000) and may not be as large a proportion of deep dust. Recharge of the carpet surface from the reservoir of deep dust may make dust mites more resistant to thorough cleaning (Roberts et al., 1999). A clinically meaningful threshold for surface loading of antigens has not been established. We used the 5 mg/m2 cutoff because our most robust data (Der p1) were essentially bimodal around this midpoint. A concentration of 2 mg/g for Der p1 in vacuumed dust is considered to be clinically relevant as a hazard threshold for atopic children (PlattsMills et al., 1997). A comparison of our concentration data with surface loading suggests that such a threshold is reasonable. Both groups improved in their use of allergy control covers, behavior summary score, and trigger knowledge score, and both groups showed improvements in symptom days and caregiver quality of life. The improvements in the lower-intensity group for a limited number of end points might be attributable to the benefit of a single CHW visit that provided allergy control covers, education about asthma triggers, and an action plan for the home describing how to reduce triggers. Behavioral changes in the lower-intensity group were also achieved and may have contributed to reduced exposures. Other possibilities include apparent improvement because of temporal trends, regression to the mean, or Hawthorne effects (Greineder et al., 1998). We did not include a no-intervention control group because our community advisory board was against this approach, and we believed it was unethical to enroll participants in the study and defer for 1 year offering allergy control covers that have proven benefit (Shapiro et al., 1999; Carter, et al., 2001; Arlian and Platts-Mills, 2001). Increased use of allergy control covers may have been the primary reason for the improvements in asthma end points in the lower-intensity group, although the impact of the home visit and education with an action plan may also be contributing to the effect. The additional benefits noted in the higher-intensity group due to the intervention are therefore in addition to any benefit due to allergy control covers. There are several limitations to the study. The antigen exposure data are from a subset of randomly selected homes and 17% of the cohort did not have complete data for other measures. Based upon demographic analysis, the antigen subset appears similar to the complete cohort and is not likely to be a biased sample. The exposure literature is inconsistent in how toxicants in dust are reported. More recently measures of surface loading (mg/m2) have been used, often reported along with concentration (mg/g) (Douwes et al., 2000; Braun-Fuhrlander et al., 2002). Studies of blood lead in children correlate better with surface loading than concentration, suggesting that at least for young children this measure may more accurately reflect exposure, albeit Journal of Exposure Analysis and Environmental Epidemiology (2004) 14(S1)


through a different exposure route (Davies et al., 1990). Strong correlations are reported between toxicant levels reported in deep carpet dust and dust sampled using the high volume sampler method reported here (Dharmage et al., 2001). Roberts et al. (1999) suggest that deep dust serves as a reservoir for surface dust, continually recharging the surfaceloading component. Surface loading might therefore be expected to be more affected by vacuuming behavior than measures of concentration as seen here depending upon the period of recharge time and activity on the carpet. However, deep dust is probably also affected by the use of dirt-finder vacuum cleaners here as demonstrated in a convenience sample of homes in this study. In 74 homes, the total time to obtain three green lights on the dirt finder on three spots three feet apart on the carpet vacuumed was recorded using a previously described three-spot test (Roberts et al., 2001). The average three-spot test dropped 61% from 59 to 23 s, and the median time dropped 23% from 15 to 11.5 s after the new vacuum cleaners were deployed. Giving the family a new vacuum cleaner with a dirt finder provided advantages by reinforcing effective cleaning and appeared to motivate and empower the caregiver. The three-spot test was low in cost and popular with the CHWs because it could be done in less than 1 min and provided immediate results that could be seen by the CHW and the caregiver and may contribute to motivation to clean deep dust from the carpet. Some study homes had significant structural damage that was beyond the scope of our intervention. The most comprehensive intervention would be to provide affordable healthy housing, but significant fiscal and political barriers limit the feasibility of that approach, necessitating more modest strategies. A subsequent study is underway in Seattle to test the effectiveness and added benefit of structural improvements over the CHW-based intervention described here. More objective end points of the biological effect of moisture are needed, especially for allergenic or irritant molds. Studies using airborne measures of fungal exposure do not show consistent correlations with health outcome or visual assessments of water and mold damage (Dales et al., 1997; Ren et al., 2001). We attempted to measure mold exposure through measurements of viable mold cultured from vacuumed dust samples, but were not successful. In our follow-up intervention study, we are including ergosterol as a biochemical fungal marker to improve mold exposure characterization. Ergosterol is a constituent of fungal cell walls and reflects total fungal load (Bush and Portnoy, 2001). Ergosterol does not distinguish between allergenic fungi and nonallergenic fungi and therefore may not accurately represent the level of health hazard. Beta-D-glucan, another component of the fungal cell wall (in addition to some plant cell walls), functions as a respiratory irritant and has been used by many to estimate fungal contamination (Rylander, 1999; Douwes et al., 2000). Further validation of these Journal of Exposure Analysis and Environmental Epidemiology (2004) 14(S1)

Takaro et al.

measures and their relationship to standardized questionnaires is needed. Our findings show that CHWs can be trained to teach and provide support for these families and achieve reductions in asthma triggers and morbidity. The individual benefit of each component of the intervention package was not assessed, but significant benefit of the intervention was demonstrated above that produced by allergy covers alone. Behavioral changes are likely to contribute significantly to intervention effect. The benefits of such change may not be reflected in measurement of exposure reduction due to high variance or error in the measures, and problems of sensitivity and specificity in the instruments used. Efforts to further validate the instruments and model the impacts of behavior on exposure and clinical outcomes, along with the combination of these effects are needed. With the exception of Der p1, none of the exposures reported here have published clinical effect thresholds. The importance of multiple triggers above individual thresholds is also unknown. More work is needed to develop and rigorously test measures of the multiple exposures encountered in daily life by asthmatics and establish optimal exposure levels for these combinations. Until those efforts are completed, it appears prudent to reduce asthma trigger exposures in the home as low as can be reasonably achieved. Health plans should focus more effort on this preventive approach. Testing the effectiveness of an intervention which is in the middle range of intensity between those used here, would also be useful to ascertain if lower cost interventions achieve a favorable cost-effectiveness ratio. Childhood asthma continues to be a significant health problem for developed countries. Environmental triggers play an important role for children with asthma, particularly in those allergic to antigens commonly found in the home environment. Clinicians need additional tools to aid in the management of these children. This study demonstrates that a global approach to environmental interventions in the homes of low-income children with asthma is easily taught to and understood by families. The intervention can change behaviors related to asthma triggers, reduce many of the offending exposures, and improve asthma morbidity and caregiver quality of life.

Acknowledgements Primary funding was provided under NIEHS Grant #5 R21 ES09095, with additional support from Seattle Partners for Healthy Communities (CDC- U48/CCU009654-07) and the Nesholm Foundation. The project community home environmental specialists, Zhoni Gilbert, Jean Jackson, Nilsa Nicholson, Matthew Nguyen and LaTanya Wilson worked devotedly with their clients. Carol Allen and Georgiana Arnold coordinated field operations. S141

Takaro et al.

The Hoover Vacuum Company provided vacuum cleaners with dirt-finders at cost. Group Health Cooperative of Puget Sound donated 10 free slots in their Free and Clear tobacco cessation program. The Local Hazardous Waste Management Program of King County donated 300 green cleaning kits and pails. Aerotech Laboratories, Inc. provided reduced cost fungal analysis. Linda Graybird, Sharon Harris, Blythe Horman, and Scott Jones provided administrative support. Tianji Yu designed the database and tracking system. Lisa Lopez, Barbara Monsey, and Liz Quinn served as research coordinators. Kristy Seidel consulted on statistical analysis. Sanders Chai, Amy Duggan, Jane Koenig, John Roberts, James Stout, and Todd Yerkes participated on the project steering committee. Harriet Amman, David Bates, Thomas Platts-Mills, and Gail Shapiro served on the Scientific Advisory Group. Carol S. Collins, Ha Vu Minh Duong, Rochelle (Toni) Gibson, Rosie Williams Gordon, Augustine Evon Hampton, Doi Le, Celese McDuffie, Kelly (Trinh) Nguyen, Son Thuy Nguyen, Lauretta Perkins, Mary Tranh Pham, Quoi V. Phung, Debbie Rosenthal, Joann Sampson, Nura Sayed, and Robin Shields participated on the Parent Advisory Group.

References Akinbami L.J., and Schoendorf K.C. Trends in childhood asthma: prevalence health care utilization mortality. Pediatrics 2002: 110: 315–322. Andriessen J.W., Brunkekreef B., and Roemer W. Home dampness respiratory health status in European children. Clin Exp Allergy 1998: 28: 1191–1200. Arlian L.G., and Platts-Mills T.A.E. The biology of dust mites and the remediation of mite allergens in allergic disease. J Allergy Clin Immunol 2001: 107(Suppl.): 406–413. ASTM (American Society of Testing Materials). Standard Practices for Collection of Dust from Carpeted Floors for Chemical Analysis, Method D5438-00. ASTM, Philadelphia, PA, 1994. Bierman C.W. Environmental control of asthma. Immunol Allergy Clin N Am 1996: 16: 753–765. Braun-Fuhrlander C., Riedler J., and Herz U., et al. Environmental exposure to endotoxin and its relation to asthma in school-age children. N Eng J Med 2002: 347: 869–877. Brunekreef B., Dockery D.W., and Speixer F.E., et al. Home dampness respiratory morbidity in children. Am Rev Respir Dis 1989: 140: 1363–1364. Burchfield C.M., Higgins M.W., and Keller J.B., et al. Passive smoking in childhood: respiratory conditions and pulmonary function in Tecemseh, Michigan. Am Rev Respir Dis 1986: 133: 966–973. Bush R.K., and Portnoy J.M. The role abatement of fungal allergens in allergic disease. J Allergy Clin Immunol 2001: 107(Suppl.): 430–440. Carswell F., Birmingham K., and Oliver J., et al. The respiratory effects of reduction of mite allergen in the bedrooms of asthmatic children: a doubleblind controlled trial. Clin Exp Allergy 1996: 26: 386–396. Carter M.C., Perzanowski M.S., and Raymond A., et al. Home intervention in the treatment of asthma among inner-city children. J Allergy Clin Immunol 2001: 108: 732–737. Chapman M.D., Aalberse R.C., and Brown M.J., et al. Monoclonal antibodies to the major feline allergen Fel d I. II. Single step affinity purification of Fel d I, N-terminal sequence analysis, and development of a sensitive two-site immunoassay to assess Fel d I exposure. J Immunol 1988: 140: 812–818. Chilmonczyk B.A., Salmun L.M., and Megathlin K.N. Association between exposure to environmental tobacco smoke and exacerbations of asthma in children. N Engl J Med 1993: 328: 1665–1669.

S142


Dales R.E. Respiratory health effects of home dampness molds among Canadian children. Am J Epidemiol 1991: 134: 196–203. Dales R.E., Miller D., and McMullen E. Indoor air quality and health: and validity determinants of reported home dampness and moulds. Int J Epidemiol 1997: 26: 120–125. Davies D.J.A., Thornton J., and Watt J.M., et al. Lead intake and blood lead in two-year old UK urban children. Sci Total Environ 1990: 90: 13–29. De Blay F., Chapman M.D., and Platts-Mills T.A. Airborne cat allergen (Fel d I): environmental control with the cat in situ. Am Rev Respir Dis 1991: 143: 1334–1339. De Blay F., Pauli G., and Velten M., et al. Influence of mite exposure on symptoms of mite-sensitive patients with asthma. J Allergy Clin Immunol 1992: 90: 135–138. de Groot H., Goei K.G., and van Swieten P., et al. Affinity purification of a major and a minor allergen from dog extract: serologic activity of affinity-purified Can fI and of Can fI-depleted extract. J Allergy Clin Immunol 1991: 87: 1056–1065. Dharmage S., Bailey M., and Raven J., et al. A reliable and valid home visit report for studies of asthma in young adults. Indoor Air 1999: 9: 188–192. Dharmage S., Bailey M., and Raven J., et al. Current indoor allergen levels of fungi cats, but not house dust mites influence allergy and asthma in adults with high dust mite exposure. Am J Respir Crit Care Med 2001: 164: 65–71. Dickey P., (ed.). Master Home Environmentalist Training Manual. American Lung Association of Washington, Seattle, WA, 1998. Douwes J., Zuidhof A., and Doekes G., et al. (1-43)–beta-D- glucan and endotoxin in house dust and peak flow variability in children. Am J Resp Crit Care Med 2000: 162: 1348–1354. Eggleston P.A., and Arruda L.K. Ecology and elimination of cockroaches allergens in the home. J Allergy Clin Immunol 2001: 107(Suppl.): 422–429. Eggleston P.A., and Bush R.K. Environmental allergen avoidance: an overview. J Allergy Clin Immunol 2001: 107(Suppl.): 403–405. Gergen P.J. The increasing problem of asthma in the United States. Am Rev Respir Dis 1992: 146: 823–824. Graves E.J., and Kozak L.J. Detailed diagnoses procedures, National Hospital Discharge Survey, 1996. Vital Health Stat 1998: 13(138): 1–151. Greenberg R.A., Strecher V.J., and Bauman K.E., et al. Evaluation of a homebased intervention program to reduce infant passive smoking and lower respiratory illness. J Behav Med 1994: 17: 273–290. Greineder D.K., Loane K.C., and Parks P. Outcomes for control patients referred to a pediatric asthma outreach program: an example of the Hawthorne effect. Am J Management Care 1998: 4: 196–202. Hardin J.W., and Hilbe J.M. Generalized Estimating Equations. Chapman & Hill/ CRC, New York, 2003. Hollander A., Doekes G., and Heederik D. Cat and dog allergy total IgE as risk factors of laboratory animal allergy. J Allergy Clin Immunol 1996: 98: 545–554. Infante-Rivard C. Childhood asthma and indoor environmental risk factors. Am J Epidemiol 1993: 137: 834–844. Institute of Medicine. Clearing the Air: Asthma and Indoor Air Exposures. National Academy Press, Washington, DC, 2000. Johnston S.L., Pattemore P.K., and Sanderson G., et al. Community study of role of viral infections in exacerbations of asthma in 9–11 year old children. BMJ 1995: 310: 1225–1229. Juniper E.F., Guyatt G.H., and Feeny D.H., et al. Measuring quality of life in the parents of children with asthma. Qual Life Res 1996: 5: 27–34. Kattan M., Mitchell H., and Eggleston P., et al. Characteristics of inner-city children with asthma: the National Cooperative Inner-City Asthma Study. Pediatr Pulmonol 1997: 24: 253–262. Kone A., Sullivan M., Senturia K., Chrisman N., Ciske S., and Krieger J. Improving collaboration between researchers and communities. Public Health Rep 2000: 115: 243–248. Krieger J.K., Takaro T.K., and Allen C., et al. The Seattle-King County Healthy Homes Project: implementation of a comprehensive approach to improving indoor environmental quality for low-income children with asthma. Environ Health Perspect 2002: 110(Suppl. 2): 311–322. Krieger J.W., Takaro T.K., Song L., and Weaver M. The Seattle-King County Healthy Homes Project: A randomized controlled trail of a community health worker intervention to decrease exposure to indoor asthma triggers (Under review March 2004). Leese K.E., Cole E.C., and Hall R.M., et al. Measurement of airborne and floor dust in a non problem building. Am Indus Hyg Assoc J 1997: 58: 432–438.



Luczynska C.M., Arruda L.K., and Platts-Mills T.A., et al. A two-site monoclonal antibody ELISA for the quantification of the major Dermatophagoides spp. allergens, Der p I and Der f I. J Immunol Methods 1989: 118: 227–235. Maestrelli P., Zanolla L., and Puccinelli P., et al. Low domestic exposure to house dust mite allergens (Der p 1) is associated with a reduced non-specific bronchial hyperresponsiveness in mite-sensitized asthmatic subjects under optimal drug treatment. Clin Exp Allergy 2001: 31: 715–721. Mannino D.M., Homa D.M., and Akinbami L.J., et al. Surveillance for asthma F United States, 1980–1999. MMWR Morb Mortal Wkly Rep, (CDC Surveillance Summary), 2002: 51: 1–13. Mannino D.M., Homa D.M., and Pertowski C.A., et al. Surveillance for asthma F United States, 1960–1995. MMWR Morb Mortal Wkly Rep, (CDC Surveillance Summary), 1998: 47: 1–27. Nelson H.S., Szefler S.J., and Jacobs J., et al. The relationships among environmental allergen sensitization allergen exposure, pulmonary function, and bronchial hyperresponsiveness in the Childhood Asthma Management Program. J Allergy Clin Immunol 1999: 104(Part I): 775–785. Nieuwenhuijsen M.J., Putcha V., and Gordon S., et al. Exposure–response relations among laboratory animal workers exposed to rats. Occup Environ Med 2003: 60: 104–108. On allergens and asthma (again). Does exposure to allergens in homes exacerbate asthma [editorial]. Clin Exp Allergy 2001: 31: 670–673. O’Meara T., and Tovey E. Monitoring personal allergen exposure. Clin Rev Allergy Immunol 2000: 18: 341–395. Ott W.R., and Roberts J.W. Everyday exposure to toxic pollutants. Sci Am 1998: 278: 86–92. Platts-Mills T.A., Vaughan J.W., and Carter M.C., et al. The role of intervention in established allergy: avoidance of indoor allergens in the treatment of chronic allergic disease. J Allergy Clin Immunol 2000: 106: 787–804. Platts-Mills T.A., Vervloet D., and Thomas W.R., et al. Indoor allergens asthma: report of the third international workshop. J Allergy Clin Immunol 1997: 100: S2–S24. Public Health: Seattle and King County. Seattle Healthy Homes Project: Environmental Interventions to Improve Children’s Health. [Public Health – Seattle and King County Web site]. Available at:http://www.metrokc.gov/ health/asthma/healthyhomes/.Accessed March 30, 2003. Ren P., Jankun T.M., and Belanger K., et al. The relation between fungal propagules in indoor air and home characteristics. Allergy 2001: 56: 419–424.


Takaro et al.

Roberts J.W., Clifford W.S., and Glass G., et al. Reducing dust, lead, dust mites, bacteria, and fungi in carpets by vacuuming. Arch Environ Contam Toxicol 1999: 36: 477–484. Roberts J.W., Glass G., Krieger J., and Song L. The 3-spot test for estimating exposure from deep dust in carpets. Proceeding of the 11th Annual Conference of the International Society of Exposure Analysis. Charleston, SC, 2001, 448 pages. Rosenstreich D.L., Eggleston P., and Kattan M., et al. The role of cockroach allergy and exposure to cockroach allergen in causing morbidity among innercity children with asthma. N Engl J Med 1997: 336: 1356–1363. Rylander R. Indoor Air-related effects and airborne (1-3)-b-d-glucan. Environ. Health Pespect 1999: 107(suppl 3): 501–503. Sears M.R., Burrows B., and Herbison G.P., et al. Atopy in childhood, II: relationship to airway responsiveness, hay fever and asthma. Clin Exp Allergy 1993: 23: 949–956. Shapiro G.G., Wighton T.G., and Chinn T., et al. House dust mite avoidance for children with asthma in homes of low-income families. J Allergy Clin Immunol 1999: 103: 1069–1074. Stata Corporation. Stata Version 7.0, 1997. Sullivan M., Kone A., Senturia K.D., Chrisman N.J., Ciske S.J., and Krieger J.W. Researcher and researched-community perspectives: toward bridging the gap. Health Educ Behav 2001: 28: 130–149. Sullivan S.D., Weiss K.B., and Lynn H., et al. for the National Cooperative InnerCity Asthma Study Investigators. The cost-effectiveness of an inner-city asthma intervention for children. J Allergy Clin Immunol 2002: 110: 576–581. Van der Heide S., de Monchy J.G.R., and de Vries K., et al. Seasonal variation in airway hyperresponsiveness and natural exposure to house dust mite allergens in patients with asthma. J Allergy Clin Immunol 1994: 93: 470–475. Verhoeff A.P., Van Strien R.T., and van Wijnen J.H., et al. Damp housing and childhood respiratory symptoms: the role of sensitization to dust mites and molds. Am J Epidemiol 1995: 141: 103–110. Weitzman M., Gortmaker S., and Walker D.K., et al. Maternal smoking and childhood asthma. Pediatrics 1990: 85: 505–511. Wooton M., and Ashley P. Residential Hazards: Asthma. Healthy Homes Initiative Background Information. US Department of Housing and Urban Development, Washington, DC, 2000. Young S., Le Souef P.N., and Geelhoed G.C., et al. The influence of a family history of asthma and parental smoking on airway responsiveness in early infancy. N Engl J Med 1991: 324: 1168–1173.

S143