Sources of household air pollution: The association with lung function and respiratory symptoms in middle-aged adult

Environmental Research 164 (2018) 140–148

Contents lists available at ScienceDirect

Environmental Research journal homepage: www.elsevier.com/locate/envres

Sources of household air pollution: The association with lung function and respiratory symptoms in middle-aged adult

T

Laurent Deviena,b, Jonathan Giovannellia,b, Damien Cunyc, Régis Matranc,d, Philippe Amouyela,b, ⁎ Sébastien Huloc,d, Jean Louis Edméc,d, Luc Daucheta,b, a

Univ. Lille, Institut Pasteur de Lille, INSERM U1167 - RID-AGE Risk factors and molecular determinants of aging-related diseases, F-59000 Lille, France CHU Lille, Epidemiology Service, Health Economics and Prevention, F-59000 Lille, France c Univ. Lille, EA4483 - IMPECS (IMPact of Environmental ChemicalS on human health), F-59000 Lille, France d Pulmonary Function Testing Department, CHU Lille, F-59000 Lille, France b

A R T I C L E I N F O

A B S T R A C T

Keywords: Lung function Mould Household air pollution Respiratory symptom Cross-sectional study

Introduction: The objective of the present study was to investigate the relationship between sources of household air pollution, respiratory symptoms and lung function. Methods: 3039 adults aged from 40 to 65 participated in the 2011–2013 ELISABET cross-sectional survey in northern France. Lung function was measured using spirometry. During a structured interview, respiratory symptoms, household fuels, exposure to moulds, and use of ventilation were recorded on a questionnaire. Results: The self-reported presence of mould in at least two rooms (not including the bathroom and the kitchen) was associated with a 2.5% lower predicted forced expiratory volume in 1 s (95% confidence interval, −4.7 to −0.29; p-trend < 0.05) and a higher risk of wheezing (p-trend < 0.001). Visible condensation was associated with wheezing (p < .05) and chronic cough (p < .05). There were no significant associations with the type of household fuel or inadequate ventilation/aeration. Similar results were found when the analyses were restricted to participants without known respiratory disease. Conclusion: Our results suggest that the presence of mould (known to be associated with more severe asthma symptoms) could also have an impact on respiratory symptoms and lung function in the general population and in populations without known respiratory disease.

1. Introduction 1.1. Context Outdoor air pollution has a major health impact on the general population (Pascal et al., 2013). It is notably associated with the incidence and exacerbations of cardiovascular and respiratory diseases (Mannucci et al., 2015), such as asthma and chronic obstructive pulmonary disease (COPD) (Berend, 2016). A number of studies have shown that exposure to air pollution has an impact on lung function (Adam et al., 2015; de Jong et al., 2016; Rice et al., 2015). Indoor air pollution is an especially important issue because many people spend two-thirds of their time at home (Brasche and Bischof, 2005), and much of the rest of their time in other buildings. Moreover, rooms that lack ventilation may have higher concentration of pollutants (including PM10 (Dorizas et al., 2015; Zhou et al., 2014)) than outdoor

environments (Kattan et al., 2007; Schneider et al., 2001). The measure of household air pollution is difficult because (i) it requires individual measures, and (ii) the sources of pollution are heterogeneous. In the literature, two approaches for evaluating household air pollution have been described. The first consists in performing quantitative measurements of particle matter, nitrogen dioxide, carbon monoxide, volatile organic compounds, spores, and so on. (Bentayeb et al., 2015). Quantitative studies often have small sample size, and the measurement are only valid for a given point in time. The second approach consists in estimating the sources of pollution in a non-quantitative manner using questionnaires. These results can be more readily exploited for prevention because modifiable sources of pollution can be identified. These studies have often a larger size sample. In developing countries, the household air pollution resulting from cooking and heating is a known risk factor for COPD (Kurmi et al., 2010). In developed countries, smoke from biomass is less of a problem. However, household air pollution

Abbreviation: COPD, chronic obstructive pulmonary disease; ECRHS, European community respiratory health survey; FEV1, volume in one second; FVC, forced vital capacity ⁎ Correspondence to: Service d’épidémiologie, économie de la santé et prévention, Maison Régionale de la Recherche Clinique, CHRU de Lille, 2 rue du Pr Laguesse, F-59037 Lille Cedex, France. E-mail address: [email protected] (L. Dauchet). https://doi.org/10.1016/j.envres.2018.02.016 Received 12 October 2017; Received in revised form 8 February 2018; Accepted 12 February 2018 0013-9351/ © 2018 Elsevier Inc. All rights reserved.

Environmental Research 164 (2018) 140–148

L. Devien et al.

because the calculation of predicted values as a function of height would have been irrelevant (Miller et al., 2005). Respiratory symptoms (such as wheezing and chronic cough) were reported on the standardized Medical Research Council questionnaire (Medical Research Council’s Committee on Environmental and Occupational Health, 1986). Participants were considered to have wheezing if they had experienced wheezing or whistling in their chest at any time in the previous 12 months. Chronic cough was defined as cough on most days for at least three months each year in the winter.

results from several other sources, such as mould, dampness (Fisk et al., 2007), outdoor air pollution (Sarnat et al., 2000), environmental tobacco smoke (IARC Working Group on the Evaluation of Carcinogenic Risks to Humans, 2004; Repace and Lowrey, 1980), building materials, household cleaning products, and so on. Furthermore, sources of household air pollution (such as mould and dampness) have been linked to the exacerbation of asthma (Mendell et al., 2011). Lastly, ventilation with outdoor air is intended to remove pollutants emitted from indoor sources. Inadequate ventilation may be associated with a greater frequency of lower respiratory tract problems and asthma symptoms (Sundell et al., 2011), and with more pronounced effects of some air pollutants on respiratory health (Bentayeb et al., 2015). Some previous studies have reported a non-significant association between exposure to mould and poorer respiratory function (Ebbehøj et al., 2005; Gunnbjörnsdottir et al., 2003; Hernberg et al., 2014; Norbäck et al., 2011).

2.3. Assessment of exposure Exposure was measured using a questionnaire. Participants were asked whether or not they had mould in their dining room, living room, kitchen, own bedroom, other bedroom, bathroom or other rooms. The mould score was defined as the stated number of rooms with mould (i.e. the number of “yes” answers, ranging from 0 to 7). In previous studies (Norbäck et al., 2013; Pekkanen et al., 2007), the association with asthma has been found to be stronger for mould exposure in the living area (i.e. the living room and the bedroom) than for exposure in the kitchen or bathroom. Therefore, we analyzed the data for the living area and the non-living area (kitchen or bathroom) separately. The mould score in the living area (ranging from 0 to 5) was defined as the number of “yes” answers to the questions concerning the dining room, living room, own bedroom, other bedroom and other rooms (excluding toilets and utility rooms, in this case). The mould score in the non-living area (ranging from 0 to 3) was defined as the number of “yes” answers to the question concerning the kitchen, the bathroom and other rooms (toilets and utility rooms only, in this case). Cellars and attics were not considered. The presence of household fuel was defined as presence in the household of a gas stove, a boiler not connected to an outside exhaust or an auxiliary source of heating with household fuel (such as an open fireplace or oil- or coal-burning stove) used on more than 14 days a year. In order to evaluate the impact of exposure time, we performed a sensitivity analysis of exposure to an auxiliary source of heating used throughout the winter. The presence of condensation was defined as an answer of “often” or “always” to the question “Is there occasional condensation on the windows in your dwelling”. Ventilation was defined in the present study as a system (whether mechanical or nonmechanical) enabling the movement of outdoor air around an indoor space. Ventilation was considered to be adequate if the accommodation had mechanical ventilation or natural ventilation that was not obstructed in the winter. Aeration was defined in the present study as opening a room’s windows. Aeration was considered to be adequate if the room was aired more than once a day in summer and in winter. Exposure to passive smoking was defined as the presence of a current smoker living and smoking in the household or the presence of smokers in the household at least once a month. Information on passive smoking was collected for non-smokers only.

1.2. Objective The objective of the present study of middle-aged adults in northern France was to investigate the relationship between sources of household air pollution (mould, window condensation, inadequate ventilation/airing, and household fuels) on one hand and lung function and respiratory symptoms (wheezing and chronic cough) on the other. 2. Methods 2.1. Study population The study included adults aged from 40 to 65 participating in the 2011–2013 Enquête Littoral Souffle Air Biologie Environnement (ELISABET) cross-sectional survey in northern France. The methodology of the ELISABET Study has been described in detail elsewhere (Clement et al., 2017; Giovannelli et al., 2016; Hulo et al., 2016; Quach et al., 2015). Briefly, the study sample is representative of the general population in the Lille and Dunkirk urban areas. The participation rate was 32.9% (3276 out of 9945 potentially eligible participants). Data were collected at home or (very occasionally) during a consultation in a healthcare establishment by 12 nurse investigators. In all cases, a trained, registered nurse administered a detailed questionnaire and performed spirometry testing. The study protocol was approved by the local independent ethics committee (CPP Nord Ouest IV, reference 2010-A00065-34; ClinicalTrials.gov identifier: NCT02490553), in compliance with the French legislation on biomedical research. All participants provided their written, informed consent to participation in the study. 2.2. Outcome assessments Spirometric lung function was assessed in terms of the forced expiratory volume in one second (FEV1), the forced vital capacity (FVC), and the FEV1/FVC ratio. Spirometry was performed according to the 2005 American Thoracic Society and European Respiratory Society guidelines (Miller et al., 2005). Values were expressed as a percentage of the predicted value (100 x observed value/predicted value) for the participant’s age, height and gender, using previously developed equations (Quanjer et al., 2012). The spirometers (Micro 6000, Medisoft, Sorinnes, Belgium) were calibrated weekly. No bronchodilators were administered. For each participant, the spirometry test was repeated (up to seven times) until three acceptable, reproducible flowvolume loops were obtained, according to the same guidelines (Miller et al., 2005). The highest acceptable values of FEV1 and FVC were selected for statistical analysis. All spirometry data were validated by an experienced, specialist physician (JLE). Participants lacking acceptable spirometry results were excluded from the analysis. Women under 145 cm in height and men under 155 cm in height were also excluded

2.4. Covariables The following variables were recorded: age, gender, educational status (the number of years of full-time education, including primary school), smoking status, height, body mass index (BMI), level of income, number of people living in the household, population density in the locality, and the investigator. Tobacco exposure was estimated from the self-reported smoking status as either a “current smoker” (i.e. at least one cigarette per day for the previous 12 months), a “former smoker” or a “never smoker”. Population density data for each locality were sourced from the French National Institute for Statistics and Economic Studies’ database (INSEE, 2013, 2011). The “level of income” variable was missing for 801 participants, and so a “missing data” modality was created in this instance. 141

Environmental Research 164 (2018) 140–148

L. Devien et al.

Table 1a characteristics of the study population, as a function of domestic exposure to mould and household fuels. Total study population n = 3039

Men, n (%) 1436 (47.3%) Centre, n (%) Dunkirk urban area 1478 (48.6%) Lille urban area 1561 (51.4%) Age (years), mean (SD) 53.15 (7.23) Height (cm), mean (SD) 169.3 (9.15) 27.07 (5.18) BMI (kg/m2), mean (SD) Smoking status, n (%) Never smoker 1527 (50.2%) Former smoker 943 (31%) Current smoker 569 (18.7%) Educational level, n (%) More than 17 years of full-time 571 (18.8%) education From 15–17 years of full-time 568 (18.7%) education From 12–15 years of full-time 543 (17.9%) education From 9–12 years of full-time 1051 (34.6%) education Less than 9 years of full-time 306 (10.1%) education Income level (euros per year), n (%) < 15,000 208 (6.8%) 15,000– 30,000 694 (22.8%) 30,000– 45,000 685 (22.5%) > 45,000 639 (21%) Didn't answer the question. 813 (26.8%) Population density, people per km2 2726 (1846.01) (SD) Number of people in the household 2.84 (1.35) (n), mean (SD) Spirometry FEV1% predicted (%), mean (SD) 96.87 (15.76) FVC % predicted (%), mean (SD) 101.1 (14.37) FEV1/FVC % predicted (%), mean 95.41 (8.67) (SD) Symptoms, n (%) Wheezing 484 (15.9%) Chronic cough 276 (9.1%) Exposure number of rooms in the living area with mould, n (%) 0 2429 (79.93%) 1 407 (13.39%) ≥2 199 (6.55%) Indoor relative humidity (%), mean (SD) Window condensation, n (%) 186 (6.1%) Inadequate ventilation or aeration, n 349 (11.5%) (%) Household fuels, n (%) 2171 (71.4%)

Number of rooms in the living area with mould

Household fuels

0 (n = 2429)

1 (n = 407)

> 2 (n = 199)

pa

No (n = 860)

Yes (n = 2171)

pb

1187 (48.9%)

164 (40.3%)

83 (41.7%)

< 0.01 < 0.05

423 (49.2%)

1008 (46.4%)

0.18 < 0.05

1168 (48.1%) 1261 (51.9%) 53.41 (7.24) 169.4 (9.23) 27.08 (5.07)

193 (47.4%) 214 (52.6%) 52.34 (7.11) 168.5 (9.03) 26.83 (5.29)

117 (58.8%) 82 (41.2%) 51.64 (7.02) 169.3 (8.3) 27.32 (6.1)

389 (45.2%) 471 (54.8%) 53.42 (7.46) 170.3 (8.89) 26.64 (4.73)

1085 (50%) 1086 (50%) 53.03 (7.14) 168.8 (9.21) 27.24 (5.34)

1227 (50.5%) 755 (31.1%) 447 (18.4%)

196 (48.2%) 130 (31.9%) 81 (19.9%)

103 (51.8%) 56 (28.1%) 40 (20.1%)

447 (52%) 297 (34.5%) 116 (13.5%)

1075 (49.5%) 643 (29.6%) 453 (20.9%)

472 (19.4%)

71 (17.4%)

28 (14.1%)

210 (24.4%)

361 (16.6%)

443 (18.2%)

84 (20.6%)

40 (20.1%)

195 (22.7%)

372 (17.1%)

440 (18.1%)

64 (15.7%)

39 (19.6%)

168 (19.5%)

372 (17.1%)

826 (34%)

151 (37.1%)

72 (36.2%)

235 (27.3%)

814 (37.5%)

248 (10.2%)

37 (9.1%)

20 (10.1%)

52 (6%)

252 (11.6%)

157 (6.5%) 528 (21.7%) 543 (22.4%) 538 (22.1%) 663 (27.3%) 2731 (1846.98) 2.76 (1.31)

35 (8.6%) 105 (25.8%) 99 (24.3%) 72 (17.7%) 96 (23.6%) 2846 (1867.28) 3.08 (1.43)

14 (7%) 61 (30.7%) 42 (21.1%) 28 (14.1%) 54 (27.1%) 2388 (1751.16) 3.36 (1.44)

166 (7.6%) 516 (23.8%) 474 (21.8%) 391 (18%) 624 (28.7%) 2685 (1841.07)

0.05

< 0.0001

42 (4.9%) 177 (20.6%) 206 (24%) 246 (28.6%) 189 (22%) 2830 (1857.37) 2.73 (1.3)

2.88 (1.36)

< 0.01

97.15 (15.75) 101.3 (14.45) 95.5 (8.62)

96.31 (15.53) 100.5 (13.9) 95.34 (8.62)

94.63 (16.38) 99.7 (14.34) 94.45 (9.47)

< 0.05 0.09 0.10

98.56 (15.28) 102.5 (14.24) 95.79 (8.01)

96.2 (15.92) 100.5 (14.39) 95.27 (8.93)

< 0.001 < 0.001 0.13

355 (14.6%) 221 (9.1%)

90 (22.1%) 35 (8.6%)

38 (19.1%) 18 (9%)

< 0.0001 0.58

108 (12.6%) 67 (7.8%)

376 (17.3%) 208 (9.6%)

< 0.01 0.14

– – – 56.87 (11.37)

– – – 58.77 (10.89)

– – – 59.56 (11.19)

< 0.001

730 (84.9%) 87 (10.1%) 41 (4.8%) 57.26 (11.57)

1693 (78%) 318 (14.6%) 158 (7.3%) 57.33 (11.23)

0.89

89 (3.7%) 257 (10.6%)

53 (13%) 56 (13.8%)

44 (22.1%) 33 (16.6%)

< 0.0001 0.07

32 (3.7%) 88 (10.2%)

154 (7.1%) 260 (12%)

< 0.001 0.19

1693 (69.7%)

318 (78.1%)

158 (79.4%)

< 0.05

–

–

–

< 0.001 0.47 0.83 0.96

0.63

< 0.0001

< 0.01

0.08

0.19 < 0.0001 < 0.01 < 0.0001

< 0.0001

< 0.001

Missing data : wheezing : 1; chronic cough : 13; mould : 5; indoor relative humidity : 860; relative humidity observed - expected : 1 179; inadequate ventilation or aeration : 3; household fuels : 8. a p-trend for binary and quantitative variables, chi-squared for qualitative variables. b chi-squared for qualitative variables, and an ANOVA for comparison of means.

rooms per household member, educational level, gender, age, height, smoking status, BMI, population density, income level, number of people living in the household, and the investigator. Height is a determinant of lung volume; even though it is taken into account in the observed and predicted volumes, a residual association with height is possible. The “height” variable was not included in the logistic regression model. The final model for lung function variables was adjusted for educational level, gender, age, height, smoking status, BMI, population density, income level, number of people living in the household, and the investigator. To comply with the logistic regression model’s validity condition, we selected no more than one variable for 10 events. The p-trend for mould scores was calculated by considering

2.5. Statistical methods An analysis of variance was used to compare mean values, and a chisquared test was used to compare percentages. Multivariate statistical analyses were adjusted for known individual confounders (Tables 1a, 1b). A linear mixed model was used for quantitative variables. The “nurse investigator” variable (with 12 modalities) was defined as the random effect variable. Logistic regression was applied to qualitative variables. Using stepwise regression, the best-fitting models were determined separately in logistic regression and linear mixed regression analyses by including the following variables: centre (Dunkirk or Lille), number of 142

Environmental Research 164 (2018) 140–148

L. Devien et al.

Table 1b characteristics of the study population, as a function of domestic exposure to condensation and inadequate ventilation or aeration. Window condensation

Men, n (%) Centre, n (%) Dunkirk urban area Lille urban area Age (years), mean (SD) Height (cm), mean (SD) BMI (kg/m2), mean (SD) Smoking status, n (%) Never smoker Former smoker Current smoker Educational level, n (%) More than 17 years of full-time education From 15–17 years of full-time education From 12–15 years of full-time education From 9–12 years of full-time education Less than 9 years of full-time education Income level (euros per year), n (%) < 15,000 15,000– 30,000 30,000– 45,000 > 45,000 Didn't answer the question. Population density, people per km2 (SD) Number of people in the household (n), mean (SD) Spirometry FEV1% predicted (%), mean (SD) FVC % predicted (%), mean (SD) FEV1/FVC % predicted (%), mean (SD) Symptoms, n (%) Wheezing Chronic cough Exposure number of rooms in the living area with mould, n (%) 0 1 ≥2 Indoor relative humidity (%), mean (SD) Window condensation, n (%) Inadequate ventilation or aeration, n (%) Household fuels, n (%)

Inadequate ventilation or aeration

No (n = 2 853)

yes (n = 186)

pa

No (n = 2 685)

Yes (n = 349)

pa

1360 (47.7%)

76 (40.9%)

0.08 0.17

1263 (47%)

170 (48.7%)

0.59 0.07

1397 (49%) 1456 (51%) 53.29 (7.19) 169.3 (9.17) 27.08 (5.15)

81 (43.5%) 105 (56.5%) 50.96 (7.5) 168.9 (8.78) 26.97 (5.56)

1289 (48%) 1396 (52%) 53.14 (7.19) 169.3 (9.12) 27.02 (5.12)

186 (53.3%) 163 (46.7%) 53.22 (7.49) 169.1 (9.41) 27.43 (5.62)

1448 (50.8%) 892 (31.3%) 513 (18%)

79 (42.5%) 51 (27.4%) 56 (30.1%)

1344 (50.1%) 842 (31.4%) 499 (18.6%)

181 (51.9%) 99 (28.4%) 69 (19.8%)

538 525 513 988 289

33 43 30 63 17

503 500 484 930 268

67 (19.2%) 65 (18.6%) 58 (16.6%) 121 (34.7%) 38 (10.9%)

< 0.0001 0.54 0.78 < 0.001

0.60 (18.9%) (18.4%) (18%) (34.6%) (10.1%)

(17.7%) (23.1%) (16.1%) (33.9%) (9.1%)

0.95 (18.7%) (18.6%) (18%) (34.6%) (10%)

< 0.001 187 (6.6%) 633 (22.2%) 648 (22.7%) 609 (21.3%) 776 (27.2%) 2709 (1844.4) 2.81 (1.32)

21 (11.3%) 61 (32.8%) 37 (19.9%) 30 (16.1%) 37 (19.9%) 2986 (1856.25) 3.24 (1.6)

97.02 (15.6) 101.2 (14.25) 95.51 (8.59)

0.85 0.73 0.17 0.52

< 0.001

< 0.05 < 0.0001

189 (7%) 581 (21.6%) 621 (23.1%) 566 (21.1%) 728 (27.1%) 2728 (1852.02) 2.85 (1.35)

19 (5.4%) 113 (32.4%) 64 (18.3%) 69 (19.8%) 84 (24.1%) 2709 (1799.21) 2.78 (1.32)

0.85 0.40

94.56 (17.89) 100.2 (16.01) 93.84 (9.78)

< 0.05 0.39 < 0.05

97.01 (15.78) 101.2 (14.46) 95.43 (8.59)

95.77 (15.66) 100.1 (13.63) 95.19 (9.35)

0.17 0.18 0.63

434 (15.2%) 246 (8.6%)

50 (26.9%) 30 (16.1%)

< 0.0001 < 0.001

423 (15.8%) 235 (8.8%)

59 (16.9%) 41 (11.7%)

0.64 0.07

2340 (82%) 354 (12.4%) 155 (5.4%) 57.16 (11.31) – 315 (11%) 2017 (70.7%)

89 (47.8%) 53 (28.5%) 44 (23.7%) 59.14 (11.42) – 34 (18.3%) 154 (82.8%)

2168 (80.7%) 350 (13%) 166 (6.2%) 1 (0%) 152 (5.7%) – 1906 (71%)

257 (73.6%) 56 (16%) 33 (9.5%) 3 (0.9%) 34 (9.7%) – 260 (74.5%)

< 0.0001

< 0.05 – < 0.01 < 0.001

< 0.05

0.76 < 0.01 – 0.19

Missing data : wheezing : 1; chronic cough : 13; mould : 5; indoor relative humidity : 860; relative humidity observed - expected : 1 179; inadequate ventilation or aeration : 5; household fuels : 8. a chi-squared for qualitative variables, and an ANOVA for comparison of means.

the 3276 participants were included in the final analysis. The characteristics of the study population are summarized in Tables 1a, 1b as a function of their exposure to mould and household fuels. With regard to mould, 407 (13.4%) participants were exposed in just one room in the living area, and 199 (6.6%) participants were exposed in two or more rooms in the living area. Exposure to mould was significantly more frequent in the city of Dunkirk and for female participants, and was also significantly associated with younger age, lower income and a greater number of people living in the household. With regard to household fuel, 2171 (71.4%) participants were exposed; this exposure was significantly associated with smoking, lower income, lower education and higher BMI. When considering the spirometry results, exposure to mould and exposure to household fuel were both associated with low FEV1 and the presence of wheezing. A total of 186 participants (6.1%) were exposed to condensation; this exposure was associated with smoking, lower income, and a greater number of people living in the household. Lastly, 285 (9.4%) participants were exposed to inadequate ventilation or aeration; this exposure was associated with lower income only. Among the non-smokers, 523 (19.7%) lived with a smoker, 184 (7.5%) were exposed to passive smoking at home, and 115 (4.3%) stated that someone smoked in the home at least once a day.

the score as a continuous variable. The final model for respiratory symptoms included age, smoking status, BMI, population density, income level and the investigator. Participants with missing data for at least one of the covariables were excluded from the analysis. We performed sensitivity analyses for participants who had stated being free of respiratory diseases because some associations between outcomes and exposures can be influenced by medical treatments and illness-related housing modifications. All statistical analyses were performed with R software (version 3.2.2, R Core Team, R Foundation for Statistical Computing, Vienna, Austria, 2014, http://www.R-project.org), using the prettyR, corrplot, lme4, lmerTest, and dfexplore packages. The threshold for statistical significance was set to p < .05. 3. Results 3.1. Characteristics of the study population and household exposure A total of 3276 participants were included in the ELISABET survey. We excluded 47 participants without spirometry data and 190 participants with unacceptable spirometry data (including 3 women under 145 cm in height, and one man under 155 cm in height). Hence, 3039 of 143

Environmental Research 164 (2018) 140–148

L. Devien et al.

Table 2 Exposure to mould, condensation, inadequate ventilation and household fuels, and respiratory symptoms. Exposure

N exposed

Wheezing

Chronic cough

Odds ratio (95% CI)

pa

Odds ratio (95% CI)

pa

2429 407 199

ref. 1.61 (1.21; 2.11) 1.33 (0.89; 1.95)

< 0.01

ref. 0.87 (0.59; 1.27) 0.96 (0.55; 1.58)

0.38

2679 314 42

ref. 1.54 (1.12; 2.08) 2.42 (1.16; 4.82)

< 0.001

ref. 1.55 (1.06; 2.22) 1.04 (0.3; 2.71)

0.06

2197 565 273 186 279 2171 2076 114 148 30 88

ref. 1.52 1.66 1.54 1.01 1.19 1.24 0.89 1.22 0.61 0.95

< 0.001

réf. 1.03 (0.74; 1.43) 1.15 (0.74; 1.75) 1.64 (1.04; 2.51) 1.4 (0.96; 2.01) 1.1 (0.82; 1.49) 1.15 (0.86; 1.54) 0.84 (0.37; 1.68) 0.61 (0.29; 1.12) 2.39 (0.84; 5.92) 1.26 (0.59; 2.41)

0.86

ref. 1.04 (0.67; 1.57) 1.04 (0.55; 1.81)

0.79

Total study population 3039 Number of rooms with mould rooms in the living area 0 1 ≥2 rooms in non-living area 0 1 ≥2 all rooms 0 1 ≥2 Window condensation Inadequate ventilation or aeration Household fuel Gas stove Open fireplace Oil Coal Water heater Participants free of respiratory disease

(1.18; 1.95) (1.19; 2.3) (1.05; 2.22) (0.73; 1.38) (0.93; 1.52) (0.98; 1.57) (0.46; 1.59) (0.78; 1.86) (0.17; 1.64) (0.5; 1.69)

< 0.05 0.95 0.17 0.08 0.70 0.37 0.37 0.86

< 0.05 0.07 0.54 0.35 0.65 0.14 0.08 0.52

2565 Number of rooms with mould rooms in the living area 0 1 ≥2 rooms in non-living area 0 1 ≥2 all rooms 0 1 ≥2 Window condensation Inadequate ventilation or aeration Household fuel Gas stove Open fireplace Oil Coal Water heater

2052 346 165

ref. 2.16 (1.55; 2.99) 1.15 (0.66; 1.90)

2269 260 34

ref. 1.62 (1.09; 2.36) 2.22 (0.83; 5.27)

1859 477 227 153 235 1818 1738 96 122 25 77

ref. 1.86 (1.36; 2.52) 1.67 (1.08; 2.51) 1.44 (0.89; 2.28) 0.91 (0.59; 1.35) 1.09 (0.81; 1.48) 1.19 (0.89; 1.6) 0.44 (0.13; 1.1) 1.1 (0.62; 1.86) 0.52 (0.08; 1.9) 0.82 (0.35; 1.69)

< 0.05

< 0.01

< 0.05 ref. 1.72 (1.12; 2.58) 1.24 (0.29; 3.69)

< 0.001

0.12 0.63 0.58 0.26 0.12 0.74 0.39 0.61

ref. 1.24 (0.85; 1.78) 1.33 (0.80; 2.12) 1.64 (0.96; 2.68) 1.60 (1.05; 2.39) 1.16 (0.83; 1.64) 1.2 (0.86; 1.68) 0.96 (0.39; 2.03) 0.55 (0.23; 1.13) 1.97 (0.45; 6.14) 1.34 (0.57; 2.74)

0.41

0.06 < 0.05 0.40 0.29 0.93 0.14 0.29 0.46

Adjusted for age, BMI, smoking status, income level, population density, and investigator, after a stepwise analysis. a p-trend for the number of rooms.

exposure to passive smoking (data not shown).

3.2. Association between exposure and respiratory symptoms For the study population as a whole (Table 2), a higher number of rooms with mould was significantly associated with a higher risk of wheezing (p-trend < 0.001; odds ratio (OR) [95% confidence interval (CI)] = 1.66 [1.19–2.3] for two rooms exposed). This association was also observed for rooms in the living area (p-trend < 0.01) and rooms in the non-living area (p-trend < 0.001). Window condensation was associated significantly with a higher risk of wheezing (OR = 1.54 [1.05–2.22] p < .05) and with chronic cough (OR = 1.64 [1.04–2.51], p < .05). For participants with no known respiratory diseases (Table 2), mould in any room (p-trend < 0.001), mould in rooms in the living area (p < .05), and mould in the non-living area (p < .01) were associated significantly with a higher risk of wheezing. Inadequate ventilation and aeration was associated significantly with a higher risk of chronic cough (OR = 1.60 [1.05–2.39], p < .05). There were no significant associations between respiratory symptoms/lung function and

3.3. Associations between exposure and lung function In the study population as a whole sample (Table 3), FEV1 was significantly associated with the total number of rooms with mould (ptrend < 0.05; mean decrement compared with non-exposed participants = −1.90% [−3.8 to 0.03] for two rooms exposed) and the number of rooms with mould in the living area (p-trend < 0.05; mean decrement = −2.5% [−4.7 to −0.29] for two rooms exposed). No association was observed for rooms in non-living area (including the kitchen and bathroom). There were no interactions between ventilation and mould variables. Window condensation was associated non-significantly with lower FEV1 values (mean decrement = −1.17% [−3.44 to 1.08] p = .31) and lower FVC values (mean decrement = -0.3% [−2.35 to 1.75] p = .78). Inadequate ventilation and aeration were associated non-significantly with lower FEV1 values (mean decrement = −1.13 [−2.82; 0.56] p = .19) and lower FEV1 values 144

Environmental Research 164 (2018) 140–148

L. Devien et al.

Table 3 Exposure to mould, condensation, inadequate ventilation and indoor combustible and lung function. Exposure

N

FEV1 (% predicted)

FVC (% predicted)

ratio FEV1/FVC (% predicted)

Exposed

Difference (95% CI)

pa

Difference (95% CI)

pa

Difference (95% CI)

pa

2429 407 199

ref. −0.66 (−2.3; 0.92) −2.5 (−4.7; −0.29)

< 0.05

ref. −0.82 (−2.3; 0.63) −1.7 (−3.7; 0.29)

0.08

ref. 0.11 (−0.78; 0.99) −0.91 (−2.1; 0.32)

0.24

2679 314 42

ref. −0.97 (−2.7; 0.81) −0.52 (−5.1; 4.1)

0.34

ref. −0.58 (−2.2; 1) 0.54 (−3.6; 4.7)

0.69

ref. −0.43 (−1.4; 0.57) −1.1 (−3.7; 1.5)

0.25

2197 565 273 186 279 2171 2076 114 148 30 88

ref. −0.57 (−2; 0.82) −1.9 (−3.8; 0.025) −1.17 (−3.44; 1.08) −1.13 (−2.82; 0.56) −0.88 (−2.1; 0.33) −1.15 (−2.33; 0.03) 1.34 (−1.54; 4.2) −1.33 (−3.84; 1.16) −0.38 (−5.82; 5.06) 1.67 (−1.52; 4.87)

< 0.05

ref. −0.77 (−2.1; 0.5) −1.2 (−3; 0.51) −0.3 (−2.35; 1.75) −0.96(−2.49; 0.58) −0.53 (−1.63; 0.57) −0.94 (−2.02; 0.13) 1.35 (−1.26; 3.96) −1.12 (−3.4; 1.16) 2.03 (−2.93; 6.97) 1.63 (−1.28; 4.53)

0.12

ref. 0.18 (−0.61; 0.96) −0.79 (−1.9; 0.28) −0.94 (−2.21; 0.32) −0.28 (−1.22; 0.66) −0.34 (−1.02; 0.33) −0.23 (−0.89; 0.43) 0.09 (−1.52; 1.69) −0.24 (−1.65; 1.15) −2.56 (−5.57; 0.53) 0.09 (−1.69; 1.87)

0.15

Total study population 3039 Number of rooms with mould rooms in living area 0 1 ≥2 rooms in non-living area 0 1 ≥2 all rooms 0 1 ≥2 Window condensation Inadequate ventilation or aeration Household fuel Gas stove Open fireplace Oil Coal Water heater Participants free of respiratory disease

0.31 0.19 0.16 0.06 0.36 0.30 0.89 0.31

0.78 0.22 0.35 0.09 0.31 0.34 0.42 0.27

0.15 0.56 0.32 0.49 0.91 0.73 0.01 0.92

2565 Number of rooms with mould rooms in living area 0 1 ≥2 rooms in non-living area 0 1 ≥2 all rooms 0 1 ≥2 Window condensation Inadequate ventilation or aeration Household fuel Gas stove Open fireplace Oil Coal Water heater

2052 346 165

ref. −1.17 (−2.78; 0.45) −2.14 (−4.4; 0.12)

< 0.05

ref. −0.91 (−2.45; 0.62) −1.59 (−3.74; 0.56)

0.12

ref. −0.35 (−1.18; 0.50) −0.73 (−1.91; 0.44)

0.14

2269 260 34

ref. 0.11 (−1.72; 1.94) 0.71 (−4.08; 5.51)

0.79

ref. 0.27 (−1.48; 2.00) 0.24 (−4.31; 4.8)

0.77

ref. −0.08 (−1.02; 0.88) 0.17 (−2.33; 2.65)

0.96

1859 477 227 153 235 1818 1738 96 122 25 77

ref. −0.72 (−2.16; 0.71) −1.35 (−3.33; 0.64) −1.28 (−3.62; 1.04) −1.15 (−2.86; 0.58) −0.74 (−1.98; 0.49) −1.14 (−2.35; 0.06) 0.85 (−2.1; 3.78) −0.52 (−3.12; 2.05) 0.58 (−5.02; 6.19) 2.13 (−1.07; 5.34)

0.14

ref. −0.68 (−2.05; 0.68) −0.92 (−2.8; 0.96) −0.36 (−2.59; 1.85) −1.17 (−2.81; 0.47) −0.79 (−1.97; 0.38) −1.29 (−2.43; −0.15) 0.8 (−2; 3.59) −0.22 (−2.68; 2.23) 2.35 (−2.99; 7.67) 0.97 (−2.09; 4.01)

0.27

ref. −0.11 (−0.85; 0.64) −0.53 (−1.56; 0.50) −0.93 (−2.14; 0.28) −0.03 (−0.93; 0.86) 0.03 (−0.62; 0.67) 0.08 (−0.55; 0.7) 0.19 (−1.34; 1.71) −0.31 (−1.66; 1.03) −1.75 (−4.64; 1.19) 1.16 (−0.51; 2.83)

0.23

0.28 0.20 0.24 0.06 0.57 0.69 0.84 0.19

0.75 0.16 0.19 < 0.05 0.58 0.86 0.39 0.54

0.13 0.94 0.93 0.81 0.80 0.65 0.24 0.18

Adjusted for gender, age, height, BMI, smoking status, educational level, income level, number of people in the household, population density, and the investigator, after stepwise analysis. a p-trend for the number of rooms.

(mean decrement = −0.96 [−2.49; 0.58] p = .22). There were no significant associations with the use of household fuels. Gas cooking was non-significantly associated with lower FEV1 values (mean decrement = −1.15% [−2.33 to 0.03]; p = .06) and lower FVC values (mean decrement = −0.94% [−2.02 to 0.13]; p = .09). For participants free of respiratory disease (Table 3), mould in rooms in the living area was associated significantly with lower FEV1 (p-trend < 0.05) and FVC (mean decrement [95%CI] = −1.41% [−2.69 to −0.13]; p < .05) values. Condensation was associated nonsignificantly with lower FEV1 values (mean decrement = −1.28% [−3.62 to 1.04] p = .28) and lower FEV1 values (mean decrement = −0.36% [−2.59 to 1.85] p = .75). Inadequate ventilation and aeration were associated non-significantly with lower FEV1 values (mean decrement = −1.15 [−2.86; 0.58] p = .20) and lower FVC values (mean decrement = −1.17 [−2.81; 0.47] p = .16). Regarding household fuels, the only significant association was thats between gas cooking and FVC (mean decrement = −1.29% [−2.43 to −0.15]; p < .05). No significant association were observed between respiratory symptom or

lung function and exposure to passive smoking (data not shown). 3.4. Sensitivity analysis Twenty-eight participants used an open fireplace all winter but none of the associations were significant. Fifty-three participants used oil heating all winter; this exposure was associated with a lower FEV1 in an adjusted analysis (OR [95%CI] = −5.3% [−9.4% to −1.2%]; p < .05) and lower FVC (OR [95%CI] = −4.9% [−8.61% to −1.15%]; p < .05). Twenty participants used coal heating all winter; this exposure was associated with a lower FEV1/FVC ratio (OR [95%CI] = [−8.72; −1.31]; p < .01) and increased chronic cough (OR [95%CI] = 3.07 [0.94–8.52]; p < .05) 4. Discussion In our cross-sectional sample of adults aged from 40 to 65, exposure to mould at home was associated with lower FEV1% predicted values 145

Environmental Research 164 (2018) 140–148

L. Devien et al.

due to water use. Hence, the conditions for microbial growth may differ from to those observed in living areas, and may modify the chemicals emitted by the moulds. Furthermore, kitchens and bathrooms may be better ventilated than living areas (Pekkanen et al., 2007). Lastly, mould may be a marker of poor-quality housing and confinement, rather than the cause per se of poor lung function. To grow and develop, mould needs humidity, high temperatures, and a carbon source. In our study, the presence of mould was associated with condensation, high relative humidity, and inadequate ventilation or aeration. However, these other housing characteristics were not associated with abnormally poor lung function.

and a higher risk of wheezing relative to non-exposed participants. In the subset of participants free of known respiratory disease, we observed a significant association between gas cooking and low FVC. 4.1. Mould According to the World Health Organization, a large body of evidence shows clearly that exposure to mould is associated with asthma development, asthma exacerbation, current asthma, hypersensitivity pneumonitis, respiratory infections, upper respiratory tract symptoms, cough, wheeze, and dyspnoea (World Health Organization, 2009). The association between mould exposure and wheeze observed in the present study is consistent with previous results. In two meta-analyses, exposure to mould and dampness has been associated with respiratory symptoms, respiratory tract infections, and bronchitis (Fisk et al., 2007, 2010). In contrast, the associations between mould exposure and decrease lung function observed in previous cross-sectional studies of adults had never achieved statistical significance. Gunnbjörnsdottir et al. found that a mean [95%CI] decrement of 36 mL [−103 to 32] in FEV1 and a mean [95%CI] decrement of 16 mL [−98 to 65] in FVC were associated with the observation of visible mould in the household in the previous 12 months (Gunnbjörnsdottir et al., 2003). Hernberg et al. found that a mean [95%CI] decrement of 0.15 L [−1.12 to 0.81] in FEV1 was associated with visible mould at home at any time, and that a mean [95%CI] decrement 0.20 L [−0.60 to 0.21] in FEV1 and a mean [95%CI] decrement of 0.46 [−0.95 to 0.03] in FVC were associated with mould odour in the previous 12 months (Hernberg et al., 2014). Ebbehøj et al.’s study of teachers exposed to high levels of mould at school evidenced a non-significant 1.1% decrement in FEV1 and a nonsignificant 2.8% decrement in FVC for males, and a non-significant 0.2% increase in FEV1 and a non-significant 3.2% decrement in FVC for females (Ebbehøj et al., 2005). Lastly, Chen et al found that adolescents exposed to mould at home displayed a non-significant 0.005 increment in a FEV1 Z-score and a non-significant 0.027 decrement in a FVC Zscore (Chen et al., 2017). However, these study populations were smaller than ours, with around 2000 participants in the studies by Gunnbjörnsdottir et al. and Chen et al., and less than 500 in the studies by Hernberg et al. and Ebbehøj et al. Hence, a lack of statistical power might explain the lack of statistically findings in the literature. Furthermore, the above-mentioned studies concerned younger adults (aged 20–40) or adolescents, in whom the effect of indoor air pollution might not yet be visible. In a longitudinal study (Norbäck et al., 2011), however, mould exposure was associated with a decline over time of 0.90 [−1.71 to −0.08] mL/year in FEV1 in women only. There are several possible explanations for the putative association between mould exposure and lung function. Mould is known for its allergenic effects and its exacerbation of asthma. However, most of the participants in the present study were healthy, and the associations with exposure to mould in the living area were still statistically significant when participants with known respiratory disease were excluded from the analysis. This association seen for non-asthmatic patients might be due to the presence of mild, non-diagnosed asthma or the involvement of a non-allergic mechanism. It is noteworthy that the associations between mould on one hand and asthma incidence and prevalence on the other have been observed in both atopic and nonatopic populations (World Health Organization, 2009). The results of toxicological studies suggest that exposure to the volatile organic compounds (VOCs) released by mould are involved in the development of non-allergic inflammatory responses (Korpi et al., 2009). In the present study, we observed an association between lung function and the presence of mould in rooms in the living area but not the presence of mould in the kitchen or bathroom. This finding is in line with the literature data on mould and asthma (Norbäck et al., 2013; Pekkanen et al., 2007). Participants may spend less time in bathrooms and kitchen, and the presence of dampness in these rooms may be mostly

4.2. Condensation, ventilation and aeration Condensation and inadequate ventilation/aeration have been associated with the presence of respiratory symptoms because they are linked to indoor confinement and (probably) increased air pollution. This association has been found in previous studies (Bentayeb et al., 2015; Engvall et al., 2001; Wang et al., 2014a, 2014b). In the present study, inadequate ventilation/aeration was not associated with wheezing. However, the self-reporting of these features on our study questionnaire might not have been sufficiently accurate. 4.3. Passive smoking The prevalence of exposure to passive smoking at home was low. When exposure was reported, it tended to be infrequent (i.e. less than daily). Therefore, exposure to passive smoking at home was not extensive enough to be studied in our population. The low prevalence measured in our study may be partly due to the reinforcement of the French legislation on smoke-free areas. Since 2007, smoking has been banned in public building and on mass transportation. These legislative changes may have prompted smokers to establish smoking bans in their homes (Mons et al., 2013). 4.4. Gas stove We found a significant association between gas stove use and a decrement in FVC among participants without known respiratory disease. Previous studies have observed a significant association between gas stove use and respiratory symptoms (Jarvis et al., 1998) but not with poorer lung function. Gas stoves are major sources of nitrogen dioxide, which is formed during the combustion process (Adamkiewicz et al., 2010). Outdoors, most of the nitrogen dioxide comes from road traffic. Although short-term exposure of humans to low concentrations of nitrogen dioxide (similar to those observed outdoors) does not have an effect on lung function, the presence of pulmonary inflammation has been linked to exposure to nitrogen dioxide levels similar to those reached when a gas stove is used in a kitchen (Hesterberg et al., 2009). Lastly, this association was only statistically significant in the subgroup of participants without known respiratory disease. This result may have been due to chance, and so warrants further investigation. 4.5. Strengths and limitations Our sample size was larger than in previous studies of the association between mould exposure and lung function (Ebbehøj et al., 2005; Gunnbjörnsdottir et al., 2003; Hernberg et al., 2014; Norbäck et al., 2011). There were some limitations with regard to the interpretation of statistical significance. The ELISABET survey was cross-sectional; it is therefore difficult to infer a chronological association between exposure to air pollution and the outcomes. Furthermore, we performed a large number of statistical tests; some findings may have been due to chance. Furthermore, many sources of pollution were involved, and it is therefore very difficult to assess the level of household air pollution. 146

Environmental Research 164 (2018) 140–148

L. Devien et al.

Development for their financial support.

The source of exposure were reported by the participants. This type of measurement is probably inaccurate, and may represent a source of classification bias. Furthermore, mould present in areas hidden from sight could not be included in this subjective approach. Accordingly, self-reporting may have decreased the study’s power, and may explain why some results concerning lung function were not significant - despite the large sample size. The direct measurement of household air pollution would have been more accurate and would have enabled us to study dose-effect relationships. However, measurements of household air pollution are difficult to achieve in large populations. Furthermore, a single, objective measurement of household air quality may not be representative of chronic exposure. A building inspection during the home visit would have provided more objective information on exposure. However, studying the household air was not the study’s main objective. A building inspection would have been time-consuming and impractical to implement as part of the study protocol, and would have discouraged people from volunteering to participate in the study. The self-reported exposure was not validated against a gold standard. However, we used variables similar to those in the European Community Respiratory Health Survey (ECRHS) questionnaire. Similar associations for both self-reported and observed measures of exposure were found in the ECRHS study (Norbäck et al., 2011, 2013). Few participants were exposed to heating with fuels all winter, making it difficult to study the impact of the duration of exposure; significant results in the sensitivity analysis should be interpreted with caution. Despite the large sample size, the number of participants who reported visible condensation or inadequate ventilation/aeration was low – thus reducing the statistical power of the analyses of these exposures. Lastly, the participation rate was low, as is frequently observed in this type of study. Mid-level professions and white-collar male workers were slightly overrepresented, relative to blue-collar workers (Quach et al., 2015); however, this difference with census data seems to have little impact on the prevalence of airway obstruction and cardiovascular risk (Quach et al., 2015; Clement et al., 2017). There are many potential sources of household air pollution: mould, dampness, household fuels, allergens, tobacco smoke, cooking, furniture, paints, cleaning agents, and domestic pets (Apte and Salvi, 2016). Furthermore, the relationships between pollutants are complex. Each source can generate several pollutants, and a given pollutant can be produced by several different sources. Pollutants may be correlated, and may have synergetic effects. Hence, the identification of a causal relationship between a specific pollutant and an effect on health is challenging.

Funding sources The ELISABET survey was sponsored by Lille University Medical Centre (CHRU de Lille, Lille, France) and funded by the Nord Pas-deCalais Regional Council and the European Regional Development Fund (ERDF-FEDER Presage No. 36034) as part of the CPER Institut de Recherche en ENvironnement Industriel (IRENI) programme. This work is a contribution to the CPER research project CLIMIBIO. Disclosure of interest the authors declare that they have no conflicts of interest concerning this work. Ethics committee The study protocol was approved by the local independent ethics committee (CPP Nord Ouest IV, reference 2010-A00065-34). References Adamkiewicz, G., World Health Organization, Regional Office for Europe, 2010. WHO Guidelines for Indoor Air Quality: Selected Pollutants. Adam, M., Schikowski, T., Carsin, A.E., Cai, Y., Jacquemin, B., Sanchez, M., Vierkötter, A., Marcon, A., Keidel, D., Sugiri, D., Al Kanani, Z., Nadif, R., Siroux, V., Hardy, R., Kuh, D., Rochat, T., Bridevaux, P.-O., Eeftens, M., Tsai, M.-Y., Villani, S., Phuleria, H.C., Birk, M., Cyrys, J., Cirach, M., de Nazelle, A., Nieuwenhuijsen, M.J., Forsberg, B., de Hoogh, K., Declerq, C., Bono, R., Piccioni, P., Quass, U., Heinrich, J., Jarvis, D., Pin, I., Beelen, R., Hoek, G., Brunekreef, B., Schindler, C., Sunyer, J., Krämer, U., Kauffmann, F., Hansell, A.L., Künzli, N., Probst-Hensch, N., 2015. Adult lung function and long-term air pollution exposure. escape: a multicentre cohort study and metaanalysis. Eur. Respir. J. 45, 38–50. http://dx.doi.org/10.1183/09031936.00130014. Apte, K., Salvi, S., 2016. Household air pollution and its effects on health. F1000Research 5. http://dx.doi.org/10.12688/f1000research.7552.1. Bentayeb, M., Norback, D., Bednarek, M., Bernard, A., Cai, G., Cerrai, S., Eleftheriou, K.K., Gratziou, C., Holst, G.J., Lavaud, F., Nasilowski, J., Sestini, P., Sarno, G., Sigsgaard, T., Wieslander, G., Zielinski, J., Viegi, G., Annesi-Maesano, I., 2015. Indoor air quality, ventilation and respiratory health in elderly residents living in nursing homes in Europe. Eur. Respir. J. 45, 1228–1238. http://dx.doi.org/10.1183/09031936. 00082414. Berend, N., 2016. Contribution of air pollution to COPD and small airway dysfunction. Respirol. Carlton Vic. 21, 237–244. http://dx.doi.org/10.1111/resp.12644. Brasche, S., Bischof, W., 2005. Daily time spent indoors in German homes–baseline data for the assessment of indoor exposure of German occupants. Int. J. Hyg. Environ. Health 208, 247–253. http://dx.doi.org/10.1016/j.ijheh.2005.03.003. Chen, Y.C., Ho, W.C., Yu, Y.H., 2017. Adolescent lung function associated with incense burning and other environmental exposures at home. Indoor Air 27, 746–752. http:// dx.doi.org/10.1111/ina.12355. Clement, G., Giovannelli, J., Cottel, D., Montaye, M., Ciuchete, A., Dallongeville, J., Amouyel, P., Dauchet, L., 2017. Changes over time in the prevalence and treatment of cardiovascular risk factors, and contributions to time trends in coronary mortality over 25 years in the Lille urban area (northern France). Arch. Cardiovasc. Dis. http:// dx.doi.org/10.1016/j.acvd.2017.03.009. de Jong, K., Vonk, J.M., Zijlema, W.L., Stolk, R.P., van der Plaat, D.A., Hoek, G., Brunekreef, B., Postma, D.S., Boezen, H.M., 2016. Air pollution exposure is associated with restrictive ventilatory patterns. Eur. Respir. J. 48, 1221–1224. http://dx.doi. org/10.1183/13993003.00556-2016. (LifeLines Cohort Study Group). Dorizas, P.V., Assimakopoulos, M.-N., Helmis, C., Santamouris, M., 2015. An integrated evaluation study of the ventilation rate, the exposure and the indoor air quality in naturally ventilated classrooms in the Mediterranean region during spring. Sci. Total Environ. 502, 557–570. http://dx.doi.org/10.1016/j.scitotenv.2014.09.060. Ebbehøj, N.E., Meyer, H.W., Würtz, H., Suadicani, P., Valbjørn, O., Sigsgaard, T., Gyntelberg, F., 2005. Molds in floor dust, building-related symptoms, and lung function among male and female schoolteachers. Indoor Air 15 (Suppl 10), S7–S16. http://dx.doi.org/10.1111/j.1600-0668.2005.00352.x. (Members of a Working Group under the Danish Mold in Buildings program) (DAMIB). Engvall, K., Norrby, C., Norbäck, D., 2001. Asthma symptoms in relation to building dampness and odour in older multifamily houses in Stockholm. Int. J. Tuberc. Lung Dis. Off. J. Int. Union Tuberc. Lung Dis. 5, 468–477. Fisk, W.J., Eliseeva, E.A., Mendell, M.J., 2010. Association of residential dampness and mold with respiratory tract infections and bronchitis: a meta-analysis. Environ. Health Glob. Access Sci. Source 9, 72. http://dx.doi.org/10.1186/1476-069X-9-72. Fisk, W.J., Lei-Gomez, Q., Mendell, M.J., 2007. Meta-analyses of the associations of respiratory health effects with dampness and mold in homes. Indoor Air 17, 284–296. http://dx.doi.org/10.1111/j.1600-0668.2007.00475.x. Giovannelli, J., Chérot-Kornobis, N., Hulo, S., Ciuchete, A., Clément, G., Amouyel, P., Matran, R., Dauchet, L., 2016. Both exhaled nitric oxide and blood eosinophil count

4.6. Conclusion Our results confirmed the previously reported association between mould and wheezing. To the best of our knowledge, the present study is the first to have observed a significant association between mould and worsened lung function. The latter finding needs to be confirmed in other cross-sectional and longitudinal studies. However, our study suggests that mould exposure can impact lung function in the general population (and not just in people with asthma). Acknowledgements The authors thank Lille University Hospital (especially the Institut de Biologie et de Pathologie), the University of Lille, the Institut Pasteur de Lille (especially the Departments of Médecine Préventive, Biologie Spécialisée and Médecine du Travail, and the Laboratoire d’Analyses Génomiques) and the Centre Hospitalier Général de Dunkerque (especially the Departments of Biology and Pneumology); they particularly thank the nurses, physicians and secretarial staff of the University of Lille and the Institut Pasteur de Lille. The authors also thank the French Ministère de l'Enseignement Supérieur et de la Recherche, the Hauts de France Region and the European Funds for Regional Economic 147

Environmental Research 164 (2018) 140–148

L. Devien et al.

325–331. http://dx.doi.org/10.1136/oemed-2012-100963. Norbäck, D., Zock, J.-P., Plana, E., Heinrich, J., Svanes, C., Sunyer, J., Künzli, N., Villani, S., Olivieri, M., Soon, A., Jarvis, D., 2011. Lung function decline in relation to mould and dampness in the home: the longitudinal European Community Respiratory Health Survey ECRHS II. Thorax 66, 396–401. http://dx.doi.org/10.1136/thx.2010. 146613. Pascal, M., Corso, M., Chanel, O., Declercq, C., Badaloni, C., Cesaroni, G., Henschel, S., Meister, K., Haluza, D., Martin-Olmedo, P., Medina, S., 2013. Assessing the public health impacts of urban air pollution in 25 European cities: results of the Aphekom project. Sci. Total Environ. 449, 390–400. http://dx.doi.org/10.1016/j.scitotenv. 2013.01.077. Pekkanen, J., Hyvärinen, A., Haverinen-Shaughnessy, U., Korppi, M., Putus, T., Nevalainen, A., 2007. Moisture damage and childhood asthma: a population-based incident case-control study. Eur. Respir. J. 29, 509–515. http://dx.doi.org/10.1183/ 09031936.00040806. Quach, A., Giovannelli, J., Chérot-Kornobis, N., Ciuchete, A., Clément, G., Matran, R., Amouyel, P., Edmé, J.-L., Dauchet, L., 2015. Prevalence and underdiagnosis of airway obstruction among middle-aged adults in northern France: the ELISABET study 2011–2013. Respir. Med. 109, 1553–1561. http://dx.doi.org/10.1016/j.rmed.2015. 10.012. Quanjer, P.H., Stanojevic, S., Cole, T.J., Baur, X., Hall, G.L., Culver, B.H., Enright, P.L., Hankinson, J.L., Ip, M.S.M., Zheng, J., Stocks, J., 2012. Multi-ethnic reference values for spirometry for the 3-95-yr age range: the global lung function 2012 equations. Eur. Respir. J. 40, 1324–1343. http://dx.doi.org/10.1183/09031936.00080312. (ERS Global Lung Function Initiative). Repace, J.L., Lowrey, A.H., 1980. Indoor air pollution, tobacco smoke, and public health. Science 208, 464–472. Rice, M.B., Ljungman, P.L., Wilker, E.H., Dorans, K.S., Gold, D.R., Schwartz, J., Koutrakis, P., Washko, G.R., O’Connor, G.T., Mittleman, M.A., 2015. Long-term exposure to traffic emissions and fine particulate matter and lung function decline in the Framingham heart study. Am. J. Respir. Crit. Care Med. 191, 656–664. http://dx.doi. org/10.1164/rccm.201410-1875OC. Sarnat, J.A., Koutrakis, P., Suh, H.H., 2000. Assessing the relationship between personal particulate and gaseous exposures of senior citizens living in Baltimore, MD. J. Air Waste Manag. Assoc. 1995 (50), 1184–1198. Schneider, P., Gebefügi, I., Richter, K., Wölke, G.W., Schneille, J., Wichmann, H.E., Heinrich, J., 2001. Indoor and outdoor BTX levels in German cities. Sci. Total Environ. 267, 41–51 (INGA Study Group. INdoor exposure and Genetics in Asthma). Sundell, J., Levin, H., Nazaroff, W.W., Cain, W.S., Fisk, W.J., Grimsrud, D.T., Gyntelberg, F., Li, Y., Persily, A.K., Pickering, A.C., Samet, J.M., Spengler, J.D., Taylor, S.T., Weschler, C.J., 2011. Ventilation rates and health: multidisciplinary review of the scientific literature. Indoor Air 21, 191–204. http://dx.doi.org/10.1111/j.16000668.2010.00703.x. Wang, J., Engvall, K., Smedje, G., Norbäck, D., 2014a. Rhinitis, asthma and respiratory infections among adults in relation to the home environment in multi-family buildings in Sweden. PLoS One 9, e105125. http://dx.doi.org/10.1371/journal.pone. 0105125. Wang, J., Li, B., Yu, W., Yang, Q., Wang, H., Huang, D., Sundell, J., Norbäck, D., 2014b. Rhinitis symptoms and asthma among parents of preschool children in relation to the home environment in Chongqing, China. PLoS One 9, e94731. http://dx.doi.org/10. 1371/journal.pone.0094731. World Health Organization, 2009. WHO Guidelines for Indoor Air Quality: Dampness and Mould. World Health Organization, Regional Office for Europe, Copenhagen. Zhou, Y., Zou, Y., Li, X., Chen, S., Zhao, Z., He, F., Zou, W., Luo, Q., Li, W., Pan, Y., Deng, X., Wang, X., Qiu, R., Liu, S., Zheng, J., Zhong, N., Ran, P., 2014. Lung function and incidence of chronic obstructive pulmonary disease after improved cooking fuels and kitchen ventilation: a 9-year prospective cohort study. PLoS Med. 11, e1001621. http://dx.doi.org/10.1371/journal.pmed.1001621.

were associated with mild allergic asthma only in non-smokers. Clin. Exp. Allergy J. Br. Soc. Allergy Clin. Immunol. 46, 543–554. http://dx.doi.org/10.1111/cea.12669. Gunnbjörnsdottir, M.I., Norbäck, D., Plaschke, P., Norrman, E., Björnsson, E., Janson, C., 2003. The relationship between indicators of building dampness and respiratory health in young Swedish adults. Respir. Med. 97, 302–307. Hernberg, S., Sripaiboonkij, P., Quansah, R., Jaakkola, J.J.K., Jaakkola, M.S., 2014. Indoor molds and lung function in healthy adults. Respir. Med. 108, 677–684. http:// dx.doi.org/10.1016/j.rmed.2014.03.004. Hesterberg, T.W., Bunn, W.B., McClellan, R.O., Hamade, A.K., Long, C.M., Valberg, P.A., 2009. Critical review of the human data on short-term nitrogen dioxide (NO2) exposures: evidence for NO2 no-effect levels. Crit. Rev. Toxicol. 39, 743–781. http://dx. doi.org/10.3109/10408440903294945. Hulo, S., de Broucker, V., Giovannelli, J., Cherot-Kornobis, N., Nève, V., Sobaszek, A., Dauchet, L., Edmé, J.-L., 2016. Global Lung Function Initiative reference equations better describe a middle-aged, healthy French population than the European Community for Steel and Coal values. Eur. Respir. J. 48, 1779–1781. http://dx.doi. org/10.1183/13993003.00606-2016. IARC Working Group on the Evaluation of Carcinogenic Risks to Humans, 2004. Tobacco smoke and involuntary smoking. IARC Monogr. Eval. Carcinog. Risks Hum. 83, 1–1438. INSEE, 2013. Recensements de la population - Etat civil - Clap, Insee-DGFiP-Cnaf-CnavCcmsa, Fichier localisé social et fiscal. INSEE, 2011. DGFiP Revenus fiscaux localisés des ménages. Jarvis, D., Chinn, S., Sterne, J., Luczynska, C., Burney, P., 1998. The association of respiratory symptoms and lung function with the use of gas for cooking. Eur. Commun. Respir. Health Surv. Eur. Respir. J. 11, 651–658. Kattan, M., Gergen, P.J., Eggleston, P., Visness, C.M., Mitchell, H.E., 2007. Health effects of indoor nitrogen dioxide and passive smoking on urban asthmatic children. J. Allergy Clin. Immunol. 120, 618–624. http://dx.doi.org/10.1016/j.jaci.2007.05.014. Korpi, A., Järnberg, J., Pasanen, A.-L., 2009. Microbial volatile organic compounds. Crit. Rev. Toxicol. 39, 139–193. http://dx.doi.org/10.1080/10408440802291497. Kurmi, O.P., Semple, S., Simkhada, P., Smith, W.C.S., Ayres, J.G., 2010. COPD and chronic bronchitis risk of indoor air pollution from solid fuel: a systematic review and meta-analysis. Thorax 65, 221–228. http://dx.doi.org/10.1136/thx.2009.124644. Mannucci, P.M., Harari, S., Martinelli, I., Franchini, M., 2015. Effects on health of air pollution: a narrative review. Intern. Emerg. Med. 10, 657–662. http://dx.doi.org/ 10.1007/s11739-015-1276-7. Medical Research Council’s Committee on Environmental and Occupational Health, 1986. Medical Research Council’s Committee on Environmental and Occupational Health. Questionnaire on Respiratory Symptoms. Medical Research Council, London, pp. 1986. Mendell, M.J., Mirer, A.G., Cheung, K., Tong, M., Douwes, J., 2011. Respiratory and allergic health effects of dampness, mold, and dampness-related agents: a review of the epidemiologic evidence. Environ. Health Perspect. 119, 748–756. http://dx.doi.org/ 10.1289/ehp.1002410. Miller, M.R., Hankinson, J., Brusasco, V., Burgos, F., Casaburi, R., Coates, A., Crapo, R., Enright, P., van der Grinten, C.P.M., Gustafsson, P., Jensen, R., Johnson, D.C., MacIntyre, N., McKay, R., Navajas, D., Pedersen, O.F., Pellegrino, R., Viegi, G., Wanger, J., 2005. Standardisation of spirometry. Eur. Respir. J. 26, 319–338. http:// dx.doi.org/10.1183/09031936.05.00034805. (ATS/ERS Task Force). Mons, U., Nagelhout, G.E., Allwright, S., Guignard, R., van den Putte, B., Willemsen, M.C., Fong, G.T., Brenner, H., Pötschke-Langer, M., Breitling, L.P., 2013. Impact of national smoke-free legislation on home smoking bans: findings from the International Tobacco Control Policy Evaluation Project Europe Surveys. Tob. Control 22, e2–e9. http://dx.doi.org/10.1136/tobaccocontrol-2011-050131. Norbäck, D., Zock, J.-P., Plana, E., Heinrich, J., Svanes, C., Sunyer, J., Künzli, N., Villani, S., Olivieri, M., Soon, A., Jarvis, D., 2013. Mould and dampness in dwelling places, and onset of asthma: the population-based cohort ECRHS. Occup. Environ. Med. 70,

148