Occupational Exposure Influences on Gender ... - Springer Link

3 downloads 0 Views 285KB Size Report
Nov 15, 2011 - META-ANALYSIS: OCCUPATIONAL LUNG DISEASE. Occupational Exposure Influences on Gender Differences in Respiratory Health.
Lung (2012) 190:147–154 DOI 10.1007/s00408-011-9344-x

META-ANALYSIS: OCCUPATIONAL LUNG DISEASE

Occupational Exposure Influences on Gender Differences in Respiratory Health Helen Dimich-Ward • Kris Beking • Anne DyBuncio Moira Chan-Yeung • Weiwei Du • Barbara Karlen • Pat G. Camp • Susan M. Kennedy



Received: 20 January 2011 / Accepted: 17 October 2011 / Published online: 15 November 2011 Ó Springer Science+Business Media, LLC 2011

Abstract Background The aim of this study was to evaluate gender differences in the respiratory health of workers exposed to organic and inorganic dusts. Methods Meta-analysis techniques incorporating logistic regression were applied to a combined file of 12 occupational health studies. Results Meta-analysis of data on 1,367 women and 4,240 men showed that women had higher odds of shortness of breath whether exposed to inorganic dust or having no occupational exposure, with an overall odds ratio (OR) of 2.07 (95% confidence interval [CI] = 1.57–2.73) adjusted for smoking status, age, body mass index (BMI), ethnic status, atopy, and job duration. Inorganic dust exposure was associated with the highest odds of asthma (adjusted OR = 8.38, 95% CI = 1.72–40.89) for women compared to men, but no differences were found for unexposed

H. Dimich-Ward  K. Beking  A. DyBuncio  M. Chan-Yeung Respiratory Division, Department of Medicine, University of British Columbia, Vancouver, BC, Canada H. Dimich-Ward (&) UBC Department of Medicine, VGH Research Pavilion, 390-828 West 10th Avenue, Vancouver, BC V5Z 1L8, Canada e-mail: [email protected] W. Du  B. Karlen  S. M. Kennedy School of Population and Public Health, University of British Columbia, Vancouver, BC, Canada P. G. Camp Department of Physical Therapy, University of British Columbia, Vancouver, BC, Canada P. G. Camp James Hogg Research Centre, University of British Columbia, Vancouver, BC, Canada

workers. With organic dust exposure, men had elevated odds for occasional wheeze and worse lung function compared to women. Conclusion Within the limitations of this analysis, gender differences in respiratory health, as suggested by population-based studies, were confirmed in our analysis of occupational health studies, with the general type of exposure, organic or inorganic, generally determining the extent of differences. The higher risks for women compared to men for shortness of breath were robust regardless of work exposure category, with the highest odds ratios found for asthma. Keywords

Occupation  Gender  Lung health

Introduction There has been increasing scientific interest about the inequities in health attributed to sex or gender. In general, gender differences are apparent in the perception and reporting of symptoms [1]. Respiratory health may be further influenced by sex differences in the toxicokinetic parameters of lung size, lung function (volumes, flows, and respiratory rates), deposition of fine particles, gas absorption, gas–blood barrier permeability, vascular response, inflammation, and airway responsiveness [2–4]. Airway hyperresponsiveness is a marker for asthma, an atopic respiratory disease showing sex and gender differences in prevalence and severity over lifetime. In general, the incidence of asthma has been found to be greater in women after puberty, with age-adjusted relative risk estimates ranging from 1.4 to 5.9 for adult women (versus men) aged 20–44 years in a community-based analysis of 16 participating countries [5]. However, in a large

123

148

Lung (2012) 190:147–154

population-based study of never smokers in Scandinavia [6], there was no difference in risk between men and women for physician-diagnosed asthma (OR = 0.9 for women versus men, 95% CI = 0.6–1.4, adjusted for age, family history of respiratory disease, and environmental tobacco smoke exposure). In the adult population, apart from smoking behavior, the extent and type of occupational exposure may contribute to gender differences in respiratory health. For instance, in a study of synthetic textile workers [7], women had a higher prevalence of dyspnea (49.0% vs. 34.8% for men), unlike that observed for unexposed workers (5.0% for women and 5.1% for men). Occupational health studies, in general, have low statistical power to determine gender differences in outcomes. This is due to primarily the relatively small number of women typically employed in primary industries. One way to overcome this limitation is by combining individual studies. A meta-analysis of 12 studies [8] of food processing workers, textile workers, and farmers concluded that women workers had significantly less chronic cough, chronic phlegm, and chronic bronchitis than their male coworkers. The purpose of this article was to evaluate whether there were differences in symptoms and lung function between employed men and women according to type of exposure (organic or inorganic dusts) using a meta-analysis of combined occupational health studies.

Methods The combined data file of occupational health studies was an amalgamation of individual studies on the respiratory health

of workers carried out by investigators from the UBC Occupational and Environmental Lung Diseases Research Unit over the past three decades. These cross-sectional workplace-based studies comprise populations at risk for airflow obstruction from exposure to airborne contaminants in the workplace; some studies also consisted of unexposed workers serving as comparison controls. Studies of workers were included only if spirometry data were available and there was a minimum of ten members of each sex. Retired subjects were not consistently ascertained and were therefore excluded. For repeated cross-sectional studies, only the most recent cross-sectional study with the largest number of female participants was included in the pooled data file. The individual occupational health studies are listed in Table 1 and covered almost a 30-year span since 1980. All individual studies and their testing procedures were approved by the University of British Columbia Clinical Research Ethics Board. Occupational exposure categories were determined principally by the job title in the targeted study population. Job titles within studies were grouped according to whether the exposure was to ‘‘inorganic’’ dusts, fumes, or vapors or to ‘‘organic’’ dusts (grain or wood), or there was ‘‘no exposure’’ (typically office and administrative jobs), as shown in Table 1. Most of the regulatory office workers were considered to be unexposed; exceptions were radiographers, lab technicians, laundry, workshop, printers, and cleaners who were classified as having inorganic exposures. Inorganic exposures were predominant in the smelter (pot rooms), trades apprentices (painters, welders, machinists, electricians, and motor winders), and in theatre workers exposed to smoke and fog used for special effects.

Table 1 Occupational respiratory health studies carried out by the UBC Occupational and Environmental Lung Diseases Research Unit since 1980, according to exposure group and sex distribution Study group (year)

Abbreviation

Exposure group Unexposed % of sample

Sex

Inorganic % of sample

Organic % of sample

Men n

Women n

% Women

Total n

Smelter plant (1980)

Smelter

11.5

88.5

0

1,933

145

7.0

2,078

Regulatory office (1983)

Office

90.5

9.5

0

441

472

51.7

913

Marine transport (1989)

Marine

91.1

8.9

0

511

120

19.0

631

Trades apprentices (1989–90)

Trades

0

0

345

11

3.1

356

Vancouver Community Survey (1993–4)

Community

2.9

191

230

54.6

421

Rural/Farm (1994)

Farm

152

148

49.3

300

Sawmill (1996)

100

81.0

16.2

0

0

Sawmill

23.0

0

77.0

204

31

13.2

235

Administration office (1996)

Admin

100.0

0

0

45

52

53.6

97

Food and beverage servers (1998)

Work ETS

34.4

65.6

0

52

44

45.8

96

Health technicians (1999)

Health Tech

0

0

16

46

74.2

62

Theatre smoke (2001)

Theatre

5.9

94.1

0

68

33

32.7

101

Grain terminal (2008) Total

Grain

1.6

0

98.4

282 4,240

35 1,367

11.0 24.4

317 5,607

123

100

100

Lung (2012) 190:147–154

Among the food and beverage servers, bar workers, unlike servers with some smoking restrictions, typically were heavily exposed to inorganic fumes from environmental tobacco smoke. Organic dust exposures were predominant among farm residents (grain dust), sawmill workers (wood dust), and grain terminal workers (grain dust). Subjects included from the Vancouver Community Survey were currently employed and typically held unexposed jobs. A positive response to a specific question on exposure to vapors, gas, dust, or fumes was used to confirm exposure to either organic or inorganic substances according to their current job description. Participation rates in all studies were high (above 80%) apart from the Vancouver Community Survey participants who were invited to attend laboratory follow-up. For all studies, trained interviewers administered questionnaires, based on the American Thoracic Society (ATS) epidemiological questionnaire [9], for the collection of extensive data on respiratory symptoms and for historical and current information on personal characteristics, occupational exposures, and tobacco smoke habits. Spirometry was conducted according to the ATS recommended protocol (ATS 1995), using a Collins mechanical spirometer in the four early studies, which was replaced subsequently in 1989 by a rolling seal spirometer (model VRS2000, S&M Instrument Company, Doylestown, PA). Measurements of forced vital capacity (FVC) and forced expiratory volume in 1 second (FEV1) were converted to percent of predicted values using reference equations for each sex based on age and height [10], using 85% of the predicted value for nonCaucasians. Low values of FEV1 and FVC were defined as being less than 80% of their predicted value. Having a ratio of FEV1/FVC \ 70% was classified as a low-lung-function value. Allergy skin prick tests were carried out on the forearm. Allergens used in common were house dust mite (Dermatophagoides farinae), grass mix, and cat fur. Histamine was used as the positive control and normal saline for the negative control. For each allergen, a wheal diameter of 3 mm greater than any response to the negative control, measured 15 min after testing, was considered to be a positive reaction. A subject was considered to be atopic if there was a positive skin reaction to at least one of the skin allergens tested. Questions on the following chronic symptoms/conditions were asked in an identical manner: Chronic cough or chronic phlegm—usually having a cough (or phlegm) for at least three consecutive months; Shortness of breath— experienced while hurrying on level ground or when walking up a slight hill; Occasional wheeze—having a wheezy or whistling chest occasionally, apart from colds; Asthma—having a doctor diagnosis of asthma and current symptoms. A never smoker had smoked less than 20 packs of cigarettes over a lifetime and was considered to have

149

never smoked a cigar (less than 1/week for a year) or pipe (less than 12 oz.). An ex-smoker had smoked tobacco but as of at least 1 month prior to testing had stopped smoking, unlike current smokers. Pack-years were calculated as the average number of packs of cigarettes smoked per day multiplied by the number of years of active smoking. Analysis was done using Excel 2007 (Microsoft Corp., Redmond, WA) and SPSS ver. 18 (SPSS, Inc., Chicago IL). Univariate analysis of the pooled unweighted data set consisted of comparison of proportions by gender and exposure groups using v2 tests for significance. Metaanalysis was undertaken using Excel calculations to derive mean effect sizes based on logistic regression output on individual studies. For each individual study, unweighted odds ratios (OR) with 95% confidence intervals (CI) were initially calculated to determine the risk of respiratory symptoms and low lung function for women compared to men. Adjusted ORs by gender were calculated for each study with the following variables entered in a multiple logistic regression analysis: sex, smoking status (current or ex-smoker versus never) and personal characteristics (older age, i.e., C40 years), overweight (C25 BMI), and atopy. Non-Caucasian ethnicity was included for the analysis of symptoms. To derive a mean effect size, each study’s OR contributed according to its precision (effect size weighted by the inverse of its variance). The random-effects model was applied with the expectation of heterogeneity between studies, confirmed by the Q test for the analysis of adjusted ORs comparing women to men overall for chronic phlegm, occasional wheeze, asthma, and the ratio of FEV1/FVC. ORs were then transformed to their natural log forms and each weighted by the inverse of a between-study variance component (v) in addition to the within-study variance component (se): w = 1/(se ? v). The 95% CI of the mean effect size was derived from the standard error of the mean effect size (square root of the inverse of the summed weights from each study) multiplied by 1.96. A forest plot was used to illustrate the pooled random-effects coefficient for gender and the individual adjusted ORs and 95% CIs for each study for which a coefficient for gender could be derived (asthma having the least number of contributing studies). The vertical line through ‘‘one’’ represented ‘‘no effect,’’ when the individual study’s confidence interval overlapped it.

Results In total, there were over 5,600 study participants recruited in 12 different occupational respiratory health studies. As seen in Table 1, there were over three times as many men (n = 4,240) as women (n = 1,367). The studies in which

123

150

Lung (2012) 190:147–154

the majority of workers were women were primarily in the ‘‘unexposed’’ office category but also included health technicians with inorganic exposures. Men dominated the industrial workplaces, such as the smelter plant, trades apprentices, sawmills, and grain terminals. A greater percentage of men than women were employed in inorganicexposed jobs in industry, had worked longer on average in their current job, and were current smokers, older, overweight, and non-Caucasian (Table 2). Figures 1, 2, 3, 4 show forest plots of the individual studies with statistically significant pooled ORs. The majority of studies showed a consistent pattern (although not all statistically significant) that favored a greater risk for either women (for shortness of breath and asthma) or men (for chronic phlegm and ratio of FEV1/FVC) Of note is that men working in smelters or employed in the community survey had significantly higher chronic phlegm but significantly lower shortness of breath symptoms. Table 3 shows the raw prevalence of symptoms and low lung function for men and women according to whether they were exposed to organic or inorganic dusts or had no occupational exposures. Men had a consistently higher prevalence of chronic phlegm and a low FEV1/FVC ratio, regardless of the type of exposure. Shortness of breath and asthma tended to be elevated for women as was low values

of percent of predicted FVC and FEV1 for men. The prevalence of occasional wheeze was higher for men working in industries with organic dust exposure. With adjustment for differences in smoking and personal characteristics in a logistic regression analysis (Table 4), there were no sex and gender differences in the overall odds of having chronic cough, occasional wheeze, or low FEV1 or FVC. Overall, men had almost twice the risk as women for chronic phlegm and for a low FEV1/FVC ratio, while women had just over twice the risk for shortness of breath and asthma. Men with organic exposures, but not overall, had an elevated OR for occasional wheeze and low FEV1. Women had about twice the risk for shortness of breath among those not occupationally exposed and for those exposed to inorganic dusts. The outstanding OR was for inorganic exposure for which the OR for asthma in women was over eight times that for men.

Discussion The meta-analyses of our occupational health studies have shown that overall women had a higher risk for shortness of breath and asthma, while men had a greater risk of chronic phlegm and lower lung function, particularly for

Table 2 Occupational exposure and duration, smoking variables, and personal characteristics by sex Women n

Men %

n

p %

Occupational exposure Unexposed

925

67.7%

1,251

29.5%

Inorganic Organic

243 199

17.8% 14.6%

2,383 606

56.2% 14.3%

Job duration (mean ± SD) (years) Median years

6.7 ± 7.0

7.7 ± 8.1

4.0

4.5

\0.001

\0.001

Smoking status Never-smoker

685

50.1%

1,472

34.7%

Ex-smoker

333

24.4%

1,319

31.1%

Smoker

348

25.5%

1,449

34.2%

\0.001

Ever smokers (mean ± SD) \0.001

Cigarettes/day

16.4 ± 10.3

19.2 ± 10.5

Age started

17.4 ± 4.5

17.0 ± 4.1

Pack-years

11.5 ± 14.1

18.6 ± 18.1

\0.001

Age (mean ± SD) (years)

37.5 ± 11.6

40.4 ± 12.8

\0.001

BMI (mean ± SD) (kg/m2)

23.5 ± 4.1

25.7 ± 3.7

0.030

Personal characteristics \0.001

Non-Caucasian

122

9.0%

555

13.1%

\0.001

Atopy Childhood asthma

386 50

29.1% 3.8%

1,097 154

26.5% 3.6%

0.069 0.811

123

Lung (2012) 190:147–154

151

Fig. 1 Meta-analysis forest plot showing individual odds ratios and 95% confidence intervals for chronic phlegm, comparing women to men by study group

Fig. 3 Meta-analysis forest plot showing individual odds ratios and 95% confidence intervals for asthma, comparing women to men by study group

Fig. 2 Meta-analysis forest plot showing individual odds ratios and 95% confidence intervals for shortness of breath, comparing women to men by study group

Fig. 4 Meta-analysis forest plot showing individual odds ratios and 95% confidence intervals for low ratio of FEV1 to FVC, comparing women to men by study group

123

152

Lung (2012) 190:147–154

Table 3 Prevalence of respiratory symptoms and low lung function (unadjusted) by sex and occupational category Occupational exposure group Unexposed

N

Inorganic

Organic

Total

Women %

Men %

Women %

Men %

Women %

Men %

Women %

Men %

925

1,251

243

2,383

199

606

1,367

4,240

Respiratory symptoms Chronic cough Chronic phlegm

10.3 8.1

10.5 11.2*

13.2 8.2

15.3 14.4*

13.1 10.1

13.0 17.3*

11.2 8.4

13.5* 13.9*

Occasional wheeze

16.7

16.5

16.0

14.2

10.7

17.9*

15.7

15.4

Shortness of breath

19.7

13.2*

19.8

12.8*

21.1

15.5

19.9

13.3*

4.6

3.7

5.1

2.5*

10.1

5.0*

5.5

3.2*

FEV1 (\80% predicted)

5.6

10.8*

6.6

9.5

3.0

14.5*

5.4

10.6*

FVC (\80% predicted)

2.7

5.2*

2.2

4.4

2.0

6.2*

2.5

4.9*

FEV1/FVC (\70%)

6.0

12.3*

3.1

9.2*

9.6

17.9*

6.1

11.4*

Asthma Low lung function

* v2 statistics for the comparison of proportions was statistically significant (p \ 0.05)

Table 4 Adjusted mean odds ratiosa for risk of respiratory symptoms and low lung function, comparing employed women to men Unexposed

Inorganic exposure

Organic exposure

Overall random

Chronic cough

1.14 (0.69–1.87)

1.44 (0.68–3.09)

1.05b (0.57–1.94)

1.04 (0.76–1.44)

Chronic phlegm

0.71

0.77

0.61*

0.59*

(0.35–1.44)

(0.40–1.50)

(0.40–0.93)

(0.37–0.92)

1.06

1.22

0.48*,

0.95

(0.93–1.21)

(0.53–2.82)

(0.25–0.91)

(0.66–1.38)

2.09*

2.56*

1.57

2.07*

(1.27–3.44)

(1.26–5.21)

(0.83–2.97)

(1.57–2.73)

1.43

8.38*

2.33*

2.16*

(0.55–3.71)

(1.72–40.89)

(1.59–3.43)

(1.04–4.48)

FEV1 \ 80% predicted

1.18

1.55

0.34*,

0.94

(0.62–2.25)

(0.54–4.45)

(0.13–0.92)

(0.55–1.64)

FVC \ 80% predicted

0.88b

0.78b

0.38b

0.76

(0.50–1.53)

(0.23–2.69)

(0.12–1.26)

(0.56–1.06)

0.85

0.47

0.55

0.55*

(0.49–1.46)

(0.17–1.27)

(0.14–2.11)

(0.33–0.92)

Occasional wheeze Shortness of breath Asthma

FEV1/FVC \ 70%

b

b

a Based on meta-analysis of combined odds ratios [with each study OR adjusted for smoking (ex- or current vs. never), older age (C40 years), overweight (C25 BMI), non-Caucasian, and atopy, job duration (C4 years)] and 95% confidence intervals derived from each study containing the exposure group. For lung function outcomes, there was no adjustment for race b

Fixed effects presented; random-effects model output was unstable

* Odds ratio for gender difference in risk was statistically significant (p \ 0.05)

the FEV1/FVC ratio. Our present findings are consistent with much of the scientific literature on community-based populations. Shortness of breath is associated with impaired ventilatory function and is consistently reported more frequently in women than men in population-based studies [4]. Data from the European Respiratory Health Study strongly supports a female predominance in asthma

123

incidence [5]. Gender difference in asthma has been postulated to be a function of biological differences in airway and parenchymal size and hormonal status, although differences in environmental exposure and medical management also may play a role [11]. In a meta-analysis of occupational organic exposure studies [1], men were consistently at a higher risk of

Lung (2012) 190:147–154

chronic phlegm, whether working in food processing, textile, or farming industries. We also found an increase in the prevalence of chronic phlegm for men exposed to organic dusts. Men in the textile industry [1] had comparatively low FVC (from expected values) and the ratio of FEV1 to FVC could not be derived. We found that men exposed to organic dusts had a greater risk of having an FEV1/FVC \ 70%. This low ratio has prognostic value for chronic obstructive pulmonary disease and is supported by the Global Initiative for COPD (GOLD) [12]. A novel finding of our study is the demonstration of sex differences in respiratory health, which differed somewhat from unexposed workers according to whether exposure was inferred to organic or inorganic substances. Men exposed to organic dusts through their employment in the grain terminal, farming, or sawmill industries had a greater risk of occasional wheeze and low FEV1 than women coworkers. Among workers exposed to inorganic dusts, the risk of current asthma was extremely high for women relative to men; this is in contrast to no gender differences for unexposed workers. It may be speculated that the high risk for asthma in women may be a result of asthmatic men avoiding employment in more highly exposed jobs. The occupational health studies combined in our metaanalyses were overseen by the same team of researchers, who used the ATS epidemiological questionnaire [9] and similar measurement protocols. Over three times as many men as women took part in the various occupational respiratory health studies over a 30-year period. This disparity in numbers is typical of occupational health surveys of industrial exposures in workplaces dominated by men [13]. The approach to evaluate gender differences separately for each study is hindered by low power due to the relatively small number of women employed in such industries. Typically, in studies on occupational and environmental health, the sex variable is adjusted for or reporting is limited to the dominant sex only [2]. The metaanalyses technique allowed for weighting of each study by its variability and calculation of a mean effect size for each type of exposure. However, the question of whether respiratory health differences were attributed to sex ‘‘susceptibility’’ due to physical or biological differences or to gender differences in social roles and exposure cannot be determined [14]. For instance, our finding that women had a greater risk of shortness of breath, regardless of exposure, would support the influence of biological sex determinants. Yet, this symptom can also be influenced by consistency of gender differences in perception and reporting [4]. There were a number of limitations with respect to our study methods. In the calculation of percent of predicted lung function values for determining low lung function, the coefficient for the sex variable in the predictive equations used [10] may not necessarily be accurate for women or

153

men. The exposure metric was limited to binary variables, which was a all studies. In determining the exposure category from the individual job titles, there was the potential for misclassification. Random misclassification would tend to attenuate the risk estimate. The exposure groups were based on current job title, without consideration of job history. Nonparticipation, particularly in the communitybased study, may have introduced selection bias. The potential for differential healthy worker bias must be considered in this aggregation of occupational health studies, whereby men or women may be more likely to avoid certain work exposures or retire early upon the development of health problems. Women and men exposed to organic or inorganic dusts may have differed from each other and from those not exposed with respect to attributes related to respiratory health status. Such differential sources of error may bias the outcome and either attenuate or exaggerate the risk estimate. Within each exposure classification, differences between male and female workers in the duration of work were statistically adjusted for; however, men and women may be assigned to different tasks that could influence their exposure levels [15]. It would be anticipated that men generally had higher exposures within the occupational exposure category. Because we lacked quantitative or even qualitative exposure measurements in many of the individual studies, we could not test for this directly. However, it was noted that among smelter workers, for example, men had a greater risk of chronic phlegm but lower odds of shortness of breath. However, a higher percentage of men worked in the pot room, which has the highest inorganic exposures in the metal refinery process [16]; thus, the level of exposure does not necessarily affect all respiratory health parameters equally for men and women. Because we compared workers with two different types of exposures to workers who were most likely unexposed, we were able to show that for some respiratory health outcomes there was consistency, regardless of exposure (therefore, job assignments within a workplace would not be relevant for unexposed men and women). On the other hand, showing differences in outcome dependent on the type of exposure does emphasize occupational exposure influences on the extent of gender differences in respiratory health. Overall, gender differences observed in respiratory health in population-based studies were confirmed in our analysis of occupational health studies. The higher risks for women for shortness of breath were robust, regardless of work exposure, and were consistent with general population studies. Furthermore, the general type of exposure, organic or inorganic, determined the extent of differences between men and women in specific respiratory symptoms and low-lung-function measures. This was particularly evident in the risk estimates for asthma, which ranged from

123

154

Lung (2012) 190:147–154

no gender differences in those without obvious occupational exposure to over an eightfold higher risk for women versus men exposed to inorganic dusts. Acknowledgments This study was funded by the Canadian Institutes of Health Research (CIHR), Institute of Gender and Health (IGH), and Institute of Circulatory and Respiratory Health (ICRH). The authors appreciate the support of members of the Interdisciplinary Capacity Enhancement: Bridging Excellence in Respiratory Disease and Gender Studies (ICEBERGS) team at the University of British Columbia for their encouragement to create the combined data set and evaluate occupational influences on gender differences in respiratory health. Conflict of interest

None

References 1. Krieger N (2003) Genders, sexes, and health: what are the connections—and why does it matter? Int J Epidemiol 32(4): 652–657 2. Clougherty JE (2010) A growing role for gender analysis in air pollution epidemiology. Environ Health Perspect 118(2):167–176 3. Silvaggio T, Mattison D (1994) Setting occupational health standards: toxicokinetic differences among and between men and women. J Occup Med 36:849–854 4. Becklake MR, Kauffmann F (1999) Gender differences in airway behaviour over the human life span. Thorax 54(12):1119–1138 5. de Marco R, Locatelli F, Sunyer J, Burney P (2000) Differences in incidence of reported asthma related to age in men and women a retrospective analysis of the data of the European Respiratory Health Survey. Am J Respir Crit Care Med 162(1):68–74

123

6. Larsson ML, Loit H-M, Meren M et al (2003) Passive smoking and respiratory symptoms in the FinEsS Study. Eur Respir J 21(4):672–676 7. Zuskin E, Mustajbegovic J, Schachter EN, Kern J, Budak A, Godnic-Cvar J (1998) Respiratory findings in synthetic textile workers. Am J Ind Med 33(3):263–273 8. Schachter EN, Zuskin E, Moshier E et al (2009) Gender and respiratory findings in workers occupationally exposed to organic aerosols: a meta analysis of 12 cross-sectional studies. Environ Health 8(1):1 9. Ferris B (1978) Epidemiology Standardization Project (American Thoracic Society). Am Rev Respir Dis 118:1–120 10. Crapo R, Morris A, Gardner R (1981) Reference spirometric values using techniques and equipment that meet ATS recommendations. Am Rev Respir Dis 123:659–664 11. Redline S, Gold D (1994) Challenges in interpreting gender differences in asthma. Am J Respir Crit Care Med 150(5):1219–1221 12. Celli BR, Halbert RJ (2010) Point: Should we abandon FEV/FVC less than 0.70 to detect airway obstruction? No. Chest 138: 1037–1040 13. Messing K, Punnett L, Bond M et al (2003) Be the fairest of them all: Challenges and recommendations for the treatment of gender in occupational health research. Am J Ind Med 43(6):618–629 14. Niedhammer I, Saurel-Cubizolles MJ, Piciotti M, Bonenfant S (2000) How is sex considered in recent epidemiological publications on occupational risks? Occup Environ Med 57(8):521–527 15. Messing K, Mager Stellman J (2006) Sex, gender and women’s occupational health: the importance of considering mechanism. Environ Res 101(2):149–162 16. Chan-Yeung M, Enarson DA, MacLean L, Irving D (1989) Longitudinal study of workers in an aluminum smelter. Arch Environ Health 44:134–139