Prevalence and associated risk factors of chronic bronchitis in First Nations people

The overall respiratory health of Aboriginal peoples (First Nations, Metis, and Inuit) is poorer than that of the general Canadian population [1, 2]. Rates of respiratory diseases including asthma, chronic bronchitis (CB), and chronic obstructive pulmonary disease (COPD) are higher in Aboriginal peoples [1, 3]. According to the First Nations Regional Longitudinal Health Survey, in 2002/03 the age-standardized prevalence of self-reported, physician-diagnosed CB was 3.7% in First Nations people living on reserves [4]; and according to the Canadian Community Health Survey (CCHS) 2005, the prevalence of CB in Aboriginal people living off-reserve was 4.9% [5]. In a recently published article, based on data from the 2006 Aboriginal Peoples Survey (APS), the prevalence of CB was 6.6% among off-reserve First Nations people (5.0% in males and 7.2% in females) [2]. These rates were higher than the prevalence of CB for the non-Aboriginal Canadian population (2.4%) [5]. CB is a respiratory disease defined as “cough productive of sputum for at least 3 months of the year for at least 2 years” [6]. Research has shown that CB is a significant cause of morbidity and an underlying condition for the development of COPD [7]. Based on limited data available for on-reserve First Nations, it was reported that First Nations adults living on reserves have higher age-adjusted rates of CB compared to other Canadians [8, 9].

Exposure to cigarette smoke, inadequate housing, low socio-economic status, and obesity are some of the factors associated with the increased prevalence and incidence of CB [5, 10, 11]. These factors are more prevalent among First Nations people as compared to the general Canadian population and might be responsible for the high prevalence of CB in Aboriginal peoples [1, 9, 12‐15]. Approximately 11% of Canada’s Aboriginal peoples live in Saskatchewan [16]. To date, knowledge about the risk factors associated with CB among on-reserve First Nations peoples is limited and has not been well established. The purpose of the baseline component of First Nations Lung Health Project (FNLHP) conducted in 2012 was to address these gaps. The objective of this manuscript was to compute the prevalence rates of CB and determine its associated risk factors in on-reserve First Nations people.

The FNLHP is a prospective cohort study being conducted using interviewer-administered surveys in two phases, the baseline, and the follow-up. The baseline survey was completed between 2012 and 2013. The details of this study are given elsewhere [17]. In this study, through the application of a Population Health Framework, we attempt to understand the association between the individual and contextual factors; as well as the interactions between them in relation to respiratory health outcomes after adjusting for important covariates [18]. In brief, two on-reserve communities were invited to participate in this study. Henceforth, we will call these communities Community A and Community B. These communities were selected based on a long-term previously established relationships. The questionnaire collected information on individual and contextual factors and a history of ever diagnosed with CB (outcome variable). Some measures of housing conditions, life-style and socio-economic status used in our questionnaires were adopted from previous research studies that had validated these measures [18, 19].

The mean age of the study population was 34.83 ± 14.47 years and 17.8% were older than 50 years. The demographics of the population were: 52.1% females; 28.8% overweight; 35.4% obese; 76.8% current smokers; and, 13.5% ex-smokers. The overall prevalence of CB was 7.8% (Fig. 1). Prevalence of CB in Community A was 8.9% (6.8% in males and 10.6% in females) and in Community B was 6.8% (6.0% in males and 7.5% in females) (Fig. 1). In both communities, the prevalence of CB in females was higher compared to the prevalence of CB in males but not statistically significant (Community A - p value = 0.21; Community B - p value = 0.57). The univariate relationships between the contextual factors and individual factors or covariates and CB using unadjusted logistic regression are shown in Table 1. The multivariable logistic regression results are presented in Table 2. The significant predictors of CB were: increasing age; the presence of mildew odor or musty smell in the house; having an air conditioner; and, having a reported allergic reaction to house dust. Exposure to environmental tobacco smoke in the house modified the relationship between BMI and chronic bronchitis (p < 0.05). That is, a person who was obese or overweight and exposed to environmental tobacco smoke in the house had an increased risk of self-reported, doctor-diagnosed CB compared to a person who was not exposed to environmental tobacco smoke in the house (Fig. 2).

A sub-analysis was completed to explore the 2-way interaction between sex and smoking (in terms of pack-years) and CB. Smoking pack-years, a continuous variable, was divided into three categories: 0 for never-smokers; > 0 and < 10 pack-years (moderate smokers); and ≥ 10 pack-years (heavy smokers). Based on crosstabs analysis, we observed (data not shown), for females there was a significant positive association between smoking pack-years and the prevalence of CB (5.4% self-reported, doctor-diagnosed CB among never-smokers, 4.3% among ex-smokers, and 21.3% among current-smokers); however, this relationship was not significant for males. The association between smoking pack-years and prevalence of CB was different for males and females, indicating that sex is an effect modifier in the relationship between smoking pack-years and the prevalence of CB. However, we were not able to test this interaction in the multivariable model because of relatively small numbers due to pack-years information missing for approximately 40% of the study population.

By using the information from the baseline component of the longitudinal study, we determined the prevalence of CB and examined the associated risk factors in First Nations people residing in two on-reserve communities in Saskatchewan. In this study, a fairly good response for household participation for completing the baseline survey via interviewer-administered questionnaire was observed (53.9% in Community A and 89.9% in Community B). The overall prevalence of CB was 7.8% in these two on-reserve First Nations communities (6.4% in males and 9.1% in females). Modifiable risk factors identified were housing conditions such as mildew odor or musty smell in home, having an air conditioner in the home, exposure to environmental tobacco smoke in the house and obesity in the communities. The other significant risk factors of CB were increasing age and allergic reaction to house dust.

Geographic location (reserve) appears to have a large impact with respect to health status and use of physician services [22‐24]. The prevalence of on-reserve First Nations peoples (7.8%, based on the current study) was higher than the off-reserves Aboriginals (6.6% based on the 2006 APS [2] and 4.9% based on the 2005 CCHS [5]). In contrast, the 2002/03 First Nations Regional Longitudinal Health Survey reported that age-standardized prevalence of self-reported, doctor-diagnosed CB was 3.7% in First Nations living on-reserve [4]. All these rates are higher than the prevalence of 2.4 and 2.5% found in the non-Aboriginal Canadian population, according to the 2005 CCHS and 2007/2008 CCHS, respectively [5, 25]. A recent study of Saskatchewan rural residents (excluding Aboriginal residents) reported a prevalence of 5.9% [26]. These differences in the prevalence of CB support the findings that geographical location (on-reserve, off-reserve and rural) has a large impact on respiratory health.

An article based on the 2006 APS including off-reserve First Nations, Metis, and Inuit people observed that older age, lower educational attainment, lower income, and urban residence were significantly associated with self-reported doctor-diagnosed CB [2]. It was also reported that smoking status and body mass index were significantly associated with CB, and their effects differed by sex [2]. A recent Saskatchewan Rural Health Study (SRHS) [26] reported a positive association between greater prevalence of CB and lower household income adequacy, increasing age, allergies, history of lung disease in a parent, exposure to stubble smoke, obesity, prenatal exposure to smoking, and female sex. The SRHS also reported that smoking modified the relationship of CB with occupational exposure to wood dust, solvents and allergic reaction to solvents [26]. Similar to these studies, the current study reported that older age was significantly associated with CB. In contrast, based on our data we did not observe any association between education level, income level, and the prevalence of CB. This could be due to small numbers and less variation in the data related to income and education. In this study, approximately 27% of households reported an annual income of < $10,000 or no income, and, 15% reported being in the $10,000–$19,999 annual income range. Thirty percent of households did not report the income. About half of them had less than a high school education. For the current study, we did not collect information on a history of lung diseases in a parent, exposure to stubble smoke, and prenatal exposure to smoking. There was a positive observable association between allergies and CB when we conducted univariate analysis, but the association was not significant when multivariate analysis was conducted.

Exposure to environmental tobacco smoke (household smoke) is the most important risk factor for the development of CB as reported by several studies [2, 25‐29]. The current study did not report a significant association between personal non-traditional use of tobacco and the prevalence of CB; however, there was a significant interaction with environmental tobacco smoke and BMI. In the participating two on-reserve communities, 77% were currently engaged in the non-traditional use of tobacco, 13% had been ex-smokers and only 6% of smoke-free individuals implying that almost everyone in the household had been exposed to secondhand smoke. Hence, we decided to investigate the association between household smoke and CB. It is possible that, due to the high proportion of those who currently or have in the past engaged in the non-traditional use of tobacco, we did not find any positive association between non-traditional use of tobacco and CB in spite of a vast literature available on the positive association between smoking and CB. Data from the CCHS 2000/2001 reported that self-reported recent secondhand smoke exposure in Canada was high and was associated with asthma, chronic bronchitis, and hypertension in never- and ex-smokers [30]. Few other studies reported the relationship between secondhand smoke and CB [31, 32].

In our study, the prevalence of CB was 6.4% for males and 9.1% for females, supporting the earlier finding of higher prevalence of CB in females in Aboriginal and rural populations [2, 26]. Based on our recent study of small town and rural people in Saskatchewan, we observed that females were significantly more likely to self-report doctor-diagnosed CB (OR 1.32, 95% CI: 1.10–1.58). In our study, the proportion of current smokers (78.5% vs. 76.0%) were higher for males compared to females, and there were higher proportions of ex-smokers (14.4% vs. 12.1%) among females compared to males. Research has shown a strong positive association between smoking and CB and a higher prevalence of CB in females [2, 25‐28]. A univariate sub-analysis of this study revealed that the association between smoking pack-years and prevalence of CB was different for males and females, indicating that sex is an effect modifier in the relationship between smoking pack-years and the prevalence of CB.

Research has shown that environmental tobacco smoke contains more than 4000 chemicals and many of these are potent respiratory irritants [33]. These chemicals pollute the environment and damage human lungs when inhaled [33]. In the current study, BMI modified the relationship between exposure to secondhand smoke in the house and the prevalence of CB. Overweight and obese individuals who were exposed to secondhand smoke in the house were more likely to self-report doctor-diagnosed CB compared to those who were not exposed to secondhand smoke in the house.

An intergenerational link has been shown between attendance at a residential school, smoking, and CB [4, 34]. We explored the association between “Have you ever attended the residential school?” and “Have either of your parents/grandparents attended the residential school?” and the prevalence of CB. We did not find a significant association between either of the variables related to residential school and the prevalence of CB.

The prevalence of CB in the First Nations people could be high because risk factors associated with the prevalence of CB are more common in these communities compared to the general Canadian population. CB occurred as a result of both non-modifiable (such as age) and modifiable risk factors. Modifiable risk factors identified were housing conditions such as mildew odor or musty smell in the home, secondhand smoke in the house, and obesity. We observed interaction effects on prevalence of CB between secondhand smoke in the house and BMI (adjusting for other variables in the model) and smoking pack-years and sex (not adjusting for any other variable) for First Nations peoples. These interaction effects have not been widely reported previously about on-reserve First Nations communities. We have established a baseline for the prevalence of CB among First Nations peoples in Saskatchewan against which future research related to the control of this disease can be developed.