Module 4: Chronic Obstructive Pulmonary Disease and Occupations - Epidemiology of COPD

EPIDEMIOLOGY OF COPD:

COPD is a leading cause of morbidity and mortality in developed countries. The epidemiology of COPD in developing countries is less well studied. It is clear that TB can also play a major role.

The table below summarises some of the known and putative causes of COPD, classified as host or environmental factors.

SOURCES OF EVIDENCE FOR OCCUPATION AS A RISK FACTOR FOR COPD:

Sources of evidence for occupational as a risk factor for COPD comes from a wide variety of studies. These vary greatly in their design and the subjects participating in them. Three examples will be given:

• Community based studies (example of the DHS).
• Cross-sectional studies (example of gold miners).
• A combination of Cross-sectional and Longitudinal studies (example of cotton dust).

Evidence from a community based study:

Consider the risk factors for COPD in South Africa. South Africa had a population of 43 million in 1999, of which 14 million are below the age of 15.

• 46% live in urban areas.
• Smoking rates: 42% males, 11% females.
• 66% of homes electrified (80% urban, 46% rural) but 20% use electricity for lights only.
• High tuberculosis incidence ? > 400/100 000.
• The most industrialised country on the continent with a diverse economy and a large mining sector.

Because South Africa has a relatively young population COPD may not seem to be of particular public health significance at present. However, large numbers of people are currently exposed to risk factors for COPD.

Community based study (example of the DHS):

The Demographic and Health Survey (DHS) (MRC, 1995) was a South African community based study that set out to evaluate a wide range of health related questionnaire responses in relation to the country’s demographics. Amongst its many findings were some that supported a role for occupational exposures as a risk factor for COAD.

The DHS was a national household survey with 2 stage sampling. 13827 adults participated. (Women: 8073, Men:5753).

Participants completed a household questionnaire providing demographic data on age, sex, race, education, urban status, wealth and household fuels and a health questionnaire and measurements : (medical history, habits, smoking, past TB, occupational exposure, height, weight, PEFR).

Respiratory questionnaire outcomes of interest were chronic bronchitis: Cough/phlegm daily at least 3 months, > 2 yrs. The following two tables summarise the important findings:

The above table gives the prevalence of chronic bronchitis in men and women with the risk factors indicated. The prevalence ratio indicates how much more common chronic bronchitis is in people with the risk factor compared to those without the risk. In this simple bivariate analysis a history TB seems to be the most important factor.

The table above the results of multivariate analysis of the DHS data. In this analysis the population attributable fraction for the burden of chronic bronchitis in SA is estimated. This is a measure of the relative contribution of each of the risk factors to the burden of disease observed in the population. The results suggested that occupational exposures in men were the most important risk factor for chronic bronchitis, whereas amongst women past TB was most important.

Summary of DHS findings:

The main risk factors for chronic bronchitis symptoms in male adult South Africans identified in the study were positive responses to the following questions (in order of relative importance):

• "ever worked in a job regularly exposed to smoke, dust fumes or strong smells?"
• "ever told by a doctor or nurse that you had TB?"
• Do you smoke?

Linking occupational risk and COPD (Cross sectional and longitudinal studies):

Many respirable dusts, smoke or fume appear to be linked to accelerated loss of lung function, primarily obstructive. The mechanisms whereby this takes place are varied, but the end result is indistinguishable. These agents include, but are not limited to the following:

• Inorganic dust;
• Silica;
• Coal;
• Organic dust;
• Cotton;
• Grain;
• Smoke or fume;
• Welding;

Although the mechanisms whereby various agent damage the airways may vary, the net effect is similar irrespective of the causative agent and COPD caused by smoking, mineral or organic dust does not have any pathologically distinguishing features. The figure on the right illustrates what has been observed in many studies (in this instance a comparison of workers in the steel industry). The figure graphs the % predicted FEV1 against years of exposure in the industry.

Given appropriate reference values for FEV1, it would be expected that these populations would, with aging, maintain the same % predicted since the prediction equation takes age into account. Instead the graph shows that in all three studies the workers have a declining FEV1 % predicted. Differently put, these workers have accelerated loss of lung function, over that which would be expected due to age alone. Accelerated loss of lung function is the single most important measurable feature accompanying occupationally and tobacco related COPD and identifying this accelerated decline is the major objective of medical surveillance programmes using lung function tests in dust, smoke or fume exposed workers.

Cross-sectional studies (example of gold miners):

Cowie and Mabena (1992) conducted a cross-sectional study of 1 197 current underground goldminers who completed a questionnaire, had lung function tests and chest x-rays.

Their main finding was that:

Expressed differently, the authors had found that the exposure to the mining environment had an effect on lung function that was thought equivalent to the effect of smoking 20 cigarettes a day.

Cross-sectional studies and Longitudinal studies (example of cotton dust):

The picture shows a cotton textile worker (CTW) cleaning a carding machine. Carding occurs early in the cotton textile production process and the raw cotton processed still contains many contaminants. These contaminants are complex and heterogeneous. From a pathogenetic point of view the most important contaminants in respirable cotton dust are thought to be the endotoxins of bacteria and fungi.

Cotton textiles environment: mechanisms:

Endotoxins present in the dust activate alveolar macrophages, resulting in an inflamatory response in the airways. This aspect of the inflamatory response to cotton dust is thought to be more important than any allergic response to the dust.

The respiratory disease resultant on prolonged cotton dust exposure is termed byssinosis (from the Greek bysson- for linen). It is best understood, not as special type of occupational asthma, but rather a cause of occupational COPD with a clearly defined cause.

Respiratory disease indistinguishable from byssinosis has also been documented in other contexts of gram-negative endotoxin exposure, such as in the context of animal confinement workers in agriculture, particularly as practiced in the Northern Hemisphere.

Exposure in the cotton textile environment has well documented acute respiratory effects.

• The figure above shows a study of lung function response to work in CTW. FEV1 is measured before and after work, Monday to Friday.
• The workers have been divided into three groups. Asymptomatic workers show little change in FEV1. Those with chest tightness at work on a Monday appear to have significant decrements in FEV1 related to their exposure at work, particularly those who regularly have such symptoms.
• These work-related declines in lung function are related to inflamatory changes in the airways. Not all individuals are equally sensitive to the dust effect.

Exposure in the cotton textile environment has been shown to result in accelerated loss of lung function.

The figure shows the result of study where CTW in the (now defunct) Lancashire cotton textile industry are compared to controls in the engineering industry.

The CTW show declining FEV1 % predicted over their years in the industry, an effect that is much less marked in the control workers.

Longitudinal studies have shown a positive correlation between annual change in FEV1 and across-shift change in FEV1. Workers who have larger changes in FEV1 across a work shift are likely to have more rapid decline in lung function when observed longitudinally.

Exposure in the cotton textile environment has been shown to result in loss of lung function both in smokers and non-smokers:

The figure above illustrates a study of CTW wherein they are compared to community controls. The non-smoking CTW and controls have been analysed separately from the total study group.

The figure depicts the distribution of FEV1 % predicted in the study sub-groups. Low FEV1 % predicted was commoner in the CTW. Low FEV1 % predicted was commoner amongst non-smoking CTW than amongst controls.

Tobacco smoking interacts with many occupational exposures. Usually the effects of this interaction are additive, rather than synergistic with respect to airways disease.