Accurate information on disease prevalence of COPD is important to understand its impact on disability, quality of life and healthcare costs, and to inform on public health planning (4). Baseline prevalence rates also allow epidemiological analysis to monitor trends, and determine the success or failure of control efforts.

Despite the worldwide high prevalence of COPD, studies of airway diseases in Caribbean territories have concentrated on asthma. The single study which looked at the prevalence of COPD in Trinidad using measurements of lung function was a 2004 study of 720 acute medical admissions at the General Hospital, Port of Spain which showed that COPD was present in 21% of acute medical admissions. The study also showed that subjects with a low FEV1 were more likely to have cardiovascular disease (5).

Dyspnoea was assessed using the Modified MRC Dyspnoea Scale (8) and post-bronchodilator spirometric indices were assessed according to the guidelines of the American Thoracic Society (ATS) (7) by open-circuit testing. Data that was included in the analysis satisfied the parameters of acceptability and reproducibility (9). Subjects were classified into smokers and never-smokers; and by number of cigarettes used per day or grams of tobacco smoked per week. Patients were also asked to recall the number of chest infections requiring antibiotics or hospital admission within the past year.

The prevalence of COPD was higher in women than men in our study but this was because more women attended the health centers sampled though there is substantial evidence to suggest that women may be predisposed to suffer adverse respiratory consequences of tobacco smoke with greater impairment of lung function and earlier COPD development (10). COPD surveillance data from the United States has shown increased mortality rate in females compared to males (11). It may be argued however that women, having higher life expectancies, may simply be living longer so as to succumb to the effects of smoking. Additionally there is substantial data to suggest that further study into gender differences in COPD should be investigated as differences exist in the clinical manifestations of COPD, as well as its gender prevalence (12-14).

It should be noted that smoking may not be totally responsible for the higher prevalence of COPD in females in our study. Other known risk factors for the development of COPD include air pollution, infections and occupational exposures. Air pollution, particularly fine particulate indoor air pollution from biomass fuels disproportionately affects women. Studies have shown that women are more susceptible to the effects of biological or organic dust exposure reflected by an increased prevalence of respiratory symptoms and chronic bronchitis (15). Exposure to occupational dust has been shown to be related to higher COPD prevalence as well, especially in women (16). The BOLD study found that Cape Town, South Africa had the highest prevalence of stage 2 or higher COPD as well as high levels of occupational dust exposure as well as smoking rates (17). In our study, most patients with COPD were in stage II (53.2%), a finding that may hint at the need for further investigation of occupational exposure in the development of COPD in Trinidad. Measures may be needed to decrease such exposures, promote cleaner fuels, improved stoves, better home ventilation and reduce toxic dust and fume exposures.

Smoking is the major known environmental risk factor for the development of COPD and as expected, it was found that a higher proportion of patients with COPD were smokers compared to the non-COPD patients. Consistent with this was the finding that FEV1% was lower in smokers. Further as previously found COPD patients smoked for a greater number of pack years than non-COPD patients. A greater percentage of COPD patients had chronic cough, linking smoking pack years and development of these symptoms. This finding is especially intriguing as one study has shown that cough may be a better predictor of airflow limitation and when used to preselect smokers for spirometry testing, the proportion with an FEV1 less than 80% was increased. Thus increased emphasis on respiratory symptoms in smokers may aid in the detection of COPD by targeting higher risk patients (18). Studies have also shown that the prevalence of undetected airflow limitation is high among asymptomatic smokers, a finding that supports the need for targeted screening of patients who present with chronic cough, and dyspnoea (19).

Comorbid conditions were very common in our study and previous studies have confirmed that pulmonary function, represented by FEV1, is an independent risk factor for Ischemic Heart Disease mortality (20). In fact, 71.4% of COPD patients in our study presented with cardiovascular disease, indicating a major link with reduced lung function suggesting that greater emphasis should be given to investigation of comorbid conditions by West Indian physicians in the management of COPD. Further, individuals with COPD are more likely to be at risk of vascular events due to preexisting cardiovascular disease (21).

We also obser ved that in patients presenting with cardiovascular conditions, FVC values were significantly lower than in patients without vascular disease. It has been postulated that diminished respiratory function as measured by FVC is associated with increased risk of cardiovascular mortality but this is largely unexplained. However previous studies by Friedman et al (22) suggest that clinical research investigating predictive value of FVC is necessary, in line with our findings.

Our study has limitations. The cross-sectional design limits our ability to describe progression of the stages of COPD. In patient recruitment, our study focused on outpatient clinics so that our sample did not represent those who do not attend these health centers and it is likely that there was a gender bias in sampling because of this. The non-COPD patients had reduced FEV1 and FVC as estimated from that predicted for age and height and gender. This is not unexpected as the patients were recruited from the chronic disease clinics of the primary care system. Diabetes, obesity and cardiac failure are associated with a restrictive ventilatory defect. Thus because of reduced FVC, we may have missed some of the COPD patients within these chronic disease clinics.