Studies on the prevalence of COPD in Latin America (LA) are scarce; the PLATINO project was a multicenter study carried out in five centers in LA, filling an important gap of knowledge in this area.1

The original aim of the PLATINO study, launched in 2002, was to describe the epidemiology of COPD in five major LA cities: São Paulo (Brazil), Santiago (Chile), Mexico City (Mexico), Montevideo (Uruguay), and Caracas (Venezuela), among adults aged 40 years or more, since subjects less than 40 years of age have low prevalence of COPD.2 A secondary objective was to evaluate the prevalence of self-reported medical co-morbidities that can occur quite often with COPD, such as cardiovascular diseases and lung cancer.

These sites represent the various different geographical areas of LA and the largest metropolitan area in each participating country. The study was conducted from 2002 to 2004 as an initiative from the Associación Latinoamericana del Tórax (ALAT).

The PLATINO baseline study revealed a high prevalence of COPD in Latin America based on the fixed ratio criteria (FEV1/FVC <70% post bronchodilator), figures previously unknown for Latin America.3,4

Some of the most relevant aspects of the PLATINO baseline study were: (a) the fact that it was a population-based study and that most of the studies on COPD in the literature are carried out among selected COPD patients; (b) the high response rate in all centers; and (c) the high quality of the spirometries.

The detailed methods of the study are published elsewhere.5

However, this was a cross sectional study comprising of a visit at one point in time and this does not allow us to analyze temporality and some outcomes of interest in COPD. Therefore, after 5 years, it was decided that a follow-up of the study in some centers from the original project should be carried out.

Several issues were raised concerning the difficulties of following people in LA, such as the high level of migration and mobility, the lack of official registries to locate the same subjects, the possibility of death with subjects aged 40 years and older since the baseline study, and the adequacy of the information on vital statistics for certifying death. Another potential important problem was the increased violence in LA, which could be a limitation for visiting households in some areas of the cities. Funding was obtained to repeat the study in three centers out of the five; they were chosen according to the limitations listed above and to the prevalence level of COPD found in the baseline.

The first follow-up visit took place in the city of Montevideo (5 years after the baseline), given that this center presented the highest prevalence of COPD in the PLATINO baseline study (19.7%; 95% CI 17.2–22.1), it is a small sized city, and it has very low mobility level among people. Santiago was chosen as the second center for the follow-up, as it occupied the second place in the ranking of prevalence of COPD (16.9%; 95% CI 14.7–19.1), it presented the highest prevalence of smoking in the baseline study (38.6%), and because it is considered a safe place in LA. Finally, the follow-up was done in a third center (Sao Paulo), since the city was ranked as the third one on COPD prevalence (15.8%; 95% CI 13.5–18.1).4 The time interval between the two visits in each of the centers was different, varying from 5, 6 to 9 years in Montevideo, Santiago, and Sao Paulo (SP), respectively. The main reason for the different times between both phases of the study was finding the resources to fund the project. The general objectives of the follow-up study were: (a) to evaluate the natural history of COPD in the same sample of subjects from the original PLATINO project; (b) to validate the reliability of original diagnosis of COPD; (c) to describe the natural history of COPD in terms of survival, morbidity, occupational history, respiratory function, respiratory symptoms, exacerbations, general health status, physical activity limitation, management, hospitalization, absenteeism, nutrition, among others; and (d) to obtain blood samples for analyzing biomarkers.

The present paper describes: (a) the methods used in a population-based longitudinal study in Latin America; (b) the preliminary results, such as follow-up rates, spirometry quality control, information on the deaths occurring during the period; (c) the procedures for collecting and storing blood samples; and (d) the main limitations for carrying out such studies in Latin America.

The design of the follow-up visit was cross sectional and the target population was the same one that had been investigated and that had performed spirometry; as we ended up with two visits over time for the same population, we can say that the design of the overall study is a cohort design.

Each census tract and household visited in the baseline study was contacted again and the same procedures were employed as in the PLATINO baseline study. All eligible adults from the original sample answered a questionnaire and performed the same procedures as before.5

The previous and new variables collected in both phases of the PLATINO study are listed in Table 1. Details of the questionnaires can be found at the PLATINO website.1

The fieldwork was done sequentially and, therefore, we had different interval times between the two visits for each center. Table 2 shows the mean time of follow-up for each center, of which Sao Paulo took the longest. The number of deaths, refusals and losses was higher in Sao Paulo than in the other centers; nevertheless, the overall response rate for spirometries (95%) was nearly the same among the three centers. The overall response rate for the blood samples was 75.3% in Montevideo, 82.7% in Santiago and 60% in Sao Paulo (Table 2).

The follow-up rates according to key baseline characteristics by country are shown in Table 3. In the period of 5–9 years, the percentage of follow-up rates did not vary according to sex, age, skin color, family history of chronic bronchitis, emphysema or COPD, passive smoking, waist circumference, and Body Mass Index (BMI). Subjects from lower schooling levels were more likely to be located in the city of Sao Paulo, although there was no difference compared to the other sites. With regard to smoking status, there was a higher percentage of follow-up rates for non-smoking subjects compared to smokers (88.5% vs 80.6%) in Santiago, and a higher percentage of traced subjects among ex-smokers (82.0%) in Sao Paulo.

Regarding the overall quality of spirometries in both phases of the study, 80% or more of the tests achieved the ATS criteria in all centers (Fig. 1); it can also be observed that baseline spirometries showed a higher quality than follow-up tests (Fig. 1).

Fig. 1.

Quality of spirometries according to the ATS criteria for the three centers in the baseline and follow-up. The PLATINO study.

(0.1MB).

The death certificates were obtained from the national death registry of each country, with 76.1% of the death certificates being located in Montevideo, 88.3% in Santiago and 84.5% in Sao Paulo.

To the best of our knowledge, the PLATINO study is the first longitudinal population-based investigation addressing COPD in Latin America and will likely provide important information on the natural history of the disease. As its sample is representative of the adult population over 40 years of age in urban areas of Latin America, it can also contribute to the evaluation of other outcomes, such as overall mortality, the impact of public health policies (particularly on smoking), comorbidities (heart disease, hypertension, diabetes and lung cancer) as well as the management of patients with COPD.

The literature offers a number of important, large studies on the natural history of COPD and associated mortality rates. However, most of these investigations are not population-based studies.24–26 In Spain, there are two cross-sectional population based-studies (IBERPOC and EPI-SCAN) aimed at comparing the trends on the prevalence of COPD.27 The authors found a decline of 50.4% in the prevalence of COPD over a 10-year period. The results of these two studies are very relevant, but they cannot be analyzed as a cohort design since they were carried out in two different populations.

The fact that we could work with the same principal investigators in the three centers and some of the technicians from the baseline study helped ensure the quality of the data. However, some limitations of the study should be mentioned, such as the high cost of follow-up studies, problems involved in taking blood samples at home and the adequacy of vital statistics in some countries of Latin America. These points are addressed below.

Since funding was obtained gradually, each of the three centers had different follow-up periods. Thus, all future analyses must take these intervals into account and censored data will be used for the analysis of mortality.28 Such strategies should minimize the potential bias caused by follow-up visits with different intervals.

Blood samples collected at home require a large amount of work and high cost. For safety reasons, the decision was made to work with technicians from well-known clinical laboratories at each center. Due to the need for fasting and delays among the laboratory staff in reaching the homes, subjects who had agreed to give blood samples sometimes became refusals. Although detailed information on the address of the households was provided, it was not always easy to locate the correct home. Moreover, due to time restraints between blood collection and processing at the laboratory (a mean of 3h at all centers), few subjects were scheduled for each day. The main complaint of the subjects (especially elderly individuals) regarded the fasting period. Despite these difficulties, the overall blood sample rate was reasonably high (around 70%). The same protocol for collecting and processing the samples was followed rigorously at all centers.29,30

Some limitations in this study should be pointed out. One of them is the evaluation of trends or the decline of lung function. We have two cross sectional studies over time carried out with the same population; therefore, the design of the study is a cohort. However, for some outcomes such as natural history of a disease or reduction of lung function over time, two points are not enough for the evaluation of temporal trends. On the other hand, they can establish a pattern of the outcomes under study. It is our intention to have a third follow-up of the PLATINO study in the future.

Another point that can be considered a limitation of the study is the dose of bronchodilator (200mcg) administered to the subjects before the second spirometry; although the new GOLD guidelines propose a dose of 400mcg we thought it would be better for comparison with the PLATINO baseline study to have the same bronchodilator doses; otherwise, any future result could be attributable to the increased doses of bronchodilator.

Mortality data were the most important limitation of the present study. This information was extracted from official death certificates, which posed some problems. It was not possible to find certificates for some of the subjects who, according to the families, had died (23.9% in Uruguay, 11.7% in Chile and 8.2% in Brazil). Moreover, a considerable number of the death certificates were not properly filled out by the physicians, mainly in Uruguay. To improve the quality of the mortality data, the decision was made to have an expert on mortality records to recode the death certificates and certify the basic cause of death.31,32 This expert was blind to whether the subjects having died were diagnosed with COPD or not. The quality of the information on the death certificates from Santiago and Sao Paulo was considered as very adequate by the expert.

Despite these limitations, we encourage researchers in Latin America to conduct similar studies. It is our duty to show health authorities that vital statistics must be improved in Latin America to allow comparisons with data from other regions. Moreover, the creation of central laboratories, such as biobanks, would optimize time, costs and safety in the storage of biological material. Surveys of this type strengthen capacity building in Latin America and joint efforts allow higher standards of scientific knowledge to be achieved.

A.M.B. Menezes was the general coordinator of the PLATINO study, wrote the draft and reviewed the final version of this paper. M.M. de Oca also reviewed the manuscript. A. Muiño, M.V. Lopez-Varela, G. Valdivia, C. Lisboa and J.R. Jardim were the local coordinators of the PLATINO study. R. Perez-Padilla was responsible for spirometry quality control. F.C. Wehrmeister performed the statistical analysis. All authors read and approved the final version of manuscript.

The study was funded by the Associación Latinoamericana de Tórax (ALAT), Boehringer Ingelheim Pharma GmbH & Co. KG, GlaxoSmithKline and Novartis.

The authors declare no conflict of interests regarding this paper.