Lung cancer is the principle cause of morbidity and mortality from cancer in the developed countries, and several epidemiological studies show its relation to environmental exposure in urban industrial areas.1-3 Studies of this type are rare in regions of South America, where there are similar urban industrial areas to those in developed countries.

With the objective of contributing to this body of knowledge, we carried out an ecological study aimed at comparing the mortality from cancer among residents in a region of São Paulo State. Within this state, which is the most developed in Brazil, the region of Baixada Santista has the greatest general mortality from cancer, highlighting lung cancer as the leading cause of death among that population.4 Baixada Santista is a coastal region with bathing beaches, and had about 1 300 000 inhabitants in 1994 distributed among eight municipalities. It is separated from the capital of the state, which is located on a plateau, by a large mountain range. Four of its municipalities (NIP), located along the coast, are geographically and economically distant from the largest industrial port of Brazil, to which the other four municipalities (IP) are linked.

In the IP area, beginning in 1950, along a stretch of 148 km² of Cubatão municipality, were installed: an oil refinery, a petrochemical plant, a steel mill, a cement plant, six fertilizer companies, and a paper mill. The raw material and supplies, as well as the final products from this area are transported over an intense network of roads to the plateau and to the three ports, one of which is the most important in South America.

Deaths from lung cancer (industrial classification of diseases revision 9 (ICD–9) 142 and 163) occurring in 1980–93 among male residents from the area of IP (533) and residents outside this area, NIP (365), were selected through the state system of data analysis (SEADE). In each of the areas, the mortality was calculated, by standardisation to the world population of 1960, for the age groups over 10 years and 30–74, for the total period, and for the periods 1980–6 and 1987–93. The ratios of mortality (were calculated with 95% of probability for the confidence interval (95% CI) were calculated. The methods of the International Agency for Research on Cancer (IARC) were used for these analyses.5

A significant excess of mortality from lung cancer in the area of IP was found in all the age groups in all the periods studied. However, the mortality ratios based on 95% CI, were greater in the first period (1980–6), with a ratio of 2.24 (95% CI 1.80 to 2.79) for the age group over 10 years and 2.24 (95% CI 1.76 to 2.86) for the age group 30–74. This corresponded to that portion of the population which is economically active. In the second period (1987–93) the ratios were 1.72 (95% CI 1.49 to 2.00) and 1.53 (95% CI 1.27 to 1.81), respectively, for the age groups over 10 and 35–74.

It is thought that this excess is related to exposure to occupational and environmental carcinogenic agents, and to smoking. Some reports from the governmental environmental institution6 7 have shown an increased concentration of carcinogenic substances in the workplace, as well as in the general environment. Among them, we highlight the following: metals (chrome and nickel), aromatic hydrocarbons (benzene and styrene), policyclic aromatic hydrocarbons (PAHs, benzo(a)pyrene, antracene, naphthalene), halogenated derivatives of hydrocarbons (tetrachloroethylene, perchloroethylene, vinyl chloride), formaldehyde, lampblack, silica, particulate material, nitrogenated composites, and derivatives of sulphur. The associations between lung cancer and exposure to such substances have been analyzed by several authors.3 8 Moreover, exposure to asbestos must be great in Baixada Santista, in view of the fact that this fibre is in widespread use as thermal insulation, not only in industries, but also in cargo ships. Steeland et al 3 found that the risk of lung cancer was five times greater in people exposed to asbestos. Also in the region, the exposure to PAHs seems to be intense and extensive, originating from the activities of industries and ports, and also as a result of the combustion of diesel oil by the lorries that pass along the network of roads.

The complexity of the exposure to these multiple carcinogenic agents makes imperative the need to take into account three factors that can be contributing greatly to the increase in concentrations in workplaces, air, soil, and water of this area of Baixada Santista: (a) the transfer, to some industries, of obsolete technology from other countries; (b) the barrier formed by the mountain range (Serra do Mar), making the dispersal of industrial pollutants difficult; (c) the lack of effective measures of industrial hygiene up to 1983. Only after this date were governmental programmes of control of local pollution implemented. It is assumed that the lower ratios of the significant excess in mortality in the area of IP could have been due in the second part of the study (1987–93) to the implementation of this programme of control.

These results reinforce the need for epidemiological case-control studies that could better characterise the relation between lung cancer and several occupational and environmental carcinogenic agents which are present in the region, possible synergism among them, and other risk factors—in particular, tobacco smoke.

Furthermore, such analyses could put the interactions of these factors into perspective, alongside genetic susceptibility—another important factor in the induction of cancer. Knowledge of this type is fundamental so that preventive measures can be taken.