[BIGEPI project: environmental and health data]. / Progetto BIGEPI: dati ambientali e dati sanitari.

ABSTRACT

OBJECTIVES: the BIGEPI project, co-funded by INAIL, has used big data to identify the health risks associated with short and long-term exposure to air pollution, extreme temperatures and occupational exposures. DESIGN: the project consists of 5 specific work packages (WP) aimed at assessing 1. the acute effects of environmental exposures over the national territory; 2. the acute effects of environmental exposures in contaminated areas, such as Sites of National Interest (SIN) and industrial sites; 3. the chronic effects of environmental exposures in 6 Italian longitudinal metropolitan studies; 4. the acute and chronic effects of environmental exposures in 7 epidemiological surveys on population samples; 5. the chronic effects of occupational exposures in the longitudinal metropolitan studies of Rome and Turin. SETTING AND PARTICIPANTS: BIGEPI analyzed environmental and health data at different levels of detail the whole Italian population (WP1); populations living in areas contaminated by pollutants of industrial origin (WP2); the entire longitudinal cohorts of the metropolitan areas of Bologna, Brindisi, Rome, Syracuse, Taranto and Turin (WP3 and WP5); population samples participating in the epidemiological surveys of Ancona, Palermo, Pavia, Pisa, Sassari, Turin and Verona (WP4). MAIN OUTCOME MEASURES: environmental exposure PM10, PM2,5, NO2 and O3 concentrations and air temperature at 1 Km2 resolution at national level. Occupational exposures employment history of subjects working in at least one of 25 sectors with similar occupational exposures to chemicals/carcinogens; self-reported exposure to dust/fumes/gas in the workplace. Health data cause-specific mortality/hospitalisation; symptoms/diagnosis of respiratory/allergic diseases; respiratory function and bronchial inflammation. RESULTS: BIGEPI analyzed data at the level of the entire Italian population, data on 2.8 million adults (>=30 yrs) in longitudinal metropolitan studies and on about 14,500 individuals (>=18 yrs) in epidemiological surveys on population samples. The population investigated in the longitudinal metropolitan studies had an average age of approximately 55 years and that of the epidemiological surveys was about 48 years; in both cases, 53% of the population was female. As regards environmental exposure, in the period 2013-2015, at national level average values for PM10, PM2.5, NO2 and summer O3 were 21.1±13.6, 15.1±10.9, 14.7±9.1 and 80.3±17.3 µg/m3, for the temperature the average value was 13.9±7.2 °C. Data were analyzed for a total of 1,769,660 deaths from non-accidental causes as well as 74,392 incident cases of acute coronary event and 45,513 of stroke. Epidemiological investigations showed a high prevalence of symptoms/diagnoses of rhinitis (range 14.2-40.5%), COPD (range 4.7-19.3%) and asthma (range 3.2-13.2%). The availability of these large datasets has made it possible to implement advanced statistical models for estimating the health effects of short- and long-term exposures to pollutants. The details are reported in the BIGEPI papers already published in other international journals and in those published in this volume of E&P. CONCLUSIONS: BIGEPI has confirmed the great potential of using big data in studies of the health effects of environmental and occupational factors, stimulating new directions of scientific research and confirming the need for preventive action on air quality and climate change for the health of the general population and the workers.