Background: Per- and polyfluoroalkyl substances (PFAS) are associated with many adverse health conditions. Among the main effects is carcinogenicity in humans, which deserves to be further clarified. An evident association has been reported for kidney cancer and testicular cancer. In 2013, a large episode of surface, ground and drinking water contamination with PFAS was uncovered in three provinces of the Veneto Region (northern Italy) involving 30 municipalities and a population of about 150,000. We report on the temporal evolution of all-cause mortality and selected cause-specific mortality by calendar period and birth cohort in the local population between 1980 and 2018. Methods: The Italian National Institute of Health pre-processed and made available anonymous data from the Italian National Institute of Statistics death certificate archives for residents of the provinces of Vicenza, Padua and Verona (males, n = 29,629; females, n = 29,518) who died between 1980 and 2018. Calendar period analysis was done by calculating standardised mortality ratios using the total population of the three provinces in the same calendar period as reference. The birth cohort analysis was performed using 20–84 years cumulative standardised mortality ratios. Exposure was defined as being resident in one of the 30 municipalities of the Red area, where the aqueduct supplying drinking water was fed by the contaminated groundwater. Results: During the 34 years between 1985 (assumed as beginning date of water contamination) and 2018 (last year of availability of cause-specific mortality data), in the resident population of the Red area we observed 51,621 deaths vs. 47,731 expected (age- and sex-SMR: 108; 90% CI: 107–109). We found evidence of raised mortality from cardiovascular disease (in particular, heart diseases and ischemic heart disease) and malignant neoplastic diseases, including kidney cancer and testicular cancer. Conclusions: For the first time, an association of PFAS exposure with mortality from cardiovascular disease was formally demonstrated. The evidence regarding kidney cancer and testicular cancer is consistent with previously reported data.

Background: Arsenic in drinking water is a global public health concern. This study aims to investigate the association between chronic low-level exposure to arsenic in drinking water and health outcomes in the volcanic area of Mt. Amiata in Italy, using a residential cohort study design.Methods: Chronic exposure to arsenic in drinking water was evaluated using monitoring data collected by the water supplier. A time-weighted average arsenic exposure was estimated for the period 2005–2010. The population-based cohort included people living in five municipalities in the Mt. Amiata area between 01/01/1998 and 31/12/2019. Residence addresses were georeferenced and each subject was matched with arsenic exposure and socio-economic status. Mortality and hospital discharge data were selected from administrative health databases. Cox proportional hazard models were used to test the associations between arsenic exposure and outcomes, with age as the temporal axis and adjusting for gender, socio-economic status and calendar period.Results: The residential cohort was composed of 30,910 subjects for a total of 407,213 person-years. Analyses reported risk increases associated with exposure to arsenic concentrations in drinking water > 10 µg/l for non-accidental mortality (HR = 1.07 95%CI:1.01–1.13) and malignant neoplasms in women (HR = 1.14 95%CI:0.97–1.35). Long-term exposure to arsenic concentrations > 10 µg/l resulted positively associated with several hospitalization outcomes: non-accidental causes (HR = 1.06 95%CI:1.03–1.09), malignant neoplasms (HR = 1.10 95%CI:1.02–1.19), lung cancer (HR = 1.85 95%CI:1.14–3.02) and breast cancer (HR = 1.23 95%CI:0.99–1.51), endocrine disorders (HR = 1.13 95%CI:1.02–1.26), cardiovascular (HR = 1.12 95%CI:1.06–1.18) and respiratory diseases (HR = 1.10 95%CI:1.03–1.18). Some risk excesses were also observed for an exposure to arsenic levels below the regulatory standard, with evidence of exposure-related trends.Conclusions: Our population-based cohort study in the volcanic area of Mt. Amiata showed that chronic exposure to arsenic concentrations in drinking water above the current regulatory limit was associated with a plurality of outcomes, in terms of both mortality and hospitalization. Moreover, some signs of associations emerge even at very low levels of exposure, ​​below the current regulatory limit, highlighting the need to monitor arsenic concentrations continuously and implement policies to reduce concentrations in the environment as far as possible.

The present work aims to identify the predictors of hemodynamic failure (HF) developed during pediatric intensive care unit (PICU) stay testing a set of machine learning techniques (MLTs), comparing their ability to predict the outcome of interest. The study involved patients admitted to PICUs between 2010 and 2020. Data were extracted from the Italian Network of Pediatric Intensive Care Units (TIPNet) registry. The algorithms considered were generalized linear model (GLM), recursive partition tree (RPART), random forest (RF), neural networks models, and extreme gradient boosting (XGB). Since the outcome is rare, upsampling and downsampling algorithms have been applied for imbalance control. For each approach, the main performance measures were reported. Among an overall sample of 29,494 subjects, only 399 developed HF during the PICU stay. The median age was about two years, and the male gender was the most prevalent. The XGB algorithm outperformed other MLTs in predicting HF development, with a median ROC measure of 0.780 (IQR 0.770–0.793). PIM 3, age, and base excess were found to be the strongest predictors of outcome. The present work provides insights for the prediction of HF development during PICU stay using machine-learning algorithms.

Background: Asbestos exposure is a recognized risk factor for ovarian cancer and malignant mesothelioma. There are reports in the literature of geographical ecological associations between the occurrence of these two diseases. Our aim was to further explore this association by applying advanced Bayesian techniques to a large population (10 million people). Methods: We specified a series of Bayesian hierarchical shared models to the bivariate spatial distribution of ovarian and pleural cancer mortality by municipality in the Lombardy Region (Italy) in 2000–2018. Results: Pleural cancer showed a strongly clustered spatial distribution, while ovarian cancer showed a less structured spatial pattern. The most supported Bayesian models by predictive accuracy (widely applicable or Watanabe–Akaike information criterion, WAIC) provided evidence of a shared component between the two diseases. Among five municipalities with significant high standardized mortality ratios of ovarian cancer, three also had high pleural cancer rates. Wide uncertainty was present when addressing the risk of ovarian cancer associated with pleural cancer in areas at low background risk of ovarian cancer. Conclusions: We found evidence of a shared risk factor between ovarian and pleural cancer at the small geographical level. The impact of the shared risk factor can be relevant and can go unnoticed when the prevalence of other risk factors for ovarian cancer is low. Bayesian modelling provides useful information to tailor epidemiological surveillance.

The geothermal power plants for electricity production currently active in Italy are all located in Mt. Amiata area in the Tuscany region. A cross-sectional survey was conducted in the framework of the regional project “InVETTA—Biomonitoring Survey and Epidemiological Evaluations for the Protection of Health in the Amiata Territories”, using objective measures of lung function to investigate the role of hydrogen sulphide (H₂S) in affecting the respiratory health of the population living in this area. 2018 adults aged 18–70 were enrolled during 2017–2019. Home and workplace addresses of participants were geocoded. Dispersion modelling was used to evaluate the spatial variability of exposure to H₂S from the geothermal power plants’ emissions. We estimated average long-term historical exposure to H₂S and more recent exposure indicators. Chronic Obstructive Pulmonary Disease (COPD) was defined according to the Global Initiative for Chronic Obstructive Lung Disease (GOLD). Multivariable logistic regressions were performed to investigate associations between outcome and exposure. Our findings did not showed any evidence of an association between increasing H₂S exposure and lung function impairments. Some risk reductions were observed: a -32.8% (p = 0.003) for FEV1<80% and a -51.7% (p = 0.001) risk decrease for FVC<80% were associated with interquartile increase (13.8 μg/m³) of H₂S levels. Our study provides no evidence that chronic exposure to low levels of H₂S is associated with decrements in pulmonary function, suggesting that ambient H₂S exposure may benefit lung function.

Space‐time analysis of mortality risk is useful to evaluate the epidemiologic transitions at the subnational level. In our study, we analysed the death certificate records for lung cancer in Italy in 1995‐2016, obtained from the Italian National Institute of Statistics. Our objective was to investigate the spatial‐temporal evolution of lung cancer mortality by sex and province of residence (n = 107) using the birth cohort as relevant time axis. We built Bayesian space‐time models with space‐time interactions. Among males (n = 554 829), mortality peaked in the 1920‐1929 cohort, followed by a generalised decline. Among females (n = 158 619), we found novel original evidence for a peak in the 1955‐1964 cohort, equivalent to a 35‐year delay, with a downward trend being observed thereafter. Over time, the documented north‐south decreasing mortality gradient has been replaced by a west‐east decreasing gradient. Naples has become the province at highest risk in Italy, both among males and females. This pattern is consistent with an epidemiologic transition of risk factors for lung cancer to the south‐west of the country and raises concern, because 5‐year age‐standardised net survival from the disease in this geographic area is lower than in northern and central Italy. The variability of mortality rates among provinces has changed over time, with an increasing homogeneity for males and an opposite trend for females in the more recent birth cohorts. These unprecedented observations provide evidence for a profound spatio‐temporal transition of lung cancer mortality in Italy.