RESUMO
The Cohort Study of Mobile Phone Use and Health (COSMOS) has repeatedly collected self-reported and operator-recorded data on mobile phone use. Assessing health effects using self-reported information is prone to measurement error, but operator data were available prospectively for only part of the study population and did not cover past mobile phone use. To optimize the available data and reduce bias, we evaluated different statistical approaches for constructing mobile phone exposure histories within COSMOS. We evaluated and compared the performance of four regression calibration (RC) methods (simple, direct, inverse, and generalized additive model for location, shape, and scale), complete-case (CC) analysis and multiple imputation (MI) in a simulation study with a binary health outcome. We used self-reported and operator-recorded mobile phone call data collected at baseline (2007-2012) from participants in Denmark, Finland, the Netherlands, Sweden, and the UK. Parameter estimates obtained using simple, direct, and inverse RC methods were associated with less bias and lower mean squared error than those obtained with CC analysis or MI. We showed that RC methods resulted in more accurate estimation of the relation between mobile phone use and health outcomes, by combining self-reported data with objective operator-recorded data available for a subset of participants.
RESUMO
Background: Environmental factors such as air pollution have been associated with Parkinson's disease (PD), but findings have been inconsistent. We investigated the association between exposure to several air pollutants, road traffic noise, and PD risk in two Dutch cohorts. Methods: Data from 50,087 participants from two Dutch population-based cohort studies, European Prospective Investigation into Cancer and Nutrition in the Netherlands and Arbeid, Milieu en Gezondheid Onderzoek were analyzed. In these cohorts, 235 PD cases were ascertained based on a previously validated algorithm combining self-reported information (diagnosis, medication, and symptoms) and registry data. We assigned the following traffic-related exposures to residential addresses at baseline: NO2, NOx, particulate matter (PM)2.5absorbance (as a marker for black carbon exposure), PM with aerodynamic diameter ≤2.5 µm (PM2.5), ≤10 µm (PM10), PMcoarse (size fraction 2.5-10 µm), ultrafine particles <0.1 µm (UFP), and road traffic noise (Lden). Logistic regression models were applied to investigate the associations with PD, adjusted for possible confounders. Results: Both single- and two-pollutant models indicated associations between exposure to NOx, road traffic noise, and increasing odds of developing PD. Odds ratios of fully adjusted two-pollutant models in the highest compared with the lowest exposure quartile were 1.62 (95% CI = 1.02, 2.62) for NOx and 1.47 (95% CI = 0.97, 2.25) for road traffic noise, with clear trends across exposure categories. Conclusions: Our findings suggest that NOx and road traffic noise are associated with an increased risk of PD. While the association with NOx has been shown before, further investigation into the possible role of environmental noise on PD is warranted.