ObjectivesIn case-control studies exposure is generally inferred from self-reported job histories coded to occupation and industry classifi cations. In the INTEROCC study, a hygienist from each of seven countries coded local jobs reported (39 613 total jobs) to assign chemical and extremely low frequency (ELF) exposures through linkage with two jobexposure matrices. To maximise the quality and comparability of coding between countries, two comparison trials were conducted. Methods After establishing guidelines for coding, and a group training exercise, a fi rst trial assessed coding on 50 randomly selected subjects (241 jobs) and differences were evaluated and discussed. A second trial of 50 jobs was completed after coding fi nished. Pair-wise agreement between each coder was assessed, and an analysis of variance (ANOVA) was used to evaluate comparability of exposures assigned to selected chemical agents and ELF. Results Pair wise agreements between coders for ISCO 68 and ISCO 88 showed improvement of at least 10% between trials. Although differences were observed in the numbers of jobs assigned with exposure to the chemical agents and ELF, and in the mean cumulative exposure estimated for given agents, none were signifi cant in either trial. Conclusions Exposure misclassifi cation in occupational epidemiology has the potential to bias results and complicate interpretations. This study showed that the reliability of occupational coding could be improved by providing clear guidelines and using a web forum to discuss diffi cult cases. The remaining variability may refl ect true inter-country differences; however, the difference in coding did not result in signifi cant differences in exposure assignment. on 2 May 2019 by guest. Protected by copyright.

ResumenEn este artículo proponemos un nuevo modelo de regresión con efectos mixtos para variables acotadas fraccionarias. Este modelo nos permite incorporar covariables directamente al valor esperado, de manera que podemos cuantficar exactamente la influencia de estas covariables en la media de la variable de interés en vez de en la media condicional. La estimación se llevó a cabo desde una perspectiva bayesiana y debido a la complejidad de la distribución aumentada a posteriori usamos un algoritmo de Monte Carlo Hamiltoniano, el muestreador No-U-Turn, que se encuentra implementado en el software Stan. Se realizó un estudio de simulación que compara, en términos de sesgo y RMSE, el modelo propuesto con otros modelos tradicionales longitudinales para variables acotadas, resultando que el primero tiene un mejor desempeño. Finalmente, aplicamos nuestro modelo de regresión Beta Inflacionada con efectos mixtos a datos reales los cuales consistían en información de la utilización de las líneas de crédito en el sistema financiero peruano.Palabras-clave: proporciones, variables fraccionarias, distribución Beta Inflacionada, inferencia bayesiana, métodos MCMC, Monte Carlo Hamiltoniano, modelos mixtos, RStan.iv AbstractIn this article we propose a new mixed effects regression model for fractional bounded response variables. Our model allows us to incorporate covariates directly to the expected value, so we can quantify exactly the influence of these covariates in the mean of the variable of interest rather than to the conditional mean. Estimation is carried out from a bayesian perspective and due to the complexity of the augmented posterior distribution we use a Hamiltonian Monte Carlo algorithm, the No-U-Turn sampler, implemented using Stan software. A simulation study for comparison, in terms of bias and RMSE, was performed showing that our model has a better performance than other traditional longitudinal models for bounded variables. Finally, we applied our Beta Inflated mixed-effects regression model to real data which consists of utilization of credit lines in the peruvian financial system.