SEPI partner started to exploit the new occupational dataset, which was built as a product of the project.

This longitudinal dataset comes from the linkage of socio-demographic information, employment and health data of a cohort of residents in Turin (Piedmont, Italy) from 2008 to 2020.

It is possible to select head and neck (H&N) cancer cases based on hospital discharges and outpatient visits, and to follow each person over time to see what happens in terms of employment in the following years.

With the aim of defining profiles of H&N cancer patients within the cohort, considering sociodemographic and health variables SEPI is testing a Latent Class Analysis (LCA).

LCA is a statistical procedure used to identify a set of mutually exclusive classes of objects (where both the number of classes and their properties are unknown), based on a set of observed variables. It can be considered an «advanced» clustering tool since it deals with all types of data, even dichotomous or categorical variables. Starting from a dataset of N units, K classes of units are identified with a similar pattern with respect to the observed variables introduced in the model.