Article Text

Download PDFPDF
P041 Cohorts without cohorts. mapping occupational cancer in milan great area
  1. Enrico Oddone1,
  2. Edoardo Bai2,
  3. Stefania Massari3,
  4. Alessandro Marinaccio3,
  5. Marcello Imbriani1,
  6. Paolo Crosignani1
  1. 1Department of Public Health, Experimental and Forensic Medicine, University of Pavia, Pavia, Italy
  2. 2International Society of the Doctors for the Environment, Milan, Italy
  3. 3INAIL, Italian Authority for Workers Compensation, Occupational and Environmental Medicine, Epidemiology and Hygiene Department, Rome, Italy


Introduction In Italy, we developed a monitoring system on occupational cancer, named OCCAM (Occupational Cancer Monitoring). This is a population-based case-control study that uses information already available in electronic archives, as previously described(Crosignani et al, 2006).

Materials and Methods This design can be viewed as a cohort study(Pierce, 2012) when a single or a set of firms is considered. In this case we consider among cases and controls only cases and controls who worked in that/those factory/ies, as “exposed” (to that firm/s). The reference “unexposed” category is the same of the case-control design, i.e. the workers of the service and retail sector. Cases are the same of a cohort study within the time slice provided by the source(Cancer Registry or Hospital Discharge Records). Controls, “exposed” and “unexposed” carry information on the population-time that produced cases among the employees of the studied factory(ies) and among the reference cohort. This design was used to estimate cancer risks among workers. An important feature of this approach is that the enumeration of the whole cohort is not required. When employees rolls are missed or incomplete, this case-control approach is the only feasible tool for estimating cancer(and possibly other diseases) risk among the employees of a(set of) factory(ies).

Using this approach we studied cancer risk among the employees of a large automotive factory in Milan great area. Cases were identified using 2000-2008 HDR and controls were concurrently sampled using Health Service population files.

Results Results showed significantly increased risks for lung cancer (143 cases, RR 1.258, 95% CI: 1.052–1.505, p = 0.012) and bladder cancer (120 cases, RR 1.538, 95% CI: 1.266–1.868, p < 0.001)

Conclusions Even though the job titles are unknown, these results show an increased cancer risk; in depth inquiries were carried out showing exposure to lung(asbestos, PAH, Ni, Cr) and bladder(PAH, organic dyes) carcinogens.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.