The main aim of this work is to explore the relationship between survival to tumors and individual socio-economic characteristics and the Census tract where the considered persons reside. Cancer is, in fact, a chronic disease whose clinical history has a strong relationship with the socio-economic factors, in particular as regards access to medical care and the chance of a timely diagnosis, i.e. the factors that mainly influence the therapeutic results. The considered data are referred to the patients resident in the Umbria region. Women with breast cancer diagnosed in the period 1/1/2001-31/12/2010 (8,317 cases) and individuals of both genders with colorectal cancer diagnosed in the same period (12,087 cases) were considered. The follow-up of cases stopped to 31/12/2012, the most recent available data. We have chosen to study these tumors, because they are the most widespread in the Italian population and the literature has shown how the socio-economic characteristics of the patients affected by these cancers are associated with significant differences in their survival. For all patients we collected: data coming from the Umbrian Cancer Registry (CR) about variables that describe age, gender, status in life, diagnosis, tumor characteristics and treatment; individual socio-economic information about marital status, educational level and employment status, provided by every Umbrian municipality Registry; the variables that describe the socio-economic characteristics of the Census tract where every considered patient used to live (Census of Population and Housing 2001), made available by the Regional Statistics Office. The Umbria region was chosen for the high quality of the data that the regional information system can make available, in particular as regards the local CR, thanks to the strong operational integration that exists between the different data sources. The datasets for the considered diseases have been built by deterministic record linkage, which put in connection information from the CR with those from the municipal Registry Offices and the Census. It was possible to automatically link the data of 8,209 women with breast cancer (98.7% of the total) and 11,749 cancer patients of colorectal cancer (97.2% of the total). Manual procedures for the identification and correction of errors allowed to recover also the unrelated cases, reaching 100%. A multilevel mixed effects parametric survival model, which allowed to use all the individual and area variables as responsible for the fixed effects and the Census tract of residence as responsible for the random effects, was used for survival analysis. In addition to the effects generated by the medical variables, the results showed that marital status at the individual level will help to change the survival, with better chances of overcoming the disease for married than unmarried or widowed. This is obviously an indirect effect, which expresses the need for social and family support, also occurring at the area level for breast cancer, where the structural dependency index decrease the chances of survival. For colorectal cancer, however, the relevant area-wide variable is the ageing index. The results seem to confirm what was found by recent national and international studies, which attached to the social and family capital a particularly significant role in having an early diagnosis, properly following the care and thus improving the chances of surviving to cancer. It is, however, reduced the role of the socio-economic characteristics that traditionally are associated with health inequalities, i.e. the education level and the professional status.

Il ruolo delle disuguaglianze socio-economiche nella sopravvivenza al cancro per i tumori della mammella e del colon-retto

LILLINI, ROBERTO
2016

Abstract

The main aim of this work is to explore the relationship between survival to tumors and individual socio-economic characteristics and the Census tract where the considered persons reside. Cancer is, in fact, a chronic disease whose clinical history has a strong relationship with the socio-economic factors, in particular as regards access to medical care and the chance of a timely diagnosis, i.e. the factors that mainly influence the therapeutic results. The considered data are referred to the patients resident in the Umbria region. Women with breast cancer diagnosed in the period 1/1/2001-31/12/2010 (8,317 cases) and individuals of both genders with colorectal cancer diagnosed in the same period (12,087 cases) were considered. The follow-up of cases stopped to 31/12/2012, the most recent available data. We have chosen to study these tumors, because they are the most widespread in the Italian population and the literature has shown how the socio-economic characteristics of the patients affected by these cancers are associated with significant differences in their survival. For all patients we collected: data coming from the Umbrian Cancer Registry (CR) about variables that describe age, gender, status in life, diagnosis, tumor characteristics and treatment; individual socio-economic information about marital status, educational level and employment status, provided by every Umbrian municipality Registry; the variables that describe the socio-economic characteristics of the Census tract where every considered patient used to live (Census of Population and Housing 2001), made available by the Regional Statistics Office. The Umbria region was chosen for the high quality of the data that the regional information system can make available, in particular as regards the local CR, thanks to the strong operational integration that exists between the different data sources. The datasets for the considered diseases have been built by deterministic record linkage, which put in connection information from the CR with those from the municipal Registry Offices and the Census. It was possible to automatically link the data of 8,209 women with breast cancer (98.7% of the total) and 11,749 cancer patients of colorectal cancer (97.2% of the total). Manual procedures for the identification and correction of errors allowed to recover also the unrelated cases, reaching 100%. A multilevel mixed effects parametric survival model, which allowed to use all the individual and area variables as responsible for the fixed effects and the Census tract of residence as responsible for the random effects, was used for survival analysis. In addition to the effects generated by the medical variables, the results showed that marital status at the individual level will help to change the survival, with better chances of overcoming the disease for married than unmarried or widowed. This is obviously an indirect effect, which expresses the need for social and family support, also occurring at the area level for breast cancer, where the structural dependency index decrease the chances of survival. For colorectal cancer, however, the relevant area-wide variable is the ageing index. The results seem to confirm what was found by recent national and international studies, which attached to the social and family capital a particularly significant role in having an early diagnosis, properly following the care and thus improving the chances of surviving to cancer. It is, however, reduced the role of the socio-economic characteristics that traditionally are associated with health inequalities, i.e. the education level and the professional status.
10-giu-2016
Italiano
LUCCHINI, MARIO
Università degli Studi di Milano-Bicocca
File in questo prodotto:
File Dimensione Formato  
phd_unimib_773941.pdf

accesso aperto

Dimensione 2.68 MB
Formato Adobe PDF
2.68 MB Adobe PDF Visualizza/Apri

I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.14242/172191
Il codice NBN di questa tesi è URN:NBN:IT:UNIMIB-172191