BACKGROUND: There are marked regional differences in breast cancer mortality rates in Italy, probably linked to factors such as diagnostic delay, therapeutic strategies, and biologic and sociodemographic differences. To investigate a possible link between sociodemographic factors (e.g. age, education, and residence) and delay in the diagnosis of breast cancer, data were evaluated from all such patients from our Institute living in the Campania Region of Southern Italy for 1991-1993. METHODS: Patients were grouped into Tis-T1/N0-N+ versus T2-4/N0-N+ and the variables examined were age (< 40, 41-50, 51-60, > 60 years), education (< or = 5 vs. > 5 school years) and residence (urban vs. rural). An analysis was made using the Pearson's Chi-square test and the multiple logistic regression. RESULTS: Statistically significant differences were found for both residence (P = 0.04) and education level (P = 0.03) in the older than 60 years age group, but only for residence (P = 0.03) in the 51-60 years age group. The risks according to Mantel-Haenszel were 1.28 for education (P = 0.08) and 1.32 for residence in rural municipalities (P = 0.05). The odds ratio for residence in rural municipalities, adjusted by education and by the education-residence interaction, was 2.26 (95% confidence interval [CI], 1.12-4.54) in the 51-60 years age group and 1.74 (95% CI, 1.01-3.00) in the older than 60 years age group. CONCLUSIONS: These data clearly indicate that residents of rural municipalities, as well as poorly educated subjects, are more likely than their respective counter-parts to have a delayed diagnosis of breast cancer.
BACKGROUND: There are marked regional differences in breast cancer mortality rates in Italy, probably linked to factors such as diagnostic delay, therapeutic strategies, and biologic and sociodemographic differences. To investigate a possible link between sociodemographic factors (e.g. age, education, and residence) and delay in the diagnosis of breast cancer, data were evaluated from all such patients from our Institute living in the Campania Region of Southern Italy for 1991-1993. METHODS:Patients were grouped into Tis-T1/N0-N+ versus T2-4/N0-N+ and the variables examined were age (< 40, 41-50, 51-60, > 60 years), education (< or = 5 vs. > 5 school years) and residence (urban vs. rural). An analysis was made using the Pearson's Chi-square test and the multiple logistic regression. RESULTS: Statistically significant differences were found for both residence (P = 0.04) and education level (P = 0.03) in the older than 60 years age group, but only for residence (P = 0.03) in the 51-60 years age group. The risks according to Mantel-Haenszel were 1.28 for education (P = 0.08) and 1.32 for residence in rural municipalities (P = 0.05). The odds ratio for residence in rural municipalities, adjusted by education and by the education-residence interaction, was 2.26 (95% confidence interval [CI], 1.12-4.54) in the 51-60 years age group and 1.74 (95% CI, 1.01-3.00) in the older than 60 years age group. CONCLUSIONS: These data clearly indicate that residents of rural municipalities, as well as poorly educated subjects, are more likely than their respective counter-parts to have a delayed diagnosis of breast cancer.