the SMR database is composed by three folders: * (on DRYAD) SMR data for the period 2009-2018 by cause ( see “cancer type legend" below) at three levels of aggregation: municipal , provincial and regional; * (on ZENODO) the Figure maps of average SMR for single cancer types and levels of aggregation; * (on ZENODO) the scripts in Python language to reproduce the elaboration along with the raw source data. Within the main folders “SMR” and “Figure_maps” there are three sub-folder, one for each level of aggregation (i.e., municipal, provicnal ad regional one) The sub-folders in “SMR”, in turn, contain the computation of ten-years SMR by cause as listed in the “cancer type legend” plus that for cancer type C and C+D for single administrative units in Comma Separated Values (CSV) files. The format of CSV files is always the same across the levels of aggregation: in the rows are the single administrative units (i.e. municipalities, provinces, or region); in the columns are: the code of the administrative unit (first column), the value of SMR for the years 2009-2018 (2th-11th columns), the ten-years average (12th column), and the 90% lower confidence level (13th column) estimated according to the Byars method. A "readme.txt" file is present in each SMR sub-folders to easily access and understand the data. The folder “SMR” also holds the SMR outputs saved as pickle file, namely an homologous Python module extent (“pkl”) that can be reloaded to produce figures and further analysis in the python language by end-users. The folder Figure_maps (on ZENODO) have three sub-folder, one for each level of aggregation (i.e., municipal, provicnal ad regional one) Each sub-folders contains the figure maps on ten-years average SMR for each level of aggregation and single cancer types as listed below. In the folder “script”(on ZENODO) there are the preprocessing source data used for the computation of SMR (i.e. raw data), the scripts used to produce all the elaborations, the shapefiles of administrative boundaries, and some intermediate outputs saved as pickle file for convenience. The Python script files are denoted through leading ordinal numbers that reflect the order of execution to reproduce the results. —----------------------------------------------------------------------------------------------- Cancer type(_n) legend to access CSV files —----------------------------------------------------------------------------------------------- 1 of which malignant tumors of the lips, oral cavity and pharynx 2 of which malignant tumors of the esophagus 3 of which are malignant stomach tumors 4 of which are malignant tumors of the colon, rectum and anus 5 of which malignant tumors of the liver and intrahepatic bile ducts 6 of which are malignant tumors of the pancreas 7 of which malignant tumors of the larynx 8 of which malignant tumors of the trachea, bronchi and lungs 9 of which skin malignant melanoma 10 of which are malignant breast tumors 11 of which are malignant tumors of the uterine cervix 12 of which malignant tumors of other parts of the uterus 13 of which malignant tumors of the ovary 14 of which malignant prostate tumors 15 of which malignant kidney tumors 16 of which are malignant bladder tumors 17 of which are malignant tumors of the brain and central nervous system 18 of which malignant thyroid tumors 19 of which hodgkin's disease and lymphomas 20 of which leukemia 21 of which other malignant tumors of lymphatic / haematopoietic tissue 22 of which other malignant tumors 23 non-malignant tumors (benign and of uncertain behavior) —----------------------------------------------------------------------------------------------- Detailed rows and column specification of SMR data —----------------------------------------------------------------------------------------------- --------------------------------------------- Rows and Columns for SMR at municipal level: SMR_n = 7903 rows [adminstrative units] x 13 columns [10 years, mean SMR, SMR90low] for cancer type "n" SMRc = 7903 rows [adminstrative units] x 13 columns [10 years, mean SMR, SMR90low] for cancer macro-type C SMRcd = 7903 rows [adminstrative units] x 13 columns [10 years, mean SMR, SMR90low] for cancer macro-type C+D --------------------------------------------- Rows and Columns for SMR at provincial level SMR_n = 107 rows [adminstrative units] x 13 columns [10 years, mean SMR, SMR90low] for cancer type "n" SMRc = 107 rows [adminstrative units] x 13 columns [10 years, mean SMR, SMR90low] for cancer macro-type C SMRcd = 107 rows [adminstrative units] x 13 columns [10 years, mean SMR, SMR90low] for cancer macro-type C+D --------------------------------------------- Rows and olumns for SMR at regional level SMR_n = 20 rows [adminstrative units] x 13 columns [10 years, mean SMR, SMR90low] for cancer type "n" SMRc = 20 rows [adminstrative units] x 13 columns [10 years, mean SMR, SMR90low] for cancer macro-type C SMRcd = 20 rows [adminstrative units] x 13 columns [10 years, mean SMR, SMR90low] for cancer macro-type C+D