CONCEPT- STROKE- Common Data Model to analyse healthcare pathways of Acute Ischaemic Stroke

Technical notes and documentation on the common data model of the project CONCEPT-STROKE

This publication corresponds to the Common Data Model (CDM) specification of the CONCEPT-STROKE project for implementing a federated network analysis of the healthcare pathway of acute ischaemic stroke.

Aims of CONCEPT-STROKE:

General aim: To estimate the effectiveness and efficiency of the care pathway followed by acute ischaemic stroke patients, using Real World Data (RWD) routinely collected by five Spanish Regional Health Systems.

Main specific aims:

• To discover the actual care pathway of Acute Ischemic Stroke, from the admission to an emergency room in a public hospital to the appearance of an event of interest after discharge.
• To compare that care pathway discovered with the theoretical pathway that a patient should follow according to the 
  regional Stroke Plans.
• To analyse the impact of specific care interventions within and across care paths, in terms of patients survival
• To compare the traditional analytical methods with process mining methods in terms of modeling quality, prediction performance and information provided.

Study Design: It is a population-based retrospective observational study centered on all patients admitted to hospital due to acute ischemic stroke in five Regional Health Services within the Spanish National Health Service. We will include all individual patients that endured an acute ischaemic stroke and were admitted to an emergency room (or with an unplanned admission to hospital) in any acute care public hospital serving in Aragon, Basque Country, Catalonia, Navarre and Valencia, since 2010 until 2022.

Cohort definition: All patients admitted to hospital due to acute ischemic stroke (from 18 to 115 years old, included).

• Inclusion criteria: All patients 18 years or older admitted to emergency care services (or with an unplanned admission to hospital) with a principal diagnosis of Acute Ischemic Stroke (AIS) during the study period.
• Exclusion criteria: Patients 17 years or younger or patients with a diagnosis of acute hemorragic stroke or with other non-specific stroke diagnoses (i.e. Transient Ischemic Attack - TIA)
• Study period: From 01-01-2010 until 31-12-2022.

Files included in this publication:

This project follow the structure build using the Common Data Model Builder (https://github.com/cienciadedatosysalud/cdmb), a tool that allows you to create common data models to facilitate interoperability and reproducibility of the analyses.

############################################################################################

* If you want to create a data extraction according to the common data model, you need:

• Diagram data model (CONCEPT_STROKE_data_query_entity_diagram_cdm.png)
• Common_datamodel_CONCEPT-STROKE_v1.2.7.xlsx
• Inside 20231205_cdmb_v1.2.7.zip, you can find:
  • docs/ (this folder contains everything required according to the specifications of the common data model 20231205_cdmb_v1.2.7.zip)
    • CDM/
      • entities/ (Folder structure where, for each defined entity, the catalogs and the established validation rules are stored)
        • patient/
          • catalogs/ (catalogs required to run the synthetic data python script)
            • cat_antiaggregants_prescription_bl.csv
            • cat_antiarrhythmics_prescription_bl.csv
            • cat_anticoagulants_prescriptions_bl.csv
            • cat_antihypertensive_prescription_bl.csv
            • cat_atrial_fibrillation_bl.csv
            • cat_cnh.csv
            • cat_diabetes_bl.csv
            • cat_fibrinolitics_prescriptions_bl.csv
            • cat_heart_failure_bl.csv
            • cat_hypertension_bl.csv
            • cat_inhospital_fibrinolysis_bl.csv
            • cat_inhospital_thrombectomy_bl.csv
            • cat_municipality.csv
            • cat_nuts2_spain.csv
            • cat_type_ischemic_stroke_bl.csv
            • cat_valvular_disease_bl.csv
            • cat_zipcode.csv
          • validation-rules/ (validation rules to be followed)
            • rules_set_patient.json
      • synthetic-data/ (Folder structure contaning an automatically generated set of 1000 synthetic records per entity included en the CDM)
        • synthetic_patient.csv
    • cdmb_config.json (Configuration file)
    • cohort_definition_inclusion.csv  (csv file that defines the criteria (i.e., codes) for inclusion in a cohort)
    • common_datamodel.xlsx (The definition of the common data model in Excel format.- Common_datamodel_CONCEPT-STROKE_v1.2.7)
    • ER.gv, ER.gv.png (an Entity-Relationship Diagram of the entities included in the CDM)
    • hashed_files_list.json: List of the files generated or used after generating the project with their md5 hash. This file must be kept hidden and should be used to cross-check with the results obtained from the analysis from the original input files.

############################################################################################

* If you want to reproduce the common data model in ASPIRE  (Analytic Software Pipeline Interface for Reproducible Execution) https://github.com/cienciadedatosysalud/aspire) with the synthetic data, you need:

• Diagram data model (CONCEPT_STROKE_data_query_entity_diagram_cdm.png)
• Common_datamodel_CONCEPT-STROKE_v1.2.7.xlsx
• Inside 20231205_cdmb_v1.2.7.zip (Outputs structure and content including the files and folders that are generated when creating a research project with the cdmb Python library). It contains everything you need to enable reproducibility with the synthesized data and ASPIRE

############################################################################################

#### CONCEPT-STROKE app v.1.2.3 changelog #### 

• Update: The order of several variables in the synthetic data file has been changed.

#### CONCEPT-STROKE app v.1.2.7 changelog #### 

• Update:  Added extra information in the observation_comments column in some variables.
•  Added extra information in the encoding column in some variables. In 'municipality_code_cd' and 'zip_code_cd' variables have been changed to string/character in variable format column, and to optional in variable requeriment column. Update the 'hospital_cd' catalogues to include leading zero as in the national hospital catalogue; update the 'municipality_code_cd' catalogues to include as a string the 5-digit INE code with leading zeros; update the 'zip_code_cd' catalogues to include as a string the 5-digit postal code with leading zeros. Update the description of the variable 'municipality_code_cd' to municipality (INE code - 5 digits - without end control digit) of patient's residence. Update end date of study period.  Added extra information in the observation_comments column in 'exitus_bl' variable. Added national hospital catalogue for 'hospital_st' variable.
•  'type_discharge_emergency_bl'  variable has been changed to 'type_discharge_emergency_cd' (having changed the format to integer). Added new comments in some variables in column 'transformations_from_origin ' to clarify.  
• 'type_discharge_emergency_cd'  variable has been changed to 'discharge_emergency_bl' (having changed the format to boolean); 'n_readmissions_30days_all_cause_bl'  variable has been changed to 'readmissions_30days_bl'. Added new comments in some variables in observation_comments column to clarify. Added crosswalk in cohort definition inclusion and a new tab named 'cohort_definition_inclusion' has been created to clarify.