Related Experiment Video
Updated: Jun 19, 2026

Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
Published on: May 15, 2020
[Multicenter experience of the Lombardy Health Protection Agencies (ATS) in the application of an algorithm to
Maria Teresa Greco1, Eliana Gabellini2, Luca Cavalieri d'Oro3
1SC Unità di Epidemiologia, Agenzia di Tutela della Salute Città Metropolitana di Milano, Milano.
Objectives:
to apply, at the regional level, the algorithm developed by the Health Protection Agency (ATS) of the Metropolitan City of Milan to identify people with disabilities through the integration of routinely collected health and social care databases; to produce annual prevalence estimates and describe their territorial distribution (across ATS jurisdictions and districts), as well as the main classification characteristics according to the International Classification of Functioning, Disability and Health (ICF-2018), a framework that characterizes an individual's health status through the coding of body functions and structures, activities, and participation.
Design:
retrospective observational study based on administrative data.
Setting And Participants:
residents in Lombardy Region (Northern Italy), which is divided into 8 ATSs and 86 districts, during the period 2018-2024.
Main Outcome Measures:
annual period prevalence of disability; sex differences; contribution of data flows; ICF-2018 classification combinations.
Results:
in 2024, 1,083,538 people with disability were identified (10.8% of the regional population). Prevalence ranged across ATSs from 9.6% to 16.9%, and across districts from 6.3% to 19.8%. The largest overall contributions came from outpatient claims (28-SAN), hospital discharge records (SDO), exemption registries, and major prosthetics, with heterogeneity across ATSs. The most frequent ICF-2018 combination was "Structures of the nervous system/Mental functions" identified exclusively in healthcare flows.
Conclusions:
integrating healthcare and sociohealthcare flows within a shared ICF framework enables comparable small-area disability estimates to support integrated planning. Clinical validation and systematic monitoring of data quality are needed to strengthen the algorithm use.
Related Concept Videos
Statistical Methods for Analyzing Epidemiological Data
Bias in Epidemiological Studies
Analysis of Population Pharmacokinetic Data
