Potential for health monitoring: Exploring SUS outpatient open data

Abstract

The Brazilian Unified Health System (SUS) has disseminated billions of administrative records cor-
responding to three decades of existence. Unlike hospital and notification data, which are fragmented as they are service-oriented rather than user-oriented, outpatient data can be linked via a pseudonymized identifier, enabling
the tracking of therapeutic pathways. This paper presents an automated microdata processing tool from the Open
Health Intelligence Platform (SABEIS) using open-source software to handle microdata provided through the
TabWin/TabNet strategy from the file transfer directory, without relying on sophisticated Big Data. It describes the
open data from the SUS Ambulatory Information System from 2008 to 2023, using modest hardware resources by
public health and health informatics specialists. A total of 8,106,361,265 records were processed, corresponding to
3,135 procedures, 16,407 diagnoses, and 51,875,308 users, according to the files of APAC-SIA High-cost ambulatory
procedures and high-cost medicines. There was a noticeable improvement in the quality of the pseudonymized
identifier, especially from 2022 onwards, with 0.8% of users recorded as having more than one sex, more than one
state of residence, over eight procedures, or five diagnoses, demonstrating the potential for public policy monitoring
and knowledge generation. This approach demonstrated the potential for monitoring public policies using outpatient
data with a pseudonymized SUS user identifier.