In Brazil, approximately 60% of cancer cases are diagnosed at an advanced stage. Early identification is one of the most effective strategies to reverse this trend. Detecting signs and symptoms before formal diagnosis allows treatment to begin in a timely manner, enabling less invasive interventions and higher therapeutic success rates. Primary Health Care (PHC) medical records within the Unified Health System (SUS) contain rich clinical information about these signs, recorded at each consultation over the years. Despite this, this source remains largely underused for that purpose.
This exploratory analysis, conducted in Recife with data from 2016 to 2024, investigated four priority cancer types: breast, cervical, prostate, and colorectal. The results indicate that signs related to these cancers are already recorded by health professionals in PHC records months or years before formal diagnosis. These signs represent a relevant source of clinical information that conventional ICD-10 coding cannot fully capture. One of the central challenges is making them visible and transforming them into qualified, reliable, and actionable information. To this end, Vital Strategies Brasil developed a semantic analysis methodology that processes free-text fields in medical records and systematically identifies these signs.
Recent Abstracts
More Data Better Health – Brazilian People’s Perceptions of Risk Factors for Cancer
Relatório de Atividades 2025 – Brasil
Mais Dados Mais Saúde – Percepções da população brasileira sobre fatores de risco…
O que os prontuários da atenção primária revelam sobre sinais e sintomas de…
CRVS Performance Measurement and Monitoring Resource Kit
Model Kentucky Ordinance Establishing a County Opioid Abatement Advisory Council
The invisible pain: Gaps in Brazil’s public health system data on Menstrual and…
Exploring the Interface Between Birth Registration and Citizenship Determination: A Scoping Study in…
A dor invisível: lacunas nos dados do SUS sobre dor menstrual e pélvica
Catalyzing Support for CRVS Improvement – Examples from the Data for Health Initiative