RESUMEN
It is critically important for stakeholders with distinct foci of attention on healthcare to understand patient evolution in the presence of an established diagnosis or with a suspected diagnosis of various diseases, specially considering death as an outcome. To study the long-term mortality of patients at a cardiovascular referral hospital. Deterministic binding (selection of pairs of registers from the hospital electronic health records and the mortality records of São Paulo state) from 2002 to 2017 was performed. Studied variables were: age, sex, hospital treatment unit where the first visit occurred (Emergency Unit, Outpatient Unit, Hospital Admissions, Diagnostics Services), treatment type, elapsed time between the first visit and death, diagnosis at first and last visits and variables related to death. Statistical Methods: descriptive, survival (with Kaplan-Meier method), correspondence and competitive risks analyses; in case of nonoccurrence of death until the end of 2017, the patients were considered alive. Statistical significance was set at values of P < .05. Median age at the first visit to the Hospital was 51.9 years. Birth locations included 4496 cities, 17.33% in São Paulo, 0.41% in Rio de Janeiro, 0.40% in Osasco, 24.04% in other cities. Sex included females (46.7%), males (44.2%), not defined (6.3%), and other (2.8%). We observed an association between diseases in ICD-10 Chapter 16 (certain conditions originating in the perinatal period) and Chapter 17 (congenital malformations, deformations, and chromosomal abnormalities), both as diagnoses and underlying causes of death, as well as between neoplasms as diagnoses and as the underlying cause of death. In this sample, there was an association between admission diagnoses and underlying causes of death, such as neoplasms, cardiovascular diseases, and congenital heart malformations. Additionally, patients who underwent a cardiac intervention had a smaller less mortality rate than those who were not operated on. There were also differences in cardiovascular mortality between distinct treatment units of the hospital ((Emergency Unit, Outpatient Unit, Hospital Admissions, Diagnostic Services).