Letalidade por dengue no município de Santos/SP: fatores associados e distribuição espacial

Abstract

INTRODUCTION: Deaths from dengue are considered preventable by the Ministry of Health with the adoption of low technological density measures, and its occurrence is considered an indicator of fragility of the health care network. However, the dengue epidemic takes place with bigger proportions and with a higher death toll every year. Since the actions proved ineffective in controlling the incidence, learning the factors that increase mortality allows the implementation of effective measures to prevent this outcome. OBJECTIVE: To analyze dengue deaths in Santos in order to identify factors associated with lethality, regarding individual, social, programmatic and prognostic factors and their correlation with incidence and vulnerability. CASE AND METHOD: Descriptive, analytical and ecological study, using secondary data from the Epidemiological Surveillance Section of the municipality of Santos/SP, with a study population of all subjects notified and confirmed as dengue, living in the municipality of Santos, from 1997 (year of the report of the first autochthonous case of the municipality) to 2018. The dependent variable was cure and death from dengue. The theoretical framework used was vulnerability and the variables were divided into categories: individual, social, programmatic and prognostic factors. Multivariate logistic regression model was performed to analyze the factors associated with lethality divided into these categories. And geoprocessing was performed using QGIS software using Moran and Lisa indexes to identify the influence between neighborhoods and clusters associated with lethality. RESULTS: Of the 44 deaths that occurred between 1997 and 2018, after comparing them with the years of higher incidence, there was no correlation between the two rates (incidence and lethality). Of these 44, 56.8% were female, 59.1% were white, 18.2% from 50 to 59 years of age, 2.3% in the third trimester of pregnancy, 65.9% classified as severe dengue, 59.1% had care in a private hospital, 100% did not receive hydration or received less than recommended, 22% received blood transfusion, 31% were 1 to 5 times in the emergency room before being hospitalized, 23% did not diagnose dengue at the time of hospitalization, 59% had comorbidities. Only one death resulted from serotype and none had a description of the initial symptoms. Among the individual factors, the following presented statistical significance with death: white skin color, being older than 80 years and pregnancy in the third trimester. Among the social ones, the region of residence presented statistical significance. Regarding the programmatic factors, having been hospitalized and attended in a private institution was noteworthy. Among the prognostic factors, the final classification as severe dengue showed significance. In the municipality of Santos, the neighborhoods Centro, Vila Mathias, Vila Nova, Marapé and José Menino should be the focus of specific actions and health teams must be strengthened to reduce current rates. And the neighborhoods Porto Alemoa, Ilhéu Alto, Alemoa, Chico de Paula, Vila Hadad, Porto Saboó and Morro Saboó need preventive interventions due to the influence of their neighbors. This way, they shall not start presenting a high incidence and lethality rate. Furthermore, these neighborhoods, when compared with the Vulnerability index, demonstrate that The Northwest Zone and part of the Hills region have a high incidence and lethality rate in a region of high vulnerability. The Centro neighborhood¿s both rates are high, and presents low vulnerability and Orla has high lethality and presents very low vulnerability. CONCLUSIONS: Incidence seems to have no correlation with lethality. White skin color, being over 80 years of age, being pregnant in the third trimester of pregnancy, progression to severe dengue, hospitalization and care in a private hospital and residence in the neighborhoods Centro, Vila Mathias, Vila Nova, Marapé and José Menino presents a higher risk for death. The data suggest that the occurrence of cases may even be correlated with vulnerability, but deaths do not seem to be related to socioeconomic indexes of vulnerability. The presence of Primary Care needs to be further studied to verify its real protective effect regarding the mortality rate.