Classification tree for the inference of the nursing diagnosis Fluid Volume Excess (00026)*

Authors

  • Micnéias Lacerda Botelho Universidade Federal de Mato Grosso, Instituto de Ciências da Saúde, Campus Sinop, Sinop, MT, Brasil. Universidade Estadual de Campinas, Faculdade de Enfermagem, Programa de Pós-Graduação em Enfermagem, Campinas, SP, Brasil. http://orcid.org/0000-0002-2806-9603
  • Marisa Dibbern Lopes Correia Universidade Estadual de Campinas, Faculdade de Enfermagem, Programa de Pós-Graduação em Enfermagem, Campinas, SP, Brasil. Universidade Federal de Viçosa, Viçosa, MG, Brasil. http://orcid.org/0000-0001-6254-233X
  • Juliana Prado Biani Manzoli Universidade Estadual de Campinas, Faculdade de Enfermagem, Programa de Pós-Graduação em Enfermagem, Campinas, SP, Brasil. http://orcid.org/0000-0002-5216-378X
  • Fábio Luis Montanari Universidade Estadual de Campinas, Faculdade de Enfermagem, Programa de Pós-Graduação em Enfermagem, Campinas, SP, Brasil. http://orcid.org/0000-0001-7155-0016
  • Luciana Aparecida Costa Carvalho Universidade Estadual de Campinas, Faculdade de Enfermagem, Programa de Pós-Graduação em Enfermagem, Campinas, SP, Brasil. http://orcid.org/0000-0001-5890-991X
  • Erika Christiane Marocco Duran Universidade Estadual de Campinas, Faculdade de Enfermagem, Campinas, SP, Brasil. http://orcid.org/0000-0002-9112-752X

DOI:

https://doi.org/10.1590/s1980-220x20190246-03682

Keywords:

Decision Trees, Decision Making, Nursing Diagnosis, Renal Insufficiency, Chronic, Classification, Validation Study

Abstract

Objective: To generate a Classification Tree for the correct inference of the Nursing Diagnosis Fluid Volume Excess (00026) in chronic renal patients on hemodialysis. Method: Methodological, cross-sectional study with patients undergoing renal treatment. The data were collected through interviews and physical evaluation, using an instrument with socio-demographic variables, related factors, associated conditions and defining characteristics of the studied diagnosis. The classification trees were generated by the Chi-Square Automation Interaction Detection method, which was based on the Chi-square test. Results: A total of 127 patients participated, of which 79.5% (101) presented the diagnosis studied. The trees included the elements “Excessive sodium intake” and “Input exceeds output”, which were significant for the occurrence of the event, as the probability of occurrence of the diagnosis in the presence of these was 0.87 and 0.94, respectively. The prediction accuracy of the trees was 63% and 74%, respectively. Conclusion: The construction of the trees allowed to quantify the probability of the occurrence of Fluid Volume Excess (00026) in the studied population and the elements “Excessive sodium intake” and “Input exceeds output” were considered predictors of this diagnosis in the sample.

Downloads

Download data is not yet available.

Published

2021-04-26

How to Cite

Botelho, M. L., Correia, M. D. L., Manzoli, J. P. B., Montanari, F. L., Carvalho, L. A. C., & Duran, E. C. M. (2021). Classification tree for the inference of the nursing diagnosis Fluid Volume Excess (00026)*. Revista Da Escola De Enfermagem Da USP, 55, e03682. https://doi.org/10.1590/s1980-220x20190246-03682

Issue

Section

Original Article