Risk adjusted hospital mortality prediction model: a case study in a Turkish training and research hospital

Creative Commons License

Gunturkun F., Vupa Çilengiroğlu Ö.

HACETTEPE JOURNAL OF MATHEMATICS AND STATISTICS, vol.48, no.3, pp.883-896, 2019 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 48 Issue: 3
  • Publication Date: 2019
  • Doi Number: 10.15672/hjms.2018.615
  • Journal Name: HACETTEPE JOURNAL OF MATHEMATICS AND STATISTICS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, TR DİZİN (ULAKBİM)
  • Page Numbers: pp.883-896
  • Keywords: Risk adjusted mortality, Logistic regression, Decision trees, Random forest, Artificial neural networks
  • Dokuz Eylül University Affiliated: Yes

Abstract

In today's world, health organizations give much importance to quality and patient safety. To this end, conservation of life and prevention of excessive deaths are one of the vital objectives for health services in all countries [22]. Although the main function of hospitals is to save lives, there is a little attention to hospital mortality. In this context; generating reliable mortality statistics and then monitoring them is a prerequisite for improvement in care and development in patient safety. In this study; a risk adjusted hospital mortality prediction model is developed by using some popular data mining techniques; logistic regression, decision trees, random forests and artificial neural networks. The data from 30182 inpatients of one of the Turkish training and research hospitals with 1155 beds is used. The data is collected from inpatients whose discharge period is January to November in 2014. At the end, the performance of these approaches are compared.