Identification of Mortality Predictors for Elderly Patients with Airway Cancer

The article analyzes the data of elderly patients with airway cancer after surgery. To reduce cancer mortality, it is necessary to identify markers that can predict worsening of a patient's condition. The study involved 83 patients, for whom 47 different parameters were recorded. The data contain not only general laboratory parameters, but also indicators of blood gas composition, hemodynamic data and autonomic indices, indicators of metabolism and cardiorespiratory exercise testing at different load levels. Statistical data analysis is carried out using the SPSS program and is based on the following methods: one-side Kolmogorov–Smirnov test, Mann–Whitney test, Chi-square test and the use of logistic regression to build a multivariate model. The target variable is the result of the operation, which can take the following values: no complications, complications after surgery, death. The result of this analysis, namely the identification of markers of mortality, will determine which of the patients need special medical supervision in the postoperative period. Goal of the investigation is to analyze the data of elderly oncology patients who underwent surgery and to identify the parameters that most affect patient mortality. This paper presents a part of the study where predictive analysis is carried out using statistical methods. In the future, the same analysis will be carried out using neural networks to compare the results and increase the reliability.