Cyndy Romarizka

Jurusan Matematika, Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Jember, Indonesia

Mohamat Fatekurohman

Jurusan Matematika, Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Jember, Indonesia

I Made Tirta

Jurusan Matematika, Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Jember, Indonesia

DOI: https://doi.org/10.19184/jid.v24i1.33074

ABSTRACT 

Globally, in 2021, there were 170,051,718 COVID-19 cases and 3,540,437 patients who died. The high mortality rate of patients infected with COVID-19 gives an idea to research the analysis of the factors that influence the death of Covid-19 patients. The data used in this study is data on Covid-19 patients obtained from the Mexican Government, with response variables namely time and status and predictor variables, namely patient laboratory results in the form of a history of illness that has been suffered by Covid-19 patients so that they adopt the extended model to evaluate the data. The data in this study are heterogeneous and large in number so that data clustering is carried out into 3 clusters, namely low emergency clusters, medium emergency clusters and high emergency clusters using K-means clustering. Because the study could not find the factors that influence the death of Covid-19 patients, two clusters were chosen, namely the medium emergency cluster and the high emergency cluster. So that the factors that influence the death of Covid-19 patients in the medium emergency cluster are sorted by the highest hazard ratio, namely pneumonia, old age, renal chronic, diabetes, Chronic Obstructive Pulmonary Disease (COPD), immune system, hypertension, cardiovascular, obesity, gender, and asthma. In the high emergency cluster, sorted by the highest hazard ratio is the immune system, renal chronic, cardiovascular, COPD, tobacco, hypertension, obesity, gender, and pneumonia.

Keywords: Covid-19, Comorbid diseases, Extended cox, K-Means Clustering.