Analysis of length of stay for patients admitted to Korean hospitals based on the Korean National Health Insurance Service Database

Prediction of patient length of stay using random forest method based on the Indonesian national health insurance

“…By randomly selecting F attributes, it is not necessary to evaluate all available attributes; only the selected F attributes are utilized. The control over the random forest's effectiveness lies in adjusting both the F value and the number of trees in the forest [4]. When the F value is too small, the trees generated will have minimal correlation, and the opposite is true as well.…”

“…Janwanishtaporn et al [10] has done study to determine the national hospitalization rate in Thailand using public health security scheme data. While An et al [4] using health checkup cohort DB data stored in the virtual server of National Health Insurance Service (NHIS) of Korea, while Srimannarayana et al [11] used Health Insurance in India. There are no studies specifically using National Health Insurance (BPJS) data in Indonesia.…”

“…As stated by [11], there is a significant positive correlation between age and the duration of hospitalization. An et al's [4] findings indicate that key factors contributing to the LoS include demographic variables such as age, insurance deductible ratio, and information related to the primary diagnosis and sub-diagnoses. Furthermore, the researchers identified that the number of doctors and beds available in hospitals, cholesterol level, body mass index (BMI), and the month of admission were also found to be significant factors influencing the LoS.…”

“…Previous research on predicting LoS in hospital using health insurance claims data has been conducted by [4] using ensemble method. This study recommends that future research should focus on exploring significant features that influence the performance of LoS prediction models, utilizing health examination data from different countries.…”

Health Insurance Premium Pricing Using Machine Learning Methods

“…[6] found that machine learning is not only helpful on predicting healthcare outcomes, but it also guides the health service researchers on generalizing data-driven estimators. [9] used four machine learning models likes the decision tree, random forest, support vector machine, and XGBoost to predict the length of stay (LOS) of patients that are admitted into the South Korean medical institutions and discovered that XGBoost is the best model to be used in this matter.…”