Penerapan Data Mining Untuk Analisis Pola Penyakit Pasien Rawat Inap Menggunakan Metode K-Means

Ella Kristiantini; Arief Hermawan; Donny Avianto

doi:10.70340/jirsi.v5i3.527

Authors

Ella Kristiantini Universitas Teknologi Yogyakarta
Arief Hermawan Universitas Teknologi Yogyakarta
Donny Avianto Universitas Teknologi Yogyakarta

DOI:

https://doi.org/10.70340/jirsi.v5i3.527

Keywords:

Data Mining, K-Means Clustering, Medical Records, Inpatient Care, BPJS Health Insurance

Abstract

The management of inpatient medical records generates large volumes of data that are often underutilized for managerial decision-making and data-driven healthcare planning. Hospitals frequently encounter difficulties in identifying patient characteristic patterns associated with healthcare costs, length of stay, intensive care unit (ICU) utilization, and discrepancies between Indonesia’s National Health Insurance (BPJS) reimbursement rates and hospital service charges. This study applied a data mining approach using the K-Means Clustering algorithm to analyze 9,889 inpatient records. The research process involved data cleaning, Z-score normalization, determination of the optimal number of clusters using the Elbow Method and Silhouette Score, and cluster characterization based on BPJS reimbursement rates, hospital charges, length of hospitalization, ICU length of stay, and patient age. The results identified three optimal clusters (k = 3) with a Silhouette Score of 0.3441. Cluster 0 (Critical Patients) consisted of 376 patients (3.8%) and was characterized by the highest healthcare costs and ICU utilization. Cluster 1 (Intermediate Patients) included 6,625 patients (67.0%), predominantly elderly individuals requiring moderate levels of care. Cluster 2 (Low-Acuity Patients) comprised 2,888 patients (29.2%), mainly younger patients with the lowest costs and shortest hospital stays. The analysis also revealed an average discrepancy of IDR 5,602,334 per patient between hospital charges and BPJS reimbursement rates. These findings demonstrate that clustering can effectively identify patient groups with distinct resource-use and financial characteristics, providing valuable insights for resource allocation, ICU capacity planning, geriatric service development, and evaluation of healthcare financing and BPJS reimbursement policies.

Downloads

Download data is not yet available.

References

D. Novaliendry, tegar wibowo, N. Ardi, T. Evi, and D. Admojo, “Optimizing Patient Medical Records Grouping through Data Mining and K-Means Clustering Algorithm: A Case Study at RSUD Mohammad Natsir Solok,” Int. J. Online Biomed. Eng., vol. 19, pp. 144–155, Aug. 2023, doi: 10.3991/ijoe.v19i12.42147.

W. Hardiyanti and A. Bachtiar, “Analisa Klaim JKN Rawat Inap di RS Universitas Indonesia Tahun 2023,” Jurnal Ilmiah Kesehatan, vol. 16, no. 1 SE-Articles. pp. 215–231, Mar. 30, 2024, doi: 10.37012/jik.v16i1.2181.

J. Han, M. Kamber, and J. Pei, “Data Mining: Concepts and Techniques (Third Edition),” in Data Mining: Concepts and Techniques (Third Edition), Third Edit., Boston: Morgan Kaufmann, 2012, pp. 1–38.

S. Juanita and R. Cahyono, “K-MEANS CLUSTERING WITH COMPARISON OF ELBOW AND SILHOUETTE METHODS FOR MEDICINES CLUSTERING BASED ON USER REVIEWS,” J. Tek. Inform., vol. 5, pp. 283–289, Feb. 2024, doi: 10.52436/1.jutif.2024.5.1.1349.

P. Fränti, S. Sieranoja, K. Wikström, and T. Laatikainen, “Clustering Diagnoses From 58 Million Patient Visits in Finland Between 2015 and 2018.,” JMIR Med. informatics, vol. 10, no. 5, p. e35422, May 2022, doi: 10.2196/35422.

C. Wongoutong, “The impact of neglecting feature scaling in k-means clustering,” PLoS One, vol. 19, no. 12, p. e0310839, Dec. 2024, doi: 10.1371/journal.pone.0310839.

A. Aljohani, “Optimizing Patient Stratification in Healthcare: A Comparative Analysis of Clustering Algorithms for EHR Data,” Int. J. Comput. Intell. Syst., vol. 17, no. 173, 2024, doi: 10.1007/s44196-024-00568-8.

Y. Sheng, R. Bond, R. Jaiswal, J. Dinsmore, and J. Doyle, “Augmenting K-Means Clustering With Qualitative Data to Discover the Engagement Patterns of Older Adults With Multimorbidity When Using Digital Health Technologies: Proof-of-Concept Trial,” J. Med. Internet Res., vol. 26, 2024, doi: https://doi.org/10.2196/46287.

V. Yfantis, A. Wagner, and M. Ruskowski, “Federated K-means clustering via dual decomposition-based distributed optimization,” Franklin Open, vol. 10, p. 100204, 2025, doi:https://doi.org/10.1016/j.fraope.2024.100204.

B. C. Laksono, Y. Syahidin, and Y. Yunengsih, “Implementasi Data Mining Klasterisasi Data Pasien Rawat Inap dengan Algoritma K-Means Clustering,” J. Teknol. Sist. Inf. dan Apl., vol. 7, no. 2, pp. 621–627, 2024, doi: 10.52436/1.jtsi.2024.7.2.

P. J. Rousseeuw, “Silhouettes: A graphical aid to the interpretation and validation of cluster analysis,” J. Comput. Appl. Math., vol. 20, pp. 53–65, 1987, doi: https://doi.org/10.1016/0377-0427(87)90125-7.

A. Nabyla; A. Rizal; and A. Setiawan, “Data mining pengelompokan biaya pengeluaran rumah sakit oleh BPJS Kesehatan menggunakan algoritma K-Means,” Smart Comp Jurnalnya Orang Pint. Komput., vol. 10, no. 2, 2021, doi: 10.30591/smartcomp.v10i2.2128.

B. Rawat, H. Pant, and A. Bist, “Clustering Medical Conditions in Patient Records Using Unsupervised Learning Techniques: A Comparative Study,” Biomed. Pharmacol. J., vol. 18, no. 3, pp. 2050–2060, 2025, doi: 10.13005/bpj/3236.