Kamu Hastanelerinin Rol Sınıflandırmasında Veriye Dayalı Bir Yaklaşım: Türkiye’den Kanıtlar

Yıl 2025, Cilt: 36 Sayı: 139, 803 - 829, 24.12.2025

Tevfik Bulut , Aziz Küçük

https://doi.org/10.52836/sayistay.1751816

Öz

Hastane sınıflandırması, modern sağlık sistemlerinde kaynakların etkin
kullanımı, sağlık hizmetlerine erişimin artırılması ve hastaların uygun bakım düzeyine
yönlendirilmesi açısından kritik bir rol oynamaktadır. Çalışmanın amacı, Türkiye’deki
kamu hastanelerinin rol sınıflarını Rastgele Orman (Random Forest, RF) sınıflandırma
algoritması kullanarak öngörmektir. Bu çalışma, Türkiye’de kamu hastanelerinin rol
sınıflandırması üzerine doğrudan yürütülen ilk çalışmadır. Çalışmada, Türkiye genelinde
sekiz farklı rol sınıfına ayrılmış 716 kamu hastanesine ait veriler analiz edilmiştir.
İki farklı RF modeli (Model 1 ve Model 2) geliştirilmiş ve bu modeller genel doğruluk,
Cohen’in Kappa katsayısı, eğri altı alan (AUC), F1 skoru ve dengelenmiş doğruluk gibi
çeşitli performans göstergeleriyle karşılaştırmalı olarak değerlendirilmiştir. Yapılan
performans değerlendirmesinde, RF Model 2’nin Model 1’e kıyasla üstün sonuçlar
verdiği görülmüştür. Model 2, Model 1’e kıyasla daha yüksek doğruluk (%96,82’ye karşı
%95,82), daha iyi Kappa katsayısı (0,9612’ye karşı 0,9489) ve daha yüksek AUC değeri
(0,9889’a karşı 0,9863) elde etmiştir. Bu nedenle, Türkiye’deki kamu hastanelerinin rol
sınıflarının öngörülmesinde Model 2’nin kullanılması önerilmektedir. Önerilen yöntem,
farklı ülkelerdeki hastanelerin sınıflandırılmasında da uyarlanabilir ve sağlık yöneticilerine
stratejik planlama ve kaynak tahsisinde veri temelli destek sağlayabilir.

Anahtar Kelimeler

Kamu Hastanesi , Rastgele Orman , Makine Öğrenmesi , Hastane Sınıflandırması , Değişken Önem Düzeyi

A Data–Driven Approach to the Role Classification of Public Hospitals: Evidence from Türkiye

Öz

Hospital classification plays a crucial role in modern healthcare systems by
promoting efficient resource utilization, enhancing access to care, and directing
patients to the most appropriate service levels. This study aimed to predict the role
classifications of public hospitals in Türkiye using the Random Forest (RF) classification
algorithm. It represents the first study conducted specifically on the role-based
classification of public hospitals in Türkiye. The dataset included 716 public hospitals,
categorized into eight distinct role classes. Two RF models (Model 1 and Model 2)
were developed and evaluated using performance metrics, including overall accuracy,
Cohen’s Kappa coefficient, area under the curve (AUC), F1 score, and balanced accuracy.
RF Model 2 consistently outperformed Model 1, achieving higher accuracy (96.82%
vs. 95.82%), Kappa (0.9612 vs. 0.9489), and AUC (0.9889 vs. 0.9863). Thus, Model 2
is recommended for classifying the roles of public hospitals in Türkiye. The proposed
approach can be adapted to classify hospitals in different healthcare systems, offering
support for data-driven strategic planning and more equitable resource allocation.

Anahtar Kelimeler

Public Hospital , Random Forest , Machine Learning , Hospital Classification , Impotance Levels

Kaynakça

• Adjei, H.P., Weyori, B., Boateng, O.N., Adekoya, A.F., Nti, I.K. (2023). Predicting Diabetes Using Cohen’s Kappa Blending Ensemble Learning. International Journal of Electronic Healthcare, 13(1):1. https://doi.org/10.1504/ijeh.2023.10052670
• Akdağ, R. (2012). Turkey Health Transformation Program Assessment Report (2003-2011), Ministry of Health of Turkey, Ankara.
• Alam, Md. Z., Rahman, M.S., Rahman, M.S. (2019). A Random Forest based predictor for medical data classification using feature ranking. Informatics in Medicine Unlocked 15:100180. https://doi.org/10.1016/j.imu.2019.100180
• An, Q., Rahman, S., Zhou, J., & Kang, J. J. (2023). A Comprehensive Review on Machine Learning in the Healthcare Industry: Classification, Restrictions, Opportunities, and Challenges. Sensors (Basel, Switzerland), 23(9), 4178. https://doi.org/10.3390/ s23094178
• Arnold, J. (2024). ggthemes: Extra Themes, Scales and Geoms for ‘ggplot2’. R package version 5.1.0.
• Atasever, M. (2021). Sağlık Kurumları İşletmeciliği ve Hastane Yönetimi, Akademisyen Kitabevi: Ankara.
• Babacan, A., & Akca, N. (2024). Kamu Hastanelerinin Finansal Performansının Oran Analizi ile Değerlendirilmesi: Türkiye Örneği. Muhasebe ve Finansman Dergisi, (101), 21-44. https://doi.org/10.25095/mufad.1313428
• Bailly, A, Blanc, C, Francis, É., Guillotin, T, Jamal, F, Wakim, B, & Roy, P. (2022). Effects of dataset size and interactions on the prediction performance of logistic regression and deep learning models. Computer methods and programs in biomedicine, 213, 106504. https://doi.org/10.1016/j.cmpb.2021.106504
• Ben-David, A. (2007). About the relationship between ROC curves and Cohen’s kappa. Engineering Applications of Artificial Intelligence. 21(6), 874-882. https://doi. org/10.1016/j.engappai.2007.09.009
• Bhadouria, A.S., Singh, R.K. (2024). Machine learning model for healthcare investments predicting the length of stay in a hospital & mortality rate. Multimed Tools Appl 83, 27121-27191. https://doi.org/10.1007/s11042-023-16474-8
• Bradshaw, T.J., Huemann, Z., Hu, J., & Rahmim, A. (2023). A Guide to Cross-Validation for Artificial Intelligence in Medical Imaging. Radiology. Artificial intelligence, 5(4), e220232. https://doi.org/10.1148/ryai.220232
• Breiman, L. (2001) Random Forests. Machine Learning, 45, 5-32. http://dx.doi. org/10.1023/A:1010933404324
• Brodersen, K.H., Ong, C.S, Stephan, K.E and Buhmann, J.M. (2010). The Balanced Accuracy and Its Posterior Distribution. 2010 20th International Conference on Pattern Recognition, Istanbul, Türkiye, 3121-3124. https://doi.org/10.1109/ICPR.2010.764
• Butt, U.M, Letchmunan, S, Ali, M, Hassan, F.H, Baqir, A, & Sherazi, H.H.R. (2021). Machine Learning Based Diabetes Classification and Prediction for Healthcare Applications. Journal of healthcare engineering, 2021, 9930985. https://doi. org/10.1155/2021/9930985
• Choi, J.H., Choi, E.S., & Park, D. (2023). In-hospital fall prediction using machine learning algorithms and the Morse fall scale in patients with acute stroke: a nested casecontrol study. BMC medical informatics and decision making, 23(1), 246. https://doi. org/10.1186/s12911-023-02330-0
• Cohen, J. (1960). A Coefficient of Agreement for Nominal Scales. Educational and Psychological Measurement. 20(1), 37-46. https://doi.org/10.1177/001316446002000104
• Corporation, M, Weston, S. (2025). doParallel: Foreach Parallel Adaptor for the ‘parallel’ Package. R package version 1.0.17, https://github.com/revolutionanalytics/doparallel
• De Hond, A.A.H., Steyerberg, E.W., & van Calster, B. (2022). Interpreting the area under the receiver operating characteristic curve. The Lancet. Digital health, 4(12), e853– e855. https://doi.org/10.1016/S2589-7500(22)00188-1
• De Oliveira, H., Prodel, M., Augusto, V. (2018). Binary Classification on French Hospital Data: Benchmark of 7 Machine Learning Algorithms. 2018 IEEE International Conference on Systems, Man, and Cybernetics, 1743-1748. https://doi.org/10.1109/ smc.2018.00301
• Dharmapala, P. (2021). A Classification of Hospitals using Performance Features and Machine Learning Algorithms in the Event of Random Surge in Inpatients and Prediction of Live Discharges. Journal of Analytics 1(1):20-41. https://doi.org/10.29020/nybg. ja.v1i1.2
• Ekinci, G, & Bakır, İ. (2021). Sağlık Kurumlarında Finansal Performans Analizi A1 Dal Hastanesi Örneği. Uluslararası Sağlık Yönetimi ve Stratejileri Araştırma Dergisi, 7(1), 1-18.
• Garnier, S, Ross, N, Rudis R, Sciaini, M, Camargo, AP et al. (2024). viridis(Lite)-Colorblind- Friendly Color Maps for R. viridis package version 0.6.5. https://doi.org/10.5281/ zenodo.4679423
• Griffin, D.J. (Ed.) 2012, Hospitals: what they are and how they work, Fourth Edition, Jones & Bartlett Learning: Sudbury.
• Habehh, H., & Gohel, S. (2021). Machine Learning in Healthcare. Current genomics, 22(4), 291–300. https://doi.org/10.2174/1389202922666210705124359
• Han, J., Kamber, M., Pei, J. (2012). Classification. In: Data Mining, Third Edition, Elsevier: Waltham, MA. 327-391. https://doi.org/10.1016/b978-0-12-381479-1.00008-3
• Healy, J. and McKee, M. (2002). The Role and Function of Hospitals. McKee, M. and Healy, J. (Eds). Hospitals in a Changing Europe. Open University Press, Maidenhead, pp. 59-82.
• Hong, W., Lu, Y., Zhou, X., Jin, S., Pan, J., Lin, Q., Yang, S., Basharat, Z., Zippi, M., & Goyal, H. (2022). Usefulness of Random Forest Algorithm in Predicting Severe Acute Pancreatitis. Frontiers in cellular and infection microbiology, 12, 893294. https://doi. org/10.3389/fcimb.2022.893294
• Jackins, V., Vimal, S., Kaliappan, M. et al. (2021). AI-based smart prediction of clinical disease using a random forest classifier and Naive Bayes. J Supercomput 77, 5198–5219. https://doi.org/10.1007/s11227-020-03481-x
• Jamal, F., Ahmadini, A. A. H., Hassan, M. M., Sami, W., Ameeq, M., & Naeem, A. (2024). Exploring critical factors in referral systems at different health-care levels. World Medical & Health Policy, 16, 729–744. https://doi.org/10.1002/wmh3.632
• Jaotombo, F., Pauly, V., Fond, G., Orleans, V., Auquier, P., Ghattas, B., & Boyer, L. (2022). Machine-learning prediction for hospital length of stay using a French medicoadministrative database. Journal of market access & health policy, 11(1), 2149318. https://doi.org/10.1080/20016689.2022.2149318
• Kassambara, A. (2022). Ggpubr: ‘Ggplot2’ Based Publication Ready Plots. https://cran.rproject. org/web/packages/ggpubr/index.html
• Keskin, H.İ. (2018). Türkiye’de Sağlıkta Dönüşüm Programı ve Kamu Hastanelerinin Etkinliği. Akdeniz İİBF Dergisi, 18(38), 124-150. https://doi.org/10.25294/auiibfd.492741
• Koca, M., and Demir Uslu, Y. (2022). Health Efficiency Scorecard Applications, Retrospective Evaluation of Hospitals’ Efficiency with Stochastic Boundary Approach: The Case of AI Role Group Hospitals, Turkiye Klinikleri J Health Sci. 7(1): 212-20. https://doi. org/10.5336/healthsci.2021-81180
• Korkmaz, E. ve Tercan, Ş. (2023). Sağlık Hizmetlerinin Sunumunda Yeni Bir Uygulama Alanı Olarak Medikal Muhasebe Üzerine Bir Araştırma. Sayıştay Dergisi, 34(130), 441-467. https://doi.org/10.52836/sayistay.1335177
• Köse, H.Ö. and Polat, N. (2021), Dijital Dönüşüm ve Denetimin Geleceğine Etkisi, Sayıştay Dergisi, 32(123): 9-41 https://doi.org/10.52836/sayistay.1068328
• Krämer, J., Schreyögg, J., & Busse, R. (2019). Classification of hospital admissions into emergency and elective care: a machine learning approach. Health care management science, 22(1), 85–105. https://doi.org/10.1007/s10729-017-9423-5 Kuhn, M. (2008). Building Predictive Models in R Using the caret Package. Journal of Statistical Software, 28(5), 1-26. https://doi.org/10.18637/jss.v028.i05
• Kuhn, M., Wickham, H. (2020). Tidymodels: a collection of packages for modeling and machine learning using tidyverse principles. https://www.tidymodels.org
• Li, L., Du, T., & Zeng, S. (2022). The Different Classification of Hospitals’ Impact on Medical Outcomes of Patients in China. Frontiers in public health, 10, 855323. https://doi. org/10.3389/fpubh.2022.855323
• McHugh, M.L. (2012). Interrater reliability: the kappa statistic. Biochemia Medica 276-282. https://doi.org/10.11613/bm.2012.031
• Ministry of Health (2011). Health Services Requiring Specialized Planning in Türkiye 2011- 2023. Publication No: 836, AG Design: Ankara.
• Ministry of Health (2023). Ministry of Health Criteria for Determining the Roles of Inpatient Health Facilities. Date of the document: March 8, 2023. Number of the document: E-83913885. Ankara. https://www.saglik.gov.tr/TR-11024/saglik-bolgeplanlamasi- hakkinda-genelge-ile-hastane-yatak-ve-rolleri-tescil-onayi-201050. html, 22.04.2025
• Müller, K. (2020). Here: A Simpler Way to Find Your Files. R package version 1.0.1. Müller, K, Wickham, H. (2023). tibble: Simple Data Frames. R package version 3.2.1, https:// tibble.tidyverse.org/
• Olsen, L., Zachariae, H. (2024). CVMS: Cross-Validation for Model Selection. R package version 1.6.2.
• Övgün, B., and Küçük, A. (2013). Rescaling in Health Services: Regional Applications. Amme Idaresı Dergısı, 46(1), 57-80.
• Polat, M. (2024). Yapay Zekânın Denetimde Kullanılması ve Etik Sorunlar. Sayıştay Dergisi, 35(134), 395-423. https://doi.org/10.52836/sayistay.1554497
• Posit Team (2024). RStudio: Integrated Development Environment for R. Posit Software, PBC, Boston, MA. http://www.posit.co/
• R Core Team (2024). R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/
• Robin, X., Turck, N., Hainard, A., Tiberti, N., Lisacek, F., Sanchez, J.C., & Müller, M. (2011). pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinformatics, 12, 77. https://doi.org/10.1186/1471-2105-12-77
• Sarker, I.H. (2021). Machine Learning: Algorithms, Real-World Applications and Research Directions. SN computer science, 2(3), 160. https://doi.org/10.1007/s42979-021- 00592-x
• Schauberger, P, Walker, A. (2024). openxlsx: Read, Write and Edit xlsx Files. R package version 4.2.7.1, https://github.com/ycphs/openxlsx
• Seraj, A, Mohammadi-Khanaposhtani M, Daneshfar R, Naseri M, Esmaeili M, Baghban A, et al. (2023). Cross-validation. In: Handbook of Hydroinformatics. Elsevier. p.89-105.
• Simsekler, M. C. E., Alhashmi, N. H., Azar, E., King, N., Luqman, R. A. M. A., & Al Mulla, A. (2021). Exploring drivers of patient satisfaction using a random forest algorithm. BMC medical informatics and decision making, 21(1), 157. https://doi.org/10.1186/s12911- 021-01519-5
• Smith, G.C, Seaman, S.R, Wood, A.M, Royston, P, & White, I.R. (2014). Correcting for optimistic prediction in small data sets. American journal of epidemiology, 180(3), 318–324. https://doi.org/10.1093/aje/kwu140
• Song, M., Jung, H., Lee, S., Kim, D., & Ahn, M. (2021). Diagnostic Classification and Biomarker Identification of Alzheimer’s Disease with Random Forest Algorithm. Brain sciences, 11(4), 453. https://doi.org/10.3390/brainsci11040453
• Tatar, M., Mollahaliloğlu, S., Şahin, B., Aydın, S., Maresso, A., HernándezQuevedo, C. (2011). Turkey: Health system review. Health Systems in Transition, 13(6):1–186.
• Thölke, P., Mantilla-Ramos, Y. J., Abdelhedi, H., Maschke, C., Dehgan, A., et al. (2023). Class imbalance should not throw you off balance: Choosing the right classifiers and performance metrics for brain decoding with imbalanced data. NeuroImage, 277, 120253. https://doi.org/10.1016/j.neuroimage.2023.120253
• Tseng, S.F., Lee, T.S., & Deng, C.Y. (2015). Cluster analysis of medical service resources at district hospitals in Taiwan, 2007–2011. Journal of the Chinese Medical Association, 78(12), 732-745. https://doi.org/10.1016/j.jcma.2015.05.013
• Vieira-Meyer, A.P.G.F., Coutinho, M.B., Santos, H.P.G., Saintrain, M.V., & Candeiro, G.T.M. (2022). Brazilian Primary and Secondary Public Oral Health Attention: Are Dentists Ready to Face the COVID-19 Pandemic?. Disaster medicine and public health preparedness, 16(1), 254-261. https://doi.org/10.1017/dmp.2020.342
• Waring, E, Quinn, M, McNamara, A, Arino de la Rubia E, Zhu H, Ellis S. (2022). skimr: Compact and Flexible Summaries of Data. R package version 2.1.5, https://github.com/ ropensci/skimr
• Wickham, H. (2016). ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York, https://doi.org/10.1007/978-3-319-24277-4
• Wickham, H, François, R, Henry, L, Müller, K. and Vaughan, D. (2023). dplyr: A Grammar of Data Manipulation. https://dplyr.tidyverse.org
• Wickham, H. and Bryan, J. (2023) Readxl: Read Excel Files R Package Version 1.4.3. https:// CRAN.R-project.org/package=readxl
• Wickham, H, Vaughan D, Girlich M (2025). tidyr: Tidy Messy Data. R package version 1.3.1.9000, https://github.com/tidyverse/tidyr
• Xie, F., Zhou, J., Lee, J.W., Tan, M., Li, S., Rajnthern, L.S., Chee, M.L., Chakraborty, B., Wong, A. I., Dagan, A., Ong, M.E.H., Gao, F., & Liu, N. (2022). Benchmarking emergency department prediction models with machine learning and public electronic health records. Scientific data, 9(1), 658. https://doi.org/10.1038/s41597-022-01782-9
• Yılmaz, F. & Şenel, İ.K. (2019). Sağlık Kurumlarının Etkinliklerinin Veri Zarflama Analizi ile Değerlendirilmesi. Sosyal Güvence (15), 63-88. https://doi.org/10.21441/ sosyalguvence.600856