Profiling Patients with Chronic Ulcers Using K-Means Clustering and Analysis of the Impact on the Consumption of Medical Resources: Retrospective Study on Hospitalized Patients in Romania
Year: 2025
Type: article
Study design: Observational Study
Abstract: Background/Objectives: Chronic ulcers represent a major public health concern, being associated with substantial morbidity, impaired quality of life, and significant costs to healthcare systems. Against the backdrop of an aging population and increasing prevalence of chronic comorbid conditions, this study aimed to profile hospitalized patients with chronic ulcers in Romania and to examine their differential patterns of healthcare resource utilization. Methods: We conducted a retrospective analysis of the national administrative hospitalization database between 2017 and 2022, including adult patients with at least two admissions coded with a primary diagnosis of chronic ulcer. Sociodemographic, clinical, and healthcare utilization indicators were extracted, standardized, and analyzed using the K-means clustering algorithm to derive utilization-based phenotypes. Results: Two distinct patient clusters were identified: the first comprised predominantly elderly patients with multiple comorbidities, prolonged hospitalizations, and frequent readmissions, representing a high-burden profile; the second included relatively younger patients with fewer comorbidities, shorter hospital stays, and lower readmission rates, reflecting a more stable clinical profile. The high-burden cluster accounted for a disproportionate share of inpatient resource consumption, underscoring its impact on the healthcare system. Conclusions: These findings highlight the importance of early identification of potential high-burden patients, enabling the implementation of personalized care strategies and more efficient allocation of hospital resources, with the potential to improve health outcomes and support healthcare system sustainability.
Source: Journal of Clinical Medicine
DOI: 10.3390/jcm14176252
Language: English
Subfield: Epidemiology
Field: Medicine
Domain: Health Sciences
Open Access status: gold
Estimated APC paid: $3,137
OpenAlex ID: W4414023538