Purpose: This study aimed to identify symptom clusters experienced by patients with stroke, as well as factors that affect each cluster, to provide a basis for nursing intervention in symptom management.Method: This secondary data analysis study used retrospective data to analyze electronic medical records from a cohort of 240 patients between 8 days and 6 months after stroke diagnosis. Data were collected from medical records using a questionnaire to examine the general characteristics, clinical characteristics, hemiplegia, dysphagia, dysarthria, and pain symptoms of patients who received inpatient treatment at P University Hospital in Y City from December 1, 2018, to November 30, 2022. Data were analyzed using SPSS/WIN 28.0 to determine descriptive statistics, chi-square test, Fisher’s exact test, Kruskal-Wallis test, two-step cluster analysis, and multinomial logistic regression analysis.Results: Among patients with stroke, 83.7% had hemiplegia, and 72.9% had dysarthria. The average PAS (penetration-aspiration score) scores for dysphagia screening and pain intensity were 3.37±2.81 and 1.73±0.84 points, respectively. The two-step cluster analysis revealed four clusters based on the symptoms of patients with stroke: “severe symptom complex group”, “hemiplegia with dysarthria group”, “hemiplegia group” and “moderate dysphagia with half-occurrence dysarthria group”. Multinomial logistic regression analysis was performed to identify factors influencing symptom clusters using the “moderate dysphagia with half-occurrence dysarthria group”. Consequently, the Nagelkerke coefficient of determination (R2), which indicates the model's explanatory power, was 23.8% (χ2=50.96, p<.001). The area of brain damage and degree of physical dysfunction affected symptom clusters.Conclusion: The symptoms experienced by patients with stroke were classified into four clusters depending on the degree of symptom burden. Among them, left or right brain damage and a high degree of physical dysfunction likely belong to the two clusters with the highest symptom burden. We hope that the findings of this study will function as fundamental data for symptom education and management tailored to the characteristics of patients with stroke.
연구목적: 본 연구는 뇌졸중 환자의 증상을 기반으로 증상클러스터를 규명하고, 각클러스터에 미치는 영향요인을 파악하여 증상관리를 위한 간호중재의 근거를 마련 하고자 시행되었다.연구방법: 본 연구는 전자의무기록 데이터를 분석한 후향적 자료를 통한 이차자료 분석 연구이며, 연구 대상은 뇌졸중 진단 후 8일에서 6개월 이내인 240명의 환자이다.자료수집은 2018년 12월 1일부터 2022년 11월 30일까지 Y시 P대학교병원에서 입원 치료를 받은 환자의 전자의무기록에서 일반적 특성, 임상적 특성, 편측마비, 연하곤란,구음장애, 통증을 선별하기 위한 조사지를 이용하였다. 자료분석은 SPSS/WIN 28.0 프로그램을 이용하여 기술통계, 교차분석, Fisher의 정확검정, 일원분산분석, Kruskal-Wallis 검정, 이단계 군집분석, 다항 로지스틱 회귀분석을 실시하였다.연구결과: 뇌졸중 환자 중 편측마비는 83.7%, 구음장애는 72.9%의 대상자가 경험하였고, 연하곤란 선별을 위한 PAS 점수는 평균 3.37±2.81점, 통증 강도는 1.73±0.84점으로 나타났다. 뇌졸중 환자의 증상을 기반으로 이단계 군집분석을 시행한결과 “심한 복합 증상군(severe symptom complex group)”, “편측마비와 구음장애군 (hemiplegia with dysarthria group)”, “편측마비군(hemiplegia group)”, “중등도 연하 곤란과 일부 구음장애군(moderate dysphagia with half-occurrence dysarthria group)”의 4개 클러스터가 도출되었다. 증상클러스터에 미치는 영향요인을 파악하기 위하여 “중등도 연하곤란과 일부 구음장애군”을 참조범주로 하여 다항 로지스틱 회귀분석을시행한 결과, 모형의 설명력을 나타내는 Nagelkerke의 결정계수(R2 )는 23.8%의설명력을 보여주었고(χ2=50.96, p