驗證簡短版柏格平衡量表在慢性思覺失調症患者的同時效度與預測效度

Validation of the Concurrent and Predictive Validity of Short-forms of the Berg Balance Scale in Patients with Chronic Schizophrenia

目的：本研究旨在驗證二種簡短版柏格平衡量表（Short-form Berg Balance Scale, SFBBS和Short form of the BBS using an Machine Learning Approach, BBS-ML）在慢性思覺失調症患者中的同時效度與預測效度，為注意力有限患者提供快速有效的評估工具。方法：研究納入61名慢性思覺失調症患者，研究者使用Pearson相關係數分析二簡短版BBS與原版BBS的同時效度，並檢驗其預測跌倒的能力。結果：SFBBS與原版BBS總分之間的相關係數為0.95，BBS-ML為0.90，顯示兩者具良好的同時效度。預測效度分析顯示：SFBBS與有無跌倒的相關係數為-0.52，BBS-ML為-0.43，均為中度負相關。結論：本研究結果證實，SFBBS與BBS-ML在慢性思覺失調症患者中有良好的同時效度與預測效度。簡短版量表操作簡便，能減少評估時間與患者負擔，具臨床應用潛力，特別適用於注意力有限的患者。

Objectives. This study aimed to examine the concurrent and predictive validity of two short-form versions of the Berg Balance Scale (Short-form Berg Balance Scale, SFBBS, and the Short Form of the BBS using a Machine Learning Approach, BBS-ML) in patients with chronic schizophrenia. The goal was to provide efficient and rapid assessment tools suitable for individuals with limited attention spans.
Methods. A total of 61 patients with chronic schizophrenia were recruited. Pearson correlation coefficients were used to assess the concurrent validity of the two short-form versions compared to the original BBS, and to evaluate their predictive validity in relation to fall risk.
Results. The SFBBS showed a strong correlation with the original BBS (r=0.95), and the BBS-ML also demonstrated a high correlation (r=0.90), indicating strong concurrent validity for both. The SFBBS was moderately negatively correlated with fall occurrence (r=-0.52), as was the BBS-ML (r=-0.43), supporting their predictive validity.
Conclusion. Both the SFBBS and BBS-ML demonstrated strong concurrent and moderate predictive validity in patients with chronic schizophrenia. These short-form scales are simple to administer, reducing assessment burden while maintaining validity. They may be particularly useful for clinical populations with reduced attention capacity.
(J Med Health. 2025;14(3):37-45)