Lowery J C, Hiller L D, Davis J A, Shore C J
VA Health Services Research and Development Field Program, Ann Arbor, MI 48113-0170, USA.
Med Care. 1998 Nov;36(11):1578-88. doi: 10.1097/00005650-199811000-00007.
The classification of a patient's nutrition status is important for identifying patients who require nutrition care, for designing effective nutrition interventions, and for measuring severity of illness. The objective of this study was to evaluate the reliability and validity of two variants of the Department of Veterans Affairs' nutrition status classification: professional judgment versus an algorithm.
The study consisted of two phases, both of which included providing a sample of approximately 60 registered dietitians and 60 clinical dietetic technicians with data on 16 (phase I) and 20 (phase II) patients, to which they assigned nutrition statuses using both professional judgment and the algorithm. Improvements in instructions and training were implemented between the two phases. Interrater reliability of the responses was calculated, and content validity was measured by comparing the staff's responses with those of an expert panel.
Reliability improved significantly between phases for both professional judgment and the algorithm. Greater reliability and validity were observed with use of the algorithm, by both dietitians and technicians, during both phases.
Classification of a patient's nutrition status is important in the delivery of cost-effective health care. The Department of Veterans Affairs' nutrition status classification is a good one for assessing nutrition status quickly and reliably, especially when an algorithm is used. The results underscore the advantages of a classification system based on an algorithm when the system is designed to be used by many different staff across multiple facilities.
对患者营养状况进行分类，对于识别需要营养护理的患者、设计有效的营养干预措施以及衡量疾病严重程度而言至关重要。本研究的目的是评估退伍军人事务部营养状况分类的两种变体的可靠性和有效性：专业判断与一种算法。
该研究包括两个阶段，在两个阶段中均向约60名注册营养师和60名临床营养技师提供16名（第一阶段）和20名（第二阶段）患者的数据，让他们分别使用专业判断和算法对营养状况进行分类。两个阶段之间对指导和培训进行了改进。计算回答的评分者间信度，并通过将工作人员的回答与专家小组的回答进行比较来衡量内容效度。
专业判断和算法在两个阶段之间的信度均显著提高。在两个阶段中，营养师和技师使用算法时均观察到更高的信度和效度。
对患者营养状况进行分类在提供具有成本效益的医疗保健方面很重要。退伍军人事务部的营养状况分类是一种快速且可靠地评估营养状况的良好方法，尤其是在使用算法时。结果强调了基于算法的分类系统在设计供多个机构的许多不同工作人员使用时的优势。
