BACKGROUND The prevalence of malnutrition in cancer patients is reported as high as 65%; however, malnutrition screening is often substandard. The Malnutrition Screening Tool (MST) has been validated for use by health care professionals to detect at-risk patients; however, there is a gap in the literature regarding validation of patient-led MST screening. OBJECTIVE The aim of the study was to assess the concurrent validity of patient-led MST against the Subjective Global Assessment (SGA) and the interrater reliability of patient-led MST against dietitian-led MST in patients attending ambulatory cancer care services for chemotherapy or supportive treatments. DESIGN/PARTICIPANTS A single-site diagnostic accuracy study of 201 patients between May and June 2017 attending the ambulatory cancer care setting at an Australian metropolitan tertiary hospital in Queensland. MAIN OUTCOME MEASUREMENTS The primary outcome measures were concurrent validity and interrater reliability of MST scores as determined by patients (patient-MST), dietitians (dietitian-MST), and SGA as completed by the dietitian. STATISTICAL ANALYSIS Concurrent validity of patient-led MST scores against the SGA was determined using specificity, sensitivity, positive predictive values, and negative predictive values. Interrater reliability of patient-MST and dietitian-MST was assessed using κ coefficient. RESULTS The ability of the patient-led MST scores (0 to 1 vs 2 to 5) to indicate nutrition status was found to have a sensitivity of 94% (95% CI 81% to 99%), a specificity of 86% (95% CI 79% to 91%), and an area under the receiver operating characteristic curve of 0.93 (95% CI 0.89 to 0.96). The positive predictive value was 59% (95% CI 45% to 71%), and the negative predictive value was 99% (95% CI 95% to 100%). A weighted κ of 0.83 (95% CI 0.82 to 0.87) between patient-MST and dietitian-MST was found. CONCLUSION Patient-led MST screening is a reliable and valid measure that can accurately identify ambulatory cancer care patients as at risk or not at risk of malnutrition.

中文翻译：

---

在门诊癌症护理门诊环境中使用营养不良筛查工具评估患者主导筛查的并发有效性和评估者间可靠性

背景 据报道，癌症患者营养不良的患病率高达 65%；然而，营养不良筛查往往不合标准。营养不良筛查工具 (MST) 已经过验证，可供卫生保健专业人员用于检测高危患者；然而，关于验证患者主导的 MST 筛查的文献存在空白。目的 本研究的目的是评估患者主导的 MST 与主观总体评估 (SGA) 的同时有效性，以及患者主导的 MST 与营养师主导的 MST 在接受化疗或支持性癌症门诊治疗的患者中的评估者间信度。治疗。设计/参与者 一项针对 2017 年 5 月至 2017 年 6 月期间在昆士兰澳大利亚大城市三级医院门诊癌症护理机构就诊的 201 名患者进行的单点诊断准确性研究。主要结果测量主要结果测量是由患者（患者-MST）、营养师（营养师-MST）和营养师完成的SGA确定的MST评分的并发有效性和评分者间信度。统计分析 使用特异性、敏感性、阳性预测值和阴性预测值来确定患者主导的 MST 评分对 SGA 的同时有效性。使用κ系数评估患者-MST和营养师-MST的评分者间可靠性。结果发现，患者主导的 MST 评分（0 至 1 比 2 至 5）指示营养状况的敏感性为 94%（95% CI 81% 至 99%），特异性为 86%（95 % CI 79% 至 91%），接受者操作特征曲线下面积为 0.93（95% CI 0.89 至 0.96）。阳性预测值为 59%（95% CI 45% 至 71%），阴性预测值为 99%（95% CI 95% 至 100%）。发现患者 MST 和营养师 MST 之间的加权 κ 为 0.83（95% CI 0.82 至 0.87）。结论 以患者为主导的 MST 筛查是一种可靠且有效的措施，可以准确地将门诊癌症护理患者识别为有营养不良风险或没有营养不良风险。阳性预测值为 59%（95% CI 45% 至 71%），阴性预测值为 99%（95% CI 95% 至 100%）。发现患者 MST 和营养师 MST 之间的加权 κ 为 0.83（95% CI 0.82 至 0.87）。结论 以患者为主导的 MST 筛查是一种可靠且有效的措施，可以准确识别非住院癌症护理患者是否存在营养不良风险。阳性预测值为 59%（95% CI 45% 至 71%），阴性预测值为 99%（95% CI 95% 至 100%）。发现患者 MST 和营养师 MST 之间的加权 κ 为 0.83（95% CI 0.82 至 0.87）。结论 以患者为主导的 MST 筛查是一种可靠且有效的措施，可以准确识别非住院癌症护理患者是否存在营养不良风险。