Smith M C, Holcombe J K, Stullenbarger E
University of Alabama at Birmingham.
Oncol Nurs Forum. 1994 Aug;21(7):1201-9; discussion 1209-10.
To describe intervention effectiveness for symptom management in oncology nursing research.
Integrative research review and meta-analysis.
28 randomized experimental and control group studies of symptom management that were identified from 428 published and unpublished oncology nursing research reports authored by nurses from 1981-1990.
Experimental studies identified and coded by four oncology nurse specialists; findings transformed into effect sizes; effect sizes combined into overall effect for 28 studies; and effects clustered by symptoms, interventions, and outcomes.
Combined effect for 28 symptom management study outcomes.
For 28 study outcomes tested for symptom management, weighted average effect was d = 0.47, the 95% confidence interval ranged from 0.31 to 0.61, statistical significance was p = 5.17 x 10(-12) with wide variability of effect (chi 2 = 66.574, df = 27, p = 3.41 x 10(-5), and success rate improved from 39% to 62% for subjects in the treatment group.
Interventions were effective in relieving symptoms in the collection of studies and in the clusters for managing nausea and vomiting, pain, anxiety, alopecia, infection, and side effects of chemotherapy.
Missing data related to subject characteristics and need for replications prior to use in practice preclude making specific recommendations at this time for research-based nursing interventions for symptom management in patients with cancer.
描述肿瘤护理研究中症状管理的干预效果。
综合研究综述与荟萃分析。
从1981年至1990年护士撰写的428篇已发表和未发表的肿瘤护理研究报告中识别出的28项症状管理随机实验组和对照组研究。
由四位肿瘤护理专家识别并编码实验研究；将研究结果转化为效应量；将28项研究的效应量合并为总体效应；并按症状、干预措施和结果进行效应聚类。
28项症状管理研究结果的合并效应。
对28项症状管理研究结果进行测试，加权平均效应为d = 0.47，95%置信区间为0.31至0.61，统计学显著性为p = 5.17×10⁻¹²，效应存在广泛差异（卡方 = 66.574，自由度 = 27，p = 3.41×10⁻⁵），治疗组受试者的成功率从39%提高到62%。
在这些研究以及管理恶心呕吐、疼痛、焦虑、脱发、感染和化疗副作用的聚类研究中，干预措施在缓解症状方面有效。
与受试者特征相关的缺失数据以及在实践中使用前需要重复研究，使得目前无法针对癌症患者症状管理的循证护理干预提出具体建议。
