Abstract: Background　For various reasons，cancer patients are at high risk of suicide.At present，the research mainly focuses on the causes and nursing of suicide in patients with malignant tumors.There are few studies on suicide theory and evaluation methods，and there is no unified standard for suicide risk assessment of patients.Objective　To develop Cancer Suicide Risk Scale（CSRS） and examine its reliability and validity.Methods　Items of CSRS was initially developed based on literature reviews，expert interview and semi-structured questionnaire survey from January to December 2017.The draft scale was formulated by expert consultation and pre-survey.A total of 355 cancer patients admitted to the Affiliated Hospital of Southwest Medical University and Traditional Chinese Medicine Hospital Affiliated to Southwest Medical University were selected for investigation by convenience sampling to test the reliability and validity of the scale，and SPSS 22.0 was used to analyze the data.Results　The variation coefficient of items were greater than 15%，and the item score of the high score group was higher than that of the low score group，and the difference were statistically significant（P<0.001）.CSRS had 19 items，including 4 factors，negative emotion，suicide attitude，suicide preparation，and suicide motivation，and the cumulative variance contribution rate was 54.796%.The correlation coefficient between each dimension was 0.370-0.556，between scale and each dimension was 0.635-0.891（P<0.05）.The score of CSRS was positively correlated with the score of Self-rating Depression Scale（SDS）（r=0.708，P<0.001）.The content validity of CSRS was 0.919，Cronbach's α coefficient was 0.876，split-half reliability was 0.874.Conclusion　Cancer suicide risk scale includes four dimensions and 19 items，which basically consistents with the theoretical assumption and has good reliability and validity，so it can be used to evaluate suicide risk of cancer patients in Chinasuicide risk of cancer patients in China.

Key words: Carcinoma, Suicide, Scale, Reliability, Validity

摘要： 背景　由于各种原因，癌症患者成为自杀的高危人群。目前研究主要围绕恶性肿瘤患者自杀的原因和护理，鲜见关于自杀理论和评估方法的研究，对患者的自杀风险评估缺乏统一标准。目的　初步构建恶性肿瘤患者自杀风险评估量表（CSRS）并进行信效度评价。方法　2017年1—12月在文献分析的基础上开展专家访谈和半结构化问卷调查并初步拟定量表初始条目，通过专家评定、预调查形成预成量表，采用便利抽样法对西南医科大学附属医院及西南医科大学附属中医医院的355例恶性肿瘤患者进行问卷调查，采用SPSS 22.0软件对数据进行统计分析，检验量表的信效度。结果　各条目变异系数（CV）均>15%，条目决断值（CR）法结果显示高分组各条目得分高于低分组（P<0.001）。CSRS包括4个因子共19个条目，4个因子分别为负性情绪、自杀态度、自杀准备和自杀动机，累积方差贡献率为54.796%；各因子得分间相关系数为0.370~0.556，各因子得分与量表总分间的相关系数为0.635~0.891（P<0.05）。CSRS总分与抑郁自评量表（SDS）总分之间呈正相关（r=0.708，P<0.001）。量表内容效度指数为0.919，Cronbach's α系数为0.876，分半信度为0.874。结论　CSRS共包括4个维度19个条目，基本符合理论构想，具有较好的信效度，可用于我国恶性肿瘤患者自杀风险的评估。

关键词: 癌, 自杀, 量表, 信度, 效度