Abstract: Background　Prevention and management of depression is essential for relapse and recurrence and rehabilitation in depression.Good prevention and management of depression are helpful for effectively reducing the morbidity，rates of relapse and recurrence.Personalized and stage-based interventions and management evaluations are important facilitators for the rehabilitation and reduction of relapse and recurrence.However，there is a lack of effective tools to evaluate the process of depression prevention and management.Objective　To sinicize the English version of Depression Prevention & Management Survey（DPMS） and test the validity and reliability of the Chinese version.Methods　Data were collected in 2017.Stratified multistage sampling with the Center for Epidemiological Studies Depression Scale（CES-D） was used to select 429 residents with depressive symptoms（CES-D score ≥16 or with a previous diagnosis of depression） from 2 000 residents randomly selected from 40 communities in 5 administrative divisions（8 communities were selected from each division，and 50 residents from each community） in 7 central urban areas of Wuhan.The Chinese version of DPMS was developed via translating the English version，and was used to survey the 429 cases，then based on the survey results，its validity was explored via content validity analysis，exploratory factor analysis and confirmatory factor analysis，and its reliability was tested using internal consistency coefficient and Pearson correlation analysis.Two weeks later，the test-retest reliability of the scale was evaluated in a convenience sample of 30 cases selected from the initial sample.Results　The Chinese version of the DPMS retained 45 items and 4 subscales of the original scale.The inter-subscale correlation coefficients of the Chinese version of DPMS ranged -0.001-0.410.Except that the change stage was not related to decisional obstacles，the correlation coefficients of other subscales were statistically different（P<0.05）.The content validity index for the scale was 0.85，and for the items ranged 0.67-0.87.Exploratory factor analysis of the decision equilibrium subscale yielded two common factors，which were named perceptual benefits and perceptual obstacles，respectively，accounting for 61.89% of the cumulative variance，and confirmatory factor analysis showed that NFI=0.97，CFI=0.99.Exploratory factor analysis of the change process produced 4 common factors named personal cognition，knowledge and social support，public environment，and positive feedback，respectively，accounting for 61.98% of the cumulative variance，and confirmatory factor analysis showed that NFI=0.83，CFI=0.91.A common factor was obtained through exploratory factor analysis for self-efficacy subscale，which was named as self-efficacy，accounting for 68.62% of the cumulative variance，and confirmatory factor analysis showed that NFI=0.98 and CFI=0.99.The Cronbach's α coefficient for the scale was 0.95，and for the subscales ranged 0.68-0.95.The test-retest reliability coefficient for the scale was 0.75，and 0.85（P<0.01） for change stage subscale，0.64（P<0.05） for decision equilibrium subscale，0.53（P>0.05） for change process subscale，and 0.94（P<0.01） for self-efficacy subscale，respectively.Conclusion　The Chinese version of DPMS showed good validity and reliability in our survey，which may be useful in evaluating behavioral changes in depression prevention and management.

Key words: Depression, Prevention, Management, Reliability, Validity, Community residents, Scale

摘要： 背景　抑郁症预防与管理是抑郁症复发及康复过程中的重要环节，良好的抑郁症预防和管理有助于减少抑郁症的发生，有效降低发病率和复发率。个性化的阶段性预防与管理评估对促进抑郁症康复及降低抑郁症复发率有重要意义。但是，目前缺乏评估抑郁症预防及管理过程的有效工具。目的　汉化抑郁症预防与管理调查量表（DPMS），并检验其信度和效度。方法　2017年度随机抽取武汉市7个中心城区5个行政区的40个社区，每个行政区抽取8个社区，每个社区抽取50名居民，共筛查社区居民2 000名，采用多阶段分层抽样法选择存在抑郁症状（流调用抑郁自评问卷得分≥16分）或曾被诊断为抑郁症的社区居民共429名进行调查。对英文版DPMS进行翻译形成中文版DPMS，通过内容效度、探索性因子分析及验证性因子分析评价其效度，通过内部一致性和Pearson相关分析评价其信度，2周后采取方便抽样从已参与正式调查的人群中抽取30例进行二次调查以确定量表重测信度。结果　中文版DPMS保留原量表的45个条目，4个维度（变化阶段、决策平衡、变化过程、自我效能），中文版DPMS各维度相关系数在-0.001~0.410之间，除了变化阶段与决策障碍不相关外，其余维度间均有相关性（P<0.05）。量表水平的内容效度指数为0.85，条目水平的内容效度指数为0.67~0.87。决策平衡维度探索性因子分析得到2个公因子，累积方差贡献率为61.89%，分别命名为知觉利益、知觉障碍，验证性因子分析正规拟合指数（NFI）=0.97，比较拟合指数（CFI）=0.99；变化过程维度探索性因子分析得到4个公因子，累积方差贡献率为61.98%，分别命名为个人认知、知识和社会支持、公众环境和正性反馈，验证性因子分析NFI=0.83，CFI=0.91；自我效能维度探索性因子分析得到1个公因子，累积方差贡献率为68.62%，命名为自我效能，验证性因子分析NFI=0.98，CFI=0.99。量表的Cronbach'sα系数为0.95，各维度的Cronbach's α系数为0.68~0.95；量表的重测信度系数为0.75，各维度的重测信度：变化阶段0.85（P<0.01）、决策平衡0.64（P<0.05）、变化过程0.53（P>0.05）和自我效能0.94（P<0.01）。结论　中文版DPMS具有良好的信效度，可用于评价抑郁症预防与管理的行为转变。

关键词: 抑郁症, 预防, 管理, 信度, 效度, 社区居民, 量表