摘要
目的研发用于精神障碍筛查与风险评估的筛查软件并检测其有效性。方法 2017年9—12月运用随机整群抽样法抽取1 883名入营应征青年,对其实施精神障碍筛查软件的受测者自评、知情者他评和计算机引导的半结构式访谈3部分软件测试。结果 (1)5类危险行为(自伤、伤人、破坏行为、私自出走、晕倒抽搐)及总危险行为与自评的阳性符合率为95.96%~98.61%,阴性符合率为80.14%~88.13%,总符合率为80.75%~89.14%;与他评的阳性符合率为33.33%~69.84%,阴性符合率为80.14%~88.13%,总符合率为91.62%~97.05%;与"自评+他评"的阳性符合率均为100.00%,阴性符合率为79.01%~87.34%,总符合率为79.77%~88.97%;自评、他评、"自评+他评"测试结果均与访谈有关联,且呈正相关(Φ=0.209~0.685,均P<0.05)。(2)自评与他评的阳性符合率、阴性符合率和总符合率分别为82.79%、81.98%和82.03%,自评与"自评+他评"的阳性符合率、阴性符合率和总符合率分别为94.94%、100.00%和98.79%,他评与"自评+他评"的阳性符合率、阴性符合率和总符合率分别为29.61%、100.00%和83.25%;前置条件测试结果之间均有关联,呈正相关(Φ=0.396~0.967,均P<0.05)。(3)城镇入营应征青年自伤、晕倒抽搐和总危险行为的发生率显著高于农村入营应征青年,差异均有统计学意义(χ~2=6.336、4.321、6.210,均P<0.05)。结论软件前置条件与判别条件设计合理,可以较准确的评估危险行为;在危险行为筛查中,应选自评+访谈、自评+他评+访谈两种模式之一,则能确保评估的准确性。
[Objective]To develop the Screening and Risk Assessment System for Mental Disorders(SRAS-MD) and test its effectiveness.[Methods]1 883 recruited male youth were collected by the random cluster sampling from September to December 2017,and they were assessed with SRAS-MD in three parts: subjects’ self-assessment,others-ratings and computer-guided semi-structured interview.[Results]① For the five kinds of risk behaviors(self-injury,injury,destruction,running away,fainting and convulsion) and total risk behaviors,the positive,negative and total coincidence rates to self-assessment were 95.96%-98.61%,80.14%-88.13% and 80.75%-89.14%,respectively. The positive,negative and total coincidence rates to others-ratings were 33.33%-69.84%,80.14%-88.13%,91.62%-97.05%,respectively. The positive,negative and total coincidence rates to selfassessment+others-ratings were 100.00%,79.01%-87.34% and 79.77%-88.97%,respectively. The test results of self-assessment,others-ratings,and self-assessment+others-ratings were positively correlated with interviews(Φ=0.209-0.685,all P<0.05).② The positive,negative and total coincidence rates between self-assessment and others-ratings were 82.79%,81.98% and82.03%,respectively. The positive,negative and tota l coincidence rates between self-assessment and self-assessment+others-ratings were 94.94%,100.00% and 98.79%,respectively. The positive,negative and total coincidence rates between others-ratings and selfassessment+others-ratings were 29.61%,100.00% and 83.25%,respectively. There were positive correlations among preconditions(Φ=0.396-0.967,all P<0.05).③The incidence rates of self-injury,fainting and convulsion and total risk behavior among urban recruited youth were higher than those among rural recruited youth(χ^2=6.336,4.321,6.210,all P<0.05).[Conclusion]The criterion and prec ondition design of software are rea sonable,which can accurately evaluate risk behavior. In the screening of risk behavior,one of the two modes,self-assessment+interview and self-assessment+others-ratings+interview,should be selected to ensure the accuracy of assessment.
作者
马宏丽
李红政
王骞
雷美英
陈海燕
MA Hong-li;LI Hong-zheng;WANG Qian;LEI Mei-ying;CHEN Hai-yan(Mental Health Center,923 Hospital of PLA ,Nanning Guangxi,530021,China)
出处
《职业与健康》
CAS
2019年第10期1333-1336,1340,共5页
Occupation and Health
基金
心理卫生应用性重点建设项目(FYJ2015W01)
全军医学科学技术"十二五"计划课题(CWS11J278)
军队心理卫生应用性科研课题(12XLZ204)
关键词
入营应征青年
精神障碍
危险行为
筛查软件
Recruited male youth
Mental disorder
Risk behavior
Screening software