Purba Fredrick Dermawan, Hunfeld Joke A M, Iskandarsyah Aulia, Fitriana Titi Sahidah, Sadarjoen Sawitri S, Passchier Jan, Busschbach Jan J V
Department of Psychiatry, Section Medical Psychology and Psychotherapy, Erasmus MC University Medical Center Rotterdam, s-Gravendijkwal 230, 3015 CE, Rotterdam, The Netherlands.
Department of Developmental Psychology, Faculty of Psychology, Padjadjaran University, Jatinangor, Indonesia.
Qual Life Res. 2017 May;26(5):1197-1208. doi: 10.1007/s11136-016-1445-9. Epub 2016 Oct 31.
In valuing health states using generic questionnaires such as EQ-5D, there are unrevealed issues with the quality of the data collection. The aims were to describe the problems encountered during valuation and to evaluate a quality control report and subsequent retraining of interviewers in improving this valuation.
Data from the first 266 respondents in an EQ-5D-5L valuation study were used. Interviewers were trained and answered questions regarding problems during these initial interviews. Thematic analysis was used, and individual feedback was provided. After completion of 98 interviews, a first quantitative quality control (QC) report was generated, followed by a 1-day retraining program. Subsequently individual feedback was also given on the basis of follow-up QCs. The Wilcoxon signed-rank test was used to assess improvements based on 7 indicators of quality as identified in the first QC and the QC conducted after a further 168 interviews.
Interviewers encountered problems in recruiting respondents. Solutions provided were: optimization of the time of interview, the use of broader networks and the use of different scripts to explain the project's goals to respondents. For problems in interviewing process, solutions applied were: developing the technical and personal skills of the interviewers and stimulating the respondents' thought processes. There were also technical problems related to hardware, software and internet connections. There was an improvement in all 7 indicators of quality after the second QC.
Training before and during a study, and individual feedback on the basis of a quantitative QC, can increase the validity of values obtained from generic questionnaires.
在使用如EQ-5D等通用问卷评估健康状态时，数据收集质量存在未被揭示的问题。本研究旨在描述评估过程中遇到的问题，并评估一份质量控制报告以及随后对访谈员进行的再培训对改善这种评估的效果。
使用了EQ-5D-5L评估研究中前266名受访者的数据。对访谈员进行了培训，并询问了他们在这些初始访谈中遇到的问题。采用了主题分析法，并提供了个人反馈。在完成98次访谈后，生成了第一份定量质量控制（QC）报告，随后开展了为期1天的再培训项目。随后，还根据后续的质量控制提供了个人反馈。使用Wilcoxon符号秩检验，根据第一次质量控制中确定的7个质量指标以及在另外168次访谈后进行的质量控制来评估改进情况。
访谈员在招募受访者方面遇到问题。提供的解决方案包括：优化访谈时间、利用更广泛的网络以及使用不同的脚本来向受访者解释项目目标。对于访谈过程中的问题，采用的解决方案包括：培养访谈员的技术和个人技能以及激发受访者的思维过程。还存在与硬件、软件和互联网连接相关的技术问题。第二次质量控制后，所有7个质量指标均有所改善。
在研究前和研究过程中进行培训，并基于定量质量控制提供个人反馈，可以提高从通用问卷中获得的值的有效性。
