EuroQol Office, EuroQol Research Foundation, Rotterdam, the Netherlands; Health Values Research and Consultancy, Amsterdam, the Netherlands.
Centre for Research on Health and Social Care Management (CERGAS), SDA Bocconi School of Management, Bocconi University, Milan, Italy.
Soc Sci Med. 2022 Jan;292:114519. doi: 10.1016/j.socscimed.2021.114519. Epub 2021 Oct 28.
To test the feasibility of using videoconferencing (VC) administered interviews and to derive an EQ-5D-5L value set for Italy.
Preferences were collected using the EuroQol standardized valuation protocol (EQ-VT) administered via VC. Two valuation methods were employed, composite time trade-off (cTTO) and discrete choice experiment (DCE). Technical, organizational and protocol feasibility were tested in a pilot of 198 interviews. Upon positive assessment, data collection continued with a target sample of 1000-1200 participants including the pilot. Quality control (QC) procedures were employed to monitor interviewers' performance during the pilot and the data collection. Data were modelled using GLS, Tobit, Logit and Hybrid models with different error specifications. Monotonicity of coefficients, statistical significance, and theoretical considerations informed the model choice.
Dropouts and technical problems occurred in less than 5% of the 198 pilot interviews. Protocol compliance was demonstrated with significant improvements in QC parameters and limited interviewers' effects, for all interviewers. Overall, interviewers were satisfied with this mode of administration, highlighting it allows flexibility and efficient scheduling. Based on these results, VC was deemed as a feasible mode of administration. The study collected preferences for 1182 responders, including the pilot interviews. The demographic characteristics of the sample were representative of the Italian general population for age, gender and geographical macro-areas. The hybrid Tobit heteroscedastic model without constant estimated on the full sample (including pilot) was selected for the derivation of the value set. Values ranged from -0.571 for the worst health state (55555) to 1 for the best health state (11111). Pain/discomfort registered the largest decrement, followed by mobility, anxiety/depression, self-care, and usual activities. 523 health states were worse than dead.
VC is viable for the conduct of valuation interviews. The Italian value set for the EQ-5D-5L can be used for value determinations of health technologies.
测试使用视频会议（VC）进行访谈的可行性，并为意大利制定 EQ-5D-5L 值集。
通过 VC 管理的 EuroQol 标准化估值协议（EQ-VT）收集偏好。采用复合时间权衡（cTTO）和离散选择实验（DCE）两种估值方法。在 198 次访谈的试点中，测试了技术、组织和协议的可行性。在对试点进行积极评估后，继续收集目标样本 1000-1200 名参与者（包括试点）的数据。采用质量控制（QC）程序，以监测试点和数据收集期间访谈者的表现。使用 GLS、Tobit、Logit 和混合模型以及不同的误差规范对数据进行建模。根据系数的单调性、统计显著性和理论考虑，选择了模型。
198 次试点访谈中，仅不到 5%的受访者出现了中途退出和技术问题。所有访谈者的 QC 参数都得到了显著改善，并且访谈者的影响有限，证明了协议的合规性。总体而言，访谈者对这种管理模式感到满意，认为它具有灵活性和高效的调度能力。基于这些结果，VC 被认为是一种可行的管理模式。该研究共收集了 1182 名受访者的偏好，包括试点访谈。样本的人口统计学特征在年龄、性别和地理大区方面代表了意大利的一般人口。在包括试点在内的整个样本上估计的无常数的混合 Tobit 异方差模型被选为值集的推导。值范围从最差健康状态（55555）的-0.571 到最佳健康状态（11111）的 1。疼痛/不适的下降幅度最大，其次是移动性、焦虑/抑郁、自我护理和日常活动。有 523 种健康状态比死亡更差。
VC 可用于进行估值访谈。可将意大利 EQ-5D-5L 值集用于健康技术的价值评估。
