Nguyen Hung Manh, Baradaran Mohammad, Daigle Gaétan, Nshimyumukiza Leon, Guertin Jason Robert, Reinharz Daniel
Département de médecine sociale et préventive Université Laval Québec Canada.
Département de génie électrique et de génie informatique Université Laval Québec Canada.
Health Sci Rep. 2023 Aug 23;6(8):e1516. doi: 10.1002/hsr2.1516. eCollection 2023 Aug.
Quantitative approaches for eliciting preferences for new interventions are mostly conducted by patients and rarely by policymakers. This study aimed to quantify the preferences of pregnant women and policymakers regarding the addition of a new test to prenatal screening programs for detecting chromosomal abnormalities.
A discrete choice experiment was conducted to measure the respondents' preferences for a new prenatal test. A seven-attribute instrument was built based on interviews with pregnant women and policymakers. The data were analyzed using robust conditional logistic regression and nested logit models.
In total, 272 pregnant women and 24 policymakers completed the questionnaire (response rates of 48% and 55%, respectively). Overall, all attributes were statistically significant in the pregnant women group, whereas only three attributes (test performance, degree of test result certainty, and cost) were statistically significant in the policymakers group. Statistically significant differences in test performance and information were observed between the two groups.
Policymakers differed from pregnant women in their appraisal of attributes related to their preference for a new prenatal screening intervention. The low response rates observed in both groups suggest that further investigation of the relevance of this approach must be conducted.
引发对新干预措施偏好的定量方法大多由患者进行，而政策制定者很少采用。本研究旨在量化孕妇和政策制定者对在产前筛查项目中增加一项用于检测染色体异常的新检测的偏好。
进行了一项离散选择实验，以测量受访者对一项新的产前检测的偏好。基于对孕妇和政策制定者的访谈构建了一个七属性工具。使用稳健条件逻辑回归和嵌套逻辑模型对数据进行分析。
共有272名孕妇和24名政策制定者完成了问卷（回复率分别为48%和55%）。总体而言，所有属性在孕妇组中均具有统计学意义，而在政策制定者组中只有三个属性（检测性能、检测结果确定性程度和成本）具有统计学意义。两组之间在检测性能和信息方面观察到具有统计学意义的差异。
政策制定者在对与新的产前筛查干预措施偏好相关属性的评估上与孕妇存在差异。两组中观察到的低回复率表明必须对这种方法的相关性进行进一步研究。
