Supplementary materials: Assessing the performance of physician’s prescribing preference as an instrumental variable in comparative effectiveness research with moderate and small sample sizes: a simulation study

These are peer-reviewed supplementary materials for the article 'Supplementary materials: Assessing the performance of physician’s prescribing preference as an instrumental variable in comparative effectiveness research with moderate and small sample sizes: a simulation study' published in the Journal of Comparative Effectiveness Research.Table S1. The performance of IV estimates and OLS estimates in the samples size of 255Table S2. The performance of IV estimates and OLS estimates in the samples size of 620Table S3. The performance of IV estimates and OLS estimates in the sample sizes of 2452TableS4. The performance of IV estimates and OLS estimates in the samples size of 5869Table S5. The performance of ‘true’ prescribing preference in four sample sizesTable S6. The performance of proportional preference in four sample sizesR CodesAim: This simulation study is to assess the utility of physician’s prescribing preference (PPP) as an instrumental variable for moderate and smaller sample sizes. Materials & methods: We designed a simulation study to imitate a comparative effectiveness research under different sample sizes. We compare the performance of instrumental variable (IV) and non-IV approaches using two-stage least squares (2SLS) and ordinary least squares (OLS) methods, respectively. Further, we test the performance of different forms of proxies for PPP as an IV. Results: The percent bias of 2SLS is around approximately 20%, while the percent bias of OLS is close to 60%. The sample size is not associated with the level of bias for the PPP IV approach. Conclusion: Irrespective of sample size, the PPP IV approach leads to less biased estimates of treatment effectiveness than OLS adjusting for known confounding only. Particularly for smaller sample sizes, we recommend constructing PPP from long prescribing histories to improve statistical power.

本数据集为发表于《比较有效性研究杂志》的论文《补充材料：评估医师处方偏好的作为比较有效性研究中中等和小样本量工具变量的性能：一项模拟研究》的同行评审补充材料。表S1：在样本量为255的情况下的工具变量估计和普通最小二乘估计的性能；表S2：在样本量为620的情况下的工具变量估计和普通最小二乘估计的性能；表S3：在样本量为2452的情况下的工具变量估计和普通最小二乘估计的性能；表S4：在样本量为5869的情况下的工具变量估计和普通最小二乘估计的性能；表S5：在四个样本量下的真实处方偏好的性能；表S6：在四个样本量下的比例偏好的性能；R代码：目的：本模拟研究旨在评估医师处方偏好（PPP）作为中等及较小样本量工具变量的效用。研究方法：我们设计了一项模拟研究，以模仿不同样本量下的比较有效性研究。我们分别使用两阶段最小二乘法（2SLS）和普通最小二乘法（OLS）比较了工具变量（IV）和非工具变量方法的表现。此外，我们还测试了不同形式的PPP作为IV的替代物的性能。结果：2SLS的百分比偏差约为20%，而OLS的百分比偏差接近60%。样本量与PPP IV方法偏差水平无关。结论：无论样本量大小，PPP IV方法相比于仅调整已知混杂因素的OLS，能够得到对治疗效果估计的更小偏差。特别对于较小的样本量，我们建议从长期处方历史中构建PPP以提高统计功效。