Hui Xie1, Weihua Gao2, Baodong Xing3, Daniel F Heitjan4, Donald Hedeker5, Chengbo Yuan2. 1. Division of Epidemiology & Biostatistics (M/C 923), School of Public Health, University of Illinois at Chicago, 1603 West Taylor Street, Room 984, Chicago, 60612-4336, IL; Faculty of Health Sciences, Simon Fraser University, Burnaby, BC, V6S0G6, Canada; Arthritis Research Canada, Richmond, BC, V6X 2C7, Canada. Electronic address: huixie@uic.edu. 2. Division of Epidemiology & Biostatistics (M/C 923), School of Public Health, University of Illinois at Chicago, 1603 West Taylor Street, Room 984, Chicago, 60612-4336, IL. 3. AbbVie Inc., Chicago, 60612, IL. 4. Department of Statistical Science, Southern Methodist University, Dallas, 75275-0332, TX; Department of Clinical Sciences, University of Texas Southwestern Medical Center, Dallas, 75390-9066, TX. Electronic address: dheitjan@smu.edu. 5. Department of Public Health Sciences, The University of Chicago, Chicago, 60637, IL.
Abstract
BACKGROUND AND OBJECTIVE: The popular assumption of ignorability simplifies analyses with incomplete data, but if it is not satisfied, results may be incorrect. Therefore it is necessary to assess the sensitivity of empirical findings to this assumption. We have created a user-friendly and freely available software program to conduct such analyses. METHOD: One can evaluate the dependence of inferences on the assumption of ignorability by measuring their sensitivity to its violation. One tool for such an analysis is the index of local sensitivity to nonignorability (ISNI), which evaluates the rate of change of parameter estimates to the assumed degree of nonignorability in the neighborhood of an ignorable model. Computation of ISNI avoids the need to estimate a nonignorable model or to posit a specific magnitude of nonignorability. Our new R package, named isni, implements ISNI analysis for some common data structures and corresponding statistical models. RESULT: The isni package computes ISNI in the generalized linear model for independent data, and in the marginal multivariate Gaussian model and the linear mixed model for longitudinal/clustered data. It allows for arbitrary patterns of missingness caused by dropout and/or intermittent missingness. Examples illustrate its use and features. CONCLUSIONS: The R package isni enables a systematic and efficient sensitivity analysis that informs evaluations of reliability and validity of empirical findings from incomplete data.
BACKGROUND AND OBJECTIVE: The popular assumption of ignorability simplifies analyses with incomplete data, but if it is not satisfied, results may be incorrect. Therefore it is necessary to assess the sensitivity of empirical findings to this assumption. We have created a user-friendly and freely available software program to conduct such analyses. METHOD: One can evaluate the dependence of inferences on the assumption of ignorability by measuring their sensitivity to its violation. One tool for such an analysis is the index of local sensitivity to nonignorability (ISNI), which evaluates the rate of change of parameter estimates to the assumed degree of nonignorability in the neighborhood of an ignorable model. Computation of ISNI avoids the need to estimate a nonignorable model or to posit a specific magnitude of nonignorability. Our new R package, named isni, implements ISNI analysis for some common data structures and corresponding statistical models. RESULT: The isni package computes ISNI in the generalized linear model for independent data, and in the marginal multivariate Gaussian model and the linear mixed model for longitudinal/clustered data. It allows for arbitrary patterns of missingness caused by dropout and/or intermittent missingness. Examples illustrate its use and features. CONCLUSIONS: The R package isni enables a systematic and efficient sensitivity analysis that informs evaluations of reliability and validity of empirical findings from incomplete data.
Authors: Jill K Murphy; Hui Xie; Vu Cong Nguyen; Leena W Chau; Pham Thi Oanh; Tran Kieu Nhu; John O'Neil; Charles H Goldsmith; Nguyen Van Hoi; Yue Ma; Hayami Lou; Wayne Jones; Harry Minas Journal: Int J Ment Health Syst Date: 2020-02-12
Authors: Nicole L A Catherine; Michael Boyle; Yufei Zheng; Lawrence McCandless; Hui Xie; Rosemary Lever; Debbie Sheehan; Andrea Gonzalez; Susan M Jack; Amiram Gafni; Lil Tonmyr; Lenora Marcellus; Colleen Varcoe; Ange Cullen; Kathleen Hjertaas; Caitlin Riebe; Nikolina Rikert; Ashvini Sunthoram; Ronald Barr; Harriet MacMillan; Charlotte Waddell Journal: CMAJ Open Date: 2020-10-27