A comparison of methods to estimate the survivor average causal effect in the presence of missing data: a simulation study
Tóm tắt
Attrition due to death and non-attendance are common sources of bias in studies of age-related diseases. A simulation study is presented to compare two methods for estimating the survivor average causal effect (SACE) of a binary exposure (sex-specific dietary iron intake) on a binary outcome (age-related macular degeneration, AMD) in this setting. A dataset of 10,000 participants was simulated 1200 times under each scenario with outcome data missing dependent on measured and unmeasured covariates and survival. Scenarios differed by the magnitude and direction of effect of an unmeasured confounder on both survival and the outcome, and whether participants who died following a protective exposure would also die if they had not received the exposure (validity of the monotonicity assumption). The performance of a marginal structural model (MSM, weighting for exposure, survival and missing data) was compared to a sensitivity approach for estimating the SACE. As an illustrative example, the SACE of iron intake on AMD was estimated using data from 39,918 participants of the Melbourne Collaborative Cohort Study. The MSM approach tended to underestimate the true magnitude of effect when the unmeasured confounder had opposing directions of effect on survival and the outcome. Overestimation was observed when the unmeasured confounder had the same direction of effect on survival and the outcome. Violation of the monotonicity assumption did not increase bias. The estimates were similar between the MSM approach and the sensitivity approach assessed at the sensitivity parameter of 1 (assuming no survival bias). In the illustrative example, high iron intake was found to be protective of AMD (adjusted OR 0.57, 95% CI 0.40–0.82) using complete case analysis via traditional logistic regression. The adjusted SACE odds ratio did not differ substantially from the complete case estimate, ranging from 0.54 to 0.58 for each of the SACE methods. On average, MSMs with weighting for exposure, missing data and survival produced biased estimates of the SACE in the presence of an unmeasured survival-outcome confounder. The direction and magnitude of effect of unmeasured survival-outcome confounders should be considered when assessing exposure-outcome associations in the presence of attrition due to death.
Tài liệu tham khảo
Hahn P, Milam AH, Dunaief JL. Maculas affected by age-related macular degeneration contain increased chelatable iron in the retinal pigment epithelium and Bruch's membrane. Arch Ophthalmol. 2003;121(8):1099–105.
Ugarte M, Osborne NN, Brown LA, Bishop PN. Iron, zinc, and copper in retinal physiology and disease. Surv Ophthalmol. 2013;58(6):585–609.
Kaarniranta K, Salminen A, Haapasalo A, Soininen H, Hiltunen M. Age-related macular degeneration (AMD): Alzheimer's disease in the eye? J Alzheimers Dis. 2011;24(4):615–31.
Wong RW, Richa DC, Hahn P, Green WR, Dunaief JL. Iron toxicity as a potential factor in AMD. Retina. 2007;27(8):997–1003.
McGuinness MB, Karahalios A, Kasza J, Guymer RH, Finger RP, Simpson JA. Survival bias when assessing risk factors for age-related macular degeneration: a tutorial with application to the exposure of smoking. Ophthalmic Epidemiol. 2017;24(4):229–38.
Frangakis CE, Rubin DB. Principal stratification in causal inference. Biometrics. 2002;58(1):21–9.
Frangakis CE, Rubin DB, An MW, MacKenzie E. Principal stratification designs to estimate input data missing due to death. Biometrics. 2007;63(3):641–9.
Hayden D, Pauler DK, Schoenfeld D. An estimator for treatment comparisons among survivors in randomized trials. Biometrics. 2005;61(1):305–10.
Rubin DB. Causal inference through potential outcomes and principal stratification: application to studies with “censoring” due to death. Stat Sci. 2006;21(3):299–309.
Zhang JL, Rubin DB. Estimation of causal effects via principal stratification when some outcomes are truncated by “death”. J Educ Behav Stat. 2016;28(4):353–68.
Robins JM, Hernan MA, Brumback B. Marginal structural models and causal inference in epidemiology. Epidemiology. 2000;11(5):550–60.
Chiba Y. Marginal structural models for estimating principal stratum direct effects under the monotonicity assumption. Biom J. 2011;53(6):1025–34.
Shardell M, Hicks GE, Ferrucci L. Doubly robust estimation and causal inference in longitudinal studies with dropout and truncation by death. Biostatistics. 2015;16(1):155–68.
Tchetgen Tchetgen EJ. Identification and estimation of survivor average causal effects. Stat Med. 2014;33(21):3601–28.
Williamson T, Ravani P. Marginal structural models in clinical research: when and how to use them? Nephrol Dial Transplant. 2017;32(s2):84–90.
Zhang JL, Rubin DB, Mealli F. Likelihood-based analysis of causal effects of job-training programs using principal stratification. J Am Stat Assoc. 2009;104(485):166–76.
Egleston BL, Scharfstein DO, Freeman EE, West SK. Causal inference for non-mortality outcomes in the presence of death. Biostatistics. 2007;8(3):526–45.
Greenland S, Pearl J, Robins JM. Causal diagrams for epidemiologic research. Epidemiology. 1999;10(1):37–48.
Shrier I, Platt RW. Reducing bias through directed acyclic graphs. BMC Med Res Methodol. 2008;8:70.
Robins JM. An analytic method for randomized trials with informative censoring: part 1. Lifetime Data Anal. 1995;1(3):241–54.
Rubin DB. Causal inference without counterfactuals - comment. J Am Stat Assoc. 2000;95(450):435–8.
Rosenbaum PR, Rubin DB. The central role of the propensity score in observational studies for causal effects. Biometrika. 1983;70(1):41–55.
Greenland S, Robins JM, Pearl J. Confounding and collapsibility in causal inference. Stat Sci. 1999;14(1):29–46.
Chiba Y. Estimation and sensitivity analysis of the survivor average causal effect under the monotonicity assumption. J Biomet Biostat. 2012;03(07):e116.
Ding P, Geng Z, Yan W, Zhou X-H. Identifiability and estimation of causal effects by principal stratification with outcomes truncated by death. J Am Stat Assoc. 2011;106(496):1578–91.
Greene T, Joffe M, Hu B, Li L, Boucher K. The balanced survivor average causal effect. Int J Biostat. 2013;9(2):291–306.
Lee K, Daniels MJ. Causal inference for bivariate longitudinal quality of life data in presence of death by using global odds ratios. Stat Med. 2013;32(24):4275–84.
Seuc AH, Peregoudov A, Betran AP, Gulmezoglu AM. Intermediate outcomes in randomized clinical trials: an introduction. Trials. 2013;14(78):78.
Sjolander A, Humphreys K, Vansteelandt S, Bellocco R, Palmgren J. Sensitivity analysis for principal stratum direct effects, with an application to a study of physical activity and coronary heart disease. Biometrics. 2009;65(2):514–20.
Frangakis C. Addressing complications of intention-to-treat analysis in the combined presence of all-or-none treatment-noncompliance and subsequent missing outcomes. Biometrika. 1999;86(2):365–79.
Rubin DB. Randomization analysis of experimental-data - the fisher randomization test - comment. J Am Stat Assoc. 1980;75(371):591–3.
Hernan MA, Robins JM. Estimating causal effects from epidemiological data. J Epidemiol Community Health. 2006;60(7):578–86.
Austin PC, Stuart EA. Moving towards best practice when using inverse probability of treatment weighting (IPTW) using the propensity score to estimate causal treatment effects in observational studies. Stat Med. 2015;34(28):3661–79.
Guo S, Fraser MW. Propensity score analysis. 2nd ed. California: Sage; 2014.
Tsiatis AA, Davidian M. Comment: demystifying double robustness: a comparison of alternative strategies for estimating a population mean from incomplete data. Stat Sci. 2007;22(4):569–73.
Austin PC. An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivar Behav Res. 2011;46(3):399–424.
Burton A, Altman DG, Royston P, Holder RL. The design of simulation studies in medical statistics. Stat Med. 2006;25(24):4279–92.
Carpenter J, Bithell J. Bootstrap confidence intervals: when, which, what? A practical guide for medical statisticians. Stat Med. 2000;19(9):1141–64.
Giles GG. The Melbourne study of diet and cancer. Proc Nutr Soc Austr. 1990;15:94–103.
Aung KZ, Robman L, Chong E, English D, Giles G, Guymer R. Non-mydriatic digital macular photography: how good is the second eye photograph? Ophthalmic Epidemiol. 2009;16(4):254–61.
Lewis J, Milligan GC, Hunt A, Zealand FSAN. NUTTAB 95: nutrient data table for use in Australia: food standards: Australia New Zealand; 1995.
Ferris FL 3rd, Wilkinson CP, Bird A, Chakravarthy U, Chew E, Csaky K, et al. Clinical classification of age-related macular degeneration. Ophthalmology. 2013;120(4):844–51.
Matsumoto M, Nishimura T. Mersenne twister: a 623-dimensionally equidistributed uniform pseudo-random number generator. ACM Transact Model Comput Simul (TOMACS). 1998;8(1):3–30.
Adams MKM, Simpson JA, Richardson AJ, Guymer RH, Williamson E, Cantsilieris S, et al. Can genetic associations change with age? CFH and age-related macular degeneration. Hum Mol Genet. 2012;21(23):5229–36.
Toops KA, Tan LX, Lakkaraju A: Apolipoprotein E isoforms and AMD. In: Advances in experimental medicine and biology: retinal degenerative diseases. Volume 854. Bowes RC, LaVail M, Anderson R, Grimm C, Hollyfield J, Ash J. Cham: Springer; 2016: 3–9.
Kasza J, Wolfe R, Schuster T. Assessing the impact of unmeasured confounding for binary outcomes using confounding functions. Int J Epidemiol. 2017;46(4):1303–11.
Ding P, VanderWeele TJ. Sensitivity analysis without assumptions. Epidemiology. 2016;27(3):368–77.
Chong EW, Simpson JA, Robman LD, Hodge AM, Aung KZ, English DR, et al. Red meat and chicken consumption and its association with age-related macular degeneration. Am J Epidemiol. 2009;169(7):867–76.
Amirul Islam FM, Chong EW, Hodge AM, Guymer RH, Aung KZ, Makeyeva GA, et al. Dietary patterns and their associations with age-related macular degeneration: the Melbourne collaborative cohort study. Ophthalmology. 2014;121(7):1428–34.
Biesemeier A, Yoeruek E, Eibl O, Schraermeyer U. Iron accumulation in Bruch's membrane and melanosomes of donor eyes with age-related macular degeneration. Exp Eye Res. 2015;137:39–49.
Austin PC, Mamdani MM, Stukel TA, Anderson GM, Tu JV. The use of the propensity score for estimating treatment effects: administrative versus clinical data. Stat Med. 2005;24(10):1563–78.
Chaix B, Evans D, Merlo J, Suzuki E. Commentary: weighing up the dead and missing: reflections on inverse-probability weighting and principal stratification to address truncation by death. Epidemiology. 2012;23(1):129–31.
Hayati Rezvan P, Lee KJ, Simpson JA. Sensitivity analysis within multiple imputation framework using delta-adjustment: application to longitudinal study of Australian children. Longitudinal and Life Course Studies. 2018;9(3):20.
White IR, Carpenter J, Evans S, Schroter S. Eliciting and using expert opinions about dropout bias in randomized controlled trials. Clinical trials (London, England). 2007;4(2):125–39.