Tags

Type your tag names separated by a space and hit enter

A comparison of methods to estimate the survivor average causal effect in the presence of missing data: a simulation study.
BMC Med Res Methodol 2019; 19(1):223BM

Abstract

BACKGROUND

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.

METHODS

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.

RESULTS

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.

CONCLUSIONS

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.

Authors+Show Affiliations

Centre for Eye Research Australia, Royal Victorian Eye and Ear Hospital, Melbourne, Australia. myra.mcguinness@unimelb.edu.au. Centre for Epidemiology and Biostatistics, Melbourne School of Population and Global Health, University of Melbourne, Melbourne, Australia. myra.mcguinness@unimelb.edu.au.Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Victoria, 3010, Australia.Centre for Epidemiology and Biostatistics, Melbourne School of Population and Global Health, University of Melbourne, Melbourne, Australia.Centre for Eye Research Australia, Royal Victorian Eye and Ear Hospital, Melbourne, Australia. Ophthalmology, Department of Surgery, University of Melbourne, Melbourne, Australia.Department of Ophthalmology, University of Bonn, Bonn, Germany.Centre for Epidemiology and Biostatistics, Melbourne School of Population and Global Health, University of Melbourne, Melbourne, Australia. Cancer Epidemiology Centre, Cancer Council Victoria, Melbourne, Australia.

Pub Type(s)

Journal Article

Language

eng

PubMed ID

31795945

Citation

McGuinness, Myra B., et al. "A Comparison of Methods to Estimate the Survivor Average Causal Effect in the Presence of Missing Data: a Simulation Study." BMC Medical Research Methodology, vol. 19, no. 1, 2019, p. 223.
McGuinness MB, Kasza J, Karahalios A, et al. A comparison of methods to estimate the survivor average causal effect in the presence of missing data: a simulation study. BMC Med Res Methodol. 2019;19(1):223.
McGuinness, M. B., Kasza, J., Karahalios, A., Guymer, R. H., Finger, R. P., & Simpson, J. A. (2019). A comparison of methods to estimate the survivor average causal effect in the presence of missing data: a simulation study. BMC Medical Research Methodology, 19(1), p. 223. doi:10.1186/s12874-019-0874-x.
McGuinness MB, et al. A Comparison of Methods to Estimate the Survivor Average Causal Effect in the Presence of Missing Data: a Simulation Study. BMC Med Res Methodol. 2019 Dec 3;19(1):223. PubMed PMID: 31795945.
* Article titles in AMA citation format should be in sentence-case
TY - JOUR T1 - A comparison of methods to estimate the survivor average causal effect in the presence of missing data: a simulation study. AU - McGuinness,Myra B, AU - Kasza,Jessica, AU - Karahalios,Amalia, AU - Guymer,Robyn H, AU - Finger,Robert P, AU - Simpson,Julie A, Y1 - 2019/12/03/ PY - 2018/07/13/received PY - 2019/11/20/accepted PY - 2019/12/5/entrez PY - 2019/12/5/pubmed PY - 2019/12/5/medline KW - Causal inference KW - Death KW - Iron KW - Macular degeneration KW - Missing data KW - Principal stratification KW - Sensitivity analysis KW - Simulation study KW - Survival bias KW - Unmeasured confounding SP - 223 EP - 223 JF - BMC medical research methodology JO - BMC Med Res Methodol VL - 19 IS - 1 N2 - BACKGROUND: 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. METHODS: 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. RESULTS: 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. CONCLUSIONS: 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. SN - 1471-2288 UR - https://www.unboundmedicine.com/medline/citation/31795945/A_comparison_of_methods_to_estimate_the_survivor_average_causal_effect_in_the_presence_of_missing_data:_a_simulation_study L2 - https://bmcmedresmethodol.biomedcentral.com/articles/10.1186/s12874-019-0874-x DB - PRIME DP - Unbound Medicine ER -