Semiparametric Regression Models for Repeated Measures of Mortal Cohorts with Non-monotone Missing Outcomes and Time-dependent Covariates

Overview

Journal Stat Med

Publisher Wiley

Specialty Public Health

Date 2010 Jun 22

PMID 20564729

Citations 10

Authors

Michelle Shardell

Gregory E Hicks

Ram R Miller

Jay Magaziner

Affiliations

Soon will be listed here.

Abstract

We propose a semiparametric marginal modeling approach for longitudinal analysis of cohorts with data missing due to death and non-response to estimate regression parameters interpreted as conditioned on being alive. Our proposed method accommodates outcomes and time-dependent covariates that are missing not at random with non-monotone missingness patterns via inverse-probability weighting. Missing covariates are replaced by consistent estimates derived from a simultaneously solved inverse-probability-weighted estimating equation. Thus, we utilize data points with the observed outcomes and missing covariates beyond the estimated weights while avoiding numerical methods to integrate over missing covariates. The approach is applied to a cohort of elderly female hip fracture patients to estimate the prevalence of walking disability over time as a function of body composition, inflammation, and age.

Citing Articles

The impact of financial incentives on mental health among adults in HIV care in Lake Zone, Tanzania.

Chitle P, Sheira L, Katabaro E, Winters S, Njau P, Sabasaba A AIDS. 2024; 38(11):1677-1685.

PMID: 38819837 PMC: 11293976. DOI: 10.1097/QAD.0000000000003948.

Sex Differences in Recovery Across Multiple Domains Among Older Adults With Hip Fracture.

Orwig D, Abraham D, Hochberg M, Gruber-Baldini A, Guralnik J, Cappola A J Gerontol A Biol Sci Med Sci. 2021; 77(7):1463-1471.

PMID: 34555162 PMC: 9255694. DOI: 10.1093/gerona/glab271.

Pain severity as a mediator of the association between depressive symptoms and physical performance in knee osteoarthritis.

Rathbun A, Shardell M, Stuart E, Yau M, Gallo J, Schuler M Osteoarthritis Cartilage. 2018; 26(11):1453-1460.

PMID: 30092262 PMC: 6397771. DOI: 10.1016/j.joca.2018.07.016.

Semi-parametric methods of handling missing data in mortal cohorts under non-ignorable missingness.

Wen L, Seaman S Biometrics. 2018; 74(4):1427-1437.

PMID: 29772074 PMC: 6481558. DOI: 10.1111/biom.12891.

Accommodating informative dropout and death: a joint modelling approach for longitudinal and semi-competing risks data.

Li Q, Su L J R Stat Soc Ser C Appl Stat. 2017; 67(1):145-163.

PMID: 29277843 PMC: 5741179. DOI: 10.1111/rssc.12210.

References

Dolan M, Hawkes W, Zimmerman S, Morrison R, Gruber-Baldini A, Hebel J . Delirium on hospital admission in aged hip fracture patients: prediction of mortality and 2-year functional outcomes. J Gerontol A Biol Sci Med Sci. 2000; 55(9):M527-34. DOI: 10.1093/gerona/55.9.m527. View

Egleston B, Scharfstein D, MacKenzie E . On estimation of the survivor average causal effect in observational studies when important confounders are missing due to death. Biometrics. 2008; 65(2):497-504. PMC: 2700847. DOI: 10.1111/j.1541-0420.2008.01111.x. View

Parzen M, Lipsitz S, Ibrahim J, Lipshultz S . A weighted estimating equation for linear regression with missing covariate data. Stat Med. 2002; 21(16):2421-36. DOI: 10.1002/sim.1195. View

Rotnitzky A, Scharfstein D, Su T, Robins J . Methods for conducting sensitivity analysis of trials with potentially nonignorable competing causes of censoring. Biometrics. 2001; 57(1):103-13. DOI: 10.1111/j.0006-341x.2001.00103.x. View

Ibrahim J, Chen M, Lipsitz S . Monte Carlo EM for missing covariates in parametric regression models. Biometrics. 2001; 55(2):591-6. DOI: 10.1111/j.0006-341x.1999.00591.x. View

Shardell M, Miller R . Weighted estimating equations for longitudinal studies with death and non-monotone missing time-dependent covariates and outcomes. Stat Med. 2007; 27(7):1008-25. PMC: 2792882. DOI: 10.1002/sim.2964. View

Kurland B, Heagerty P . Directly parameterized regression conditioning on being alive: analysis of longitudinal data truncated by deaths. Biostatistics. 2005; 6(2):241-58. DOI: 10.1093/biostatistics/kxi006. View

Minini P, Chavance M . Sensitivity analysis of longitudinal binary data with non-monotone missing values. Biostatistics. 2004; 5(4):531-44. DOI: 10.1093/biostatistics/kxh006. View

Xie F, Paik M . Generalized estimating equation model for binary outcomes with missing covariates. Biometrics. 1998; 53(4):1458-66. View

10.

Wilkins K, Fitzmaurice G . A marginalized pattern-mixture model for longitudinal binary data when nonresponse depends on unobserved responses. Biostatistics. 2006; 8(2):297-305. DOI: 10.1093/biostatistics/kxl010. View

11.

Miller R, Cappola A, Shardell M, Hawkes W, Yu-Yahiro J, Hebel J . Persistent changes in interleukin-6 and lower extremity function following hip fracture. J Gerontol A Biol Sci Med Sci. 2006; 61(10):1053-8. DOI: 10.1093/gerona/61.10.1053. View

12.

Dufouil C, Brayne C, Clayton D . Analysis of longitudinal studies with death and drop-out: a case study. Stat Med. 2004; 23(14):2215-26. DOI: 10.1002/sim.1821. View

13.

Miller R, Shardell M, Hicks G, Cappola A, Hawkes W, Yu-Yahiro J . Association between interleukin-6 and lower extremity function after hip fracture--the role of muscle mass and strength. J Am Geriatr Soc. 2008; 56(6):1050-6. PMC: 2801424. DOI: 10.1111/j.1532-5415.2008.01708.x. View