Semiparametric log-linear regression for longitudinal measurements subject to outcome-dependent follow-up

• Published in: Journal of statistical planning and inference Vol. 138; no. 8; pp. 2450 - 2461
• Main Authors: Bůžková, Petra, Lumley, Thomas
• Format: Journal Article
• Language: English
• Published: Lausanne Elsevier B.V 01.08.2008; New York,NY Elsevier Science; Amsterdam
• Subjects: Distribution theory; Exact sciences and technology; General topics; Linear inference, regression; Log-linear regression; Longitudinal data; Marginal regression; Mathematics; Multivariate analysis; Outcome-dependent follow-up; Probability and statistics; Sciences and techniques of general use; Semiparametric regression; Statistics; Time-varying covariates; Log-linear regression; Time-varying covariates; Outcome-dependent follow-up; Marginal regression; Semiparametric regression; Longitudinal data; Biometrics; Statistical distribution; Statistical simulation; Asymptotic convergence; Medical science; Consistent estimator; Marginal distribution; Regression function; Approximation theory; Data analysis; Linear regression; Finite sample; Loglinear model; Statistical estimation; Statistical decision; Semiparametric method; Statistical method; Statistical regression; Regression model; Asymptotic normality; Estimating equation; Arbitrary function; Biased estimation
• ISSN: 0378-3758; 1873-1171
• DOI: 10.1016/j.jspi.2007.10.013

A common problem for longitudinal data analyses is that subjects follow-up is irregular, often related to the past outcome or other factors associated with the outcome measure that are not included in the regression model. Analyses unadjusted for outcome-dependent follow-up yield biased estimates. We propose a longitudinal data analysis that can provide consistent estimates in regression models that are subject to outcome-dependent follow-up. We focus on semiparametric marginal log-link regression with arbitrary unspecified baseline function. Based on estimating equations, the proposed class of estimators are root n consistent and asymptotically normal. We present simulation studies that assess the performance of the estimators under finite samples. We illustrate our approach using data from a health services research study.