Stuart R. Lipsitz
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Stuart R. Lipsitz has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
Joseph Ibrahim
University of North Carolina at Chapel Hill , USA
4 shared publicationsGeert Molenberghs
Center for Statistics, Limburgs Universitair Centrum, Belgium
3 shared publications- Geert Molenberghs
Universiteit Hasselt, Hasselt, Belgium
3 shared publications Garrett M. Fitzmaurice
Harvard School of Public Health, Boston, Massachusetts, U.S.A.
2 shared publicationsMichael Parzen
University of Chicago , USA
2 shared publications- Sundar Natarajan
VA New York Harbor Healthcare System
2 shared publications
Latest Publications
- 31 August 2010
- Published by Elsevier in Journal of Cardiac Failure
- Vol. 16 (8) , S77
Prevalence of Functional Iron Deficiency in Adults With Heart Failure and Associations With Anemia, Inflammation, and Mortality
- 31 August 2010
- Published by Elsevier in Journal of Cardiac Failure
- Vol. 16 (8) , S77
Prevalence of Functional Iron Deficiency in Adults With Heart Failure and Associations With Anemia, Inflammation, and Mortality
- 31 July 2006
- Published by Elsevier in The Journal of Pediatrics
- Vol. 149 (1) , 146
Reply
- 1 April 2004
- Published by SAGE Publications in Statistical Modelling
- Vol. 4 (1) , 3-17
Protective estimator for linear regression with nonignorably missing Gaussian outcomes
We propose a method for estimating the regression parameters in a linear regression model for Gaussian data when the outcome variable is missing for some subjects and missingness is thought to be nonignorable. Throughout, we assume that missingness is restricted to the outcome variable and that the covariates are fully observed. Although maximum likelihood estimation of the regression parameters is possible once joint models for the outcome variable and the nonignorable missing data mechanism have been specified, these models are fundamentally nonidentifiable unless unverifiable modeling assumptions are imposed. In this paper, rather than explicitly modeling the nonignorable missingness mechanism, we consider the use of a ‘protective’ estimator of the regression parameters (Brown, 1990). To implement the proposed method, it is necessary to assume that the outcome variable and one of the covariates have an approximate bivariate normal distribution, conditional on the remaining covariates. In addition, it is assumed that the missing data mechanism is conditionally independent of this covariate, given the outcome variable and the remaining covariates; the latter is referred to as the ‘protective’ assumption. A method of moments approach is used to obtain the protective estimator of the regression parameters; the jackknife (Quenouille, 1956) is used to estimate the variance. The method is illustrated using data on the persistence of maternal smoking from the Six Cities Study of the health effects of air pollution (Ware et al., 1984). The results of a simulation study are presented that examine the magnitude of any finite sample bias.
- 1 April 2004
- Published by SAGE Publications in Statistical Modelling
- Vol. 4 (1) , 3-17
Protective estimator for linear regression with nonignorably missing Gaussian outcomes
We propose a method for estimating the regression parameters in a linear regression model for Gaussian data when the outcome variable is missing for some subjects and missingness is thought to be nonignorable. Throughout, we assume that missingness is restricted to the outcome variable and that the covariates are fully observed. Although maximum likelihood estimation of the regression parameters is possible once joint models for the outcome variable and the nonignorable missing data mechanism have been specified, these models are fundamentally nonidentifiable unless unverifiable modeling assumptions are imposed. In this paper, rather than explicitly modeling the nonignorable missingness mechanism, we consider the use of a ‘protective’ estimator of the regression parameters (Brown, 1990). To implement the proposed method, it is necessary to assume that the outcome variable and one of the covariates have an approximate bivariate normal distribution, conditional on the remaining covariates. In addition, it is assumed that the missing data mechanism is conditionally independent of this covariate, given the outcome variable and the remaining covariates; the latter is referred to as the ‘protective’ assumption. A method of moments approach is used to obtain the protective estimator of the regression parameters; the jackknife (Quenouille, 1956) is used to estimate the variance. The method is illustrated using data on the persistence of maternal smoking from the Six Cities Study of the health effects of air pollution (Ware et al., 1984). The results of a simulation study are presented that examine the magnitude of any finite sample bias.
- 19 March 2004
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series C: Applied Statistics
- Vol. 53 (2) , 279-292
Generalized Linear Models with a Coarsened Covariate
Summary: We consider generalized linear models with a coarsened covariate. The term ‘coarsened’ is used here to refer to the case where the exact value of the covariate of interest is not fully observed. Instead, only some set or grouping that contains the exact value is observed. In particular, we propose a likelihood-based method for estimating regression parameters in a generalized linear model relating the mean of the outcome to covariates. We outline Newton–Raphson and EM algorithms for obtaining maximum likelihood estimates of the regression parameters. We also compare and contrast this likelihood-based approach with two somewhat ad hoc procedures: a complete-case analysis in which individuals with coarsened data are excluded and estimation is based on the remaining ‘complete cases’, and a coarsened data regression model in which the covariate values for all the complete cases are coarsened and then included in a regression model relating the mean to the coarsened covariate. The methodology that is presented is motivated by coarsened data on the racial–ethnicity categorization of patients in the US’s National Ambulatory Medical Care Survey, a study to examine the medical care that is provided to a patient in a physician’s office. In this study, the outcome of interest is the level of tests (none, non-invasive tests or invasive tests) ordered for the patient at the doctor’s visit. One of the covariates of interest is the patient’s four-level discrete covariate comprised of four racial–ethnicity categories: white–Hispanic, white–non-Hispanic, African-American–Hispanic and African-American–non-Hispanic. However, of the 19 095 patients in the sample, 14 955 (or 78%) have the exact category of race–ethnicity recorded and 4140 (or 22%) have race–ethnicity coarsened. For the latter group of 4140 individuals, the ethnicity is not recorded, but we know that 3683 are white and 457 are African-American.
- 19 March 2004
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series C: Applied Statistics
- Vol. 53 (2) , 279-292
Generalized Linear Models with a Coarsened Covariate
Summary: We consider generalized linear models with a coarsened covariate. The term ‘coarsened’ is used here to refer to the case where the exact value of the covariate of interest is not fully observed. Instead, only some set or grouping that contains the exact value is observed. In particular, we propose a likelihood-based method for estimating regression parameters in a generalized linear model relating the mean of the outcome to covariates. We outline Newton–Raphson and EM algorithms for obtaining maximum likelihood estimates of the regression parameters. We also compare and contrast this likelihood-based approach with two somewhat ad hoc procedures: a complete-case analysis in which individuals with coarsened data are excluded and estimation is based on the remaining ‘complete cases’, and a coarsened data regression model in which the covariate values for all the complete cases are coarsened and then included in a regression model relating the mean to the coarsened covariate. The methodology that is presented is motivated by coarsened data on the racial–ethnicity categorization of patients in the US’s National Ambulatory Medical Care Survey, a study to examine the medical care that is provided to a patient in a physician’s office. In this study, the outcome of interest is the level of tests (none, non-invasive tests or invasive tests) ordered for the patient at the doctor’s visit. One of the covariates of interest is the patient’s four-level discrete covariate comprised of four racial–ethnicity categories: white–Hispanic, white–non-Hispanic, African-American–Hispanic and African-American–non-Hispanic. However, of the 19 095 patients in the sample, 14 955 (or 78%) have the exact category of race–ethnicity recorded and 4140 (or 22%) have race–ethnicity coarsened. For the latter group of 4140 individuals, the ethnicity is not recorded, but we know that 3683 are white and 457 are African-American.
- 1 September 2002
- Published by Oxford University Press (OUP) in Biometrics
- Vol. 58 (3) , 621-630
Parameter Estimation in Longitudinal Studies with Outcome-Dependent Follow-Up
Summary. In many observational studies, individuals are measured repeatedly over time, although not necessarily at a set of prespecified occasions. Instead, individuals may be measured at irregular intervals, with those having a history of poorer health outcomes being measured with somewhat greater frequency and regularity; i.e., those individuals with poorer health outcomes may have more frequent follow‐up measurements and the intervals between their repeated measurements may be shorter. In this article, we consider estimation of regression parameters in models for longitudinal data where the follow‐up times are not fixed by design but can depend on previous outcomes. In particular, we focus on general linear models for longitudinal data where the repeated measures are assumed to have a multivariate Gaussian distribution. We consider assumptions regarding the follow‐up time process that result in the likelihood function separating into two components: one for the follow‐up time process, the other for the outcome process. The practical implication of this separation is that the former process can be ignored when making likelihood‐based inferences about the latter; i.e., maximum likelihood (ML) estimation of the regression parameters relating the mean of the longitudinal outcomes to covariates does not require that a model for the distribution of follow‐up times be specified. As a result, standard statistical software, e.g., SAS PROC MIXED (Littell et al., 1996, SAS System for Mixed Models), can be used to analyze the data. However, we also demonstrate that misspecification of the model for the covariance among the repeated measures will, in general, result in regression parameter estimates that are biased. Furthermore, results of a simulation study indicate that the potential bias due to misspecification of the covariance can be quite considerable in this setting. Finally, we illustrate these results using data from a longitudinal observational study (Lipshultz et al., 1995, New England Journal of Medicine332, 1738–1743) that explored the cardiotoxic effects of doxorubicin chemotherapy for the treatment of acute lymphoblastic leukemia in children.
- 1 September 2002
- Published by Oxford University Press (OUP) in Biometrics
- Vol. 58 (3) , 621-630
Parameter Estimation in Longitudinal Studies with Outcome-Dependent Follow-Up
Summary. In many observational studies, individuals are measured repeatedly over time, although not necessarily at a set of prespecified occasions. Instead, individuals may be measured at irregular intervals, with those having a history of poorer health outcomes being measured with somewhat greater frequency and regularity; i.e., those individuals with poorer health outcomes may have more frequent follow‐up measurements and the intervals between their repeated measurements may be shorter. In this article, we consider estimation of regression parameters in models for longitudinal data where the follow‐up times are not fixed by design but can depend on previous outcomes. In particular, we focus on general linear models for longitudinal data where the repeated measures are assumed to have a multivariate Gaussian distribution. We consider assumptions regarding the follow‐up time process that result in the likelihood function separating into two components: one for the follow‐up time process, the other for the outcome process. The practical implication of this separation is that the former process can be ignored when making likelihood‐based inferences about the latter; i.e., maximum likelihood (ML) estimation of the regression parameters relating the mean of the longitudinal outcomes to covariates does not require that a model for the distribution of follow‐up times be specified. As a result, standard statistical software, e.g., SAS PROC MIXED (Littell et al., 1996, SAS System for Mixed Models), can be used to analyze the data. However, we also demonstrate that misspecification of the model for the covariance among the repeated measures will, in general, result in regression parameter estimates that are biased. Furthermore, results of a simulation study indicate that the potential bias due to misspecification of the covariance can be quite considerable in this setting. Finally, we illustrate these results using data from a longitudinal observational study (Lipshultz et al., 1995, New England Journal of Medicine332, 1738–1743) that explored the cardiotoxic effects of doxorubicin chemotherapy for the treatment of acute lymphoblastic leukemia in children.
- 1 March 2002
- Published by Wiley in Canadian Journal of Statistics
- Vol. 30 (1) , 55-78
Bayesian methods for generalized linear models with covariates missing at random
The authors propose methods for Bayesian inference for generalized linear models with missing covariate data. They specify a parametric distribution for the covariates that is written as a sequence of one‐dimensional conditional distributions. They propose an informative class of joint prior distributions for the regression coefficients and the parameters arising from the covariate distributions. They examine the properties of the proposed prior and resulting posterior distributions. They also present a Bayesian criterion for comparing various models, and a calibration is derived for it. A detailed simulation is conducted and two real data sets are examined to demonstrate the methodology.
- 1 March 2002
- Published by Wiley in Canadian Journal of Statistics
- Vol. 30 (1) , 55-78
Bayesian methods for generalized linear models with covariates missing at random
The authors propose methods for Bayesian inference for generalized linear models with missing covariate data. They specify a parametric distribution for the covariates that is written as a sequence of one‐dimensional conditional distributions. They propose an informative class of joint prior distributions for the regression coefficients and the parameters arising from the covariate distributions. They examine the properties of the proposed prior and resulting posterior distributions. They also present a Bayesian criterion for comparing various models, and a calibration is derived for it. A detailed simulation is conducted and two real data sets are examined to demonstrate the methodology.
- Open Access
- 1 September 2001
- Published by Oxford University Press (OUP) in Biostatistics
- Vol. 2 (3) , 295-307
Testing for bias in weighted estimating equations
It is very common in regression analysis to encounter incompletely observed covariate information. A recent approach to analyse such data is weighted estimating equations (Robins, J. M., Rotnitzky, A. and Zhao, L. P. (1994), JASA, 89, 846–866, and Zhao, L. P., Lipsitz, S. R. and Lew, D. (1996), Biometrics, 52, 1165–1182). With weighted estimating equations, the contribution to the estimating equation from a complete observation is weighted by the inverse of the probability of being observed. We propose a test statistic to assess if the weighted estimating equations produce biased estimates. Our test statistic is similar to the test statistic proposed by DuMouchel and Duncan (1983) for weighted least squares estimates for sample survey data. The method is illustrated using data from a randomized clinical trial on chemotherapy for multiple myeloma.
- 1 September 2001
- Published by Oxford University Press (OUP) in Biostatistics
- Vol. 2 (3) , 295-307
Open AccessTesting for bias in weighted estimating equations
It is very common in regression analysis to encounter incompletely observed covariate information. A recent approach to analyse such data is weighted estimating equations (Robins, J. M., Rotnitzky, A. and Zhao, L. P. (1994), JASA, 89, 846–866, and Zhao, L. P., Lipsitz, S. R. and Lew, D. (1996), Biometrics, 52, 1165–1182). With weighted estimating equations, the contribution to the estimating equation from a complete observation is weighted by the inverse of the probability of being observed. We propose a test statistic to assess if the weighted estimating equations produce biased estimates. Our test statistic is similar to the test statistic proposed by DuMouchel and Duncan (1983) for weighted least squares estimates for sample survey data. The method is illustrated using data from a randomized clinical trial on chemotherapy for multiple myeloma.
- 1 May 2001
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series A: Statistics in Society
- Vol. 164 (3) , 449-465
A Simple Method for Estimating a Regression Model for κ Between a Pair of Raters
Summary: Agreement studies commonly occur in medical research, for example, in the review of X-rays by radiologists, blood tests by a panel of pathologists and the evaluation of psychopathology by a panel of raters. In these studies, often two observers rate the same subject for some characteristic with a discrete number of levels. The κ-coefficient is a popular measure of agreement between the two raters. The κ-coefficient may depend on covariates, i.e. characteristics of the raters and/or the subjects being rated. Our research was motivated by two agreement problems. The first is a study of agreement between a pastor and a co-ordinator of Christian education on whether they feel that the congregation puts enough emphasis on encouraging members to work for social justice (yes versus no). We wish to model the κ-coefficient as a function of covariates such as political orientation (liberal versus conservative) of the pastor and co-ordinator. The second example is a spousal education study, in which we wish to model the κ-coefficient as a function of covariates such as the highest degree of the father of the wife and the father of the husband. We propose a simple method to estimate the regression model for the κ-coefficient, which consists of two logistic (or multinomial logistic) regressions and one linear regression for binary data. The estimates can be easily obtained in any generalized linear model software program.
- 1 May 2001
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series A: Statistics in Society
- Vol. 164 (3) , 449-465
A Simple Method for Estimating a Regression Model for κ Between a Pair of Raters
Summary: Agreement studies commonly occur in medical research, for example, in the review of X-rays by radiologists, blood tests by a panel of pathologists and the evaluation of psychopathology by a panel of raters. In these studies, often two observers rate the same subject for some characteristic with a discrete number of levels. The κ-coefficient is a popular measure of agreement between the two raters. The κ-coefficient may depend on covariates, i.e. characteristics of the raters and/or the subjects being rated. Our research was motivated by two agreement problems. The first is a study of agreement between a pastor and a co-ordinator of Christian education on whether they feel that the congregation puts enough emphasis on encouraging members to work for social justice (yes versus no). We wish to model the κ-coefficient as a function of covariates such as political orientation (liberal versus conservative) of the pastor and co-ordinator. The second example is a spousal education study, in which we wish to model the κ-coefficient as a function of covariates such as the highest degree of the father of the wife and the father of the husband. We propose a simple method to estimate the regression model for the κ-coefficient, which consists of two logistic (or multinomial logistic) regressions and one linear regression for binary data. The estimates can be easily obtained in any generalized linear model software program.
- 1 June 2000
- Published by Oxford University Press (OUP) in Biometrics
- Vol. 56 (2) , 528-536
GEE with Gaussian Estimation of the Correlations When Data Are Incomplete
Summary.This paper considers a modification of generalized estimating equations GEE) for handling missing binary response data. The proposed method uses Gaussian estimation of the correlation parameters, i.e., the estimating function that yields an estimate of the correlation parameters is obtained from the multivariate normal likelihood. The proposed method yields consistent estimates of the regression parameters when data are missing completely at random (MCAR). However, when data are missing at random (MAR), consistency may not hold. In a simulation study with repeated binary outcomes that are missing at random, the magnitude of the potential bias that can arise is examined. The results of the simulation study indicate that, when the working correlation matrix is correctly specified, the bias is almost negligible for the modified GEE. In the simulation study, the proposed modification of GEE is also compared to the standard GEE, multiple imputation, and weighted estimating equations approaches. Finally, the proposed method is illustrated using data from a longitudinal clinical trial comparing two therapeutic treatments, zidovudine (AZT) and didanosine (ddI), in patients with HIV.
- 1 June 2000
- Published by Oxford University Press (OUP) in Biometrics
- Vol. 56 (2) , 528-536
GEE with Gaussian Estimation of the Correlations When Data Are Incomplete
Summary.This paper considers a modification of generalized estimating equations GEE) for handling missing binary response data. The proposed method uses Gaussian estimation of the correlation parameters, i.e., the estimating function that yields an estimate of the correlation parameters is obtained from the multivariate normal likelihood. The proposed method yields consistent estimates of the regression parameters when data are missing completely at random (MCAR). However, when data are missing at random (MAR), consistency may not hold. In a simulation study with repeated binary outcomes that are missing at random, the magnitude of the potential bias that can arise is examined. The results of the simulation study indicate that, when the working correlation matrix is correctly specified, the bias is almost negligible for the modified GEE. In the simulation study, the proposed modification of GEE is also compared to the standard GEE, multiple imputation, and weighted estimating equations approaches. Finally, the proposed method is illustrated using data from a longitudinal clinical trial comparing two therapeutic treatments, zidovudine (AZT) and didanosine (ddI), in patients with HIV.
- 1 May 1999
- Published by JSTOR in The American Statistician
- Vol. 53 (2) , 110
Nonrandom Missingness in Categorical Data: Strengths and Limitations
There have recently been substantial developments in the analysis of incomplete data. Modeling tools are now available for nonrandom missingness and these metho...
- 1 May 1999
- Published by JSTOR in The American Statistician
- Vol. 53 (2) , 110
Nonrandom Missingness in Categorical Data: Strengths and Limitations
There have recently been substantial developments in the analysis of incomplete data. Modeling tools are now available for nonrandom missingness and these metho...