Applied Longitudinal Analysis 2nd

Edition
Visit to download the full and correct content document:
https://ebookmass.com/product/applied-longitudinal-analysis-2nd-edition/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Threat assessment and risk analysis : an applied

approach 1st Edition Allen

https://ebookmass.com/product/threat-assessment-and-risk-
analysis-an-applied-approach-1st-edition-allen/

Data Science Applied to Sustainability Analysis

Jennifer B. Dunn

https://ebookmass.com/product/data-science-applied-to-
sustainability-analysis-jennifer-b-dunn/

Interaction between residential greenness and air

pollution mortality: analysis of the Chinese
Longitudinal Healthy Longevity Survey John S Ji

https://ebookmass.com/product/interaction-between-residential-
greenness-and-air-pollution-mortality-analysis-of-the-chinese-
longitudinal-healthy-longevity-survey-john-s-ji/

Applied Veterinary Clinical Nutrition, 2nd Edition

Andrea J. Fascetti

https://ebookmass.com/product/applied-veterinary-clinical-
nutrition-2nd-edition-andrea-j-fascetti/
(eTextbook PDF) for Applied Regression Analysis and
Other Multivariable Methods 5th Edition

https://ebookmass.com/product/etextbook-pdf-for-applied-
regression-analysis-and-other-multivariable-methods-5th-edition/

Machines and mechanisms : applied kinematic analysis

4th ed Edition David H Myszka

https://ebookmass.com/product/machines-and-mechanisms-applied-
kinematic-analysis-4th-ed-edition-david-h-myszka/

Primer of applied regression and analysis of variance

3rd Edition Glantz S.A.

https://ebookmass.com/product/primer-of-applied-regression-and-
analysis-of-variance-3rd-edition-glantz-s-a/

Applied Data Analysis and Modeling for Energy Engineers

and Scientists

https://ebookmass.com/product/applied-data-analysis-and-modeling-
for-energy-engineers-and-scientists/

Applied Analysis of Composite Media: Analytical and

Computational Approaches Dryga■

https://ebookmass.com/product/applied-analysis-of-composite-
media-analytical-and-computational-approaches-drygas/
Applied Longitudinal Analysis
 CONTENTS ix

5.8 Strengths and Weaknesses of Analyzing Response

Profiles 134
5.9 Computing: Analyzing Response Profiles
Using PROC MIXED in SAS 136
5.10 Further Reading 140
Problems 140

6 Modeling the Mean: Parametric Curves 143

6.1 Introduction 143
6.2 Polynomial Trends in Time 144
6.3 Linear Splines 149
6.4 General Linear Model Formulation 152
6.5 Case Studies 154
6.6 Computing: Fitting Parametric Curves
Using PROC MIXED in SAS 161
6.7 Further Reading 162
Problems 163

7 Modeling the Covariance 165

7.1 Introduction 165
7.2 Implications of Correlation among Longitudinal Data 166
7.3 Unstructured Covariance 168
7.4 Covariance Pattern Models 169
7.5 Choice among Covariance Pattern Models 175
7.6 Case Study 180
7.7 Discussion: Strengths and Weaknesses of Covariance
Pattern Models 183
7.8 Computing: Fitting Covariance Pattern Models
Using PROC MIXED in SAS 184
7.9 Further Reading 186
Problems 186

8 Linear Mixed Effects Models 189

8.1 Introduction 189
8.2 Linear Mixed Effects Models 194
8.3 Random Effects Covariance Structure 201
8.4 Two-Stage Random Effects Formulation 203
8.5 Choice among Random Effects Covariance Models 208
8.6 Prediction of Random Effects 209
X CONTENTS

8. 7 Prediction and Shrinkage* 211

8.8 Case Studies 213
8.9 Computing: Fitting Linear Mixed Effects Models
Using PROC MIXED in SAS 234
8.10 Further Reading 237
Problems 237

9 Fixed Effects versus Random Effects Models 241

9.1 Introduction 241
9.2 Linear Fixed Effects Models 241
9.3 Fixed Effects versus Random Effects:
Bias-Variance Trade-off 246
9.4 Resolving the Dilemma of Choosing Between
Fixed and Random Effects Models 249
9.5 Longitudinal and Cross-sectional Information 252
9.6 Case Study 255
9.7 Computing: Fitting Linear Fixed Effects Models
Using PROC GLM in SAS 258
9.8 Computing: Decomposition of Between-Subject and
Within-Subject Effects Using PROC MIXED in SAS 260
9.9 Further Reading 262
Problems 262

10 Residual Analyses and Diagnostics 265

10.1 Introduction 265
10.2 Residuals 265
10.3 Transformed Residuals 266
10.4 Aggregating Residuals 269
10.5 Semi-Variogram 272
10.6 Case Study 273
10.7 Summary 285
10.8 Further Reading 286
Problems 287
 CONTENTS xi

Part III Generalized Linear Models for Longitudinal Data

11 Review of Generalized Linear Models 291

11.1 Introduction 291
11.2 Salient Features of Generalized Linear Models 292
11.3 Illustrative Examples 297
11.4 Ordinal Regression Models 310
11.5 Overdispersion 319
11.6 Computing: Fitting Generalized Linear Mode ls
Using PROC GENMOD in SAS 324
11.7 Overview of Generalized Linear Models* 327
11.8 Further Reading 335
Problems 336

12 Marginal Models: Introduction and Overview 341

12.1 Introduction 341
12.2 Marginal Models for Longitudinal Data 342
12.3 Illustrative Examples of Marginal Models 346
12.4 Distributional Assumptions for Marginal Models* 351
12.5 Further Reading 352

13 Marginal Models: Generalized Estimating

Equations (GEE) 353
13.1 Introduction 353
13.2 Estimation of Marginal Models: Generalized
Estimating Equations 354
13.3 Residual Analyses and Diagnostics 36 I
13.4 Case Studies 364
13.5 Marginal Models and Time- Varying Covariates 381
13.6 Computing: Generalized Estimating Equations
Using PROC GENMOD in SAS 385
13.7 Further Reading 390
Problems 39 l

14 Generalized Linear Mixed Effects Models 395

14.1 Introduction 395
14.2 Incorporating Random Effects in Generalized
Linear Models 396
xii CONTENTS

14.3 Interpretation of Regression Parameters 402

14.4 Overdispersion 409
14.5 Estimation and Inference 410
14.6 A Note on Conditional Maximum Likelihood 412
14.7 Case Studies 414
14.8 Computing: Fitting Generalized Linear Mixed Models
Using PROC GLIMMIX in SAS 429
14.9 Further Reading 433
Problems 434

15 Generalized Linear Mixed Effects Models: Approximate

Methods of Estimation 441
15.1 Introduction 441
15.2 Penalized Quasi-Likelihood 443
15.3 Marginal Quasi-Likelihood 445
15.4 Cautionary Remarks on the Use of PQL and MQL 446
15.5 Case Studies 452
15.6 Computing: Fitting GLMMs Using PROC GLIMMIX
in SAS 459
15.7 Basis of PQL and MQL Approximations* 466
15.8 Further Reading 470
Problems 471

16 Contrasting Marginal and Mixed Effects Models 473

16.1 Introduction 473
16.2 Linear Models: A Special Case 473
16.3 Generalized Linear Models 474
16.4 Simple Numerical Illustration 479
16.5 Case Study 480
16.6 Conclusion 484
16.7 Further Reading 486
 CONTENTS xiii

Part IV Missing Data and Dropout

17 Missing Data and Dropout: Overview of Concepts

and Methods 489
17.1 Introduction 489
17.2 Hierarchy of Missing Data Mechanisms 491
17.3 Implications for Longitudinal Analysis 499
17.4 Dropout 500
17.5 Common Approaches for Handling Dropout 506
17.6 Bias of Last Value Carried Forward Imputation* 511
17.7 Further Reading 513

18 Missing Data and Dropout: Multiple Imputation

and Weighting Methods 515
18.1 Introduction 515
18.2 Multiple Imputation 516
18.3 Inverse Probability Weighted Methods 526
18.4 Case Studies 531
18.5 "Sandwich" Variance Estimator Adjusting for
Estimation of Weights* 541
18.6 Computing: Multiple Imputation Using PROC MI
in SAS 542
18.7 Computing: Inverse Probability Weighted (/PW)
Methods in SAS 547
18.8 Further Reading 550

Part V Advanced Topics for Longitudinal and Clustered Data

19 Smoothing Longitudinal Data: Semiparametric Regression

Models 553
19.1 Introduction 553
19.2 Penalized Splines for a Univariate Response 554
19.3 Case Study 558
19.4 Penalized Splines for Longitudinal Data 563
19.5 Case Study 565
xiv CONTENTS

19.6 Fitting Smooth Curves to Individual

Longitudinal Data 570
19.7 Case Study 572
19.8 Computing: Fitting Smooth Curves
Using PROC MIXED in SAS 576
19.9 Further Reading 579

20 Sample Size and Power 581

20.1 Introduction 581
20.2 Sample Size for a Univariate Continuous Response 582
20.3 Sample Size for a Longitudinal ContinuousResponse 584
20.4 Sample Size for a Longitudinal Binary Response 598
20.5 Summary 604
20.6 Computing: Sample Size Calculation
Using Pseudo-Data 605
20. 7 Further Reading 609

21 Repeated Measures and Related Designs 611

21.1 Introduction 611
21.2 Repeated Measures Designs 612
21.3 Multiple Source Data 616
21.4 Case Study 1: Repeated Measures Experiment 617
21.5 Case Study 2: Multiple Source Data 620
21.6 Summary 625
21.7 Further Reading 626

22 Multilevel Models 627

22.1 Introduction 627
22.2 Multilevel Data 628
22.3 Multilevel Linear Models 630
22.4 Multilevel Generalized Linear Models 641
22.5 Summary 651
22.6 Further Reading 652
 CONTENTS xv

Appendix A Gentle Introduction to Vectors and Matrices 655

Appendix B Properties of Expectations and Variances 665

Appendix C Critical Points for a 50:50 Mixture

of Chi-Squared Distributions 669

References 671

Index 695
 Preface

The first edition of Applied Longitudinal Analysis was designed to serve as a textbook
for a course on modern statistical methods for longitudinal data analysis, and subse-
quently, as a reference resource for students and researchers. The book was targeted
at a broad audience: graduate students in statistics, statisticians working in the health
sciences, pharmaceutical industry, and governmental health-related agencies, as well
as researchers and graduate students from a variety of substantive fields. In the seven
years that have elapsed since publication of the first edition, Applied Longitudinal
Analysis has been used extensively in university classrooms throughout the United
States and abroad. We are grateful to many colleagues, course instructors, students,
and readers who have offered constructive suggestions on how the book could be
improved. This feedback has been invaluable and helped shape the content of the
second edition.
The feedback we received has encouraged us to retain the general structure and
format of the first edition while taking the opportunity to introduce a number of new
and important topics. Although there is much new material in this second edition, the
principles that guided us in writing the first edition have not changed. Our primary
goal is to present a rigorous and comprehensive description of modern statistical meth-
ods for the analysis of longitudinal data that is accessible to a wide range of readers.
A strong emphasis is placed on the application of these methods to longitudinal data
and the interpretation of results. Although the methods are presented in the setting of
numerous applications to actual data sets drawn from studies in health-related fields,
reflecting our own research interests in the health sciences, they apply equally to other
areas of application, for example, education, psychology, and other branches of the
behavioral and social sciences.
xvii
xviii PREFACE

How does this edition differ from its predecessor? The major changes in this
edition have resulted from the addition of six new chapters:
1. A chapter (Chapter 9) on "fixed effects models," in which subject-specific effects
are treated as fixed rather than random, has been added. This chapter complements
the existing chapter on mixed effects models (Chapter 8) and includes a discussion
of the relative advantages of these two classes of models.
2. In the first edition, a single chapter was devoted to marginal models and generalized
estimating equations (GEE) that focused exclusively on binary and count data. We
now devote two chapters (Chapters 12 and 13) to marginal models and GEE, with
new material on models for ordinal responses, residual diagnostics, and issues
that arise when modeling time-varying covariates.
3. A chapter (Chapter 15) on approximate methods for generalized linear mixed
effects models discusses penalized quasi-likelihood (PQL) and marginal quasi-
likelihood (MQL) methods. We highlight settings where these approximations
are unlikely to be accurate and can yield biased estimates of effects.
4. A second chapter (Chapter 18) on missing data and dropout, focusing on multiple
imputation and inverse probability weighting (IPW) methods, has been added. To
give greater prominence to methods for accounting for missing data and dropout
in longitudinal analyses, the two companion chapters (Chapters 17 and 18) now
appear before the Advanced Topics part of the book.
5. A chapter (Chapter 19) on smoothing longitudinal data has been added to the
Advanced Topics. This chapter focuses on the connection between penalized
splines and linear mixed effects models.
6. A chapter on sample size and power (Chapter 20) has been added to the Advanced
Topics. This chapter considers issues of sample size, power, number of repeated
measurements, and study duration for longitudinal study designs.
In addition the chapter on residual analyses and diagnostics (Chapter 10) has been re-
vised to include material on recently developed model-checking techniques based on
cumulative sums of residuals and the chapters that review generalized linear models
(Chapter 11) and generalized linear mixed effects models (Chapter 14) have been up-
dated to include new material on models for ordinal data and on methods for handling
overdispersion. Finally, extra problem sets have been added to many of the chapters.
As in the first edition, the prerequisites for a course based on this book are an in-
troductory course in statistics and a strong background in regression analysis. Some
previous exposure to generalized linear models (e.g., logistic regression) would be
helpful, although these models are reviewed in detail in the text. An understanding of
matrix algebra or calculus is not assumed. Although we do not assume a high level of
mathematical preparation, we have written this book for the motivated reader who is
willing to consider mathematical ideas. The more technical or mathematical sections
of the book are signposted with asterisks and may be omitted at first reading without
loss of continuity.
 PREFACE xix

The methods described in this book require the use of appropriate statistical
software. As before, we include illustrative SAS commands for performing the analy-
ses presented throughout the text at the end of many chapters, with basic descriptions
of their usage. Because many of the analyses we discuss can be performed using
alternative software packages (e.g., R, S-Plus, Stata, and SPSS), this book can be sup-
plemented with any one of them. Readers are encouraged to perform and verify the
results of analyses using statistical software of their choice. Programming statements
and computer output for selected examples, prepared using SAS, Stata, and R, can
be downloaded from the website: www.biostat.harvard.edu/~fi tzmaur /ala2e.
Because statistical software is constantly evolving, we will endeavor to update the
website as new procedures become available in the major statistical software pack-
ages. The thirty-two real data sets used throughout the text and problem sets to
illustrate the applications of longitudinal methods also can be downloaded from the
website.
We hope this second edition of Applied Longitudinal Analysis provides a broader
foundation in modem methods for the analysis of longitudinal data and will prove a
worthy successor to the first edition. The original impetus for writing this book arose
from teaching a graduate-level course on "Applied Longitudinal Analysis" at the
Harvard School of Public Health. We are especially grateful to the students who have
participated in the course since its inception almost twenty years ago; we have learned
much from these extraordinary students. The collection of individuals who gave us
useful feedback on the first edition is far too long to list. However, we would like to
thank the many friends and colleagues who have helped us with this project. A special
word of thanks to Amy Herring and Russell Localio. We thank Amy for her many
helpful and constructive suggestions on how the book could be improved. We thank
Russell for reading a draft of the new chapters and for providing invaluable feedback
and suggestions that improved their content. Thanks also to Nick Horton, Stu Lipsitz,
and Caitlin Ravichandran for their helpful suggestions and insightful comments on
several chapters. Finally, we thank Steve Quigley and Susanne Steitz-Filler of Wiley,
for their advice and encouragement during all stages of this project.

GARRETT M. FITZMAURICE
NANM. LAIRD
JAMES H. WARE

Boston, Massachusetts
May, 2011
 Preface to First Edition

Our goal in writing this book is to provide a rigorous and systematic description of
modem methods for analyzing data from longitudinal studies. In recent years there
have been remarkable developments in methods for longitudinal analysis. Despite
these important advances, the methods have been somewhat slow to move into the
mainstream. Applied Longitudinal Analysis bridges the gap between theory and
application by presenting a comprehensive account of these methods in a way that is
accessible to a wide range of readers.
The impetus for this book arose from teaching a graduate-level course on "Applied
Longitudinal Analysis" at the Harvard School of Public Health. As course instructors,
we were frustrated by the lack of a suitable textbook that adequately covered modem
statistical methods for longitudinal analysis at a level accessible to a broad audience
of researchers and graduate students in the health and medical sciences. We envision
this book as a textbook for such a course and, subsequently, as a reference resource
for researchers and graduate students. It is also suitable for graduate students in
statistics and for statisticians already working in the health sciences, governmental
health-related agencies, and the pharmaceutical industry. It is intended to allow a
diverse group of statisticians, researchers, and graduate students in substantive fields
to master modem methods for longitudinal data analysis.
The scope of this book is broad, covering methods for the analysis of diverse
types of longitudinal data arising in the health sciences. The methods are pre-
sented in the setting of numerous applications to real data sets. Our main em-
phasis is on the practical rather than the theoretical aspects of longitudinal anal-
ysis. Twenty-five real data sets, drawn from studies in health-related fields, are

xxi
xxii PREFACE TO FIRST EDITION

used throughout the text and problem sets to illustrate the applications of longitu-
dinal methods. These data sets can be downloaded from the website for the book:
www.biostat.harvard.edu/~fitzmaur/ala. Although the methods are applied
to data sets drawn from the health sciences, they apply equally to other areas of ap-
plication, for example, education, psychology, and other branches of the behavioral
and social sciences.
Because longitudinal data are a special case of clustered data, albeit with a natural
ordering of the measurements within a cluster, we include also a description of modem
methods for analyzing clustered data, more broadly defined. Indeed, one of our
goals is to demonstrate that methods for longitudinal analysis are, more or less,
special cases of more general regression methods for clustered data. As a result a
comprehensive understanding of longitudinal data analysis provides the basis for a
broader understanding of methods for analyzing the wide range of clustered data that
commonly arises in studies in the biomedical and health sciences.
The prerequisites for a course based on this book are an introductory course in
statistics and a strong background in regression analysis. Some previous exposure to
generalized linear models (e.g., logistic regression) would be helpful, although these
models are reviewed in the text. An understanding of matrix algebra or calculus is
not assumed; the reader will be gently introduced to only those aspects of vector and
matrix notation necessary for understanding the matrix representation of regression
models for longitudinal data. Because vectors and matrices are used to simplify
notation, the reader is required to attain some basic facility with the addition and
multiplication of vectors and matrices. Although we do not assume a high level
of mathematical preparation, a willingness to read and consider mathematical ideas
is required. More technical or mathematical sections of the book are marked with
asterisks and may be omitted at first reading without loss of continuity.
To use the methods described in this book, appropriate statistical software is re-
quired. In general, the methods available via commercially available software lag
behind the recent advances in statistical methods; longitudinal data analysis is not
exceptional in this regard. Recently the introduction of new programs for analyzing
multivariate and longitudinal data has made these methods far more accessible to
practitioners and students. We use SAS, which is widely available, to perform the
analyses presented throughout the text. Illustrative SAS commands are included at
the end of many of the chapters, with basic descriptions of their usage. Program-
ming statements and computer output for the examples, prepared using SAS, can be
downloaded from the website: www.biostat.harvard.edu/~fitzmaur/ala. We
selected SAS because all of the analyses we discuss can be performed using its pro-
cedures. Many of the methods can be carried out using alternative software packages
(e.g., S-Plus and Stata) or special purpose programs (e.g., BMDP5-V) and this book
can be supplemented with any one of them. Readers are encouraged to perform and
verify the results of analyses using software of their choice. Because statistical soft-
ware is constantly evolving, we anticipate that all of the methods we discuss will soon
be available within most of the major statistical packages.
Throughout the text references have been kept to an absolute minimum. Instead,
at the end of each chapter we include suggestions for further readings that provide
 PREFACE TO FIRST EDITION xxiii

more in-depth coverage of certain topics. We also include "bibliographic notes" that
highlight key references in the mainstream statistical literature. Although many of
our readers may find the latter references to be too technical, they are included to give
due credit to those who have contributed to the statistical methods described in each
chapter.
Finally, we would like to thank the many friends and colleagues who have helped us
to write this book. A special word of thanks to Misha Salganik, for preparation of the
diagrams and many helpful suggestions for improvement of graphical displays. We
are especially grateful to Joe Hogan and Russell Localio, for reading a first draft and
providing invaluable feedback, comments, and suggestions that improved the book.
We would also like to thank Rino Bellocco, Brent Coull, Nick Horton, Sharon-Lise
Normand, Misha Salganik, Judy Singer, S. V. Subramanian, and Florin Vaida, for their
insightful comments on several chapters. We are grateful to the students who have
participated in the course on "Applied Longitudinal Analysis" at the Harvard School
of Public Health since its inception; they have provided the impetus and motivation
for writing this book. We gratefully acknowledge support from grant GM 29745 from
the National Institutes of Health. The first author gratefully acknowledges support
from the Junior Faculty Sabbatical Program at the Harvard School of Public Health;
the support provided by a sabbatical created a unique opportunity to begin writing
this book. Last, but not least, we thank Steve Quigley and Susanne Steitz of Wiley,
for their advice and encouragement during all stages of this project.

GARRETT M. FITZMAURICE
NANM. LAIRD
JAMES H. WARE
Boston, Massachusetts
March, 2004
 Acknowledgments

Throughout this book we have used data sets drawn from published studies in health-
related fields to exemplify important concepts in the analysis of longitudinal and
clustered data. We are grateful to the following investigators for sharing their data with
us: Graham Bentham, Doug Dockery, Brian Flay, Robert Greenberg, Keith Henry,
Aviva Must, Elena Naumova, George Rhoads, Jan Schouten, Linda Van Marter, and
Gwen Zahner.
We also thank the following publishers for permission to reproduce published data
sets in print and electronic format: The American Statistical Association, Blackwell
Publishing, Brooks/Cole (a division of Thomson Learning), CRC Press, Elsevier,
Iowa State Press, Oxford University Press, and SAS Institute, Inc.
Finally, in all data sets used throughout this book, the original subject identification
(ID) numbers have been deleted and replaced with new subject ID numbers, to ensure
that the data sets cannot be linked to the original records.

XXV
 Part/

Introduction to Longitudinal
and Clustered Data
 1
Longitudinal and
Clustered Data

1.1 INTRODUCTION

Research on statistical methods for the design and analysis of human investigations
expanded explosively in the second half of the twentieth century. Beginning in the
early 1950s, the U.S. government shifted a substantial part of its research support from
military to biomedical research. The legislative foundation for the modem National
Institutes of Health (NIH), the Public Health Service Act, was passed in 1944 and
NIH grew rapidly throughout the 1950s and 1960s. During these "golden years" of
NIH expansion, the entire NIH budget grew from $8 million in 1947 to more than
$1 billion in 1966. The NIH sponsored many of the important epidemiologic studies
and clinical trials of that period, including the influential Framingham Heart Study
(Dawber et al., 1951; Dawber, 1980).
The typical focus of these early studies was morbidity and, especially, mortality.
Investigators sought to identify the causes of early death and to evaluate the effective-
ness of treatments for delaying death and morbidity. In the Framingham Heart Study,
participants were seen at two-year intervals. Survival outcomes during successive
two-year periods were treated as independent events and modeled using multiple lo-
gistic regression. The successful use of multiple logistic regression in this setting,
and the recognition that it could be applied to case-control data, led to widespread use
of this methodology beginning in the 1960s. The analysis of time-to-event data was
revolutionized by the seminal 1972 paper of D.R. Cox, describing the proportional
hazards model (Cox, 1972). This paper was followed by a rich and important body
of work that established the conceptual basis and the computational tools for modem
survival analysis.

Applied Longitudinal Analysis, Second Edition. By Garrett M. Fitzmaurice, Nan M. Laird, 1

and James H. Wate. Copyright© 201 I John Wiley & Sons, Inc.
2 LONGITUDINAL AND CLUSTERED DATA

Although the design of the Framingham Heart Study and other cohort studies called
for periodic measurement of the patient characteristics thought to be determinants of
chronic disease, interest in the levels and patterns of change of those characteristics
over time was initially limited. As the research advanced, however, investigators
began to ask questions about the behavior of these risk factors. In the Framingham
Heart Study, for example, investigators began to ask whether blood pressure levels in
childhood were predictive of hypertension in adult life. In the Coronary Artery Risk
Development in Young Adults (CARDIA) Study, investigators sought to identify the
determinants of the transition from normotensive or normocholesterolemic status in
early adult life to hypertension and hypercholesterolemia in middle age (Friedman
et al., 1988). In the treatment of arthritis, asthma, and other diseases that are not
typically life-threatening, investigators began to study the effects of treatments on the
level and change over time in measures of severity of disease. Similar questions were
being posed in every disease setting. Investigators began to follow populations of
all ages over time, both in observational studies and clinical trials, to understand the
development and persistence of disease and to identify factors that alter the course of
disease development.
This interest in the temporal patterns of change in human characteristics came at
a period when advances in computing power made new and more computationally
intensive approaches to statistical analysis available at the desktop. Thus, in the early
1980s, Laird and Ware proposed the use of the EM algorithm to fit a class of linear
mixed effects models appropriate for the analysis of repeated measurements (Laird
and Ware, 1982); Jennrich and Schluchter (1986) proposed a variety of alternative
algorithms, including Fisher-scoring and Newton-Raphson algorithms. Later in the
decade, Liang and Zeger introduced the generalized estimating equations in the bio-
statistical literature and proposed a family of generalized linear models for fitting
repeated observations of binary and counted data (Liang and Zeger, 1986; Zeger
and Liang, 1986). Many other investigators writing in the biomedical, educational,
and psychometric literature contributed to the rapid development of methodology for
the analysis of these "longitudinal" data. The past 30 years have seen considerable
progress in the development of statistical methods for the analysis of longitudinal
data. Despite these important advances, methods for the analysis of longitudinal data
have been somewhat slow to move into the mainstream. This book bridges the gap be-
tween theory and application by presenting a comprehensive description of methods
for the analysis of longitudinal data accessible to a broad range of readers.

1.2 LONGITUDINAL AND CLUSTERED DATA

The defining feature of longitudinal studies is that measurements of the same individ-
uals are taken repeatedly through time, thereby allowing the direct study of change
over time. The primary goal of a longitudinal study is to characterize the change in
response over time and the factors that influence change. With repeated measures
on individuals, one can capture within-individual change. Indeed, the assessment of
within-subject changes in the response over time can only be achieved within a Ion-
 LONGITUDINAL AND CLUSTERED DATA 3

gitudinal study design. For example, in a cross-sectional study, where the response is
measured at a single occasion, one can only obtain estimates of between-individual
differences in the response. That is, a cross-sectional study may allow comparisons
among sub-populations that happen to differ in age, but it does not provide any infor-
mation about how individuals change during the corresponding period.
To highlight this important distinction between cross-sectional and longitudinal
study designs, consider the following simple example. Body fatness in girls is thought
to increase just before or around menarche, leveling off approximately 4 years after
menarche. Suppose that investigators are interested in determining the increase in
body fatness in girls after menarche. In a cross-sectional study design, investigators
might obtain measurements of percent body fat on two separate groups of girls: a
group of 10-year-old girls (a pre-menarcheal cohort) and a group of 15-year-old girls
(a post-menarcheal cohort). In this cross-sectional study design, direct comparison of
the average percent body fat in the two groups of girls can be made using a two-sample
(unpaired) t-test. This comparison does not provide an estimate of the change in body
fatness as girls age from 10 to 15 years. The effect of growth or aging, an inherently
within-individual effect, simply cannot be estimated from a cross-sectional study that
does not obtain measures of how individuals change with time. In a cross-sectional
study the effect of aging is potentially confounded with possible cohort effects. Put
in a slightly different way, there are many characteristics that differentiate girls in
these two different age groups that could distort the relationship between age and
body fatness. On the other hand, a longitudinal study that measures a single cohort
of girls at both ages 10 and 15 can provide a valid estimate of the change in body
fatness as girls age. In the longitudinal study the analysis is based on a paired t-test,
using the difference or change in percent body fat within each girl as the outcome
variable. This within-individual comparison provides a valid estimate of the change
in body fatness as girls age from 10 to 15 years. Moreover, since each girl acts as
her own control, changes in percent body fat throughout the duration of the study are
estimated free of any between-individual variation in body fatness.
A distinctive feature of longitudinal data is that they are clustered. In longitudi-
nal studies the clusters are composed of the repeated measurements obtained from a
single individual at different occasions. Observations within a cluster will typically
exhibit positive correlation, and this correlation must be accounted for in the analysis.
Longitudinal data also have a temporal order; the first measurement within a clus-
ter necessarily comes before the second measurement, and so on. The ordering of
the repeated measures has important implications for analysis. There are, however,
many studies in the health sciences that are not longitudinal in this sense but which
give rise to data that are clustered or cluster-correlated. For example, clustered data
commonly arise when intact groups are randomized to health interventions or when
naturally occurring groups in the population are randomly sampled. An example of
the former is group-randomized trials. In a group-randomized trial, also known as
a cluster-randomized trial, groups of individuals, rather than each individual alone,
are randomized to different treatments or health interventions. Data on the health
outcomes of interest are obtained on all individuals within a group. Alternatively,
clustered data can arise from random sampling of naturally occurring groups in the
4 LONGITUDINAL AND CLUSTERED DATA

population. Families, households, hospital wards, medical practices, neighborhoods,

and schools are all instances of naturally occurring clusters in the population that
might be the primary sampling units in a study. Finally, clustered data can arise
when data on the health outcome of interest are simultaneously obtained either from
multiple raters or from different measurement instruments.
In all these examples of clustered data, we might reasonably expect that measure-
ments on units within a cluster are more similar than the measurements on units in
different clusters. The degree of clustering can be expressed in terms of correlation
among the measurements on units within the same cluster. This correlation invalidates
the crucial assumption of independence that is the cornerstone of so many standard
statistical techniques. Instead, statistical models for clustered data must explicitly
describe and account for this correlation. Because longitudinal data are a special
case of clustered data, albeit with a natural ordering of the measurements within a
cluster, this book includes a description of modem methods of analysis for clustered
data, more broadly defined. Indeed, one of the goals of this book is to demonstrate
that methods for the analysis of longitudinal data are, more or less, special cases of
more general regression methods for clustered data. As a result a comprehensive
understanding of methods for the analysis of longitudinal data provides the basis for
a broader understanding of methods for analyzing the wide range of clustered data
that commonly arises in studies in the biomedical and health sciences.
The examples described above consider only a single level of clustering, for ex-
ample, repeated measurements on individuals. More recently investigators have de-
veloped methodology for the analysis of multilevel data, in which observations may
be clustered at more than one level. For example, the data may consist of repeated
measurements on patients clustered by clinic. Alternatively, the data may consist of
observations on children nested within classrooms, nested within schools. Although
the analysis of multilevel data is not the primary focus of this book, multilevel data
are discussed in Chapter 22.
Interest in the analysis oflongitudinal and multilevel data continues to grow. New
and more flexible models have been developed and advances in computation, such
as Markov chain Monte Carlo (MCMC) methods, have allowed greater flexibility
in model specification. Moreover, improvements in statistical software packages,
especially SAS, Stata, SPSS, R, and S-Plus, have made these models much more
accessible for use in routine data analysis. Despite these advances, however, methods
for the analysis of longitudinal data are not widely used and are seen to be accessible
only to statisticians with specialized expertise.
We believe that the methodology for the analysis of longitudinal data can be much
more widely understood and applied. It is our hope that this book will help make
that possible. It provides a comprehensive introduction to methods for the analysis
of longitudinal data, written for a reader with a basic knowledge of statistics and a
strong background in regression analysis. The book does not require a high level
of mathematical preparation but does assume a willingness to read and consider
mathematical ideas.
Another random document with
no related content on Scribd:
 DEATH OF HENRY W. GRADY.

From the “Savannah News.”

Georgia mourns for one of her most distinguished sons. Henry W.
Grady, who, a week ago last Thursday, held entranced, and at times
moved to enthusiastic applause, by his eloquence, an audience
composed of Boston’s prominent citizens, and whose name on the
following day was on the lips of millions of people, is cold in death in
his Atlanta home. He died before he had reached the meridian of life
or the zenith of his fame. His mind was steadily broadening, and he
was constantly giving evidence of the possession of still greater
ability than he had yet displayed. In his Boston speech he handled
the race question in a way that showed that he was not a mere
rhetorician, but a genuine orator, who could direct the minds of men
as well as touch their hearts and dazzle their imaginations. Had he
lived, he would have won a name that would have had a permanent
place in the history of his country. As it is, he will be remembered as
a brilliant young man whom death claimed before he had time to
show that he was fully capable of meeting the expectations which
were entertained with regard to him.
Mr. Grady was full of resources and a tireless worker. He entered
the profession of journalism very early in life, and such was the
energy and intensity with which he devoted himself to it, that even if
he had not possessed extraordinary talents, he could hardly have
failed to succeed; but, having a special fitness for his work and ability
of a very high order, it was not strange that he quickly made a
reputation that was not confined by the lines of his State.
Mr. Grady was never satisfied with what he had accomplished. He
felt that he was capable of still better things, and he strove constantly
to reach a higher mark of excellence. No sooner was he done with
one undertaking than his busy brain was engaged with another; and
it can be said of him that his aims were not selfish ones. No doubt he
had the ambitions which every man of marked ability has, but the
good of others entered largely into his thoughts and plans. Atlanta
owes to his memory a debt she can never repay. During all the time
he was a resident within her limits he kept her interests steadily in
view. He contributed to her prosperity in a hundred ways, and when
her people were lukewarm in enterprises which he or others
suggested, he pointed out to them their duty, and urged them to
perform it so eloquently and strongly that they fell into line and won
success when many thought success was impossible.
Mr. Grady was not apparently anxious to accumulate wealth.
Money did not remain with him long. His purse was always open to
his friends, and those who had claims never had to ask him twice for
assistance when he was able to render it. Doubtless there are
hundreds in Atlanta who are able to speak from personal knowledge
of his free-handed liberality.
Mr. Grady never held public office. Had he lived, however, it is
probable that he would have entered the political arena. He was
gradually being drawn in that direction, and during the last two or
three years his name was frequently mentioned in connection with
the offices of Senator and Governor. His triumphs were won as a
journalist and an orator. In the latter character he first achieved a
national reputation at the dinner of the New England Society in 1886.
Georgians loved Mr. Grady and were proud of him. The death of
very few other men could have so filled their hearts with sorrow.
 HENRY W. GRADY DEAD.

From the “Albany News and Advertiser.”

The flash that announced over the wires the death of Henry W.
Grady shocked the country, for it was a national calamity.
It is seldom that a people are called upon in so short a space of
time to mourn the loss of two such men as Jefferson Davis and
Henry W. Grady. The first was a blow for which we were prepared,
for like ripened grain, Mr. Davis fell, full of years and honor, before
the scythe of the reaper; but the death of Mr. Grady comes to us as a
sorrow with all the force of a painful surprise. He was cut down in the
bloom of a robust physical manhood, in the full enjoyment of his
magnificent mental powers by which he had just ascended to the
very pinnacle of fame. The eyes of the country were fixed upon him,
the son of the South, whose transcendent genius inspired the hope
of the blessed realization of promises with which his brief but brilliant
career was so full. But in the death of this illustrious journalist and
matchless orator the lesson is enforced that “The path of glory leads
but to the grave.”
Mr. Grady grew up in the refined atmosphere of cultured Athens,
and his mental nature treasured the classic light of that seat of
learning, and it glowed with attractive radiance in all of his editorial
work. In his death the press of the country loses its brightest
ornament, and the South loses a champion without compare, whose
pen was a trenchant blade in fighting her battles, and a shield when
used to defend her from the hurtling arrows of envy and malice. His
luminous pen made the path of the South’s progress glow, as with
unflagging zeal he devoted his best endeavors to the amelioration of
her war-ruined condition.
Mr. Grady, as the representative of what people are pleased to call
the “New South,” but which is the “Old South” rehabilitated, was, in
the providence of God, calculated to do for his country what Hill,
Gordon and other brilliant lights of the old régime could never have
compassed. As David, “the man of war,” was not permitted to build
the temple, but that glory was reserved for Solomon, so Grady, the
exponent of present principles, was permitted to gather the
fragments and broken columns of the South’s ruined fortunes and
begin the erection of a temple of prosperity so grand in proportion,
so symmetrical in outline, as to attract, in its incomplete state, the
admiration of the world.
In the extremity of our grief we are apt to magnify our loss, but
this, indeed, seems irreparable, and we can take no comfort in the
assurance of the philosopher who codified the experience of the past
into the assurance that great ability is always found equal to the
demand. On whom will Grady’s mantle fall? There really seems to be
none worthy to wear what he so easily graced. And every Southern
heart weighed down with a sense of its woe cannot but ask,

O death, why arm with cruelty thy power

To spare the idle weed yet lop the flower?
 STILLED IS THE ELOQUENT TONGUE.

From the “Brunswick Times.”

Henry W. Grady is dead!
Hushed forever is the voice of the South’s most wonderful orator!
With the laurel upon his brow, with the plaudits of a nation ringing
in his ears, with the love of his people freshly spoken, with a crown
of glory about him, the matchless defender of the South has passed
from earth, and beyond the silence of the stars his soul dwells in the
companionship of the great who have gone before.
With his sorrow fresh upon the South, this death and loss following
so closely upon that other in New Orleans but a few days ago, the
heart is not in keeping with the brain, and not now can the pen
dipped only in tears write.
Henry Grady had not reached the zenith of his fame, for the circle
was widening for him and there were still brighter flowers for him to
pluck, and in her hand Honor held out still richer prizes. But the
mystery of death is upon him, and from his hand has dropped the
forceful, graceful pen, and in silence and peace he sleeps for the
grave.
With a superb intellect, with an eloquence rivalling the golden-
tonged Chrysostom, with a love almost unapproached by any other
for the South and her people, he stood peerless and matchless as
his land’s defender and leader in all that made for her peace,
prosperity and happiness.
But his sun has set. It matters not that in all brightness it went
down; it matters not that he died full of honors; about that grave a
people will gather with tears fast flowing and hearts crushed and
bleeding. It is hard to give up one so grand of mind, so wonderful of
tongue, so magnetic of personality, so richly endowed in all that
equips the great leader.
And such was Henry W. Grady.
 Atlanta will mourn him, Georgia will weep for him, and the South
will sorrow indeed.
Upon his bier the Times lays this tribute and stands reverent and
uncovered by the grave of Georgia’s most brilliant son.
 A SHINING CAREER.

From the “Macon Telegraph.”

Henry Grady is dead. This announcement carried sorrow all over
Georgia yesterday, for there were few men in whom the people of
this State felt so much interest or for whom they cherished such a
warm affection as they did for this gifted and lovable man. He had
not attained his thirty-ninth year when “God’s finger touched him”
and closed his remarkable career, but his name was familiar from
one limit of this Union to the other. Georgia had no more famous
citizen, and perhaps there never was a man in this State in private
station who was so widely known or so much admired. Mr. Grady
never held a public office, and yet he was a recognized force in
Georgia politics almost before he had reached the years of statutory
manhood. He devoted his life to journalism, and in his chosen field
achieved a national fame. He began his career as a boy editor in
Rome, and at an age when most men are merely selecting their
standards and shaping themselves for the real work of life, he
became a prominent and influential figure, a leader of thought, and a
promoter of public enterprises. Eighteen years ago he moved to
Atlanta to pursue his profession in a broader field, and immediately
made himself felt as a positive force in the community. The debt
which Atlanta owes him is great indeed. No man did more to inspire
the pride of community, to set on foot and carry to success great
enterprises for the welfare and progress of the city, to rally its people
to an enthusiastic unanimity on all questions affecting local
prosperity than did Henry W. Grady. These public services would
have endeared him to the people of his adopted city, but they were
not so admirable as his private benefactions. He was first and
foremost in many good works, the fame of which never went beyond
the homes of the poor and unfortunate who were relieved by his
ministrations. His hand was open always to the stricken and needy.
He gave to the afflicted with a generosity which was oblivious to his
own circumstances. Of his influence in promoting public enterprises
there are enduring monuments. By his eloquence of tongue and pen
he raised in less than two weeks $85,000 for the erection of the
beautiful Young Men’s Christian Association building which now
adorns one of the principal streets of Atlanta. He was the moving
spirit in the building of the Chamber of Commerce and the
enlargement of its membership until it reached proportions that made
it a power not only in matters of business but in all the public
concerns of the city. The Confederate Soldiers’ Home of Georgia is a
monument to him, for he seized mere suggestions and made them
the text of an appeal which stirred the hearts of the people of
Georgia and evoked a long delayed tribute of gratitude to the broken
veterans of the lost cause. The Cotton Exposition of 1880 and the
Piedmont Expositions of 1887 and 1889, from which Atlanta reaped
immense benefits, were largely due to his persistent labors.
While Mr. Grady became prominent in Atlanta, and justly
esteemed by his fellow-citizens on account of works and triumphs
like these, he rose into national prominence by reason of other
evidences of his genius. His address to the New England Society in
New York in December, 1886, was one of the most famous
occasional speeches ever delivered in this country. The morning
after its delivery he literally awoke to find himself famous throughout
the country. Since that time he made various public addresses which
commanded the attention of the United States and became subjects
of common conversation among the people. His speech at the Dallas
Exposition last year and his address to the legislatures of Georgia
and South Carolina at the Augusta Exposition a few weeks later,
were themes of the public press of the entire country. But the best
and ablest public speech of his life was his last. It was that which he
delivered two weeks ago at Boston in the performance of a mission
which proved fatal to him. In this, as in all his famous public
addresses, he seemed to strive with a passionate ardor and a most
persuasive eloquence to bring the North and the South to a better
understanding of each other, to foster the spirit of mutual respect and
mutual forbearance, to inculcate the great idea that this is a re-united
country and that the duty of every good citizen in its every section is
to strive for its domestic peace, for its moral, social and material
progress, and for its glory among the nations of the earth. He
handled these great themes with a master hand and invested his
exposition of them with a most fascinating eloquence. Few men in
Georgia ever accomplished so much in so few years. Few men in
Georgia were even the object of such affection at home and such
admiration beyond the bounds of the State. The career which has
been so suddenly cut off was shining with golden promise. The
future seemed to be full of honors and there was everything
surrounding the present that could make life sweet. But the end has
come. The most eloquent tongue in Georgia has been smitten into
everlasting silence in this world. A great, generous heart has been
stilled.
A useful citizen, after a brief but busy and momentous life, which
was productive of many enterprises of public importance and
beneficent tendency, has folded his hands in the eternal rest. God’s
peace be with him!
 THE GREATEST CALAMITY.

From the “Augusta News.”

Can it be possible? Can it be that the brightest star in the galaxy of
our great luminaries is blotted out and stricken from its orbit just as it
was rising in full career to the zenith of usefulness, influence and
splendor? Can it be that the most brilliant meteor which has flashed
across our sky for a generation has fallen to earth literally burned to
ashes by its own fiery contact with the grosser air and elements of
the natural world? Can it be that the light has gone out of the most
magnetic mind and the spirit gone from the most resistless
personality in this sovereign State? Can it be that the South has lost
the man who has been first and foremost in representing its real and
progressive needs and issues, and who has done more for this
section than all the young men of his day combined? Can it be that
the kindly heart has ceased to beat which throbbed in love first for a
devoted family, and next and always for his native State?
Even so, for while still the shadows of the night hung in mournful
pall about his home and dawn lingered as if loth to look upon the
lifeless form of one whom all his people loved, his spirit soared away
to greet the dawning of an eternal day and the mortal part of Henry
Woodfin Grady lay cold in death.
Dead, did we say? Was ever the coming of Death’s angel more
untimely? So it seems to us, with our poor mortal vision, but there is
an eye above, all-seeing; a Providence, all-timely; a Power, almighty;
and to His will we bow this day. In His sight the stricken star is not
blotted out but borne aloft to a brighter realm. In His providence the
brilliant meteor of a day is not fallen, but simply shorn of all its dross
and burnished in beauty and splendor for its flight through all the
ages. In His power the spark which no longer animates the mortal
man glows again in glory and sends a ray of loving light from Heaven
to cheer and console the broken hearts on earth, and remind us that
his influence and work are not lost, but will live and bear blessed fruit
for generations yet to come.
 Henry Grady has gone from earth ere yet the dew of youth has
been drunk up by the midday sun of maturity, but in the brief span of
life allotted to him what a world of work he has done, and what a
name he made for himself! Not two-score years had passed over his
head, and yet he had attained all the substantial success and honor
which mortal man might wish. He was not only loved all over
Georgia, but he was famous all over the country, and no public
occasion of national import was deemed complete without his
presence and his eloquent voice. He was a magician in his mastery
of men, and the witchery of his voice was enchantment to any
audience in any section. He was coming to be regarded as the
representative of the whole South in the editor’s chair and on the
rostrum, and it is truly said of him that he has done more for the
material advancement of this section than any other man for the past
fifteen years. His death is the greatest calamity which has befallen
the South since the late war, and Israel may indeed mourn this day
as for her first-born.
The name of Henry W. Grady will not be forgotten, for it will live in
the affectionate regard of Georgians and grow greater in the good
results which will follow his life-work. The fact that he literally died in
the service of the South, as a result of cold contracted just after the
impassioned delivery of his recent grand oration in Boston, will bind
his name and memory nearer and dearer to Southern hearts; for to
warrior or hero was never given a better time or a nobler way to die
than to the man who gave his voice, his heart, his reputation and his
life to healing the wounds of a fratricidal war, and to the harmonious
building up of his own beloved South as the fairest and richest
domain of our common country.
God bless his name and his memory, and be a strong and abiding
support to his broken-hearted widow and household this day!
 NO ORDINARY GRIEF.

From the “Columbus Ledger.”

A great loss has befallen the South in the death of Henry W.
Grady, and deep sorrow rests upon the hearts of her people.
He was no ordinary man, and his death calls forth no ordinary
grief. Brilliant in intellect, strong in his convictions, untiring in his
efforts to promote the welfare of his country, genial, courteous, kind-
hearted, ever ready to help the unfortunate, the loss of such a man
cannot be estimated. When results were to be achieved, when
encouragement was needed, his eloquent tongue, his ready pen, his
helping hand were used with telling effect. His creed was to build up
and not to tear down; to encourage and not to discourage; to help
and not to hurt. His efforts were ever directed to the promotion of his
State and the South, and no other man has accomplished so much
for them as he. His last effort was for his country and his people, and
the good which will result from his eloquent speech at Boston, will be
a lasting monument. It would have been impossible for any man to
have attained to Mr. Grady’s position without coming into contact
with those who disagreed with him on many points, but even these
acknowledged his greatness. To read of him was to admire him; to
know him was to love him. In the midst of our sorrow let us thank
God that He lends to earth such men.
 A PLACE HARD TO FILL.

From the “Griffin News.”

Henry W. Grady died at his home in Atlanta late Sunday night of
pneumonia, contracted during his recent trip North. His illness was
very short and his untimely death is a shock not only to his many
friends and admirers, but to the whole State in which he was so well
known, and will be received with regret outside its borders. He was a
beautiful writer and a brilliant orator, as well as a prominent factor in
the development of Atlanta. He will be greatly missed in that city, and
his place in the Constitution, of which he was easily the head, will be
hard to fill. Peace to his ashes.
 “JUST HUMAN.”

From the “Thomasville Enterprise.”

Thackeray, the greatest of English novelists, in the concluding
words of Pendennis, says: “I have not painted a hero, only a man
and a brother.” When Henry W. Grady made his first appearance
before the public as a lecturer, his subject was the words that begin
this article—“Just Human.” This was years ago, when he was only
known to the world as a brilliant young journalist, and even then his
fame for quick perception, incisive utterance and felicitous manner,
was only begun. Later years added to that fame, and with each year,
there seemed to come to him a wider range of ideas, and a bolder
conception of the most effectual way to put those ideas into burning,
glowing language.
After he had made his memorable speech before the New
England Society in New York, each succeeding one only raised him
higher in public esteem as a matchless, a magnetic orator, who
could wield human hearts as he would. Through all these speeches,
and in all that he ever wrote, there lingers, like a sweet incense, this
thought, that he recognized that men were “Just Human,” and
entitled to all that charity could offer in extenuation of their faults.
There is not a heart in all the world that has received one pang
from aught that Henry Grady ever wrote or said; his utterances,
whether from the rostrum or through the columns of his paper,
always tended to make the world better, and his ambition seemed to
be to smooth away the differences that annoy, and the bitternesses
that gall. There is no man in all the country that can take up his work
where he left it.
Where can we find the same impassioned eloquence that swayed,
despite its force, as gently as the summer breezes that come across
fields of ripe grain?
 Where can we find the same acute feeling for the sorrows and
sufferings of men and women, “Just Human,” the same sweet
pleading for their extenuation or their amelioration?
When the epitaph over his grave comes to be written, no better
rendering of the true greatness of the departed could be made than
is contained in the suggestive name of his first lecture, “Just
Human,” for the noble instinct that taught him to plead so eloquently
for the failings of his fellow men, taught him to enter the Divine
presence, asking for himself that mercy he had asked for others.
 GEORGIA WEEPS.

From the “Union News.”

Hon. Henry W. Grady, of the Constitution, died at his home in
Atlanta this morning at 3:40.
This cruel blow shivers every heart with agony, even as the
thunderbolt of heaven rends the mighty monarch of the forest.
His death is a loss to Georgia. Every man feels it as a personal
bereavement. He has done more for the material development of the
State than any other one man in it. He was an enthusiast in the
cause of education, an upholder of the church, an advocate of
industrial training, a promoter of every enterprise calculated to
benefit Georgia and her people. He was a friend to humanity, true to
himself, to his country and to his God.
The most brilliant light in Southern journalism is veiled in darkness
—a manly heart has ceased to beat; the tongue that has electrified
thousands with magic eloquence is silent forever; the fingers that
wielded the pen of genius and never traced a line in bitterness or
malice, but was always uplifted in behalf of charity, love and good
will, in behalf of progress, industry and enterprise, in behalf of the
South and her institutions, his State and her people, are cold in
death; the once warm hand of benevolence and fraternal greeting is
chilled forever; a golden life is ended, but his works live after him, as
a priceless heritage to his State, a boon to his people. The influence
of his example pervades the State as a delightful aroma.
The dispensations of Providence are mysterious. It is strange fate,
past all human understanding, why so excellent a spirit, a man of so
much influence, should be cut down in the glory of his life, in the
richest prime of his royal manhood.
Only a few days ago he stood in a blaze of glory in a Northern city
and electrified thousands by his matchless oratory, in the
presentation of a question that did the South great good and justice,
and did much to soften the animosities of the North toward the
South, and establish more fraternal relations between the two
sections. But even while the plaudits of the admiring multitude were
ringing in his ears, and the press of the country was singing his
praises, the fatal hand of disease was laid upon him, and he was
brought back to his own sunny and beloved Southland to die.
Mr. Grady was a popular idol. He was destined to reap the highest
political honors in the State. His name was being prominently
mentioned in connection with the Governorship and Senatorship of
Georgia. Democratic leaders sought his favor. His influence was felt
throughout the entire State. His support was an omen of success.
Ben Hill died, and his place has never been supplied in Georgia.
Mr. Grady approached nearer to it than any other man. Now Mr.
Grady is gone, and his duplicate cannot be found in the State. No
man in recent years could so attract the eye and fasten the attention
of the North. The death of no other Georgian at this time would have
been so calamitous.
The star was rapidly hastening to the zenith of its brilliancy and
greatest magnitude when suddenly it went out in darkness, but
across the industrial and political firmament of the country it has left
an effulgent track whose reflection illumines the world.
 A GRAND MISSION.

From the “West Point Press.”

So much has been said about the lamented Grady that we may
not be able to offer anything new. But as we feel that his untimely
death is an irreparable loss we must offer our heartfelt tribute.
He was the most unselfish slave to friends, and to duty. As an
editor he was brilliant and at all times as fearless as a Spartan; as an
orator, age considered, he stood without a peer within the broad
realm of his native land, and although but in the full vigor of
manhood he has left upon record speeches that compare favorably
with the master efforts of Calhoun and Webster. As a companion he
was genial, jovial and untiring in his efforts to entertain; as a friend
there was no bound to his fidelity.
If you would know the beauty and grandeur of Henry Grady’s
character, go and learn it at the homes of poverty where he delighted
to turn in the light, by his many offices of love and charity. If you
would know the kindness of his generous heart go to those whom he
has lifted from the vale of poverty and given encouragement to look
up. Ask the army of newsboys for a chapter upon the life of Henry
Grady and you will hear words to convince you that a philanthropist
has been called hence. It seemed to us the other day while in
Atlanta, as they said “Paper, sir,” that there was a sadness in the
tone, and that a great sorrow was upon their hearts. Yes, those
newsboys miss Henry Grady, for he was their friend and protector.
Words of eulogy cannot restore those who cross the dark river; if
they could there has been enough said to recall Henry Grady to the
high position he honored by a life of unselfishness. His mission, only
begun, was a grand one, and we trust his mantle may fall upon some
one who will carry on his work.
 THE SOUTH LOVED HIM.

From the “Darien Timber Gazette.”

Seldom has the nation’s heart been so saddened as by the news
of Henry W. Grady’s death. Henry W. Grady, although comparatively
young, has conquered this vast continent—east and west, north and
south—and his many victories have been bloodless. He has truly
demonstrated that the pen is mightier than the sword. An intellect
exceptionally brilliant, an indomitable courage, a judgment keen,
clear and cool, a character unspotted and unassailable—these are
the weapons with which Henry W. Grady captured the nation.
The South loves him for his unflinching devotion to its interests;
the North admires him for the conservatism which always
characterized his political actions. The brilliancy of his intellect all
admit. We venture to say that there lives not a man in the United
States to-day whose death would be more sincerely or more
universally mourned.
That a career so unusually promising should have been so
suddenly cut off is sad indeed—sad especially for the South, whose
claims he so ably advocated and so successfully furthured. The
severing of the still more tender ties between wife and husband,
mother and son, while in the youth of his glory, adds another gloomy
chapter to the death of Southland’s most patriotic and brilliant son.
Millions will bow their heads in grief with the loving wife and devoted
mother.
We read and re-read the words of Henry W. Grady’s last speech
with a strange fascination. They are like the last notes of the dying
swan and will doubtless have much more weight under the sad
circumstances. He has literally laid down his life that the colored man
might enjoy his in peace and prosperity.
 NO SADDER NEWS.

From the “Marietta Journal.”

No sadder news ever fell upon the ears of this people than the
announcement that “Henry Grady is dead!” It staggered our people
like a bolt of lightning from a clear sky.
His death took place at the family residence in Atlanta at 3:40
o’clock Monday morning, December 22. While on a visit to Boston,
where he delivered the grandest speech of his life, he took cold, and
being ill before he left home, he was prostrated on his return home,
his sickness culminating in pneumonia and death. He was thirty-
eight years old at the time of his death, and no private citizen at that
age ever attained the renown that Grady had. As an orator and
journalist he was without a peer; gifted above his fellows to sway
men by his pen or his voice, he won the applause and admiration
and love of his countrymen wherever he came in contact with them.
His young life and genius had been devoted to deeds of kindness,
peace, unity and charity. Selfishness did not enter his heart, that
always beat in response to the woes and sufferings of his fellow
men.
There was a charm and sparkle about his writings that never failed
to captivate the senses, please and entertain. The South lost one of
her brightest minds and stanchest champions in the death of Henry
Grady. There is no man that can take his place in the rare gifts that
so befittingly endowed him in the grand work in which he was
engaged. His loss is an irreparable one. Sorrow and gloom pervade
the hearts of our people over this sad event. We may not understand
how one so superbly gifted, with capacities for the accomplishment
of so much good in the world, is taken, and many who cumber the
earth and are stumbling blocks, are left, but we know the hand of
Providence is behind it all, and He is too wise to err, too good to be
unkind.