Professional Documents
Culture Documents
Improving Population Health Using Electronic Health Records: Methods For Data Management and Epidemiological Analysis 1st Edition Goldstein
Improving Population Health Using Electronic Health Records: Methods For Data Management and Epidemiological Analysis 1st Edition Goldstein
Improving Population Health Using Electronic Health Records: Methods For Data Management and Epidemiological Analysis 1st Edition Goldstein
https://textbookfull.com/product/secondary-analysis-of-
electronic-health-records-1st-edition-mit-critical-data-auth/
https://textbookfull.com/product/biostatistics-and-computer-
based-analysis-of-health-data-using-r-1st-edition-christophe-
lalanne/
https://textbookfull.com/product/public-health-administration-
principles-for-population-based-management-3rd-edition-leiyu-shi/
https://textbookfull.com/product/the-digitalization-of-health-
care-electronic-records-and-the-disruption-of-moral-orders-first-
edition-garrety/
Public health leadership : strategies for innovation in
population health and social determinants 1st Edition
Bhattacharya
https://textbookfull.com/product/public-health-leadership-
strategies-for-innovation-in-population-health-and-social-
determinants-1st-edition-bhattacharya/
https://textbookfull.com/product/transformation-and-your-new-ehr-
the-communications-and-change-leadership-playbook-for-
implementing-electronic-health-records-1st-edition-dennis-r-
delisle-author/
https://textbookfull.com/product/nursing-health-education-and-
improving-patient-self-management-1st-edition-barbara-sassen-
auth/
https://textbookfull.com/product/systems-science-and-population-
health-1st-edition-el-sayed/
https://textbookfull.com/product/personalised-health-care-
fostering-precision-medicine-advancements-for-gaining-population-
health-impact-stefania-boccia/
Improving Population
Health Using
Electronic Health
Records
Methods for Data Management
and Epidemiological Analysis
Neal D. Goldstein
Improving Population
Health Using Electronic
Health Records
Improving Population
Health Using Electronic
Health Records
Methods for Data Management
and Epidemiological Analysis
Neal D. Goldstein
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
This book contains information obtained from authentic and highly regarded sources. Reasonable efforts
have been made to publish reliable data and information, but the author and publisher cannot assume
responsibility for the validity of all materials or the consequences of their use. The authors and publishers
have attempted to trace the copyright holders of all material reproduced in this publication and apologize to
copyright holders if permission to publish in this form has not been obtained. If any copyright material has
not been acknowledged please write and let us know so we may rectify in any future reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmit-
ted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented,
including photocopying, microfilming, and recording, or in any information storage or retrieval system,
without written permission from the publishers.
For permission to photocopy or use material electronically from this work, please access www.copyright.
com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood
Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and
registration for a variety of users. For organizations that have been granted a photocopy license by the CCC,
a separate system of payment has been arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used
only for identification and explanation without intent to infringe.
Preface ..............................................................................................................xiii
Acknowledgments ............................................................................................. xv
About the Author ............................................................................................xvii
Abbreviations and Acronyms ..........................................................................xix
1 Research in the Era of Electronic Health Records..................................1
Using EHRs for Research ............................................................................2
EHRs and Big Data .....................................................................................3
Stakeholders in the Research Process ...........................................................4
The Need for a Cohesive View of Research ..................................................6
References ....................................................................................................7
2 How to Use This Book for Research .......................................................9
Concepts and Terminology ........................................................................10
Technology Concepts ............................................................................11
Research Concepts ................................................................................14
Statistical Concepts ...............................................................................15
Epidemiological Concepts .....................................................................16
Hardware and Software Requirements.......................................................17
Data Management.................................................................................18
Data Analysis and Presentation .............................................................19
Reference ...................................................................................................19
vii
viii ◾ Contents
This book is designed as a guide for working with electronic health record (EHR)
data in epidemiological analysis. After working for many years with health data,
and being asked to recommend a hands-on guide to performing research, I realized
that no such text existed. Therefore, this book fills a void in the field known as clini-
cal epidemiology. In medicine, students, residents, fellows, physicians, and other
health-care professionals may be called upon to perform research from the EHR
to answer a pressing clinical question. This book is designed to take the researcher
through the entire spectrum of EHR-derived research: obtaining data, organizing
it in a useful fashion for research, performing the statistical analyses, and presenting
the results. But it is also useful for the public health practitioner who has practical
experience in conducting research and wishes to use data obtained from an EHR
to answer population health questions. Recognizing that modern-day research is
cross-discipline and collaborative, this book can be used in a piece meal fashion,
allowing the reader to select which portions or chapters are most relevant to their
immediate needs.
Most of this book is presented in a software-agnostic manner. Examples in
the appendices are given using the R system for statistical computing. R is a cross-
platform and freely available environment that is powerful and flexible, and has
rich graphical capabilities. Yet by no means will the reader be held back by not
using R. The statistical methods are broadly applicable to any software. For users
of R, the examples are entirely reproducible using the datasets and example codes
from the appendices.
I hope this book proves useful for both novice and experienced researchers who
are interested in working with EHR data. Trying to anticipate all research designs
from the EHR is simply not possible, so this book focuses on the most common
types of research used in public health and epidemiology. I welcome the comments
on the utility of this book and the suggestions for improvements and additional
topics in subsequent editions.
xiii
Acknowledgments
I am indebted to several people who helped to make a book of this scope possible.
I was fortunate to have three wonderful colleagues review drafts of the manuscript
for technical accuracy: Dr. Aimee Palumbo, Dr. Brett Maiese, and Jon Brolund. In
the later stages, Dr. Paul Yerkes provided a critical review of this book for cohesive-
ness and readability. All of their assistance has been invaluable, and any errors that
remain are mine alone.
I also thank Dr. Seth Welles, from the Drexel University Dornsife School of
Public Health, Philadelphia, Pennsylvania, for his motivation and encouragement
throughout this process, and Dr. David Paul, from the Christiana Care Health
System Department of Pediatrics, Wilmington, Delaware, for supporting this proj-
ect while I was a postdoc.
Select portions of this book have previously appeared as blog posts on the
author’s personal blog.
xv
About the Author
xvii
Abbreviations and
Acronyms
xix
Chapter 1
The collection of health information on a large-scale basis may not be new, but
the availability of tools and techniques to analyze “big data” makes working with
electronic health records (EHRs) possible. Today’s personal computers are capable
of astounding feats that a mere generation ago would have necessitated not only
large, air-conditioned rooms, but also capital investments that all but a few of the
largest corporations could afford. No longer is technology the limiting factor in
population health research; it is the interconnection of systems and the expertise of
cross-discipline researchers that represents the current shortcoming.
With the ultimate goal of improving the public’s health from an epidemiologi-
cal analysis of big data sources including EHRs, researchers need to be cognizant of
the entire spectrum of data acquisition, management, analysis, and interpretation.
Unfortunately, due to the variety of stakeholders in the process (Figure 1.1), research
is disjointed, with many individuals of varied backgrounds handling discrete com-
ponents. This is not ideal; the research process is sequential, dependent upon previ-
ous steps, and the end result of improving population health can only be as good as
the preceding steps enable it to be.
What follows in this book are two overarching aims. First, the book presents the
research process as a continuum, with the ultimate aim of enabling an individual
(or research group) to proceed through each step with an improved understanding
of the data. Second, this book demystifies the process at each step (i.e., preparing the
data, performing the analysis, and interpretation and implementation). For example,
extracting and linking data is often thought of as an information technology (IT)
or informatics role, with minimal appreciation for the underlying architecture. Yet,
by examining this architecture, the researcher can be better prepared to analyze
1
2 ◾ Improving Population Health Using Electronic Health Records
Patient Population
Figure 1.1 From patient to population: The public health research process.
the data properly, minimizing the chance for errors due to a miscommunication or
misunderstanding of what a data point represents.
2. Velocity. The pace at which data are acquired has also increased markedly.
Therefore, in addition to more data being captured, it is being captured quicker.
3. Variety. Data are heterogeneous, from primary care provider office notes to
medical images to laboratory reports and so on, and will continue to be as more
systems are incorporated together. Big data stores must be able to capture a
wide variety of data formats and continually expand with their diversity.
This definition was appropriate in 2001 when it was proposed and has held up in the
years since. EHRs (and HIEs) certainly fulfill the three Vs of big data, and in fact have
fulfilled the definition well before their existence. The EHR is here to stay, and few,
if any, clinicians will want to revert to a paper-based system. But what should one do
with the increasing volume of disparate health data?
This question seeks to define the architecture of big data sources, and it is the
answer to this question that will have some big data purists deviating from the archi-
tecture proposed in this book. However, before casting this book aside, the reader
is asked to keep in mind the end goal: improving the public’s health using EHRs.
That is, we are interested in a specific type of data, and given a diverse skill set of
researchers, a rather simplistic architecture is presented, with an eye toward the most
efficient pathway to epidemiological analysis.
The intended audience of this text further necessitates a practical, rather than
a theoretical, framework for big data. The primary readers of this book will be the
public health researcher who lacks the expertise and skill set of someone who works
in IT; the clinician who is interested in research and needs a concise how-to guide
for obtaining data and designing the proper study; the IT worker who is looking to
expand his or her work into clinical epidemiology; and health-care administrators who
have been tasked with improving population health, yet do not know how to define
the process. The text is written toward these readers who may have limited resources, and
therefore wish to take on multiple roles in the research process. Aside from this primary
audience, the text will also prove useful for anyone who interacts with health data,
as a better understanding of its use beyond clinical care could only improve data and
research quality. Thus, to avoid getting mired down in big data methodology that
requires both capital expense and an expansive collaborative group, we keep the data
management methods streamlined and propose an architecture that serves a specific
application using EHR data: epidemiology.
even be aware that the data generated as part of that encounter ultimately may be
used in research despite the Health Insurance Portability and Accountability Act’s
Privacy Rule that contains explicit provisions for research from patient data col-
lected during the encounter. From the patient’s perspective, the interest is getting
well (or staying well) with the expectation that the provider seen has the best treat-
ments available. From the clinician’s perspective, his or her interest also reflects the
patient’s interest: how can an illness be cured or prevented, or sequelae mitigated.
The clinician (e.g., physician, nurse, medical assistant, or other health-care profes-
sional) acts as the gatekeeper to the system. His or her knowledge of therapeutics
or prophylactics is drawn from experience, literature, and colleagues. Therefore,
the physician likely represents the first opportunity for research, often in the form
of a clinical question: “Will this intervention affect my patient?” Unfortunately,
the clinician may lack time, expertise, and data to answer this question, instead
relying on the existing (and possibly outdated) literature. The data generated from
each patient encounter is entered (by the clinician or ancillary staff) into the EHR.
At this point, the IT group assumes the responsibility for the integrity and
accessibility of the health data entered by the provider. IT and its more recent
colleagues informaticists (trained in both IT and health care) maintain the infor-
mation systems and therefore have the most comprehensive knowledge of the
structure of the data. The clinician interested in conducting research therefore
needs to effectively communicate with IT to receive the correct data to answer
the research question at hand. This step cannot be trivialized and will figuratively
make or break the research aims. The informaticist serves as the liaison between
groups but is likely a luxury only the larger institutions have on staff. In addition,
the IT group may have its own policies and procedures in place that can, occasion-
ally and unfortunately, inhibit the research process; therefore, one of the goals of
this work is to empower the clinician in the research process, making the research
data request as efficient as possible and minimizing the burden on IT.
Assuming data are available to answer the original research question, a new
stakeholder enters the process: the public health researcher. Their background has
them well suited to design and analyze the study, yet it is often dependent upon the
successful handoff of data. This handoff of data, channeled through IT, originates
with the clinician assessing which patient characteristics are most meaningful to
study. Working hand in hand with the public health researcher can further eluci-
date these characteristics. Finally, after the analysis is completed and the results
generated, the findings need to be effectively communicated to all stakeholders in
the process, as well as the community at large, though this is often an afterthought.
Yet this may be the most crucial point in the research process: findings, whether
positive or negative, need to be relayed clearly and concisely to others for incorpo-
ration into the greater body of scientific knowledge. The entire process becomes
a continuum as the community becomes influenced by the outcomes of research
(e.g., smoking cessation campaigns) and shapes the scope and nature of the patients
reentering the system. Last, but certainly not least, overseeing this entire process
6 ◾ Improving Population Health Using Electronic Health Records
information extraneous to the question at hand. The end result of completing Part I
will be a research dataset with a comprehensive data dictionary usable for any sec-
ondary data analyses, the focus of Part II.
Part II presents the core epidemiology, introducing essential study designs and
statistical techniques needed to analyze “secondary” data, in other words, data
retrieved from the EHR. Part III addresses the question, “Where do you go once
the research has been conducted?” This question is often overlooked, deliberately
or not, by the pure researcher who is interested only in the methods of research
and not the end result. However, by discussing the interpretation and implemen-
tation challenges and opportunities with the science, this book reinforces the end
result of research: improving the public’s health. Part III details the preferred
methods for presenting results and provides many concrete examples. This book
concludes with a discussion of population health and moving from knowledge
generation to knowledge dissemination through translational epidemiology.
For readers wishing to skip ahead, or use this book in a piecemeal fashion,
Chapter 2 presents a flowchart of this book, and Appendix 1 includes a research
planner with specific chapters and sections most relevant to specific points in the
research process.
References
1. Oregon Health & Science University Clinfowiki. 2015. Computer Stored Ambulatory
Record (COSTAR). http://www.clinfowiki.org/wiki/index.php/Computer_Stored_
Ambulatory_Record_(COSTAR) (accessed September 4, 2015).
2. Oregon Health & Science University Clinfowiki. 2015. Regenstrief Medical Record
System (RMRS). http://www.clinfowiki.org/wiki/index.php/RMRS (accessed
September 4, 2015).
3. Oregon Health & Science University Clinfowiki. 2015. Veterans Health Information
Systems and Technology Architecture (VistA). http://www.clinfowiki.org/wiki/
index.php/Vista (accessed September 4, 2015).
4. Oregon Health & Science University Clinfowiki. 2015. Historically Important
Electronic Medical Record Systems. http://www.clinfowiki.org/wiki/index.php/
Historically_Important_Electronic_Medical_Record_Systems (accessed September 4,
2015).
5. Centers for Disease Control and Prevention. 2012. Meaningful Use. http://www.cdc.
gov/ehrmeaningfuluse/introduction.html (accessed September 4, 2015).
6. Centers for Disease Control and Prevention. 2016. Adoption of Certified Electronic
Health Record Systems and Electronic Information Sharing in Physician Offices:
United States, 2013 and 2014. http://www.cdc.gov/nchs/data/databriefs/db236.htm
(accessed February 1, 2015).
7. U.S. Department of Health & Human Services. 2015. Meaningful Use Definition &
Objectives. http://healthit.gov/providers-professionals/meaningful-use-definition-
objectives (accessed September 4, 2015).
8 ◾ Improving Population Health Using Electronic Health Records
8. Oregon Health & Science University Clinfowiki. 2015. Regenstrief Medical Record
System (RMRS). http://www.clinfowiki.org/wiki/index.php/Health_Information_
Exchange (accessed September 4, 2015).
9. Laney D. 2012. Deja VVVu: Others Claiming Gartner’s Construct for Big Data. Garner
Blog Network. http://blsogs.gartner.com/doug-laney/deja-vvvue-others-claiming-
gartners-volume-velocity-variety-construct-for-big-data (accessed September 4, 2015).
Another random document with
no related content on Scribd:
Gallo.—Preguntaselo a Grilo,
noble varon griego, el qual
boluiendo de la guerra de Troya
passando por la ysla de Candia le
conuertio la maga Cyrçes en
puerco, y despues por ruego de
Ulixes le quisiera boluer honbre, y
tanta ventaja halló Grilo en la
naturaleza de puerco, y tanta
mejora y bondad que escogio
quedarse ansi, y menospreçió
boluerse a su natural patria.
Miçilo.—Por cierto cosas me
cuentas que avn a los hombres
de mucha esperiençia cansassen
admiraçion, quanto más a vn
pobre çapatero como yo.
Gallo.—Pues porque no me
tengas por mentiroso, y que
quiero ganar opinion contigo
contandote fabulas, sabras que
esta historia auctorizó Plutarco el
historiador griego de más
auctoridad.
Miçilo.—Pues, valame dios, que
bondad halló ese Grilo en la
naturaleza de puerco, por la qual
a nuestra naturaleza de hombre la
prefirio?
Gallo.—La que yo hallé.
Miçilo.—Eso deseo mucho saber
de ti.
Gallo.—A lo menos vna cosa
trabajaré mostrarte como aquel
que de ambas naturalezas por
esperiençia sabra dezir. Que
comparada la vida y inclinacion
de muchos hombres al comun
viuir de vn puerco, es mas perfeto
con gran ventaja en su natural.
Prinçipalmente quando de viçios
tiene el hombre ocupada la razon.
Y agora pues es venido el dia
abre la tienda y yo me passearé
con mis gallinas por la casa y
corral en el entretanto que nos
aparejas, el manjar que emos de
comer. Y en el canto que se sigue
verás claramente la prueba de mi
intinçion.
Miçilo.—Sea ansi.