Improving population health using

electronic health records : methods for

data management and epidemiological
analysis 1st Edition Goldstein
Visit to download the full and correct content document:
https://textbookfull.com/product/improving-population-health-using-electronic-health-r
ecords-methods-for-data-management-and-epidemiological-analysis-1st-edition-golds
tein/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Secondary Analysis of Electronic Health Records 1st

Edition Mit Critical Data (Auth.)

https://textbookfull.com/product/secondary-analysis-of-
electronic-health-records-1st-edition-mit-critical-data-auth/

Biostatistics and Computer-based Analysis of Health

Data using R 1st Edition Christophe Lalanne

https://textbookfull.com/product/biostatistics-and-computer-
based-analysis-of-health-data-using-r-1st-edition-christophe-
lalanne/

Public Health Administration Principles for Population

Based Management 3rd Edition Leiyu Shi

https://textbookfull.com/product/public-health-administration-
principles-for-population-based-management-3rd-edition-leiyu-shi/

The digitalization of health care : electronic records

and the disruption of moral orders First Edition
Garrety

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/

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/transformation-and-your-new-ehr-
the-communications-and-change-leadership-playbook-for-
implementing-electronic-health-records-1st-edition-dennis-r-
delisle-author/

Nursing: Health Education and Improving Patient Self-

Management 1st Edition Barbara Sassen (Auth.)

https://textbookfull.com/product/nursing-health-education-and-
improving-patient-self-management-1st-edition-barbara-sassen-
auth/

Systems science and population health 1st Edition El-

Sayed

https://textbookfull.com/product/systems-science-and-population-
health-1st-edition-el-sayed/

Personalised Health Care: Fostering Precision Medicine

Advancements for Gaining Population Health Impact
Stefania Boccia

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

© 2017 by Taylor & Francis Group, LLC

CRC Press is an imprint of Taylor & Francis Group, an Informa business

No claim to original U.S. Government works

Printed on acid-free paper

International Standard Book Number-13: 978-1-4987-5910-6 (Hardback)

International Standard Book Number-13: 978-1-1381-9637-7 (Paperback)

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.

Library of Congress Cataloging‑in‑Publication Data

Names: Goldstein, Neal (Neal D.), author.

Title: Improving population health using electronic health records : methods
for data management and epidemiological analysis / Neal Goldstein.
Description: Milton Park, Abingdon, Oxon ; New York, NY : Routledge, 2017. |
Includes bibliographical references.
Identifiers: LCCN 2016040245 | ISBN 9781498759106 (hbk) | ISBN 9781138196377
(pbk) | ISBN 9781498759120 (ebk)
Subjects: | MESH: Public Health Practice | Data Mining--methods | Electronic
Health Records--organization & administration | Research Design |
Epidemiologic Methods | Quality Improvement
Classification: LCC R858.A2 | NLM WA 100 | DDC 362.10285--dc23
LC record available at https://lccn.loc.gov/2016040245

Visit the Taylor & Francis Web site at

http://www.taylorandfrancis.com

and the CRC Press Web site at

http://www.crcpress.com
To Joanna
Contents

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

SeCtion i UnDeRStAnDinG tHe DAtA

3 Planning the Research..........................................................................23
The Research Process .................................................................................24
Study Funding, Sponsorship, and Ethics ...................................................26
Funding ................................................................................................26
Sponsorship ...........................................................................................28
Research Ethics .....................................................................................29
Data Availability ........................................................................................30

vii
viii ◾ Contents

Data Accessibility ......................................................................................32

Research Planner .......................................................................................32
References ..................................................................................................35
4 Accessing Health Data .........................................................................37
Which Data Belong in the Research Dataset? ............................................38
Which Observations? ............................................................................39
Which Variables?...................................................................................39
The Rolling Snapshot Model ................................................................ 40
Methods for Data Extraction .....................................................................41
Live Query Export and Subset ............................................................. 42
Manual Export and Subset ....................................................................47
Database ...........................................................................................47
Spreadsheet .......................................................................................48
Proprietary or Other Format.............................................................49
Manual Subset without Export .............................................................49
References ..................................................................................................49
5 Organizing, Merging, and Linking Data .............................................51
Data Organization .....................................................................................52
Wide Format .........................................................................................53
Long Format .........................................................................................54
Transforming from a Long to a Wide Format .......................................55
Data Merging ............................................................................................56
Data Linkage .............................................................................................57
Exact Matching .....................................................................................59
Fuzzy Matching ....................................................................................62
Fine-Grained Record Integration and Linkage Tool ..............................63
References ................................................................................................. 64
6 Data Management and the Research Dataset.......................................67
Manipulating Variables ..............................................................................68
Assigning Data Types ............................................................................68
Renaming and Recoding Variables ........................................................70
Removing Variables ...............................................................................73
Manipulating Observations .......................................................................73
Duplicate Observations .........................................................................73
Manipulating the Dataset ..........................................................................74
Unique Identifiers and PHI ...................................................................74
Missing Data .........................................................................................76
Data Dictionary and Audit Log ........................................................... 77
 Contents ◾ ix

SeCtion ii ConDUCtinG tHe ReSeARCH

7 Study Design and Sampling .................................................................81
Study Design Overview .............................................................................82
Cross-Sectional Study ............................................................................... 84
Cohort Study .............................................................................................88
Case–Control Study ..................................................................................92
Other Observational Study Types ..............................................................95
Study Size and Power .................................................................................97
References ..................................................................................................98
8 Measures of Frequency and Risk ..........................................................99
Measures of Frequency.............................................................................100
Prevalence ...........................................................................................100
Incidence .............................................................................................102
Choice of Cumulative Incidence versus Incidence Rate .......................103
Relationship of Prevalence and Incidence ............................................105
Measures of Association ...........................................................................106
Cross-Sectional Measures ....................................................................107
Cohort Measures .................................................................................108
Case–Control Measures ...................................................................... 111
Statistical Significance .........................................................................113
References ................................................................................................113
9 Threats to Validity ..............................................................................115
Bias .......................................................................................................... 116
Selection Bias ...................................................................................... 116
Information Bias ................................................................................. 118
Confounding ...........................................................................................120
Generalizability .......................................................................................124
References ................................................................................................125
10 The Analytic Dataset ..........................................................................127
A Motivating Example .............................................................................128
Which Observations? ..........................................................................128
Which Variables?.................................................................................128
Which Study Design?..........................................................................129
Cross-Sectional Study Design ..................................................................130
Cohort Study Design ...............................................................................130
Longitudinal Cohorts..........................................................................132
Case–Control Study Design ....................................................................133
Reference .................................................................................................134
x ◾ Contents

11 Epidemiological Analysis I .................................................................135

An Example Analytic Dataset ..................................................................136
Describing the Variables and Verifying the Assumptions .........................136
Missing Data .......................................................................................143
Regression Analysis ..................................................................................143
An Example from the birthwt Analytic Dataset................................... 145
Model Diagnostics ..............................................................................148
Sensitivity Analysis ..................................................................................149
References ................................................................................................150
12 Epidemiological Analysis II ...............................................................151
Confounder Selection .............................................................................. 152
Statistical Perspective .......................................................................... 153
Theoretical Perspective ........................................................................ 153
Multivariable Procedures for Epidemiological Study Designs ..................156
Cross-Sectional Analysis......................................................................156
Cohort Analysis................................................................................... 159
Case–Control Analysis ........................................................................165
Summary ............................................................................................168
References ................................................................................................169

SeCtion iii inteRPRetAtion to iMPLeMentAtion

13 Interpreting the Results......................................................................173
Interpreting the Statistics ......................................................................... 174
Crude Statistics ................................................................................... 174
t-Test ............................................................................................... 174
Analysis of Variance........................................................................175
Wilcoxon Rank-Sum or Mann–Whitney Test ................................177
Chi-Squared Test ............................................................................178
Adjusted Statistics ...............................................................................179
Linear Regression ...........................................................................179
Logistic Regression ......................................................................... 181
Cox Proportional Hazards Regression (Survival Analysis) ..............183
Statistical Inference ..................................................................................185
What Does the p-Value Indicate? ........................................................186
Supplementing the p-Value ..................................................................187
Statistically versus Clinically Meaningful............................................188
Causal Inference ......................................................................................189
References ................................................................................................ 191
 Contents ◾ xi

14 Publication and Presentation .............................................................193

Publication...............................................................................................194
Formatting Research for Publication ...................................................194
Strengthening the Reporting of Observational
Studies in Epidemiology Statement .....................................................196
Tables and Figures ...............................................................................196
Open Access ....................................................................................... 200
Submission Process to a Peer-Reviewed Journal...................................202
Presentation .............................................................................................203
Communicating with a Broader Audience .............................................. 204
Novel Methods of Public Health Communication ..................................205
References ................................................................................................207
15 Improving Population Health ............................................................209
Success Stories .........................................................................................210
Public Health versus Individual Health ................................................... 211
Measures of Population Health ................................................................212
Big Data for Population Health Research ................................................ 214
The Individual’s Role ............................................................................... 215
References ................................................................................................216
Appendix 1: Secondary Data Research Planner ..........................................219
Appendix 2: Example Code Using R ...........................................................227
Index ...........................................................................................................247
Preface

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

Neal D. Goldstein, PhD, MBI is an infectious dis-

ease epidemiologist at Christiana Care Health System,
Newark, Delaware, and holds a faculty appointment
in the Dornsife School of Public Health at Drexel
University, Philadelphia, Pennsylvania. He has an
extensive experience in epidemiological analyses
from secondary data sources, particularly electronic
health records. His research spans several disciplines
including vaccine-preventable diseases, sexual minor-
ity health, pediatric infectious diseases, and women’s
health surrounding pregnancy. He also possesses a
background in biomedical informatics with a detailed
knowledge of hardware and software in the health-care domain. Most recently,
he has focused on translational epidemiology, or moving from knowledge gen-
eration to application and advocacy. He writes a science blog, which is avail-
able at www.goldsteinepi.com/blog.

xvii
Abbreviations and
Acronyms

CDC Centers for Disease Control and Prevention

CDM common data model, as proposed by the Observational Medical
Outcomes Partnership
CI confidence interval
CSV comma-separated value
EHR electronic health record
FRIL fine-grained record integration and linkage
IRB institutional review board
IT information technology
HIE health information exchange
HIPAA Health Insurance Portability and Accountability Act
HITECH Health Information Technology for Economic and Clinical Health
MRN medical record number
NICU neonatal intensive care unit
ODBC Open Database Connectivity
OR odds ratio
PHI protected health information
RR relative risk or risk ratio
SES socioeconomic status
STROBE Strengthening the reporting of observational studies in epidemiology
SQL structured query language

xix
Chapter 1

Research in the Era of

Electronic Health Records

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

Stakeholders in the public health research continuum

Clinician IT and informatics Researcher Community

Data Analyze Interpret

Patient Population

Part I: Part II: Part III:

Understanding Conducting Interpretation and
the data the research implementation

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.

Using EHRs for Research

A key driver of large-scale health research has been the adoption of EHRs by hospi-
tal systems (as well as small practices) and the connection of these systems together
in health information exchanges (HIEs). EHRs represent a suite of applications and
tools that enable electronic management of health. Initially, these applications were
mere data repositories (termed clinical data repositories) and, in some crude cases,
represented scanned versions of paper medical charts. Early adopters of clinical data
repositories included notable institutions such as Massachusetts General Hospital in
the late 1960s (Computer Stored Ambulatory Record),1 Indiana University Medical
Center in the early 1970s (Regenstrief Medical Record System),2 and the Veterans
Health Administration (Veterans Health Information Systems and Technology
Architecture) in the late 1970s.3 Concurrent with these efforts were a variety of
other small- and large-scale practices both nationally and internationally.4
One of the primary drivers toward the widespread use of EHRs came from the
Health Information Technology for Economic and Clinical Health (HITECH)
Act portion of the 2009 American Reinvestment & Recovery Act, which demanded
the evolution of these systems to their current incarnation of interconnected health
tools that exist first and foremost to improve health outcomes.5 Although some pro-
viders balked at the notion (mainly from expense and early inefficiency) of EHRs,
as of 2014 nearly 75% of U.S-based providers have implemented them.6 This trend
will only continue; combined with the financial incentives for adopting EHRs as
put forth in the HITECH Act and the requirements of Medicare and Medicaid,
reverting to a paper-based system is highly unlikely.
 Research in the Era of Electronic Health Records ◾ 3

Population health research was a natural extension of HITECH, and specific

sections of the act called for “improved population health outcomes” and “more
robust research data on health systems.”7 As EHRs largely demand coded data entry,
research from these data is greatly improved, in terms of both accuracy and efficiency.
In addition, even among noncoded free-text data, as warranted to capture a specific
nuance or subtlety common in certain medical domains such as behavioral health,
natural language parsing makes it possible to transform free-text data to coded data.
Although this field may be in its infancy, it shows great promise.
The interconnection of the various EHRs was the logical next step. Again,
spurred on by the HITECH Act, a joint health record was needed that acknowl-
edged patients interact with a variety of providers, and many of these providers
stored their data in “silos,” inaccessible by other providers. By connecting data silos
together in networks, known as health information exchanges (HIEs), providers and
researchers now have a longitudinal view of patient outcomes over time.8 The data
linkage and management has been done for them, and therefore, a much richer set
of data is available to analyze population health trends.
This book mainly considers EHRs, and by extension HIEs, although the meth-
ods discussed are broadly applicable to myriad health data sources that present sim-
ilar challenges to working with a multitude of records. For example, large electronic
registries, such as immunization records, notifiable diseases, cancer diagnosis, and
billing and insurance, provide population-level data useful to researchers and are
often under the charter of local or regional health departments (except billing and
insurance that may be from private entities). Although registries are not explicitly
discussed, they can be thought of as special cases of EHRs: the patient has a specific
encounter with a provider that leads to data generated in an electronic system and
is a process dependent upon the provider encounter.

EHRs and Big Data

Big data is an abstract concept. It is not new, although the term may be. Data tend
to become “big” when they have certain properties and, depending on the institu-
tional EHR, may or may not fulfill the theoretical definition. Although this book is not
intended to offer a new definition of big data or explore its theoretical underpinnings,
there is a need to ground big data in theory, so as researchers we can be consistent in
its definition.
The META Group, now part of the Garner Group (Stamford, Connecticut),
offered a definition in 2001 centered around three “Vs” of big data: volume, velocity,
and variety.9 The definition has since been adopted by many and altered to incorporate
additional Vs. In the definition’s pure form, the three Vs are defined as follows:
1. Volume. In combination with data becoming cheap, it is also a commodity
necessary for business. Therefore, the volume of data has increased (and will
continue to increase) substantially.
4 ◾ Improving Population Health Using Electronic Health Records

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.

Stakeholders in the Research Process

Perhaps one of the biggest barriers to effective population health research involves
the complexity of the health-care system. There are a variety of stakeholders in
the research process, and each has his or her own agenda (Figure 1.1). Beginning
from the initial encounter into the system, the patient is usually concerned with
an acute or chronic health problem, or preventive maintenance. He or she may not
 Research in the Era of Electronic Health Records ◾ 5

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

is the institutional review board routinely made up of all identified stakeholders,

ensuring the research is ethical, lawful, and necessary.
As can be seen, this is a process that is highly dependent upon each previous
step, and initial errors, such as an ambiguous data request, can be further magni-
fied as the research progresses, akin to the telephone game played by children,
jeopardizing the results of the research. Further, each stakeholder in the process
has a highly specialized skill set. Efficiency in the process can be gained by hav-
ing public health researchers cross-trained in the core disciplines; for example,
clinicians who obtain epidemiology training, IT personnel who obtain clinical
training, and pubic health researchers who obtain IT training.
The goal of research in medicine and public health may differ depending on the
stakeholders included in the process, for example, a clinician wants to know how a
therapy or intervention will benefit his or her patient; a public health practitioner
wants to know an average population effect rather than a single patient effect; and
a health administrator within a hospital may want to know how a quality improve-
ment initiative have changed rates over time. Often our end goal is to establish
causality: a particular risk factor, or a set of factors, directly results in this outcome.
Yet, as will be discussed in Part III, causality is not always an attainable and practi-
cal goal; further, it may not matter for most interventions.

The Need for a Cohesive View of Research

Given the complexity of the data and variety of stakeholders in the system, the
traditional paradigm of research as its own stand-alone process needs to change.
Collaborations are given in scientific inquiry and should be viewed as crucial to
success of a given line of research. Without a clear understanding of disease pro-
cess (or any public health outcome) and direction for the research, the right data
will not be retrieved from the underlying electronic database. Once an initial
research question is formed and testable hypotheses are constructed, the correct
stakeholders need to be identified, as they may be best suited to assess the feasibil-
ity of the research.
The remainder of this book serves as a practical, how-to guide assuming a
research question is available and health data exist to answer this question. The
approach to the subject matter is divided into three distinct components: under-
standing the data (Part I), conducting the research (Part II), and interpretation and
implementation (Part III). Part I reviews the basic architecture of EHRs, focus-
ing on the database as the underlying entity, and then moves into challenges in
accessing the data including data extraction and linkage issues. This is a crucial
step in the research continuum, and unfortunately is only viewed as a “black box,”
that is, a request for data is made and data are provided that may or may not be
correct. This results in a dataset missing data on some key factors while containing
 Research in the Era of Electronic Health Records ◾ 7

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.

Fin del primer canto del gallo.

 NOTAS:
[298] En el códice de Gayangos aguadero.
[299] En el códice de La Romana se añade, á modo de apostilla,
pero de la misma letra: «y agora que son lutheranos no diffieren
de la gentilidad».
[300] La indicación del año que parece un paréntesis está en el
códice de Gayangos, pero falta en el de La Romana.
[301] Falta la palabra tálamo en el códice de La Romana.
[302] En el códice de La Romana ataviados.
[303] En el mismo códice agraciaban.
[304] En el códice de Gayangos dice sólo que «colgaban de los
ramos».
[305] En la Romana «y de allí la truxieron los de esta ciudad por
cosa admirable, y la daban agora al que fuese triunfoso en esta
fiesta y desafío».
[306] Así en La Romana. En Gayangos «vive Dios».
 ARGUMENTO
DEL SEGUNDO
CANTO DEL
GALLO

En el segundo canto que se

sigue, el auctor imita a
Plutarco en vn dialogo que
hizo entre Ulixes y vn griego
llamado Grilo; el qual auia
Cyrçes conuertido en puerco.
En esto el auctor quiere dar a
entender, que quando los
hombres estan encenagados
en los viçios y prinçipalmente
de la carne son muy peores
que brutos, y avn ay muchas
fieras que sin comparaçion los
exceden en el vso de la virtud.

Gallo.—Ya parece, Miçilo, que

es hora conueniente para
començar a vibir, dando gracias a
dios que ha tenido por bien de
passar la noche sin nuestro
peligro, y traernos al dia para que
con nuestra buena industria nos
podamos todos mantener.
Miçilo.—Bendito sea dios que
ansi lo ha permitido. Pero dime,
gallo, es esta tu primera cancion?
Porque holgaria de dormir vn
poco más hasta que cantes
segunda vez.
Gallo.—No te engañes, Miçilo,
que ya canté a la media noche
como acostumbramos, y como
estauas sepultado en la
profundidad y dulçura del primer
sueño, no te bastaron despertar
mis bozes, puesto caso que
trabajé por cantar lo mas
templado y bien comedido que
pude por no te desordenar en tu
suave dormir. Por la fortaleza
deste primer sueño creo yo que
llamaron los antiguos al dormir
ymagen de la muerte, y por su
dulçura le dixeron los poetas,
apazible holganza de los dioses.
Agora ya será casi el dia, que no
ay dos horas de la noche por
passar, despierta que yo quiero
prosseguir en mi obligaçion.
Miçilo.—Pues dizes ser essa
hora yo me quiero leuantar al
trabajo, porque proueyendo a
nuestro remedio y hambre, oyrte
me sera solaz. Agora di tu.
Gallo.—En el canto passado
quedé de te mostrar la bondad y
sosiego de la vida de las fieras, y
avn la ventaja que en su natural
hazen a los hombres. Esto
mostraré ser verdad en tanta
manera que podria ser, que si
alguna dellas diessen libertad de
quedar en su ser, o venir a ser
hombre como vos, escogeria
quedar fiera, puerco, lobo o leon
antes que venir a ser hombre, por
ser entre todos los animales la
especie mas trabajada y infeliz.
Mostrarte he el órden y conçierto
de su vibir, tanto que te
conuenças afirmar ser en ellas
verdadero vso de razon, por lo
qual las fieras sean dignas de ser
en mas tenidas, elegidas y
estimadas que los hombres.
Miçilo.—Parece, gallo, que con
tu eloquençia y manera de dezir
me quieres encantar, pues te
profieres a me mostrar vna cosa
tan lexos de verdadera y natural
razon. Temo me que en eso te
atreues a mi presumiendo que
facilmente como a pobre çapatero
qualquiera cosa me podras
persuadir. Agora pues
desengañate de oy mas que
confiado de mi naturaleza yo me
profiero a te lo defender. Di, que
me plazerá mucho oyr tus
sophisticos argumentos.
Gallo.—Por çierto yo espero que
no te parezcan sophisticos, sino
muy en demostraçion.
Prinçipalmente que no me podras
negar que yo mejor que quantos
ay en el mundo lo sabré mostrar,
pues de ambas naturalezas de
fiera y hombre tengo hecha
esperiencia. Pues agora
pareceme a mi que el prinçipio de
mi prueba se deue tomar de las
virtudes, justiçia, fortaleza,
prudençia, continençia y castidad,
de las quales vista la perfeçion
con que las vsan y tratan las
fieras conoçeras claramente no
ser manera de dezir lo que he
propuesto, mas que es muy
aueriguada verdad. Y quanto a lo
primero quiero que me digas; si
huviesse dos tierras, la vna de las
quales sin ser arada, cabada ni
sembrada, ni labrada, por sola su
bondad y generosidad de buena
naturaleza lleuasse todas las
frutas, flores y miesses muy en
abundancia? Dime, no loarias
más a esta tal tierra, y la
estimarias y antepornias a otra, la
qual por ser montuosa y para solo
pasto de cabras avn siendo
arada, muy rompida, cabada y
labrada con dificultad diesse fruto
poco y miserable?
Miçilo.—Por çierto avnque toda
tierra que da fruto avnque
trabajadamente es de estimar, de
mucho mas valor es aquella que
sin ser cultivada, o aquella que
con menos trabajo nos comunica
su fruto.
Gallo.—Pues de aqui se puede
sacar y colegir como de sentençia
de prudente y cuerdo, que ay
cosas que se han de loar y
aprobar por ser buenas, y otras
por muy mejores se han de
abraçar, amar y elegir. Pues ansi
de esta manera verdaderamente
y con necesidad me conçederas
que avnque el ánima del hombre
sea de gran valor, el ánima de la
fiera es mucho más; pues sin ser
rompida, labrada, arada ni
cabada; quiero dezir, sin ser
enseñada en otras escuelas ni
maestros que de su mesma
naturaleza es mas abil, presta y
aparejada a produçir en
abundançia el fruto de la virtud.
Miçilo.—Pues dime agora tú,
gallo, de qual virtud se pudo
nunca adornar el alma del bruto,
porque pareze que contradize a la
naturaleza de la misma virtud?
Gallo.—Y eso me preguntas?
Pues yo te probaré que la vsan
mejor que el más sabio varon.
Porque lo veas vengamos primero
a la virtud de fortaleza de la qual
vosotros, y principalmente los
españoles entre todas las
naciones, os gloriais y honrrais.
Quan vfanos y por quan gloriosos
os teneis quando os oys nombrar
atreuidos saqueadores de
çiudades, violadores de templos,
destruidores de hermosos y
sumptuosos edifiçios, disipadores
y abrasadores de fertiles campos
y miesses? Con los quales
exerçiçios de engaños y cautelas
aueis adquirido falso titulo y
renombre entre los de vuestro
tiempo de animosos y esforçados,
y con semejantes obras os aueis
usurpado el nombre de virtud.
Pero no son ansi las contiendas
de las fieras, porque si han de
pelear entre si o con vosotros,
muy sin engaños y cautelas lo
hazen, abierta y claramente las
verás pelear con sola confiança
de su esfuerço. Prinçipalmente
porque sus batallas no estan
subjetas a leyes que obliguen a
pena al que desamparare el
campo en la pelea. Pero como
por sola su naturaleza temen ser
vencidos trabajan quanto pueden
hasta vencer a su enemigo avn
que no obligan el cuerpo ni sus
animos a subjeçion ni vasallaje
siendo vencidas. Y ansi la
vençida siendo herida cayda en el
suelo es tan grande su esfuerço
que recoxe el animo en vna
pequeña parte de su cuerpo y
hasta que es del todo muerta
resiste a su matador. No hay
entre ellas los ruegos que le
otorgue la vida; no suplicaciones
lagrimas ni petiçiones de
misericordia; ni el rendirse al
vençedor confesandole la vitoria,
como vosotros hazeis quando os
tiene el enemigo a sus pies
amenaçandoos degollar. Nunca tú
viste que vn leon vençido sirua a
otro leon vençedor, ni vn cauallo a
otro, ni entre ellos ay temor de
quedar con renombre de
cobardes. Qualesquiera fieras
que por engaños o cautelas
fueron alguna vez presas en lazos
por los caçadores, si de edad
razonable son, antes se dexarán
de hambre y de sed morir que ser
otra vez presas y captiuas si en
algun tiempo pudieran gozar de la
libertad. Aunque algunas vezes
aconteçe que siendo algunas
presas siendo pequeñas se
vienen a amansar con regalos y
apazibles tratamientos, y ansi
aconteçe darseles por largos
tiempos en seruidumbre a los
hombres. Pero si son presas en
su vejez o edad razonable antes
moriran que subjetarseles. De lo
qual todo claramente se muestra
ser las fieras naturalmente
naçidas para ser fuertes y vsar de
fortaleza, y que los hombres vsan
contra verdad de titulo de fuertes
que ellos tienen usurpado
diziendo que les venga de su
naturaleza, y avn esto façilmente
se verá si consideramos vn
prinçipio de philosophia que es
vniuersalmente verdadero; y es,
que lo que conuiene por
naturaleza a vna especie
conuiene a todos los indiuiduos y
particulares igual y
indiferentemente. Como acontece
que conuiene a los hombres por
su naturaleza la risa, por la qual a
qualquiera honbre en particular
conuiene reyrse. Dime agora,
Miçilo, antes que passe adelante,
si ay aqui alguna cosa que me
puedas negar?
Miçilo.—No porque veo por
esperiençia que no ay honbre en
el mundo que no se rya y pueda
reyr; y solo el honbre propiamente
se rye. Pero yo no sé a que
proposito lo dizes.
Gallo.—Digolo porque pues esto
es verdad y vemos que
igualmente en las fieras en
fortaleça y esfuerço no diffieren
machos y hembras, pues
igualmente son fuertes para se
defender de sus enemigos, y para
sufrir los trabajos neçesarios por
defender sus hijos, o por vuscar
su mantenimiento, que
claramente pareçe conuenirles de
su naturaleza. Porque ansi
hallarás de la hembra tigre, que si
a caso fue a vuscar de comer
para sus hijos que los tenia
pequeños y en el entretanto que
se ausentó de la cueua vinieron
los cazadores y se los lleuaron;
diez y doze leguas sigue a su
robador y hallado haze con él tan
cruda guerra que veynte honbres
no se le igualaran en esfuerço. Ni
tampoco para esto aguardan
favorecerse de sus maridos, ni
con lagrimas se les quexan
contándoles su cuyta como hazen
vuestras hembras. Ya creo que
habrás oydo de la puerca de
Calidonia quantos trabajos y
fatigas dio al fuerte Theseo con
sus fuertes peleas. Que dire de
aquel sphinge de Pheniçia y de la
raposa telmesia? Que de aquella
famosa serpiente que con tanto
esfuerço peleó con Apolo?
Tambien creo que tú abrás visto
muchas leonas y osas mucho
mas fuertes que los machos en su
naturaleza. Y no se han como
vuestras mugeres las quales
quando vosotros estais en lo mas
peligroso de la guerra estan ellas
muy descuidadas de vuestro
peligro sentadas al fuego, o en el
regalo de sus camas y deleytes.
Como aquella Reyna Clithenestra,
que mientra su marido Agamenon
estaua en la guerra de troya
gozaua ella de los bessos y
abraços de su adultero Egisto. De
manera que de lo que tengo dicho
pareçeme no ser verdad, no ser
natural la fortaleza a los hombres,
porque si ansi fuesse igualmente
conuernia el esfuerço a las
henbras de vuestra espeçie, y se
hallaria como en los machos
como aconteçe en las fieras. Ansi
que podemos dezir, que los
honbres no de su voluntad, mas
forjados de vuestras leyes y de
vuestros principes y mayores
venis a exercitaros en esfuerço,
porque no osais yr contra su
mandado temiendo grandes
penas. Y estando los honbres en
el peligro más fragoso del mar, el
que primero en la tenpestad se
mueue no es para tomar el mas
pesado remo y trabajar doblado;
pero cada qual procura yr primero
por escoger el mas ligero y dexar
para los de la postre la mayor
carga, y avn del todo la reusarian
sino fuesse por miedo del castigo,
o peligro en que se ven. Y ansi
este tal no se puede dezir
esforzado, ni este se puede
gloriar ser doctado desta virtud,
porque aquel que se defiende de
su enemigo con miedo de reçebir
la muerte este tal no se deue
dezir magnanimo ni esforçado
pero cobarde y temeroso. Desta
manera aconteçe en vosotros
llamar fortaleza lo que bien
mirado con prudencia es
verdadera cobardia. Y si vosotros
os hallais ser mas esforçados que
las fieras, por qué vuestros
poetas y historiadores quando
escriuen y decantan vuestras
hazañas y hechos en la guerra os
comparan con los leones, tigres y
onzas, y por gran cosa dizen que
igualastes en esfuerço con ellos?
Y por el contrario nunca en las
batallas de las fieras fueran en su
ánimo comparadas con algun
hombre. Pero ansi como aconteçe
que comparamos los ligeros con
los vientos, y a los hermosos con
los angeles, queriendo hazer
semejantes los nuestros con las
cosas que exceden sin alguna
medida ni tasa: ansi parece que
desta manera comparais los
honbres en vuestras historias en
fortaleza con las fieras como a
cosas que exceden sin
comparaçion. Y la causa desto es,
porque como la fortaleza sea vna
virtud que consiste en el buen
gouierno de las passiones y
impetus del animo, el qual más
sincero y perfecto se halla en las
peleas que entre si tienen las
fieras. Porque los hombres
turbada la razon con la yra y la
soberuia los ciega y desbarata
tanto la colera que ninguna cosa
hazen con libertad que merezca
nombre de virtud. Avn con todo
esto quiero dezir que no teneis
porqué os quexar de naturaleza
porque no os diese vñas,
colmillos, conchas y otras armas
naturales que dio a las fieras para
su defensa, pues que vn
entendimiento de que os armó
para defenderos de vuestros
enemigos le enbotais y
entorpeçeis por vuestra culpa y
negligençia.
Miçilo.—O gallo, quan admirable
maestro me has sido oy de
Retorica, pues con tanta
abundançia de palabras has
persuadido tu proposito avn en
cosa tan seca y esteril. Forçado
me has a creer que hayas sido en
algun tiempo vno de los famosos
philosophos que obo en las
escuelas de athenas.
Gallo.—Pues mira, Micilo, que
por pensar yo que querias
redarguirme lo que tengo dicho
con algunos argumentos, o con
algun genero de contradiçion no
pasaua adelante en mi dezir. Y ya
que veo que te vas conuenciendo
quiero que pasemos a otra virtud,
y luego quiero que tratemos de la
castidad. En la qual te mostraré
que las fieras exçeden a los
hombres sin alguna comparaçion.
Mucho se preçian vuestras
mugeres tener de su parte por
exemplo de castidad vna
Penelope, vna Lucreçia Porçia,
Doña Maria de Toledo, y doña
Ysabel Reyna de Castilla; porque
dezis que estas menospreçiauan
sus vidas por no violar la virtud de
su castidad. Pues yo te mostraré
muchas fieras castas mil vezes
mas que todas esas vuestras, y
no quiero que comencemos por la
castidad de la corneja, ni Croton,
admirables fieras en este caso,
que despues de sus maridos
muertos guardan la viudez no
qualquiera tiempo, pero nueue
hedades de hombres sin ofender
su castidad. Por lo qual
neçesariamente me deues
conçeder ser estas fieras nueue
vezes mas castas que las
vuestras mugeres que por
exemplo teneis. Pero porque
tienes entendido de mí, Miçilo,
que soy retorico, quiero que
procedamos en el discurso desta
virtud segun las leyes de
Retorica, porque por ellas espero
vençerte con mas façilidad, Y ansi
primero veamos la difiniçion desta
virtud continençia, y despues
deçenderemos a sus inferiores
espeçies. Suelen dezir los
philosophos, que la virtud de
continençia es vna buena y çierta
dispusiçion y regla de los
deleytes, por la qual se desechan
y huyen los malos, vedados y
superfluos y se faboreçen y
allegan los neçesarios y naturales
en sus conuenientes tiempos.
Quanto a lo primero vosotros los
hombres todos los sentidos
corporales corrompeis y
deprabais con vuestros malos
vsos y costumbres y
inclinaciones, endereçandolos
sienpre a vuestro viçioso deleyte
y luxuria. Con los ojos todas las
cosas que veis endereçais para
vuestra laçiuia y cobdiçia. lo qual
nosotras las fieras no hazemos
ansi. Porque quando yo era
hombre me holgaua y regoçijaua
con gran deleyte viendo el oro,
joyas y piedras preçiosas, a tanto
que me andaua bobo y
desbaneçido vn dia tras vn Rey o
principe si anduuiesse vestido y
adornado de jaezes y atauios de
seda, oro, purpura y hermosos
colores. Pero agora, como lo
hacen las otras fieras, no estimo
yo en más todo eso que al lodo y
a otras comunes piedras que ay
por las pedregosas y asperas
syerras y montañas. Y ansi
quando yo era puerco estimaua
mucho más sin comparaçion
hallar algun blando y humido
cieno, o piçina en que me
refrescasse rebolcandome. Pues
si venimos al sentido del oler, si
consideramos aquellos olores
suaues de gomas, espeçias y
pastillas de que andais siempre
oliendo, regalando y afeminando
vuestras personas. En tanta
manera que ningun varon de
vosotros viene a gozar de su
propia muger si primero no se
vnta con vnçiones delicadas y
odoriferas, con las quales
procurais inçitar y despertar en
vosotros a venus. Y esto todo avn
seria sufridero en vuestras
hembras por daros deleyte usar
de aquellos olores laboratorios,
afeytes y vnturas; pero lo que
peor es que lo vsais vosotros los
varones para incitaros a luxuria.
Pero nosotras las fieras no lo
vsamos ansi, sino el lobo con la
loba, y el leon con la leona, y ansi
todos los machos con sus
hembras en su genero y espeçie
gozan de sus abraços y açessos
solamente con los olores
naturales y proprios que a sus
cuerpos dio su naturaleza sin
admistion de otro alguno de fuera.
Quando mas ay, y con que ellas
mas se deleytan es al olor que
produçen de si los olorosos
prados quando en el tiempo de su
brama, que es quando vsan sus
bodas, estan verdes y floridos y
hermosos. Y ansi ninguna hembra
de las nuestras tiene necesidad
para sus ayuntamientos de
afeytes ni vnturas para engañar y
traer al macho de su especie. Ni
los machos tienen neçesidad de
las persuadir con palabras,
requiebros, cautelas ni
ofreçimientos. Pero todos ellos en
su propio tiempo sin engaños ni
intereses hazen sus
ayuntamientos atsaydos por
naturaleza con las dispusiçiones y
concurso del tiempo, como los
quales son inçitados y llamados a
aquello. Y ansi este tiempo siendo
passado, y hechas sus preñezes,
todos se aseguran y mortiguan en
su incentiuo deleyte, y hasta la
buelta de aquel mesmo tiempo
ninguna hembra cobdiçia ni
consiente al macho, ni el macho
la acomete. Ningun otro interese
se pretende en las fieras sino el
engendrar y todo lo guiamos y
ordenamos como nuestra
naturaleza lo dispone. Y añade á
esto que entre las fieras en
ningun tiempo se cobdiçia ni
soliçita ni acomete hembra a
hembra, ni macho con macho en
açesso carnal. Pero vosotros los
hombres no ansi, porque no os
perdonais vnos a otros; pero
muger con muger, y hombre con
hombre contra las leyes de
vuestra naturaleza, os juntais, y
en vuestros carnales açessos os
toman vuestros juezes cada dia.
Ni por esto temeis la pena, quanto
quiera que sea cruel, por
satisfazer y cumplir uuestro
deleyte y luxuria. En tanta manera
es esto aborreçido de las fieras,
que si vn gallo cometiese açesso
con otro gallo, avn que le faltasse
gallina, con los picos y vñas le
hariamos en breue pedaços.
Pareçe, micilo, que te bas
conuençiendo y haciendote de mi
sentencia, pues tanto callas sin
me contradezir.
Miçilo.—Es tan efficaz, gallo, tu
persuasion, que como vna
cadena me llevas tras ti sin poder
resistir.
Gallo.—Dexemos de contar
quantos varones han tenido sus
ayuntamientos con cabras, ouejas
y perras; y las mugeres que han
effectuado su lexuria con gimios,
asnos, cabrones y perros: de los
quales açessos se han
engendrado çentauros, sphinges,
minotauros y otros admimirables
monstruos de prodigioso aguero.
Pero las fieras nunca vsaron ansi,
como lo muestra por exemplo la
continencia de aquel famoso
mendesio, cabron egipcio, que
siendo encerrado por muchas
damas hermosas para que
holgase con ellas, ofreçiéndosele
desnudas delante, las
menospreçio, y quando se pudo
soltar se fué huyendo á la
montaña á tener sus plazeres con
las cabras sus semejantes. Pues
quanto ves que son mas
inferiores en la castidad los
hombres que las fieras, ansi lo