Quick Guide to Bivariate Statistical Tests
Acquisitions Editor: Christina Burns Product Development Editor: Katherine Burland Editorial Assistant: Cassie Berube Marketing Manager: Dean Karampelas Production Project Manager: Cynthia Rudy Design Coordinator: Joan Wendt Manufacturing Coordinator: Karin Duffield Prepress Vendor: Absolute Service, Inc.
Copyright © 2017 Wolters Kluwer.
Copyright © 2012 Wolters Kluwer Health | Lippincott Williams & Wilkins. Copyright © 2008, 2004, 1999 by Lippincott Williams & Wilkins. Copyright © 1995, 1991, 1987, 1983, 1978 by J. B. Lippincott Company.
All rights reserved. This book is protected by copyright. No part of this book may be reproduced or transmitted in any form or by any means, including as photocopies or scanned-in or other electronic copies, or utilized by any information storage and retrieval system without written permission from the copyright owner, except for brief quotations embodied in critical articles and reviews. Materials appearing in this book prepared by individuals as part of their official duties as U.S. government employees are not covered by the above-mentioned copyright. To request permission, please contact Wolters Kluwer at Two Commerce Square, 2001 Market Street, Philadelphia, PA 19103, via email at [email protected], or via our website at lww.com (products and services).
9 8 7 6 5 4 3 2 1
Printed in China
Library of Congress Cataloging-in-Publication Data
Polit, Denise F., author. Nursing research : generating and assessing evidence for nursing practice / Denise F. Polit, Cheryl Tatano Beck. — Tenth edition.
p. ; cm. Includes bibliographical references and index. ISBN 978-1-4963-0023-2 I. Beck, Cheryl Tatano, author. II. Title. [DNLM: 1. Nursing Research—methods. WY 20.5]
This work is provided “as is,” and the publisher disclaims any and all warranties, express or implied, including any warranties as to accuracy, comprehensiveness, or currency of the content of this work.
This work is no substitute for individual patient assessment based on healthcare professionals’ examination of each patient and consideration of, among other things, age, weight, gender, current or prior medical conditions, medication history, laboratory data, and other factors unique to the patient. The publisher does not provide medical advice or guidance, and this work is merely a reference tool. Healthcare professionals, and not the publisher, are solely responsible for the use of this work including all medical judgments and for any resulting diagnosis and treatments.
Given continuous, rapid advances in medical science and health information, independent professional verification of medical diagnoses, indications, appropriate pharmaceutical selections and dosages, and treatment options should be made and healthcare professionals should consult a variety of sources. When prescribing medication, healthcare professionals are advised to consult the product information sheet (the manufacturer’s package insert) accompanying each drug to verify, among other things, conditions of use, warnings, and side effects and identify any changes in dosage schedule or contraindications, particularly if the medication to be administered is new, infrequently used, or has a narrow therapeutic range. To the maximum extent permitted under applicable law, no responsibility is assumed by the publisher for any injury and/or damage to persons or property, as a matter of products liability, negligence law or otherwise, or from any reference to or use by any person of this work.
Our Beloved Family: Our Husbands, Our Children (Spouses/Fiancés), and Our Grandchildren
Husbands: Alan Janosy and Chuck Beck
Children: Alex (Maryanna), Alaine (Jeff), Lauren (Vadim), and Norah (Chris); and Curt and Lisa
Grandchildren: Cormac, Julia, Maren, and Ronan
This 10th edition, like the previous nine editions, depended on the contribution of dozens of people. Many faculty and students who used the text have made invaluable suggestions for its improvement, and to all of you we are very grateful. In addition to all those who assisted us during the past 35 years with the earlier editions, the following individuals deserve special mention.
We would like to acknowledge the comments of reviewers of the previous edition of this book, anonymous to us initially, whose feedback influenced our revisions. Faculty at Griffith University in Australia made useful suggestions and also inspired the inclusion of some new content. Valori Banfi, reference librarian at the University of Connecticut, provided ongoing assistance. Dr. Deborah Dillon McDonald was extraordinarily generous in giving us access to her NINR grant application and related material for the Resource Manual.
We also extend our thanks to those who helped to turn the manuscript into a finished product. The staff at Wolters Kluwer has been of great assistance to us over the years. We are indebted to Christina Burns, Kate Burland, Cynthia Rudy, and all the others behind the scenes for their fine contributions.
Finally, we thank our family and friends. Our husbands Alan and Chuck have become accustomed to our demanding schedules, but we recognize that their support involves a lot of patience and many sacrifices.
Ellise D. Adams, PhD, CNM Associate Professor The University of Alabama in Huntsville Huntsville, Alabama
Jennifer Bellot, PhD, RN, MHSA Associate Professor and Director, DNP Program Thomas Jefferson University Philadelphia, Pennsylvania
Kathleen D. Black, PhD, RNC Assistant Professor, Jefferson College of Nursing Thomas Jefferson University Philadelphia, Pennsylvania
Dee Campbell, PhD, APRN, NE-BC, CNL Professor, Graduate Department Felician College, School of Nursing Lodi, New Jersey
Patricia Cannistraci, DNS, RN, CNE Assistant Dean Excelsior College Albany, New York
Julie L. Daniels, DNP, CNM Assistant Professor Frontier Nursing University
Rebecca Fountain, PhD, RN Associate Professor University of Texas at Tyler Tyler, Texas
Teresa S. Johnson, PhD, RN Associate Professor, College of Nursing University of Wisconsin—Milwaukee Milwaukee, Wisconsin
Jacqueline Jones, PhD, RN, FAAN Associate Professor, College of Nursing University of Colorado, Anschutz Medical Campus Aurora, Colorado
Mary Lopez, PhD, RN Associate Dean, Research Western University of Health Sciences Pomona, California
Audra Malone, DNP, FNP-BC Assistant Professor Frontier Nursing University Hyden, Kentucky
Sharon R. Rainer, PhD, CRNP Assistant Professor, Jefferson College of Nursing Thomas Jefferson University Philadelphia, Pennsylvania
Maria A. Revell, PhD, RN Professor of Nursing Middle Tennessee State University Murfreesboro, Tennessee
Stephanie Vaughn, PhD, RN, CRRN Interim Director, School of Nursing California State University, Fullerton Fullerton, California
Research methodology is not a static enterprise. Even after writing nine editions of this book, we continue to draw inspiration and new material from groundbreaking advances in research methods and in nurse researchers’ use of those methods. It is exciting and uplifting to share many of those advances in this new edition. We expect that many of the new methodologic and technologic advances will be translated into powerful evidence for nursing practice. Five years ago, we considered the ninth edition as a watershed edition of a classic textbook. We are persuaded, however, that this 10th edition is even better. We have retained many features that made this book a classic textbook and resource, including its focus on research as a support for evidence-based nursing, but have introduced important innovations that will help to shape the future of nursing research.
NEW TO THIS EDITION
New Chapters We have added two new chapters on “cutting-edge” topics that are not well covered in any major research methods textbook, regardless of discipline. The first is a chapter on an issue of critical importance to health professionals and yet inadequately addressed in the nursing literature: the clinical significance of research findings. In Chapter 20, we discuss various conceptualizations of clinical significance and present methods of operationalizing those conceptualizations so that clinical significance can be assessed at both the individual and group level. We believe that this is a “must-read” chapter for nurses whose research is designed to inform clinical practice. The second new chapter in this edition concerns the
design and conduct of pilot studies. In recent years, experts have written at length about the poor quality of many pilot studies. Chapter 28 provides guidance on how to develop pilot study objectives and draw conclusions about the appropriate next step—that is, whether to proceed to a full-scale study, make major revisions, or abandon the project. This chapter is included in Part 5 of this book, which is devoted to mixed methods research, because pilots can benefit from both qualitative and quantitative evidence.
New Content Throughout the book, we have included material on methodologic innovations that have arisen in nursing, medicine, and the social sciences during the past 4 to 5 years. The many additions and changes are too numerous to describe here, but a few deserve special mention. In particular, we have totally revised the chapters on measurement (Chapter 14) and scale development (Chapter 15) to reflect emerging ideas about key measurement properties and the assessment of newly developed instruments.
The inclusion of two new chapters made it challenging to keep the textbook to a manageable length. Our solution was to move some content in the ninth edition to supplements that are available online. In fact, every chapter has an online supplement, which gave us the opportunity to add a considerable amount of new content. For example, one supplement is devoted to evidence-based methods to recruit and retain study participants. Other supplements include a description of various randomization methods, an overview of item response theory, guidance on wording proposals to conduct pilot studies, and a discussion of quality improvement studies. Following is a complete list of the supplements for the 31 chapters of this textbook:
1. The History of Nursing Research 2. Evaluating Clinical Practice Guidelines—AGREE II 3. Deductive and Inductive Reasoning 4. Complex Relationships and Hypotheses 5. Literature Review Matrices 6. Prominent Conceptual Models of Nursing Used by Nurse Researchers,
and a Guide to Middle-Range Theories 7. Historical Background on Unethical Research Conduct 8. Research Control 9. Randomization Strategies 10. The RE-AIM Framework 11. Other Specific Types of Research 12. Sample Recruitment and Retention 13. Other Types of Structured Self-Reports 14. Cross-Cultural Validity and the Adaptation/Translation of Measures 15. Overview of Item Response Theory 16. SPSS Analysis of Descriptive Statistics 17. SPSS Analysis of Inferential Statistics 18. SPSS Analysis and Multivariate Statistics 19. Some Preliminary Steps in Quantitative Analysis Using SPSS 20. Clinical Significance Assessment with the Jacobson-Truax Approach 21. Historical Nursing Research 22. Generalizability and Qualitative Research 23. Additional Types of Unstructured Self-Reports 24. Transcribing Qualitative Data 25. Whittemore and Colleagues’ Framework of Quality Criteria in
Qualitative Research 26. Converting Quantitative and Qualitative Data 27. Complex Intervention Development: Exploratory Questions 28. Examples of Various Pilot Study Objectives 29. Publication Bias in Meta-Analyses 30. Tips for Publishing Reports on Pilot Intervention Studies 31. Proposals for Pilot Intervention Studies
Another new feature of this edition concerns our interest in readers’ access to references we cited. To the extent possible, the studies we have chosen as examples of particular research methods are published as open- access articles. These studies are identified with an asterisk in the reference list at the end of each chapter, and a link to the article is included in the Toolkit section of the Resource Manual. We hope that these revisions will help users of this book to maximize their learning experience.
ORGANIZATION OF THE TEXT The content of this edition is organized into six main parts.
• Part I—Foundations of Nursing Research and Evidence-Based Practice introduces fundamental concepts in nursing research. Chapter 1 briefly summarizes the history and future of nursing research, discusses the philosophical underpinnings of qualitative research versus quantitative research, and describes major purposes of nursing research. Chapter 2 offers guidance on utilizing research to build an evidence- based practice. Chapter 3 introduces readers to key research terms and presents an overview of steps in the research process for both qualitative and quantitative studies.
• Part II—Conceptualizing and Planning a Study to Generate Evidence further sets the stage for learning about the research process by discussing issues relating to a study’s conceptualization: the formulation of research questions and hypotheses (Chapter 4), the review of relevant research (Chapter 5), the development of theoretical and conceptual contexts (Chapter 6), and the fostering of ethically sound approaches in doing research (Chapter 7). Chapter 8 provides an overview of important issues that researchers must attend to during the planning of any type of study.
• Part III—Designing and Conducting Quantitative Studies to Generate Evidence presents material on undertaking quantitative nursing studies. Chapter 9 describes fundamental principles and applications of quantitative research design, and Chapter 10 focuses on methods to enhance the rigor of a quantitative study, including mechanisms of research control. Chapter 11 examines research with different and distinct purposes, including surveys, outcomes research, and evaluations. Chapter 12 presents strategies for sampling study participants in quantitative research. Chapter 13 describes using structured data collection methods that yield quantitative information. Chapter 14 discusses the concept of measurement and then focuses on methods of assessing the quality of formal measuring instruments. In this edition, we describe methods to assess the properties of point-in- time measurements (reliability and validity) and longitudinal measurements—change scores (reliability of change scores and
responsiveness). Chapter 15 presents material on how to develop high- quality self-report instruments. Chapters 16, 17, and 18 present an overview of univariate, bivariate, and multivariate statistical analyses, respectively. Chapter 19 describes the development of an overall analytic strategy for quantitative studies, including material on handling missing data. Chapter 20, a new chapter, discusses the issue of interpreting results and making inferences about clinical significance.
• Part IV—Designing and Conducting Qualitative Studies to Generate Evidence presents material on undertaking qualitative nursing studies. Chapter 21 is devoted to research designs and approaches for qualitative studies, including material on critical theory, feminist, and participatory action research. Chapter 22 discusses strategies for sampling study participants in qualitative inquiries. Chapter 23 describes methods of gathering unstructured self-report and observational data for qualitative studies. Chapter 24 discusses methods of analyzing qualitative data, with specific information on grounded theory, phenomenologic, and ethnographic analyses. Chapter 25 elaborates on methods qualitative researchers can use to enhance (and assess) integrity and quality throughout their inquiries.
• Part V—Designing and Conducting Mixed Methods Studies to Generate Evidence presents material on mixed methods nursing studies. Chapter 26 discusses a broad range of issues, including asking mixed methods questions, designing a study to address the questions, sampling participants in mixed methods research, and analyzing and integrating qualitative and quantitative data. Chapter 27 presents innovative information about using mixed methods approaches in the development of nursing interventions. In Chapter 28, a new chapter, we provide guidance for designing and conducting a pilot study and using data from the pilot to draw conclusions about how best to proceed.
• Part VI—Building an Evidence Base for Nursing Practice provides additional guidance on linking research and clinical practice. Chapter 29 offers an overview of methods of conducting systematic reviews that support EBP, with an emphasis on meta-analyses, metasyntheses, and mixed studies reviews. Chapter 30 discusses dissemination of evidence —how to prepare a research report (including theses and dissertations)
and how to publish research findings. The concluding chapter (Chapter 31) offers suggestions and guidelines on developing research proposals and getting financial support and includes information about applying for NIH grants and interpreting scores from NIH’s new scoring system.
KEY FEATURES This textbook was designed to be helpful to those who are learning how to do research as well as to those who are learning to appraise research reports critically and to use research findings in practice. Many of the features successfully used in previous editions have been retained in this 10th edition. Among the basic principles that helped to shape this and earlier editions of this book are (1) an unswerving conviction that the development of research skills is critical to the nursing profession, (2) a fundamental belief that research is intellectually and professionally rewarding, and (3) a steadfast opinion that learning about research methods need be neither intimidating nor dull. Consistent with these principles, we have tried to present the fundamentals of research methods in a way that both facilitates understanding and arouses curiosity and interest. Key features of our approach include the following:
• Research Examples. Each chapter concludes with one or two actual research examples designed to highlight critical points made in the chapter and to sharpen the reader’s critical thinking skills. In addition, many research examples are used to illustrate key points in the text and to stimulate ideas for a study. Many of the examples used in this edition are open-access articles that can be used for further learning and classroom discussions.
• Critiquing Guidelines. Most chapters include guidelines for conducting a critique of each aspect of a research report. These guidelines provide a list of questions that draw attention to specific aspects of a report that are amenable to appraisal.
• Clear, “user-friendly” style. Our writing style is designed to be easily digestible and nonintimidating. Concepts are introduced carefully and systematically, difficult ideas are presented clearly, and readers are assumed to have no prior exposure to technical terms.
• Specific practical tips on doing research. This textbook is filled with
practical guidance on how to translate the abstract notions of research methods into realistic strategies for conducting research. Every chapter includes several tips for applying the chapter’s lessons to real-life situations. These suggestions are in recognition of the fact that there is often a large gap between what gets taught in research methods textbooks and what a researcher needs to know to conduct a study.
• Aids to student learning. Several features are used to enhance and reinforce learning and to help focus the student’s attention on specific areas of text content, including the following: succinct, bulleted summaries at the end of each chapter; tables and figures that provide examples and graphic materials in support of the text discussion; study suggestions at the end of each chapter; a detailed glossary; and a comprehensive index for accessing information quickly.
TEACHING–LEARNING PACKAGE Nursing Research: Generating and Assessing Evidence for Nursing Practice, 10th edition, has an ancillary package designed with both students and instructors in mind.
• The Resource Manual augments the textbook in important ways. The manual itself provides students with exercises that correspond to each text chapter, with a focus on opportunities to critique actual studies. The appendix includes 12 research journal articles in their entirety, plus a successful grant application for a study funded by the National Institute of Nursing Research. The 12 reports cover a range of nursing research ventures, including qualitative, quantitative, and mixed methods studies, an instrument development study, an evidence-based practice translation project, and two systematic reviews. Full critiques of two of the reports are also included and can serve as models for a comprehensive research critique.
• The Toolkit to the Resource Manual is a “must-have” innovation that will save considerable time for both students and seasoned researchers. Included on thePoint, the Toolkit offers dozens of research resources in Word documents that can be downloaded and used directly or adapted. The resources reflect best-practice research material, most of which have been pretested and refined in our own research. The Toolkit
originated with our realization that in our technologically advanced environment, it is possible to not only illustrate methodologic tools as graphics in the textbook but also to make them directly available for use and adaptation. Thus, we have included dozens of documents in Word files that can readily be used in research projects, without requiring researchers to “reinvent the wheel” or tediously retype material from this textbook. Examples include informed consent forms, a demographic questionnaire, content validity forms, and a coding sheet for a meta-analysis—to name only a few. The Toolkit also has lists of relevant and useful websites for each chapter, which can be “clicked” on directly without having to retype the URL and risk a typographical error. Links to open-access articles cited in the textbook, as well as other open-access articles relevant to each chapter, are included in the Toolkit.
• The Instructor’s Resources on the Point include PowerPoint slides summarizing key points in each chapter, test questions that have been placed into a program that allows instructors to automatically generate a test, and an image bank.
It is our hope that the content, style, and organization of this book continue to meet the needs of a broad spectrum of nursing students and nurse researchers. We also hope that this book will help to foster enthusiasm for the kinds of discoveries that research can produce and for the knowledge that will help support an evidence-based nursing practice.
DENISE F. POLIT, PhD, FAAN
CHERYL TATANO BECK, DNSc, CNM, FAAN
PART 1: FOUNDATIONS OF NURSING RESEARCH Chapter 1: Introduction to Nursing Research in an Evidence-Based
Practice Environment Chapter 2: Evidence-Based Nursing: Translating Research Evidence into
Practice Chapter 3: Key Concepts and Steps in Qualitative and Quantitative
PART 2: CONCEPTUALIZING AND PLANNING A STUDY TO GENERATE EVIDENCE FOR NURSING
Chapter 4: Research Problems, Research Questions, and Hypotheses Chapter 5: Literature Reviews: Finding and Critiquing Evidence Chapter 6: Theoretical Frameworks Chapter 7: Ethics in Nursing Research Chapter 8: Planning a Nursing Study
PART 3: DESIGNING AND CONDUCTING QUANTITATIVE STUDIES TO GENERATE EVIDENCE FOR NURSING
Chapter 9: Quantitative Research Design Chapter 10: Rigor and Validity in Quantitative Research Chapter 11: Specific Types of Quantitative Research Chapter 12: Sampling in Quantitative Research Chapter 13: Data Collection in Quantitative Research Chapter 14: Measurement and Data Quality
Chapter 15: Developing and Testing Self-Report Scales Chapter 16: Descriptive Statistics Chapter 17: Inferential Statistics Chapter 18: Multivariate Statistics Chapter 19: Processes of Quantitative Data Analysis Chapter 20: Clinical Significance and Interpretation of Quantitative
PART 4: DESIGNING AND CONDUCTING QUALITATIVE STUDIES TO GENERATE EVIDENCE FOR NURSING
Chapter 21: Qualitative Research Design and Approaches Chapter 22: Sampling in Qualitative Research Chapter 23: Data Collection in Qualitative Research Chapter 24: Qualitative Data Analysis Chapter 25: Trustworthiness and Integrity in Qualitative Research
PART 5: DESIGNING AND CONDUCTING MIXED METHODS STUDIES TO GENERATE EVIDENCE FOR NURSING
Chapter 26: Basics of Mixed Methods Research Chapter 27: Developing Complex Nursing Interventions Using Mixed
Methods Research Chapter 28: Feasibility Assessments and Pilot Tests of Interventions Using
PART 6: BUILDING AN EVIDENCE BASE FOR NURSING PRACTICE
Chapter 29: Systematic Reviews of Research Evidence: Meta-Analysis, Metasynthesis, and Mixed Studies Review
Chapter 30: Disseminating Evidence: Reporting Research Findings Chapter 31: Writing Proposals to Generate Evidence
Appendix: Statistical Tables Index
Check Out the Latest Book Authored by Research Expert Dr. Polit If you want to make thoughtful but practical decisions about the measurement of health constructs, check out Dr. Polit and Dr. Yang’s latest book, a “gentle” introduction to and overview of complex measurement content, called Measurement and the Measurement of Change.
This book is for researchers and clinicians from all health disciplines because measurement is vital to high-quality science and to excellence in clinical practice. The text focuses on the measurement of health constructs, particularly those constructs that are not amenable to quantification by means of laboratory analysis or technical instrumentation. These health
constructs include a wide range of human attributes, such as quality of life, functional ability, self-efficacy, depression, and pain. Measures of such constructs are proliferating at a rapid rate and often without adequate attention paid to ensuring that standards of scientific rigor are met.
In this book, the authors offer guidance to those who develop new instruments, adapt existing ones, select instruments for use in a clinical trial or in clinical practice, interpret information from measurements and changes in scores, or undertake a systematic review on instruments. This book offers guidance on how to develop new instruments using both “classical” and “modern” approaches from psychometrics as well as methods used in clinimetrics. Much of this book, however, concerns the evaluation of instruments in relation to three key measurement domains: reliability, validity, and responsiveness.
This text was designed to be useful in graduate-level courses on measurement or research methods and will also serve as an important reference and resource for researchers and clinicians.
FOUNDATIONS OF NURSING RESEARCH
1 Introduction to Nursing Research in an Evidence-Based Practice Environment
NURSING RESEARCH IN PERSPECTIVE In all parts of the world, nursing has experienced a profound culture change. Nurses are increasingly expected to understand and conduct research and to base their professional practice on research evidence—that is, to adopt an evidence-based practice (EBP). EBP involves using the best evidence (as well as clinical judgment and patient preferences) in making patient care decisions, and “best evidence” typically comes from research conducted by nurses and other health care professionals.
What Is Nursing Research? Research is systematic inquiry that uses disciplined methods to answer questions or solve problems. The ultimate goal of research is to develop and expand knowledge.
Nurses are increasingly engaged in disciplined studies that benefit nursing and its clients. Nursing research is systematic inquiry designed to generate trustworthy evidence about issues of importance to the nursing profession, including nursing practice, education, administration, and informatics. In this book, we emphasize clinical nursing research, that is, research to guide nursing practice and to improve the health and quality of life of nurses’ clients.
Nursing research has experienced remarkable growth in the past three decades, providing nurses with a growing evidence base from which to practice. Yet many questions endure and much remains to be done to incorporate research innovations into nursing practice.
Examples of Nursing Research Questions:
• How effective is pressurized irrigation, compared to a swabbing method, in cleansing wounds, in terms of time to wound healing, pain, patients’ satisfaction with comfort, and costs? (Mak et al., 2015)
• What are the experiences of women in Zimbabwe who are living with advanced HIV infection? (Gona & DeMarco, 2015)
The Importance of Research in Nursing Research findings from rigorous studies provide especially strong evidence for informing nurses’ decisions and actions. Nurses are accepting the need to base specific nursing actions on research evidence indicating that the actions are clinically appropriate, cost-effective, and result in positive outcomes for clients.
In the United States, research plays an important role in nursing in terms of cred entialing and status. The American Nurses Credentialing Center (ANCC)—an arm of the American Nurses Association and the largest and most prestigious credentialing organization in the United States —developed a Magnet Recognition Program to acknowledge health care organizations that provide high-quality nursing care. As Reigle and her colleagues (2008) noted, “the road to Magnet Recognition is paved with EBP” (p. 102) and the 2014 Magnet application manual incorporated revisions that strengthened evidence-based requirements (Drenkard, 2013). The good news is that there is growing confirmation that the focus on research and evidence-based practice may have important payoffs. For example, McHugh and co-researchers (2013) found that Magnet hospitals have lower risk-adjusted mortality and failure to rescue than non-Magnet hospitals, even when differences among the hospitals in nursing credentials and patient characteristics are taken into account.
Changes to nursing practice now occur regularly because of EBP efforts. Practice changes often are local initiatives that are not publicized, but broader clinical changes are also occurring based on accumulating research evidence about beneficial practice innovations.
Example of Evidence-Based Practice: Numerous clinical practice changes
reflect the impact of research. For example, “kangaroo care” (the holding of diaper-clad infants skin to skin by parents) is now practiced in many neonatal intensive care units (NICUs), but this is a relatively new trend. As recently as the 1990s, only a minority of NICUs offered kangaroo care options. Expanded adoption of this practice reflects mounting evidence that early skin-to-skin contact has benefits without negative side effects (e.g., Ludington-Hoe, 2011; Moore et al., 2012). Some of that evidence came from rigorous studies conducted by nurse researchers in several countries (e.g., Chwo et al., 2002; Cong et al., 2009; Cong et al., 2011; Hake-Brooks & Anderson, 2008). Nurses continue to study the potential benefits of kangaroo care in important clinical trials (e.g., Campbell-Yeo et al., 2013).
The Consumer–Producer Continuum in Nursing Research In our current environment, all nurses are likely to engage in activities along a continuum of research participation. At one end of the continuum are consumers of nursing research, who read research reports or research summaries to keep up-to-date on findings that might affect their practice. EBP depends on well-informed nursing research consumers.
At the other end of the continuum are the producers of nursing research: nurses who design and conduct research. At one time, most nurse researchers were academics who taught in schools of nursing, but research is increasingly being conducted by nurses in health care settings who want to find solutions to recurring problems in patient care.
Between these end points on the continuum lie a variety of research activities that are undertaken by nurses. Even if you never personally undertake a study, you may (1) contribute to an idea or a plan for a clinical study; (2) gather data for a study; (3) advise clients about participating in research; (4) solve a clinical problem by searching for research evidence; or (5) discuss the implications of a new study in a journal club in your practice setting, which involves meetings (in groups or online) to discuss research articles. In all possible research001-related activities, nurses who have some research skills are better able than those without them to make a contribution to nursing and to EBP. An understanding of nursing research can improve the depth and breadth of every nurse’s professional practice.
Nursing Research in Historical Perspective Table 1.1 summarizes some of the key events in the historical evolution of nursing research. (An expanded summary of the history of nursing research appears in the Supplement to this chapter on ).
Most people would agree that research in nursing began with Florence Nightingale in the 1850s. Her most well-known research contribution involved an analysis of factors affecting soldier mortality and morbidity during the Crimean War. Based on skillful analyses, she was successful in effecting changes in nursing care and, more generally, in public health. After Nightingale’s work, research was absent from the nursing literature until the early 1900s, but most early studies concerned nurses’ education rather than clinical issues.
In the 1950s, research by nurses began to accelerate. For example, a
nursing research center was established at the Walter Reed Army Institute of Research. Also, the American Nurses Foundation, which is devoted to the promotion of nursing research, was founded. The surge in the number of studies conducted in the 1950s created the need for a new journal; Nursing Research came into being in 1952. As shown in Table 1.1, dissemination opportunities in professional journals grew steadily thereafter.
In the 1960s, nursing leaders expressed concern about the shortage of research on practice issues. Professional nursing organizations, such as the Western Interstate Council for Higher Education in Nursing, established research priorities, and practice-oriented research on various clinical topics began to emerge in the literature.
During the 1970s, improvements in client care became a more visible research priority and nurses also began to pay attention to the clinical utilization of research findings. Guidance on assessing research for application in practice settings became available. Several journals that focus on nursing research were established in the 1970s, including Advances in Nursing Science, Research in Nursing & Health, and the Western Journal of Nursing Research. Nursing research also expanded internationally. For example, the Workgroup of European Nurse Researchers was established in 1978 to develop greater communication and opportunities for partnerships among 25 European National Nurses Associations.
Nursing research continued to expand in the 1980s. In the United States, the National Center for Nursing Research (NCNR) at the National Institutes of Health (NIH) was established in 1986. Several forces outside of nursing also helped to shape the nursing research landscape. A group from the McMaster Medical School in Canada designed a clinical learning strategy that was called evidence-based medicine (EBM). EBM, which promulgated the view that research findings were far superior to the opinions of authorities as a basis for clinical decisions, constituted a profound shift for medical education and practice, and has had a major effect on all health care professions.
Nursing research was strengthened and given more visibility when NCNR was promoted to full institute status within the NIH. In 1993, the
National Institute of Nursing Research (NINR) was established, helping to put nursing research more into the mainstream of health research. Funding opportunities for nursing research expanded in other countries as well.
Current and Future Directions for Nursing Research Nursing research continues to develop at a rapid pace and will undoubtedly flourish in the 21st century. Funding continues to grow. For example, NINR funding in fiscal year 2014 was more than $140 million compared to $70 million in 1999—and the competition for available funding is increasingly vigorous as more nurses seek support for testing innovative ideas for practice improvements.
Broadly speaking, the priority for future nursing research will be the promotion of excellence in nursing science. Toward this end, nurse researchers and practicing nurses will be sharpening their research skills and using those skills to address emerging issues of importance to the profession and its clientele. Among the trends we foresee for the early 21st century are the following:
• Continued focus on EBP. Encouragement for nurses to engage in evidence-based patient care is sure to continue. In turn, improvements will be needed both in the quality of studies and in nurses’ skills in locating, understanding, critiquing, and using relevant study results. Relatedly, there is an emerging interest in translational research— research on how findings from studies can best be translated into practice. Translation potential will require researchers to think more strategically about long-term feasibility, scalability, and sustainability when they test solutions to problems.
• Development of a stronger evidence base through confirmatory strategies. Practicing nurses are unlikely to adopt an innovation based on weakly designed or isolated studies. Strong research designs are essential, and confirmation is usually needed through the replication (i.e., the repeating) of studies with different clients, in different clinical settings, and at different times to ensure that the findings are robust.
• Greater emphasis on systematic reviews. Systematic reviews are a cornerstone of EBP and will take on increased importance in all health disciplines. Systematic reviews rigorously integrate research
information on a topic so that conclusions about the state of evidence can be reached. Best practice clinical guidelines typically rely on such systematic reviews.
• Innovation. There is currently a major push for creative and innovative solutions to recurring practice problems. “Innovation” has become an important buzzword throughout NIH and in nursing associations. For example, the 2013 annual conference of the Council for the Advancement of Nursing Science was “Innovative Approaches to Symptom Science.” Innovative interventions—and new methods for studying nursing questions—are sure to be part of the future research landscape in nursing.
• Expanded local research in health care settings. Small studies designed to solve local problems will likely increase. This trend will be reinforced as more hospitals apply for (and are recertified for) Magnet status in the United States and in other countries. Mechanisms will need to be developed to ensure that evidence from these small projects becomes available to others facing similar problems, such as communication within and between regional nursing research alliances.
• Strengthening of interdisciplinary collaboration. Collaboration of nurses with researchers in related fields is likely to expand in the 21st century as researchers address fundamental health care problems. In turn, such collaborative efforts could lead to nurse researchers playing a more prominent role in national and international health care policies. One of four major recommendations in a 2010 report on the future of nursing by the Institute of Medicine was that nurses should be full partners with physicians and other health care professionals in redesigning health care.
• Expanded dissemination of research findings. The Internet and other electronic communication have a big impact on disseminating research information, which in turn helps to promote EBP. Through technologic advances, information about innovations can be communicated more widely and more quickly than ever before.
• Increased focus on cultural issues and health disparities. The issue of health disparities has emerged as a central concern in nursing and other health disciplines; this in turn has raised consciousness about the
cultural sensitivity of health interventions and the cultural competence of health care workers. There is growing awareness that research must be sensitive to the health beliefs, behaviors, and values of culturally and linguistically diverse populations.
• Clinical significance and patient input. Research findings increasingly must meet the test of being clinically significant, and patients have taken center stage in efforts to define clinical significance. A major challenge in the years ahead will involve getting both research evidence and patient preferences into clinical decisions, and designing research to study the process and the outcomes.
Broad research priorities for the future have been articulated by many nursing organizations, including NINR and Sigma Theta Tau International. Expert panels and research working groups help NINR to identify gaps in current knowledge that require research. The primary areas of research funded by NINR in 2014 were health promotion/disease prevention, eliminating health disparities, caregiving, symptom management, and self- management. Research priorities that have been expressed by Sigma Theta Tau International include advancing healthy communities through health promotion; preventing disease and recognizing social, economic, and political determinants; implementation of evidence-based practice; targeting the needs of vulnerable populations such as the poor and chronically ill; and developing nurses’ capacity for research. Priorities also have been developed for several nursing specialties and for nurses in several countries—for example, Ireland (Brenner et al., 2014; Drennan et al., 2007), Sweden (Bäck-Pettersson et al., 2008), Australia (Wynaden et al., 2014), and Korea (Kim et al., 2002).
SOURCES OF EVIDENCE FOR NURSING PRACTICE Nurses make clinical decisions based on knowledge from many sources, including coursework, textbooks, and their own clinical experience. Because evidence is constantly evolving, learning about best practice nursing perseveres throughout a nurse’s career.
Some of what nurses learn is based on systematic research, but much of it is not. What are the sources of evidence for nursing practice? Where
does knowledge for practice come from? Until fairly recently, knowledge primarily was handed down from one generation to the next based on experience, trial and error, tradition, and expert opinion. Information sources for clinical practice vary in dependability, giving rise to what is called an evidence hierarchy, which acknowledges that certain types of evidence are better than others. A brief discussion of some alternative sources of evidence shows how research001-based information is different.
Tradition and Authority Decisions are sometimes based on custom or tradition. Certain “truths” are accepted as given, and such “knowledge” is so much a part of a common heritage that few seek verification. Tradition facilitates communication by providing a common foundation of accepted truth, but many traditions have never been evaluated for their validity. There is concern that some nursing interventions are based on tradition, custom, and “unit culture” rather than on sound evidence. Indeed, a recent analysis suggests that some “sacred cows” (ineffective traditional habits) persist even in a health care center recognized as a leader in evidence-based practice (Hanrahan et al., 2015).
Another common source of information is an authority, a person with specialized expertise. We often make decisions about problems with which we have little experience; it seems natural to place our trust in the judgment of people with specialized training or experience. As a source of evidence, however, authority has shortcomings. Authorities are not infallible, particularly if their expertise is based primarily on personal experience; yet, like tradition, their knowledge often goes unchallenged.
Example of “Myths” in Nursing Textbooks: A study suggests that even nursing textbooks may contain “myths.” In their analysis of 23 widely used undergraduate psychiatric nursing textbooks, Holman and colleagues (2010) found that all books contained at least one unsupported assumption (myth) about loss and grief—that is, assumptions not supported by research evidence. Moreover, many evidence-based findings about grief and loss failed to be included in the textbooks.
Clinical Experience, Trial and Error, and Intuition Clinical experience is a familiar, functional source of knowledge. The ability to generalize, to recognize regularities, and to make predictions is an important characteristic of the human mind. Nevertheless, personal experience is limited as a knowledge source because each nurse’s experience is too narrow to be generally useful. A second limitation is that the same objective event is often experienced and perceived differently by two nurses.
A related method is trial and error in which alternatives are tried successively until a solution to a problem is found. We likely have all used this method in our professional work. For example, many patients dislike the taste of potassium chloride solution. Nurses try to disguise the taste of the medication in various ways until one method meets with the approval of the patient. Trial and error may offer a practical means of securing knowledge, but the method tends to be haphazard and solutions may be idiosyncratic.
Intuition is a knowledge source that cannot be explained based on reasoning or prior instruction. Although intuition and hunches undoubtedly play a role in nursing—as they do in the conduct of research—it is difficult to develop nursing policies and practices based on intuition.
Logical Reasoning Solutions to some problems are developed by logical thought processes. As a problem-solving method, logical reasoning combines experience, intellectual faculties, and formal systems of thought. Inductive reasoning involves developing generalizations from specific observations. For example, a nurse may observe the anxious behavior of (specific) hospitalized children and conclude that (in general) children’s separation from their parents is stressful. Deductive reasoning involves developing specific predictions from general principles. For example, if we assume that separation anxiety occurs in hospitalized children (in general), then we might predict that (specific) children in a hospital whose parents do not room-in will manifest symptoms of stress. Both systems of reasoning are useful for understanding and organizing phenomena, and both play a role in research. Logical reasoning in and of itself, however, is limited because
the validity of reasoning depends on the accuracy of the premises with which one starts.
Assembled Information In making clinical decisions, health care professionals rely on information that has been assembled for a variety of purposes. For example, local, national, and international benchmarking data provide information on such issues as infection rates or the rates of using various procedures (e.g., cesarean births) and can facilitate evaluations of clinical practices. Cost data—information on the costs associated with certain procedures, policies, or practices—are sometimes used as a factor in clinical decision making. Quality improvement and risk data, such as medication error reports, can be used to assess the need for practice changes. Such sources are useful, but they do not provide a good mechanism for determining whether improvements in patient outcomes result from their use.
Disciplined Research Research conducted in a disciplined framework is the most sophisticated method of acquiring knowledge. Nursing research combines logical reasoning with other features to create evidence that, although fallible, tends to yield the most reliable evidence. Carefully synthesized findings from rigorous research are at the pinnacle of most evidence hierarchies. The current emphasis on EBP requires nurses to base their clinical practice to the greatest extent possible on rigorous research001-based findings rather than on tradition, authority, intuition, or personal experience— although nursing will always remain a rich blend of art and science.
PARADIGMS AND METHODS FOR NURSING RESEARCH A paradigm is a worldview, a general perspective on the complexities of the world. Paradigms for human inquiry are often characterized in terms of the ways in which they respond to basic philosophical questions, such as, What is the nature of reality? (ontologic) and What is the relationship between the inquirer and those being studied? (epistemologic).
Disciplined inquiry in nursing has been conducted mainly within two
broad paradigms, positivism and constructivism. This section describes these two paradigms and outlines the research methods associated with them. In later chapters, we describe the transformative paradigm that involves critical theory research (Chapter 21), and a pragmatism paradigm that involves mixed methods research (Chapter 26).
The Positivist Paradigm The paradigm that dominated nursing research for decades is known as positivism (also called logical positivism). Positivism is rooted in 19th century thought, guided by such philosophers as Mill, Newton, and Locke. Positivism reflects a broader cultural phenomenon that, in the humanities, is referred to as modernism, which emphasizes the rational and the scientific.
As shown in Table 1.2, a fundamental assumption of positivists is that there is a reality out there that can be studied and known (an assumption is a basic principle that is believed to be true without proof or verification). Adherents of positivism assume that nature is basically ordered and regular and that reality exists independent of human observation. In other words, the world is assumed not to be merely a creation of the human mind. The related assumption of determinism refers to the positivists’ belief that phenomena are not haphazard but rather have antecedent causes. If a person has a cerebrovascular accident, the researcher in a positivist tradition assumes that there must be one or more reasons that can be potentially identified. Within the positivist paradigm, much research activity is directed at understanding the underlying causes of phenomena.
Positivists value objectivity and attempt to hold personal beliefs and biases in check to avoid contaminating the phenomena under study. The positivists’ scientific approach involves using orderly, disciplined procedures with tight controls of the research situation to test hunches about the phenomena being studied.
Strict positivist thinking has been challenged, and few researchers adhere to the tenets of pure positivism. In the postpositivist paradigm, there is still a belief in reality and a desire to understand it, but postpositivists recognize the impossibility of total objectivity. They do, however, see objectivity as a goal and strive to be as neutral as possible. Postpositivists also appreciate the impediments to knowing reality with certainty and therefore seek probabilistic evidence—that is, learning what the true state of a phenomenon probably is, with a high degree of
likelihood. This modified positivist position remains a dominant force in nursing research. For the sake of simplicity, we refer to it as positivism.
The Constructivist Paradigm The constructivist paradigm (often called the naturalistic paradigm) began as a countermovement to positivism with writers such as Weber and Kant. Just as positivism reflects the cultural phenomenon of modernism that burgeoned after the industrial revolution, naturalism is an outgrowth of the cultural transformation called postmodernism. Postmodern thinking emphasizes the value of deconstruction—taking apart old ideas and structures—and reconstruction—putting ideas and structures together in new ways. The constructivist paradigm represents a major alternative system for conducting disciplined research in nursing. Table 1.2 compares the major assumptions of the positivist and constructivist paradigms.
For the naturalistic inquirer, reality is not a fixed entity but rather is a construction of the individuals participating in the research; reality exists within a context, and many constructions are possible. Naturalists thus take the position of relativism: If there are multiple interpretations of reality that exist in people’s minds, then there is no process by which the ultimate truth or falsity of the constructions can be determined.
The constructivist paradigm assumes that knowledge is maximized when the distance between the inquirer and those under study is minimized. The voices and interpretations of study participants are crucial to understanding the phenomenon of interest, and subjective interactions are the primary way to access them. Findings from a constructivist inquiry are the product of the interaction between the inquirer and the participants.
Paradigms and Methods: Quantitative and Qualitative Research Research methods are the techniques researchers use to structure a study and to gather and analyze information relevant to the research question. The two alternative paradigms correspond to different methods for developing evidence. A key methodologic distinction is between quantitative research, which is most closely allied with positivism, and qualitative research, which is associated with constructivist inquiry—
although positivists sometimes undertake qualitative studies, and constructivist researchers sometimes collect quantitative information. This section provides an overview of the methods associated with the two paradigms.
The Scientific Method and Quantitative Research The traditional, positivist scientific method refers to a set of orderly, disciplined procedures used to acquire information. Quantitative researchers use deductive reasoning to generate predictions that are tested in the real world. They typically move in a systematic fashion from the definition of a problem and the selection of concepts on which to focus to the solution of the problem. By systematic, we mean that the investigator progresses logically through a series of steps, according to a specified plan of action.
Quantitative researchers use various control strategies. Control involves imposing conditions on the research situation so that biases are minimized and precision and validity are maximized. Control mechanisms are discussed at length in this book.
Quantitative researchers gather empirical evidence—evidence that is rooted in objective reality and gathered through the senses. Empirical evidence, then, consists of observations gathered through sight, hearing, taste, touch, or smell. Observations of the presence or absence of skin inflammation, patients’ anxiety level, or infant birth weight are all examples of empirical observations. The requirement to use empirical evidence means that findings are grounded in reality rather than in researchers’ personal beliefs.
Evidence for a study in the positivist paradigm is gathered according to an established plan, using structured methods to collect needed information. Usually (but not always) the information gathered is quantitative—that is, numeric information that is obtained from a formal measurement and is analyzed statistically.
A traditional scientific study strives to go beyond the specifics of a research situation. For example, quantitative researchers are typically not as interested in understanding why a particular person has a stroke as in understanding what factors influence its occurrence in people generally.
The degree to which research findings can be generalized to individuals other than those who participated in the study is called the study’s generalizability.
The scientific method has enjoyed considerable stature as a method of inquiry and has been used productively by nurse researchers studying a range of nursing problems. This is not to say, however, that this approach can solve all nursing problems. One important limitation—common to both quantitative and qualitative research—is that research cannot be used to answer moral or ethical questions. Many persistent, intriguing questions about human beings fall into this area—questions such as whether euthanasia should be practiced or abortion should be legal.
The traditional research approach also must contend with problems of measurement. To study a phenomenon, quantitative researchers attempt to measure it by attaching numeric values that express quantity. For example, if the phenomenon of interest is patient stress, researchers would want to assess if patients’ stress is high or low, or higher under certain conditions or for some people. Physiologic phenomena such as blood pressure and temperature can be measured with great accuracy and precision, but the same cannot be said of most psychological phenomena, such as stress or resilience.
Another issue is that nursing research focuses on humans, who are inherently complex and diverse. Traditional quantitative methods typically concentrate on a relatively small portion of the human experience (e.g., weight gain, depression) in a single study. Complexities tend to be controlled and, if possible, eliminated, rather than studied directly, and this narrowness of focus can sometimes obscure insights. Finally, quantitative research within the positivist paradigm has been accused of an inflexibility of vision that does not capture the full breadth of human experience.
Constructivist Methods and Qualitative Research Researchers in constructivist traditions emphasize the inherent complexity of humans, their ability to shape and create their own experiences, and the idea that truth is a composite of realities. Consequently, constructivist studies are heavily focused on understanding the human experience as it is lived, usually through the careful collection and analysis of qualitative
materials that are narrative and subjective. Researchers who reject the traditional scientific method believe that it
is overly reductionist—that is, it reduces human experience to the few concepts under investigation, and those concepts are defined in advance by the researcher rather than emerging from the experiences of those under study. Constructivist researchers tend to emphasize the dynamic, holistic, and individual aspects of human life and attempt to capture those aspects in their entirety, within the context of those who are experiencing them.
Flexible, evolving procedures are used to capitalize on findings that emerge in the course of the study. Constructivist inquiry usually takes place in the field (i.e., in naturalistic settings), often over an extended time period. In constructivist research, the collection of information and its analysis typically progress concurrently; as researchers sift through information, insights are gained, new questions emerge, and further evidence is sought to amplify or confirm the insights. Through an inductive process, researchers integrate information to develop a theory or description that helps illuminate the phenomenon under observation.
Constructivist studies yield rich, in-depth information that can elucidate varied dimensions of a complicated phenomenon. Findings from in-depth qualitative research are typically grounded in the real-life experiences of people with first-hand knowledge of a phenomenon. Nevertheless, the approach has several limitations. Human beings are used directly as the instrument through which information is gathered, and humans are extremely intelligent and sensitive—but fallible—tools. The subjectivity that enriches the analytic insights of skillful researchers can yield trivial and obvious “findings” among less competent ones.
Another potential limitation involves the subjectivity of constructivist inquiry, which sometimes raises concerns about the idiosyncratic nature of the conclusions. Would two constructivist researchers studying the same phenomenon in similar settings arrive at similar conclusions? The situation is further complicated by the fact that most constructivist studies involve a small group of participants. Thus, the generalizability of findings from constructivist inquiries is an issue of potential concern.
Multiple Paradigms and Nursing Research
Paradigms should be viewed as lenses that help to sharpen our focus on a phenomenon, not as blinders that limit intellectual curiosity. The emergence of alternative paradigms for studying nursing problems is, in our view, a healthy and desirable path that can maximize the breadth of evidence for practice. Although researchers’ worldview may be paradigmatic, knowledge itself is not. Nursing knowledge would be thin if there were not a rich array of methods available within the two paradigms —methods that are often complementary in their strengths and limitations. We believe that intellectual pluralism is advantageous.
We have emphasized differences between the two paradigms and associated methods so that distinctions would be easy to understand— although for many of the issues included in Table 1.2, differences are more on a continuum than they are a dichotomy. Subsequent chapters of this book elaborate further on differences in terminology, methods, and research products. It is equally important, however, to note that the two main paradigms have many features in common, only some of which are mentioned here:
• Ultimate goals. The ultimate aim of disciplined research, regardless of the underlying paradigm, is to gain understanding about phenomena. Both quantitative and qualitative researchers seek to capture the truth with regard to an aspect of the world in which they are interested, and both groups can make meaningful—and mutually beneficial— contributions to evidence for nursing practice.
• External evidence. Although the word empiricism has come to be allied with the classic scientific method, researchers in both traditions gather and analyze evidence empirically, that is, through their senses. Neither qualitative nor quantitative researchers are armchair analysts, depending on their own beliefs and worldviews to generate knowledge.
• Reliance on human cooperation. Because evidence for nursing research comes primarily from humans, human cooperation is essential. To understand people’s characteristics and experiences, researchers must persuade them to participate in the investigation and to speak and act candidly.
• Ethical constraints. Research with human beings is guided by ethical principles that sometimes interfere with research goals. As we discuss
in Chapter 7, ethical dilemmas often confront researchers, regardless of paradigms or methods.
• Fallibility of disciplined research. Virtually all studies have some limitations. Every research question can be addressed in many ways, and inevitably, there are trade-offs. The fallibility of any single study makes it important to understand and critique researchers’ methodologic decisions when evaluating evidence quality.
Thus, despite philosophic and methodologic differences, researchers using traditional scientific methods or constructivist methods share overall goals and face many similar challenges. The selection of an appropriate method depends on researchers’ personal philosophy and also on the research question. If a researcher asks, “What are the effects of cryotherapy on nausea and oral mucositis in patients undergoing chemotherapy?” the researcher needs to examine the effects through the careful measurement of patient outcomes. On the other hand, if a researcher asks, “What is the process by which parents learn to cope with the death of a child?” the researcher would be hard pressed to quantify such a process. Personal worldviews of researchers help to shape their questions.
In reading about the alternative paradigms for nursing research, you likely were more attracted to one of the two paradigms. It is important, however, to learn about both approaches to disciplined inquiry and to recognize their respective strengths and limitations. In this textbook, we describe methods associated with both qualitative and quantitative research in an effort to assist you in becoming methodologically bilingual. This is especially important because large numbers of nurse researchers are now undertaking mixed methods research that involves gathering and analyzing both qualitative and quantitative data (Chapters 26–28).
THE PURPOSES OF NURSING RESEARCH The general purpose of nursing research is to answer questions or solve problems of relevance to nursing. Specific purposes can be classified in various ways. We describe three such classifications—not because it is important for you to categorize a study as having one purpose or the other but rather because this will help us to illustrate the broad range of
questions that have intrigued nurses and to further show differences between qualitative and quantitative inquiry.
Applied and Basic Research Sometimes a distinction is made between basic and applied research. As traditionally defined, basic research is undertaken to enhance the base of knowledge or to formulate or refine a theory. For example, a researcher may perform an in-depth study to better understand normal grieving processes, without having explicit nursing applications in mind. Some types of basic research are called bench research, which is usually performed in a laboratory and focuses on the molecular and cellular mechanisms that underlie disease.
Example of Basic Nursing Research: Kishi and a multidisciplinary team of researchers (2015) studied the effect of hypo-osmotic shock of epidermal cells on skin inflammation in a rat model, in an effort to understand the physiologic mechanism underlying aquagenic pruritus (disrupted skin barrier function) in the elderly.
Applied research seeks solutions to existing problems and tends to be of greater immediate utility for EBP. Basic research is appropriate for discovering general principles of human behavior and biophysiologic processes; applied research is designed to indicate how these principles can be used to solve problems in nursing practice. In nursing, the findings from applied research may pose questions for basic research, and the results of basic research often suggest clinical applications.
Example of Applied Nursing Research: S. Martin and colleagues (2014) studied whether positive therapeutic suggestions given via headphones to children emerging from anesthesia after a tonsillectomy would help to lower the children’s pain.
Research to Achieve Varying Levels of Explanation Another way to classify research purposes concerns the extent to which studies provide explanatory information. Although specific study goals can
range along an explanatory continuum, a fundamental distinction (relevant especially in quantitative research) is between studies whose primary intent is to describe phenomena, and those that are cause-probing—that is, designed to illuminate the underlying causes of phenomena.
Within a descriptive/explanatory framework, the specific purposes of nursing research include identification, description, exploration, prediction/control, and explanation. For each purpose, various types of question are addressed—some more amenable to qualitative than to quantitative inquiry and vice versa.
Identification and Description Qualitative researchers sometimes study phenomena about which little is known. In some cases, so little is known that the phenomenon has yet to be clearly identified or named or has been inadequately defined. The in-depth, probing nature of qualitative research is well suited to the task of answering such questions as, “What is this phenomenon?” and “What is its name?” (Table 1.3). In quantitative research, by contrast, researchers begin with a phenomenon that has been previously studied or defined— sometimes in a qualitative study. Thus, in quantitative research, identification typically precedes the inquiry.
Qualitative Example of Identification: Wojnar and Katzenmeyer (2013) studied the experiences of preconception, pregnancy, and new motherhood for lesbian nonbiologic mothers. They identified, through in-depth interviews with 24 women, a unique description of a pervasive feeling they called otherness.
Description is another important research purpose. Examples of phenomena that nurse researchers have described include patients’ pain, confusion, and coping. Quantitative description focuses on the incidence, size, and measurable attributes of phenomena. Qualitative researchers, by contrast, describe the dimensions and meanings of phenomena. Table 1.3 shows descriptive questions posed by quantitative and qualitative researchers.
Quantitative Example of Description: Palese and colleagues (2015) conducted a study to describe the average healing time of stage II pressure ulcers. They found that it took approximately 23 days to achieve complete reepithelialization.
Qualitative Example of Description: Archibald and colleagues (2015) undertook an in-depth study to describe the information needs of parents of children with asthma.
Exploration Exploratory research begins with a phenomenon of interest, but rather than simply observing and describing it, exploratory research investigates the full nature of the phenomenon, the manner in which it is manifested, and the other factors to which it is related. For example, a descriptive quantitative study of patients’ preoperative stress might document the degree of stress patients feel before surgery and the percentage of patients who are stressed. An exploratory study might ask: What factors diminish or increase a patient’s stress? Are nurses’ behaviors related to a patient’s stress level? Qualitative methods are especially useful for exploring the full nature of a little-understood phenomenon. Exploratory qualitative research is designed to shed light on the various ways in which a phenomenon is manifested and on underlying processes.
Quantitative Example of Exploration: Lee and colleagues (2014) explored the association between physical activity in older adults and their level of depressive symptoms.
Qualitative Example of Exploration: Based on in-depth interviews with adults living on a reservation in the United States, D. Martin and Yurkovich (2014) explored American Indians’ perception of a healthy family.
Explanation The goals of explanatory research are to understand the underpinnings of natural phenomena and to explain systematic relationships among them. Explanatory research is often linked to theories, which are a method of integrating ideas about phenomena and their interrelationships. Whereas descriptive research provides new information and exploratory research provides promising insights, explanatory research attempts to offer
understanding of the underlying causes or full nature of a phenomenon. In quantitative research, theories or prior findings are used deductively to generate hypothesized explanations that are then tested. In qualitative studies, researchers search for explanations about how or why a phenomenon exists or what a phenomenon means as a basis for developing a theory that is grounded in rich, in-depth evidence.
Quantitative Example of Explanation: Golfenshtein and Drach001-Zahavy (2015) tested a theoretical model (attribution theory) to understand the role of patients’ attributions in nurses’ regulation of emotions in pediatric hospital wards.
Qualitative Example of Explanation: Smith-Young and colleagues (2014) conducted an in-depth study to develop a theoretical understanding of the process of managing work-related musculoskeletal disorders while remaining at the workplace. They called this process constant negotiation.
Prediction and Control Many phenomena defy explanation. Yet it is frequently possible to make predictions and to control phenomena based on research findings, even in the absence of complete understanding. For example, research has shown that the incidence of Down syndrome in infants increases with the age of the mother. We can predict that a woman aged 40 years is at higher risk of bearing a child with Down syndrome than is a woman aged 25 years. We can partially control the outcome by educating women about the risks and offering amniocentesis to women older than 35 years of age. The ability to predict and control in this example does not depend on an explanation of why older women are at a higher risk of having an abnormal child. In many quantitative studies, prediction and control are key objectives. Although explanatory studies are powerful in an EBP environment, studies whose purpose is prediction and control are also critical in helping clinicians make decisions.
Quantitative Example of Prediction: Dang (2014) studied factors that
predicted resilience among homeless youth with histories of maltreatment. Social connectedness and self-esteem were predictive of better mental health.
Research Purposes Linked to Evidence-Based Practice The purpose of most nursing studies can be categorized on a descriptive– explanatory dimension as just described, but some studies do not fall into such a system. For example, a study to develop and rigorously test a new method of measuring patient outcomes cannot easily be classified on this continuum.
In both nursing and medicine, several books have been written to facilitate evidence-based practice, and these books categorize studies in terms of the types of information needed by clinicians (DiCenso et al., 2005; Guyatt et al., 2008; Melnyk & Fineout-Overholt, 2011). These writers focus on several types of clinical concerns: treatment, therapy, or intervention; diagnosis and assessment; prognosis; prevention of harm; etiology; and meaning. Not all nursing studies have one of these purposes, but most of them do.
Treatment, Therapy, or Intervention Nurse researchers undertake studies designed to help nurses make evidence-based treatment decisions about how to prevent a health problem or how to manage an existing problem. Such studies range from evaluations of highly specific treatments or therapies (e.g., comparing two types of cooling blankets for febrile patients) to complex multisession interventions designed to effect major behavioral changes (e.g., nurse-led smoking cessation interventions). Such intervention research plays a critical role in EBP.
Example of a Study Aimed at Treatment/Therapy: Ling and co-researchers (2014) tested the effectiveness of a school-based healthy lifestyle intervention designed to prevent childhood obesity in four rural elementary schools.
Diagnosis and Assessment A burgeoning number of nursing studies concern the rigorous development
and evaluation of formal instruments to screen, diagnose, and assess patients and to measure important clinical outcomes. High-quality instruments with documented accuracy are essential both for clinical practice and for further research.
Example of a Study Aimed at Diagnosis/Assessment: Pasek and colleagues (2015) developed a prototype of an electronic headache pain diary for children and evaluated the clinical feasibility of the diary for assessing and documenting concussion headache.
Prognosis Studies of prognosis examine outcomes associated with a disease or health problem, estimate the probability they will occur, and predict the types of people for whom the outcomes are most likely. Such studies facilitate the development of long-term care plans for patients. They provide valuable information for guiding patients to make lifestyle choices or to be vigilant for key symptoms. Prognostic studies can also play a role in resource allocation decisions.
Example of a Study Aimed at Prognosis: Storey and Von Ah (2015) studied the prevalence and impact of hyperglycemia on hospitalized leukemia patients, in terms of such outcomes as neutropenia, infection, and length of hospital stay.
Prevention of Harm and Etiology (Causation) Nurses frequently encounter patients who face potentially harmful exposures as a result of environmental agents or because of personal behaviors or characteristics. Providing useful information to patients about such harms and how best to avoid them depends on the availability of accurate evidence about health risks. Moreover, it can be difficult to prevent harms if we do not know what causes them. For example, there would be no smoking cessation programs if research had not provided firm evidence that smoking cigarettes causes or contributes to a wide range of health problems. Thus, identifying factors that affect or cause illness,
mortality, or morbidity is an important purpose of many nursing studies.
Example of a Study Aimed at Identifying and Preventing Harms: Hagerty and colleagues (2015) undertook a study to identify risk factors for catheter- associated urinary tract infections in critically ill patients with subarachnoid hemorrhage. The risk factors examined included patients’ blood sugar levels, patient age, and levels of anemia requiring transfusion.
Meaning and Processes Designing effective interventions, motivating people to comply with treatments and health promotion activities, and providing sensitive advice to patients are among the many health care activities that can greatly benefit from understanding the clients’ perspectives. Research that provides evidence about what health and illness mean to clients, what barriers they face to positive health practices, and what processes they experience in a transition through a health care crisis are important to evidence-based nursing practice.
Example of a Study Aimed at Studying Meaning: Carlsson and Persson (2015) studied what it means to live with intestinal failure caused by Crohn disease and the influence it has on daily life.
TIP: Several of these EBP-related purposes (except diagnosis and meaning) fundamentally call for cause-probing research. For example, research on interventions focuses on whether an intervention causes improvements in key outcomes. Prognosis research asks if a disease or health condition causes subsequent adverse outcomes, and etiology research seeks explanations about the underlying causes of health problems.
ASSISTANCE FOR USERS OF NURSING RESEARCH This book is designed primarily to help you develop skills for conducting research, but in an environment that stresses EBP, it is extremely
important to hone your skills in reading, evaluating, and using nursing studies. We provide specific guidance to consumers in most chapters by including guidelines for critiquing aspects of a study covered in the chapter. The questions in Box 1.1 are designed to assist you in using the information in this chapter in an overall preliminary assessment of a research report.
BOX 1.1 Questions for a Preliminary Overview of a Research Report
1. How relevant is the research problem in this report to the actual practice of nursing? Does the study focus on a topic that is a priority area for nursing research?
2. Is the research quantitative or qualitative? 3. What is the underlying purpose (or purposes) of the study—
identification, description, exploration, explanation, or prediction and control? Does the purpose correspond to an EBP focus such as treatment, diagnosis, prognosis, harm/etiology, or meaning?
4. Is this study fundamentally cause-probing? 5. What might be some clinical implications of this research? To what type
of people and settings is the research most relevant? If the findings are accurate, how might I use the results of this study?
TIP: The Resource Manual that accompanies this book offers particularly rich opportunities to practice your critiquing skills. The Toolkit on thePoint with the Resource Manual includes Box 1.1 as a Word document, which will allow you to adapt these questions, if desired, and to answer them directly into a Word document without having to retype the questions.
RESEARCH EXAMPLES Each chapter of this book presents brief descriptions of studies conducted
by nurse researchers, focusing on aspects emphasized in the chapter. Reading the full journal articles would prove useful for learning more about the studies, their methods, and the findings.
Research Example of a Quantitative Study Study: The effects of a community-based, culturally tailored diabetes
prevention intervention for high-risk adults of Mexican descent (Vincent et al., 2014)
Study Purpose: The purpose of the study was to evaluate the effectiveness of a 5-month nurse-coached diabetes prevention program (Un Estilo de Vida Saludable or EVS) for overweight Mexican American adults.
Study Methods: A total of 58 Spanish-speaking adults of Mexican descent were recruited to participate in the study. Some of the participants, at random, were in a group that received the EVS intervention, while others in a control group did not receive it. The EVS intervention used content from a previously tested diabetes prevention program, but the researchers created a community-based, culturally tailored intervention for their population. The intervention, which was offered in community rooms of churches, consisted of an intensive phase of eight weekly 2-hour sessions, followed by a maintenance phase of 1-hour sessions for the final 3 months. Those in the group not receiving the intervention received educational sessions broadly aimed at health promotion in general. The researchers compared the two groups with regard to several important outcomes, such as weight loss, waist circumference, body mass index, and self-efficacy. Outcome information was gathered three times—at the outset of the study (prior to the intervention), 8 weeks later, and then after the program ended.
Key Findings: The analysis suggested that those in the intervention group had several better outcomes, such as greater weight loss, smaller waist circumference, and lower body mass index, than those in the control group.
Conclusions: Vincent and her colleagues (2014) concluded that implementing the culturally tailored program was feasible, was well- received among participants (e.g., high rates of program retention), and
was effective in decreasing risk factors for type 2 diabetes.
Research Example of a Quantitative Study Study: Silent, invisible, and unacknowledged: Experiences of young
caregivers of single parents diagnosed with multiple sclerosis (Bjorgvinsdottir & Halldorsdottir, 2014)
Study Purpose: The purpose of this study was to study the personal experience of being a young caregiver of a chronically ill parent diagnosed with multiple sclerosis (MS).
Study Methods: Young adults in Iceland whose parents were diagnosed with MS were recruited through the Icelandic National Multiple Sclerosis Society, and 11 agreed to be included in the study. Participants were interviewed in their own homes or in the home of the lead researcher, whichever they preferred. In-depth questioning was used to probe the experiences of the participants. The main interview question was: “Can you tell me about your personal experience being a young caregiver of a chronically ill parent with MS?” Several participants were interviewed twice to ensure rich and deep descriptions for a total of 21 interviews.
Key Findings: The young caregivers felt that they were invisible and unacknowledged as caregivers and received limited support and assistance from professionals. Their responsibilities led to severe personal restrictions and they felt they had lived without a true childhood because they were left to manage adult-like responsibilities at a young age. Their role as caregiver was demanding and stressful, and they felt unsupported and abandoned.
Conclusions: The researchers concluded that health professionals should be more vigilant about the needs for support and guidance for children and adolescents caring for chronically ill parents.
• Nursing research is systematic inquiry to develop knowledge about issues of importance to nurses. Nurses are adopting an evidence-
based practice (EBP) that incorporates research findings into their clinical decisions.
• Nurses can participate in a range of research-related activities that span a continuum from being consumers of research (those who read and evaluate studies) and producers of research (those who design and undertake studies).
• Nursing research began with Florence Nightingale but developed slowly until its rapid acceleration in the 1950s. Since the 1970s, nursing research has focused on problems relating to clinical practice.
• The National Institute of Nursing Research (NINR), established at the U.S. National Institutes of Health in 1993, affirms the stature of nursing research in the United States.
• Contemporary emphases in nursing research include EBP projects, replications of research, research integration through systematic reviews, multisite and interdisciplinary studies, expanded dissemination efforts, and increased focus on health disparities.
• Disciplined research is a better evidence source for nursing practice than other sources, such as tradition, authority, personal experience, trial and error, intuition, and logical reasoning.
• Nursing research is conducted mainly within one of two broad paradigms—worldviews with underlying assumptions about reality: the positivist paradigm and the constructivist paradigm.
• In the positivist paradigm, it is assumed that there is an objective reality and that natural phenomena are regular and orderly. The related assumption of determinism is the belief that phenomenas result from prior causes and are not haphazard.
• In the constructivist (naturalistic) paradigm, it is assumed that reality is not fixed but is rather a construction of human minds; thus, “truth” is a composite of multiple constructions of reality.
• The positivist paradigm is associated with quantitative research— the collection and analysis of numeric information. Quantitative research is typically conducted within the traditional scientific method, which is a systematic, controlled process. Quantitative researchers gather and analyze empirical evidence (evidence collected through the human senses) and strive for generalizability of
their findings beyond the study setting. • Researchers within the constructivist paradigm emphasize
understanding the human experience as it is lived through the collection and analysis of subjective, narrative materials using flexible procedures that evolve in the field; this paradigm is associated with qualitative research.
• Basic research is designed to extend the knowledge base for the sake of knowledge itself. Applied research focuses on discovering solutions to immediate problems.
• A fundamental distinction, especially relevant in quantitative research, is between studies whose primary intent is to describe phenomena and those that are cause-probing—that is, designed to illuminate underlying causes of phenomena. Specific purposes on the description/explanation continuum include identification, description, exploration, prediction/control, and explanation.
• Many nursing studies can also be classified in terms of a key EBP aim: treatment/therapy/intervention; diagnosis and assessment; prognosis; harm and etiology; and meaning and process.
Chapter 1 of the Resource Manual for Nursing Research: Generating and Assessing Evidence for Nursing Practice, 10th edition, offers study suggestions for reinforcing concepts presented in this chapter. In addition, the following questions can be addressed in classroom or online discussions:
1. Is your worldview closer to the positivist or the constructivist paradigm? Explore the aspects of the two paradigms that are especially consistent with your worldview.
2. Answer the questions in Box 1.1 about the Vincent et al. (2014) study described at the end of this chapter. Could this study have been undertaken as a qualitative study? Why or why not?
3. Answer the questions in Box 1.1 about the Bjorgvinsdottir and Halldorsdottir (2014) study described at the end of this chapter. Could
this study have been undertaken as a quantitative study? Why or why not?
STUDIES CITED IN CHAPTER 1
Archibald, M. M., Caine, V., Ali, S., Hartling, L., & Scott, S. (2015). What is left unsaid: An interpretive description of the information needs of parents of children with asthma. Research in Nursing & Health, 38, 19–28.
Bäck-Pettersson, S., Hermansson, E., Sernert, N., & Bjökelund, C. (2008). Research priorities in nursing—A Delphi study among Swedish nurses. Journal of Clinical Nursing, 17, 2221–2231.
Bjorgvinsdottir, K., & Halldorsdottir, S. (2014). Silent, invisible and unacknowledged: Experiences of young caregivers of single parents diagnosed with multiple sclerosis. Scandinavian Journal of the Caring Sciences, 28, 38– 48.
Brenner, M., Hilliard, C., Regan, G., Coughlan, B., Hayden, S., Drennan, J., & Kelleher, D. (2014). Research priorities for children’s nursing in Ireland. Journal of Pediatric Nursing, 29, 301–308.
*Campbell-Yeo, M., Johnston, C., Benoit, B., Latimer, M., Vincer, M., Walker, C., . . . Caddell, K. (2013). Trial of repeated analgesia with kangaroo mother care (TRAKC trial). BMC Pediatrics, 13, 182.
Carlsson, E., & Persson, E. (2015). Living with intestinal failure by Crohn disease: Not letting the disease conquer life. Gastroenterology Nursing, 38, 12–20.
Chwo, M. J., Anderson, G. C., Good, M., Dowling, D. A., Shiau, S. H., & Chu, D. M. (2002). A randomized controlled trial of early kangaroo care for preterm infants: Effects on temperature, weight, behavior, and acuity. Journal of Nursing Research, 10, 129–142.
*Cong, X., Ludington-Hoe, S., McCain, G., & Fu, P. (2009). Kangaroo care modifies preterm infant heart rate variability in response to heel stick pain. Early Human Development, 85, 561–567.
Cong, X., Ludington-Hoe, S., & Walsh, S. (2011). Randomized crossover trial of kangaroo care to reduce behavioral pain responses in preterm infants. Biological Research for Nursing, 13, 204–216.
Dang, M. T. (2014). Social connectedness and self-esteem: Predictors of resilience in mental health among maltreated homeless youth. Issues in Mental Health Nursing, 35, 212–219.
DiCenso, A., Guyatt, G., & Ciliska, D. (2005). Evidence-based nursing: A guide to clinical practice. St. Louis, MO: Elsevier Mosby.
Drenkard, K. (2013). Change is good: Introducing the 2014 Magnet Application Manual. Journal of Nursing Administration, 43, 489–490.
Drennan, J., Meehan, T., Kemple, M., Johnson, M., Treacy, M., & Butler, M. (2007). Nursing research priorities for Ireland. Journal of Nursing Scholarship, 39, 298–305.
Golfenshtein, N., & Drach001-Zahavy, A. (2015). An attribution theory perspective on emotional labour in nurse-patient encounters: A nested cross- sectional study in paediatric settings. Journal of Advanced Nursing, 71(5), 1123–1134.
Gona, C., & DeMarco, R. (2015). The context and experience of becoming HIV infected for Zimbabwean women: Unheard voices revealed. Journal of the Association of Nurses in AIDS Care, 26, 57–68.
Guyatt, G., Rennie, D., Meade, M., & Cook, D. (2008). Users’ guide to the medical literature: Essentials of evidence-based clinical practice (2nd ed.). New York: McGraw Hill.
Hagerty, T., Kertesz, L., Schmidt, J., Agarwal, S., Claassen, J., Mayer, S., . . . Shang, K. (2015). Risk factors for catheter-associated urinary tract infections in critically ill patients with subarachnoid hemorrhage. Journal of Neuroscience Nursing, 47, 51–54.
Hake-Brooks, S., & Anderson, G. (2008). Kangaroo care and breastfeeding of mother-preterm dyads 0–18 months: A randomized controlled trial. Neonatal Network, 27, 151–159.
Hanrahan, K., Wagner, M., Matthews, G., Stewart, S., Dawson, C., Greiner, J., . . . Williamson, A. (2015). Sacred cows gone to pasture: A systematic evaluation and integration of evidence-based practice. Worldview on Evidence-Based Nursing, 12, 3–11.
Holman, E., Perisho, J., Edwards, A., & Mlakar, N. (2010). The myths of coping with loss in undergraduate psychiatric nursing books. Research in Nursing & Health, 33, 486–499.
*Institute of Medicine. (2010). The future of nursing: Leading change, advancing health. Washington, DC: The National Academies Press.
Kim, M. J., Oh, E. G., Kim, C. J., Yoo, J. S., & Ko, I. S. (2002). Priorities for nursing research in Korea. Journal of Nursing Scholarship, 34, 307–312.
Kishi, C., Minematsu, T., Huang, L., Mugita, Y., Kitamura, A., Nakagami, G., . . . Sanada, H. (2015). Hypo-osmotic shock-induced subclinical inflammation of skin in a rat model of disrupted skin barrier function. Biological Research for Nursing, 17, 135–141.
Lee, H., Lee, J., Brar, J., Rush, E., & Jolley, C. (2014). Physical activity and depressive symptoms in older adults. Geriatric Nursing, 35, 37–41.
Ling, J., King, K., Speck, B., Kim, S., & Wu, D. (2014). Preliminary assessment
of a school-based healthy lifestyle intervention among rural elementary school children. Journal of School Health, 84, 247–255.
Ludington-Hoe, S. M. (2011). Thirty years of kangaroo care science and practice. Neonatal Network, 30, 357–362.
Mak, S., Lee, M., Cheung, J., Choi, K., Chung, T., Wong, T., . . . & Lee, D. (2015). Pressurised irrigation versus swabbing method in cleansing wounds healed by secondary intention: A randomized controlled trial with cost effectiveness analysis. International Journal of Nursing Studies, 52, 88–101.
Martin, D., & Yurkovich, E. (2014). “Close knit” defines a healthy native American Indian family. Journal of Family Nursing, 20, 51–72.
Martin, S., Smith, A., Newcomb, P., & Miller, J. (2014). Effects of therapeutic suggestion under anesthesia on outcomes in children post-tonsillectomy. Journal of Perianesthesia Nursing, 29, 94–106.
*McHugh, M. D., Kelly, L. A., Smith, H. L., Wu, E. S., Vanak, J., & Aiken, L. H. (2013). Lower mortality in Magnet hospitals. Medical Care, 51, 382–388.
Melnyk, B. M., & Fineout-Overholt, E. (2011). Evidence-based practice in nursing and healthcare: A guide to best practice (2nd ed.). Philadelphia: Lippincott Williams & Wilkins.
*Moore, E., Anderson, G., Bergman, N., & Dowswell, T. (2012). Early skin-to- skin contact for mothers and their health newborn infants. Cochrane Database of Systematic Reviews, (3), CD0003519.
Palese, A., Luisa, S., Ilenia, P., Laquintana, D., Stinco, G., & DeLiulio, P. (2015). What is the healing time of stage II pressure ulcers? Findings from a secondary analysis. Advances in Skin and Would Care, 28, 79–75.
Pasek, T., Locasto, L., Reichard, J., Fazio Sumrok, V., Johnson, E., & Kontos, A. (2015). The headache electronic diary for children with concussion. Clinical Nurse Specialist, 29, 80–88.
Reigle, B. S., Stevens, K., Belcher, J., Huth, M., McGuire, E., Mals, D., & Volz, T. (2008). Evidence-based practice and the road to Magnet status. The Journal of Nursing Administration, 38, 97–102.
Smith-Young, J., Solberg, S., & Gaudine, A. (2014). Constant negotiating: Managing work-related musculoskeletal disorders while remaining in the workplace. Qualitative Health Research, 24, 217–231.
Storey, S., & Von Ah, D. (2015). Prevalence and impact of hyperglycemia on hospitalized leukemia patients. European Journal of Oncology Nursing, 19, 13–17.
Vincent, D., McEwen, M., Hepworth, J., & Stump, C. (2014). The effects of a community-based, culturally tailored diabetes prevention intervention for high- risk adults of Mexican descent. The Diabetes Educator, 40, 202–213.
Wojnar, D. M., & Katzenmeyer, A. (2013). Experiences of preconception,
pregnancy, and new motherhood for lesbian nonbiological mothers. Journal of Obstetric, Gynecologic, and Neonatal Nursing, 43, 50–60.
Wynaden, D., Heslop, K., Omari, O., Nelson, D., Osmond, B., Taylor, M., & Gee, T. (2014). Identifying mental health nursing priorities: A Delphi study. Contemporary Nurse, 47, 16–26.
*A link to this open-access journal article is provided in the Toolkit for this chapter in the accompanying Resource Manual.
2 Evidence-Based Nursing: Translating Research Evidence into Practice
his book will help you to develop the skills you need to generate and evaluate research evidence for nursing practice. Before we delve into
the “how-tos” of research, we discuss key aspects of evidence-based practice (EBP) to clarify the key role that research plays in nursing.
BACKGROUND OF EVIDENCE-BASED NURSING PRACTICE This section provides a context for understanding evidence-based nursing practice and two closely related concepts, research utilization and knowledge translation.
Definition of Evidence-Based Practice Pioneer David Sackett defined evidence as “the integration of best research evidence with clinical expertise and patient values” (Sackett et al., 2000, p. 1). Scott and McSherry (2009), in their review of evidence-based nursing concepts, identified 13 overlapping but distinct definitions of evidence- based nursing and EBP. The definition proposed by Sigma Theta Tau International (2008) is as follows: “The process of shared decision-making between practitioner, patient, and others significant to them based on research evidence, the patient’s experiences and preferences, clinical expertise or know-how, and other available robust sources of information” (p. 57). A key ingredient in EBP is the effort to personalize “best evidence” to a specific patient’s needs within a particular clinical context.
A basic feature of EBP as a clinical problem-solving strategy is that it
de-emphasizes decisions based on custom, authority, or ritual. The emphasis is on identifying the best available research evidence and integrating it with other factors. In many areas of clinical decision making, research has demonstrated that “tried and true” practices taught in basic nursing education are not always best. For example, although many nurses not so long ago were taught to place infants in the prone sleeping position to prevent aspiration, there is strong evidence that the supine (back) sleeping position decreases the risk of sudden infant death syndrome (SIDS).
TIP: The consequences of not using research evidence can be devastating. For example, from 1956 through the 1980s, Dr. Benjamin Spock published several editions of a top-selling book, Baby and Child Care, which advised putting babies on their stomachs to sleep. In their systematic review of evidence, Gilbert and colleagues (2005) wrote, “Advice to put infants to sleep on the front for nearly half a century was contrary to evidence from 1970 that this was likely to be harmful” (p. 874). They estimated that if medical advice had been guided by research evidence, over 60,000 infant deaths might have been prevented.
Because research evidence can provide valuable insights about human health and illness, nurses must be lifelong learners who have the skills to search for, understand, and evaluate new information about patient care— as well as the capacity to adapt to change.
Research Utilization Research utilization (RU) is the use of findings from a study or set of studies in a practical application that is unrelated to the original research. In RU, the emphasis is on translating new knowledge into real-world applications. EBP is a broader concept than RU because it integrates research findings with other factors, as just noted. Also, whereas RU begins with the research itself (How can I put this new knowledge to good use in my clinical setting?), the start point in EBP is a clinical question (What does the evidence say is the best approach to solving this clinical problem?).
TIP: Theorists who have studied the diffusion of ideas recognize a continuum of knowledge utilization. At one end of the continuum are identifiable attempts to base specific actions on research findings (e.g., placing infants in supine instead of prone sleeping position). Research findings can also be used in a less focused manner—in a way that reflects awareness or enlightenment. Thus, a qualitative study might provide a rich description of courage among individuals with long-term health problems as a process that includes efforts to develop problem- solving skills. The study may make nurses more observant and sensitive in working with patients with long-term illnesses, but it may not lead to formal changes in clinical actions.
During the 1980s, research utilization emerged as an important buzzword. In education, nursing schools began to include courses on research methods so that students would become skillful research consumers. In research, there was a shift in focus toward clinical nursing problems. Yet, concerns about the limited use of research evidence in the delivery of nursing care continued to mount.
The need to reduce the gap between research and practice led to formal RU projects, including the groundbreaking Conduct and Utilization of Research in Nursing (CURN) Project, a 5-year project undertaken by the Michigan Nurses Association in the 1970s. CURN’s objectives were to increase the use of research findings in nurses’ daily practice by disseminating current findings and facilitating organizational changes needed to implement innovations (Horsley et al., 1978). The CURN project team concluded that RU by practicing nurses was feasible but only if the research is relevant to practice and if the results are broadly disseminated.
During the 1980s and 1990s, RU projects were undertaken by numerous hospitals and organizations. These projects were institutional attempts to implement changes in nursing practice based on research findings. During the 1990s, however, the call for research utilization began to be superseded by the push for EBP.
The Evidence-Based Practice Movement
The Cochrane Collaboration was an early contributor to the EBP movement. The collaboration was founded in the United Kingdom based on the work of British epidemiologist Archie Cochrane. Cochrane published an influential book in the 1970s that drew attention to the dearth of solid evidence about the effects of health care. He called for efforts to make research summaries of clinical trials available to health care providers. This eventually led to the development of the Cochrane Center in Oxford in 1993, and an international partnership called the Cochrane Collaboration, with centers established in locations throughout the world. Its aim is to help providers make good decisions about health care by preparing and disseminating systematic reviews of the effects of health care interventions.
At about the same time, a group from McMaster Medical School in Canada (including Dr. David Sackett) developed a clinical learning strategy they called evidence-based medicine. The evidence-based medicine movement has shifted to a broader conception of using best evidence by all health care practitioners (not just physicians) in a multidisciplinary team. EBP is considered a major shift for health care education and practice. In the EBP environment, a skillful clinician can no longer rely on a repository of memorized information but rather must be adept in accessing, evaluating, and using new evidence.
The EBP movement has advocates and critics. Supporters argue that EBP is a rational approach to providing the best possible care with the most cost-effective use of resources. Advocates also note that EBP provides a framework for self-directed lifelong learning that is essential in an era of rapid clinical advances and the information explosion. Critics worry that the advantages of EBP are exaggerated and that individual clinical judgments and patient inputs are being devalued. They are also concerned that insufficient attention is being paid to the role of qualitative research. Although there is a need for close scrutiny of how the EBP journey unfolds, an EBP path is the one that health care professions will almost surely follow in the years ahead.
TIP: A debate has emerged concerning whether the term “evidence- based practice” should be replaced with evidence-informed practice
(EIP). Those who advocate for a different term have argued that the word “based” suggests a stance in which patient values and preferences are not sufficiently considered in EBP clinical decisions (e.g., Glasziou, 2005). Yet, as noted by Melnyk (2014), all current models of EBP incorporate clinicians’ expertise and patients’ preferences. She argued that “changing terms now . . . will only create confusion at a critical time where progress is being made in accelerating EBP” (p. 348). We concur and we use EBP throughout this book.
Knowledge Translation Research utilization and EBP involve activities that can be undertaken at the level of individual nurses or at a higher organizational level (e.g., by nurse administrators), as we describe later in this chapter. In the early part of this century, a related movement emerged that mainly concerns system- level efforts to bridge the gap between knowledge generation and use. Knowledge translation (KT) is a term that is often associated with efforts to enhance systematic change in clinical practice.
It appears that the term was coined by the Canadian Institutes of Health Research (CIHR) in 2000. CIHR defined KT as “the exchange, synthesis, and ethically-sound application of knowledge—within a complex system of interactions among researchers and users—to accelerate the capture of the benefits of research for Canadians through improved health, more effective services and products, and a strengthened health care system” (CIHR, 2004, p. 4).
Several other definitions of KT have been proposed. For example, the World Health Organization (WHO) (2005) adapted the CIHR’s definition and defined KT as “the synthesis, exchange and application of knowledge by relevant stakeholders to accelerate the benefits of global and local innovation in strengthening health systems and improving people’s health.” Institutional projects aimed at KT often use methods and models that are similar to institutional EBP projects.
TIP: Translation science (or implementation science) has emerged as a discipline devoted to developing methods to promote knowledge translation. In nursing, the need for translational research
was an important impetus for the development of the Doctor of Nursing Practice degree. Several journals have emerged that are devoted to this field (e.g., the journal Implementation Science).
EVIDENCE-BASED PRACTICE IN NURSING Before describing procedures relating to EBP in nursing, we briefly discuss some important issues, including the nature of “evidence” and challenges to pursuing EBP, and resources available to address some of those challenges.
Types of Evidence and Evidence Hierarchies There is no consensus about the definition of evidence nor about what constitutes usable evidence for EBP, but most commentators agree that findings from rigorous research are paramount. Debate continues, however, about what constitutes rigorous research and what qualifies as best evidence.
At the outset of the EBP movement, there was a strong bias toward reliance on information from studies called randomized controlled trials (RCTs). This bias stemmed from the fact that the Cochrane Collaboration initially focused on the effectiveness of therapies rather on other types of health care questions. RCTs are, in fact, very well suited for drawing conclusions about the effects of health care interventions (Chapter 9). The bias in ranking sources of evidence in terms of questions about effective treatments led to some resistance to EBP by nurses who felt that evidence from qualitative and non-RCT studies would be ignored.
Positions about the contribution of various types of evidence are less rigid than previously. Nevertheless, many published evidence hierarchies rank evidence sources according to the strength of the evidence they provide, and in most cases, RCTs are near the top of these hierarchies. We offer a modified evidence hierarchy that looks similar to others, but ours illustrates that the ranking of evidence-producing strategies depends on the type of question being asked.
Figure 2.1 shows that systematic reviews are at the pinnacle of the hierarchy (Level I), regardless of the type of question, because the strongest evidence comes from careful syntheses of multiple studies. The
next highest level (Level II) depends on the nature of inquiry. For Therapy questions regarding the efficacy of an intervention (What works best for improving health outcomes?), individual RCTs constitute Level II evidence (systematic reviews of multiple RCTs are Level I). Going down the “rungs” of the evidence hierarchy for Therapy questions results in less reliable evidence—for example, Level III evidence comes from a type of study called quasi-experimental. In-depth qualitative studies are near the bottom, in terms of evidence regarding intervention effectiveness. (Terms in Figure 2.1 will be discussed in later chapters.)
For a Prognosis question, by contrast, Level II evidence comes from a single prospective cohort study, and Level III is from a type of study called case control (Level I evidence is from a systematic review of cohort studies). Thus, contrary to what is often implied in discussions of evidence hierarchies, there really are multiple hierarchies. If one is interested in best evidence for questions about Meaning, an RCT would be a poor source of evidence, for example. We have tried to portray the notion of multiple hierarchies in Figure 2.1, with information on the right indicating the type
of individual study that would offer the best evidence (Level II) for different questions. In all cases, appropriate systematic reviews are at the pinnacle. Information about different hierarchies for different types of cause-probing questions is addressed in Chapter 9.
Of course, within any level in an evidence hierarchy, evidence quality can vary considerably. For example, an individual RCT could be well designed, yielding strong Level II evidence for Therapy questions, or it could be so flawed that the evidence would be weak.
Thus, in nursing, best evidence refers to research findings that are methodologically appropriate, rigorous, and clinically relevant for answering persistent questions—questions not only about the efficacy, safety, and cost-effectiveness of nursing interventions but also about the reliability of nursing assessment tests, the causes and consequences of health problems, and the meaning and nature of patients’ experiences. Confidence in the evidence is enhanced when the research methods are compelling, when there have been multiple confirmatory studies, and when the evidence has been carefully evaluated and synthesized.
Of course, there continue to be clinical practice questions for which there is relatively little research evidence. In such situations, nursing practice must rely on other sources—for example, pathophysiologic data, chart review, quality improvement data, and clinical expertise. As Sackett and colleagues (2000) have noted, one benefit of the EBP movement is that a new research agenda can emerge when clinical questions arise for which there is no satisfactory evidence.
Evidence-Based Practice Challenges Nurses have completed many studies about the use of research in practice, including research on barriers to EBP. Studies on EBP barriers, conducted in several countries, have yielded similar results about constraints on clinical nurses. Most barriers fall into one of three categories: (1) quality and nature of the research, (2) characteristics of nurses, and (3) organizational factors.
With regard to the research, one problem is the limited availability of high-quality research evidence for some practice areas. There remains an ongoing need for research that directly addresses pressing clinical
problems, for replication of studies in a range of settings, and for greater collaboration between researchers and clinicians. Another issue is that nurse researchers need to improve their ability to communicate evidence, and the clinical implications of evidence, to practicing nurses.
Nurses’ attitudes and education are also potential barriers to EBP. Studies have found that some nurses do not value or know much about research, and others simply resist change. Fortunately, many nurses do value research and want to be involved in research-related activities. Nevertheless, many nurses do not know how to access research evidence and do not possess the skills to critically evaluate research findings—and even those who do may not know how to effectively incorporate research evidence into clinical decision making. Among nurses in non-English- speaking countries, another impediment is that most research evidence is reported in English.
Finally, many of the challenges to using research in practice are organizational. “Unit culture” can undermine research use, and administrative and other organizational barriers also play a major role. Although many organizations support the idea of EBP in theory, they do not always provide the necessary supports in terms of staff release time and availability of resources. Nurses’ time constraints are a crucial deterrent to the use of evidence at the bedside. Strong leadership in health care organizations is essential to making evidence-based practice happen.
RESOURCES FOR EVIDENCE-BASED PRACTICE IN NURSING The translation of research evidence into nursing practice is an ongoing challenge, but resources to support EBP are increasingly available. We urge you to explore other ideas with your health information librarian because the list of resources is growing as we write.
Preappraised Evidence Research evidence comes in various forms, the most basic of which is in individual studies. Primary studies published in professional journals are not preappraised for quality or use in practice. Chapter 5 discusses how to access primary studies for a literature review.
Preprocessed (preappraised) evidence is evidence that has been selected from primary studies and evaluated for use by clinicians. DiCenso and colleagues (2005) have described a hierarchy of preprocessed evidence. On the first rung above primary studies are synopses of single studies, followed by systematic reviews, and then synopses of systematic reviews. Clinical practice guidelines are at the top of the hierarchy. At each successive step in the hierarchy, the ease in applying the evidence to clinical practice increases. We describe several types of preappraised evidence sources in this section.
Systematic Reviews Evidence-based practice relies on meticulous integration of research evidence on a topic. Systematic reviews are a pivotal component of EBP: Their “bottom line” is a summary of what the best evidence is at the time the review was written. A systematic review is not just a literature review, such as ones we describe in Chapter 5. A systematic review is in itself a methodical, scholarly inquiry that follows many of the same steps as those for primary studies. Chapter 29 offers guidance on conducting and critiquing systematic reviews.
Systematic reviews can take various forms. A type of systematic review called a meta-analysis has emerged as an important EBP tool. Meta- analysis is a method of integrating quantitative findings statistically. In essence, meta-analysis treats the findings from a study as one piece of information. The findings from multiple studies on the same topic are combined and analyzed statistically. Instead of individual people being the unit of analysis (the basic entity of the analysis), individual studies are the unit of analysis in a meta-analysis. Meta-analysis is a convenient, objective method of integrating a body of findings and of observing patterns that might otherwise have gone undetected.
Example of a Meta-Analysis: Du and colleagues (2015) conducted a meta- analysis of the effectiveness of Tai Chi exercise for improving sleep quality in older people. The researchers integrated evidence from five clinical trials. The evidence from these studies suggests that Tai Chi could be an effective alternative and complementary approach to existing therapies for older people
with sleep disorders, but the researchers concluded that better confirmatory evidence is needed.
Integrative reviews of qualitative studies often take the form of metasyntheses, which are rich resources for EBP (Beck, 2009). A metasynthesis, which involves integrating qualitative research findings on a topic, is distinct from a quantitative meta-analysis: A metasynthesis is less about reducing information and more about amplifying and interpreting it. Strategies are also being developed in the area of mixed methods synthesis, which are efforts to integrate and synthesize both quantitative and qualitative evidence (Sandelowski et al., 2013; Thorne, 2009).
Example of a Metasynthesis: Tao and colleagues (2014) did a metasynthesis of 16 studies of the experiences of individuals living with a stoma. They identified three themes concerning patients’ personal awareness and behavioral choices on having a stoma: altered self, restricted life, and overcoming restrictions.
Some systematic reviews are published in professional journals that can be accessed using standard literature search procedures; others are available in dedicated databases. In particular, the Cochrane Database of Systematic Reviews contains thousands of systematic reviews (mostly meta-analyses) relating to health care interventions. Cochrane reviews are done with great rigor and have the advantage of being checked and updated regularly.
Example of Cochrane Review: Gillespie and colleagues (2014) conducted a Cochrane review that summarized evidence on the effects of repositioning on the prevention of pressure ulcers in adult patients. The team found evidence from only four studies and concluded that there remains a need for further high-quality research to “assess the effects of position and optimal frequency of repositioning on pressure ulcer incidence” (p. 1).
Many other resources are available for locating systematic reviews as
well as synopses of such reviews. Here is information about a few of them:
• The Agency for Healthcare Research and Quality (AHRQ) awarded contracts to establish Evidence-Based Practice Centers that issue evidence reports (www.ahrq.gov).
• The Centre for Reviews and Dissemination at the University of York (England) produces useful systematic reviews (http://www.york.ac.uk/inst/crd/index.htm).
• The Joanna Briggs Collaboration, centered in Australia with affiliates worldwide, is another useful source for systematic reviews in nursing and other health fields (http://joannabriggs.org/).
• The Campbell Collaboration includes reviews of interventions that are socially or behaviorally oriented (www.campbellcollaboration.org).
TIP: Websites cited in this chapter, plus additional websites with useful content relating to EBP, are listed in the Toolkit with the accompanying Resource Manual. This will allow you to simply use the “Control/Click” feature to go directly to the website, without having to type in the URL and risk a typographical error.
Clinical Practice Guidelines and Care Bundles Evidence-based clinical practice guidelines, like systematic reviews, represent an effort to distill a large body of evidence into a manageable form, but guidelines differ in a number of respects. First, clinical practice guidelines, which are usually based on systematic reviews, give specific recommendations for evidence-based decision making. Their intent is to influence what clinicians do. Second, guidelines attempt to address all of the issues relevant to a clinical decision, including the balancing of benefits and risks. Third, systematic reviews are evidence-driven—that is, they are undertaken when a body of evidence has been produced and needs to be synthesized. Guidelines, by contrast, are “necessity-driven” (Sackett et al., 2000), meaning that guidelines are developed to guide clinical practice—even when available evidence is limited or of unexceptional quality. Fourth, systematic reviews are done by researchers, but guideline development typically involves the consensus of a group of researchers,
experts, and clinicians. For this reason, guidelines based on the same evidence may result in different recommendations that take into account contextual factors—for example, guidelines appropriate in the United States may be unsuitable in India.
Also, organizations are developing and adopting care bundles—a concept developed by the Institute for Healthcare Improvement—that encompass a set of interventions to treat or prevent a specific cluster of symptoms (www.ihi.org). There is growing evidence that a combination or bundle of strategies produces better outcomes than a single intervention.
Example of a Care Bundle Project: Bates and colleagues (2014) explored the effect of implementing a set of interventions in incremental bundles to patients discharged after coronary artery bypass surgery. Patients who received the bundled interventions had a substantially lower rate of 30-day hospital readmissions than patients discharged prior to implementing the bundles.
Guidelines and bundles are available for many diagnostic and therapeutic decisions. Typically, they define a minimum set of services and actions appropriate for certain clinical conditions. Most guidelines allow for a flexible approach in their application to individual patients who fall outside the scope of their guideline (e.g., those with significant comorbidities).
It can be challenging to find clinical practice guidelines because there is no single guideline repository. One useful approach is to search for guidelines in comprehensive guideline databases, or through specialty organizations that have sponsored guideline development (e.g., the Association of Women’s Health, Obstetric and Neonatal Nurses or AWHONN). It would be impossible to list all possible sources, but a few deserve special mention.
• In the United States, nursing and other health care guidelines are maintained by the National Guideline Clearinghouse (www.guideline.gov).
• In Canada, information about clinical practice guidelines can be found through the Registered Nurses’ Association of Ontario (RNAO) (www.rnao.org/bestpractices).
• In the United Kingdom, two sources for clinical guidelines are the Translating Research Into Practice (TRIP) database (http://www.tripdatabase.com) and the National Institute for Health and Care Excellence (www.nice.org.uk).
• Another resource is the EBM-Guidelines, which offer recommendations relative to primary care in several languages (www.ebm- guidelines.com).
• The Guidelines International Network makes available guidelines from around the world (www.g-i-n.net).
In addition to looking for guidelines in national clearinghouses and in the websites of professional organizations, you can search bibliographic databases such as MEDLINE or EMBASE. Search terms such as the following can be used: practice guideline, clinical practice guideline, best practice guideline, evidence-based guideline, standards, and consensus statement. Be aware, though, that a standard search for guidelines in bibliographic databases will yield many references—but often a frustrating mixture of citations to not only the actual guidelines but also to commentaries, anecdotes, case studies, and so on.
Example of a Nursing Clinical Practice Guideline: In 2013, the Registered Nurses’ Association of Ontario (RNAO) published the second edition of a best practice guideline called Assessment and Management of Foot Ulcers for People with Diabetes. The guideline provides “direction to all nurses and the interprofessional team who provide care in all health care settings to people with type 1 and/or type 2 diabetes and who have established diabetic foot ulcers” (www.rnao.org).
There are many topics for which practice guidelines have not yet been developed, but the opposite problem is also true: The dramatic increase in the number of guidelines means that there are sometimes multiple guidelines on the same topic. Worse yet, because of variation in the rigor of guideline development and in interpreting the evidence, different guidelines sometimes offer different and even conflicting recommendations. Thus, those who wish to adopt clinical practice guidelines are urged to critically appraise them to identify ones that are
based on the strongest and most up-to-date evidence, have been meticulously developed, are user-friendly, and are appropriate for local use. We offer some assistance with these tasks later in this chapter.
Clinical Decision Support Tools Clinical decision support tools are designed to help nurses and other health care professionals to organize information, guide their assessments, and apply appropriate interventions. Among such decision support tools are clinical decision rules, which synthesize the best available evidence into convenient guides for practice (Shapiro, 2005). Such decision rules, by standardizing aspects of patient assessments and prescribing specific evidence-based actions, can minimize clinical uncertainty and reduce variations in practice at the bedside.
It has been argued that, to be useful, decision support tools must offer speedy guidance in real time. Technologic advances are making such point-of-care decision-making assistance possible. Computerized decisional support (on computers, tablets, and smart phones) is now available for various clinical settings and specific clinical problems (e.g., Doran, 2009; Doran et al., 2010); DiPietro and colleagues (2012) offer advice on how to make such support useful for point-of-care decisions.
Other Preappraised Evidence Several other types of preprocessed evidence are useful for EBP. These include the following:
• Synopses of systematic reviews and of single studies are available in evidence-based abstract journals such as Clinical Evidence (www.clinicalevidence.com) and Evidence-Based Nursing (www.evidencebasednursing.com). Evidence-Based Nursing presents critical summaries of studies and systematic reviews from more than 150 journals.
• An “evidence digest” feature appears in each issue of Worldviews on Evidence-Based Nursing. These digests offer concise summaries of clinically important studies, along with practice implications.
• AHRQ launched its Health Care Innovations Exchange program in 2008, which offers a repository of hundreds of effective health care
innovations (www.innovations.ahrq.gov). • The American Association of Critical-Care Nurses regularly publishes
“practice alerts,” which are evidence-based recommendations for practice changes (www.aacn.org).
Models for Evidence-Based Practice Several models of EBP have been developed and are important resources for designing and implementing EBP projects in practice settings. Some models focus on the use of research from the perspective of individual clinicians (e.g., the Stetler Model), but most focus on institutional EBP efforts (e.g., the Iowa Model). Another way to categorize existing models is to distinguish models that are process-oriented models (e.g., the Iowa Model) and models that are explicitly mentor models, such as the Clinical Nurse Scholar model and the ARCC model. Some of these models (e.g., the Ottawa Model of Research Use) have played a prominent role in KT efforts.
The many worthy EBP models are too numerous to list comprehensively, but a few are shown in Box 2.1. For those wishing to follow a formal EBP model, the cited references should be consulted. Gawlinski and Rutledge (2008) and Schaffer and colleagues (2013) offer further descriptions of EBP models and identify features to consider in selecting one to plan an EBP project.
BOX 2.1 Selected Models for Evidence- Based Practice
• ACE Star Model of Knowledge Transformation (Kring, 2008) • Advancing Research and Clinical Practice Through Close
Collaboration (ARCC) Model (Melnyk & Fineout-Overholt, 2015) • Diffusion of Innovations Theory (Rogers, 1995) • Framework for Adopting an Evidence-Based Innovation (DiCenso et
al., 2005) • Iowa Model of Evidence-Based Practice to Promote Quality Care
(Titler, 2010; Titler et al., 2001) • Johns Hopkins Nursing Evidence-Based Practice Model (Newhouse
et al., 2005) • Model for Change to Evidence-Based Practice (Rosswurm &
Larabee, 1999) • Ottawa Model of Research Use (Logan & Graham, 1998) • Promoting Action on Research Implementation in Health Services
(PARIHS) Model, (Rycroft-Malone, 2010; Rycroft-Malone et al., 2002)
• Stetler Model of Research Utilization (Stetler, 2001, 2010)
Although each model offers different perspectives on how to translate research findings into practice, several of the steps and procedures are similar across the models. In nursing, the most prominent of these models have been the nurse-developed PARIHS Model, the Stetler Model, and the Iowa Model.
We provide an overview of key activities and issues in EBP initiatives, based on a distillation of common elements from EBP models, in a subsequent section of this chapter. We rely especially heavily on the Iowa Model, a diagram for which is shown in Figure 2.2.
Example of Using an Evidence-Based Practice Model: C. G. Brown (2014) described how the Iowa Model was used to identify evidence-based strategies to reduce patient falls in an oncology unit. An interdisciplinary team worked to develop and implement a practice change and to evaluate effects on decreasing patient falls.
EVIDENCE-BASED PRACTICE IN INDIVIDUAL NURSING PRACTICE This section and the following one, which are based on the various models of EBP, provide an overview of how research can be put to use in clinical settings. More extensive guidance is available in textbooks devoted to evidence-based nursing (e.g., S. J. Brown, 2013; Craig & Smyth, 2012; DiCenso et al., 2005; Melnyk & Fineout-Overholt, 2015; Schmidt & Brown, 2011). We first discuss strategies and steps for individual clinicians and then describe activities used by teams of nurses.
Clinical Scenarios and the Need for Evidence Individual nurses make many decisions and are called upon to provide health care advice, and so they have ample opportunity to put research into practice. Here are four clinical scenarios that provide examples of such opportunities:
• Clinical Scenario 1. You work on an intensive care unit and notice that Clostridium difficile infection has become more prevalent among surgical patients in your hospital. You want to know if there is a reliable screening tool for assessing the risk of infection so that preventive measures could be initiated in a more timely and effective manner.
• Clinical Scenario 2. You work in an allergy clinic and notice how difficult it is for many children to undergo allergy scratch tests. You wonder if an interactive distraction intervention would help reduce children’s anxiety when they are being tested for allergens.
• Clinical Scenario 3. You work in a rehabilitation hospital and one of your elderly patients, who had total hip replacement, tells you she is planning a long airplane trip to visit her daughter after rehabilitation treatments are completed. You know that a long plane ride will increase her risk of deep vein thrombosis and wonder if compression stockings are an effective in-flight treatment. You decide to look for the best possible evidence to answer this question.
• Clinical Scenario 4. You are caring for a hospitalized cardiac patient who tells you that he has sleep apnea. He confides in you that he is reluctant to undergo continuous positive airway pressure (CPAP)
treatment because he worries it will hinder intimacy with his wife. You wonder if there is any evidence about what it is like to undergo CPAP treatment so that you can better understand how to address your patient’s concerns.
In these and thousands of other clinical situations, research evidence can be put to good use to improve the quality of nursing care. Some situations might lead to unit-wide or institution-wide scrutiny of current practices, but in other situations, individual nurses can personally investigate the evidence to help them address specific problems.
For individual-level EBP efforts, the major steps in EBP include the following:
1. Asking clinical questions that can be answered with research evidence 2. Searching for and retrieving relevant evidence 3. Appraising the evidence 4. Integrating the evidence with your own clinical expertise, patient
preferences, and local context 5. Assessing the effectiveness of the decision, intervention, or advice
Asking Well-Worded Clinical Questions A crucial first step in EBP involves converting information needs into well-worded clinical questions that can be answered with research evidence. Some EBP writers distinguish between background and foreground questions. Background questions are general, foundational questions about a clinical issue, for example: What is cancer cachexia (progressive body wasting), and what is its pathophysiology? Answers to such background questions are typically found in textbooks. Foreground questions, by contrast, are those that can be answered based on current best research evidence on diagnosing, assessing, or treating patients, or on understanding the meaning or prognosis of their health problems. For example, we may wonder, is a fish oil–enhanced nutritional supplement effective in stabilizing weight in patients with advanced cancer? The answer to such a question may offer guidance on how best to address the needs of patients with cachexia.
Most guidelines for EBP use the acronyms PIO or PICO to help practitioners develop well-worded questions that facilitate a search for
evidence. In the most basic PIO form, the clinical question is worded to identify three components:
1. P: the population or patients (What are the characteristics of the patients or people?)
2. I: the intervention, influence, or exposure (What are the interventions or therapies of interest? or, What are the potentially harmful influences/exposures of concern?)
3. O: the outcomes (What are the outcomes or consequences in which we are interested?)
Applying this scheme to our question about cachexia, our population (P) is cancer patients with cachexia; the intervention (I) is fish oil– enhanced nutritional supplements; and the outcome (O) is weight stabilization. As another example, in the second clinical scenario about scratch tests cited earlier, the population (P) is children being tested for allergies; the intervention is interactive distraction (I); and the outcome is anxiety levels (O).
For questions that can best be answered with qualitative information (e.g., about the meaning of an experience or health problem), two components are most relevant:
1. the population (What are the characteristics of the patients or clients?); and
2. the situation (What conditions, experiences, or circumstances are we interested in understanding?).
For example, suppose our question was, What is it like to suffer from cachexia? In this case, the question calls for rich qualitative information; the population is patients with advanced cancer and the situation is the experience of cachexia.
In addition to the basic PIO components, other components are sometimes important in an evidence search. In particular, a comparison (C) component may be needed, when the intervention or influence of interest is contrasted with a specific alternative. For example, we might be interested in learning whether fish oil–enhanced supplements (I) are better than melatonin (C) in stabilizing weight (O) in cancer patients (P). When a specific comparison is of interest, a PICO question is required, but if we
were interested in uncovering evidence about all alternatives to the intervention of primary interest, then PIO components are sufficient. (By contrast, when asking questions to undertake an actual study, the “C” must always be specified).
TIP: Other components may be relevant, such as a time frame in which an intervention might be appropriate (adding a “T” for PICOT questions) or a setting (adding an “S” for PICOS questions).
Table 2.1 offers question templates for asking well-framed clinical questions in selected circumstances. The right hand column includes questions with an explicit comparison (PICO questions), while the middle column does not. The questions are categorized in a manner similar to that discussed in Chapter 1 (EBP purposes), as featured in Table 1.3. One exception is that we have added description as a question type. Note that although there are some differences in components across question types, there is always a P component.
TIP: The Toolkit section of Chapter 2 in the accompanying Resource Manual includes Table 2.1 in a Word file that can be adapted for your use, so that the template questions can be readily “filled in.” Additional EBP resources from this chapter are also in the Toolkit.
Finding Research Evidence By asking clinical questions in the forms suggested, you should be able to more effectively search the research literature for the information you need. Using the templates in Table 2.1, the information inserted into the
blanks constitutes keywords for undertaking an electronic search. For an individual EBP endeavor, the best place to begin is to search for
evidence in a systematic review or other preappraised source because this leads to a quicker answer—and potentially a superior answer as well if your methodologic skills are limited. Researchers who prepare reviews and clinical guidelines usually have strong research skills and use exemplary standards in evaluating the evidence. Moreover, preappraised evidence is usually developed by teams of researchers, which means that the conclusions are cross-validated. Thus, when preprocessed evidence is available to answer a clinical question, you may not need to look any farther, unless the review is not recent. When preprocessed evidence cannot be located or is old, you will need to look for best evidence in primary studies, using strategies we describe in Chapter 5.
TIP: In Chapter 5, we provide guidance on using the free Internet resource, PubMed, for searching the bibliographic database MEDLINE. Of special interest to those engaged in an EBP search, PubMed offers a special tool for those seeking evidence for clinical decisions. The “Clinical Queries” link appears under the heading “PubMed Tools” on the PubMed Home Page. In another important database, CINAHL, it is now also possible to delimit a search with a “Clinical Queries” or “Evidence-Based Practice” limiter.
Appraising the Evidence After locating relevant evidence, it should be appraised before taking clinical action. Critical appraisal for EBP may involve several types of assessments (Box 2.2).
BOX 2.2 Questions for Appraising the Evidence
What is the quality of the evidence—that is, how rigorous and reliable is it? What is the evidence—what is the magnitude of effects? How precise is the estimate of effects?
What evidence is there of any side effect or side benefits? What is the financial cost of applying (and not applying) the evidence? Is the evidence relevant to my particular clinical situation?
The thoroughness of your appraisal depends on several factors, the most important of which is the nature of the clinical action for which evidence is being sought. Some actions have implications for patient safety, while others are more relevant to patient satisfaction. Using best evidence to guide nursing practice is important for a wide range of outcomes, but appraisal standards would be especially strict for evidence that could affect patient safety and morbidity.
Evidence Quality The first appraisal issue is the extent to which the findings are valid. That is, were the study methods sufficiently rigorous that the evidence is credible? We offer guidance on critiquing studies and evaluating the strength of evidence from primary studies throughout this book. If there are several primary studies and no existing systematic review, you would need to draw conclusions about the body of evidence taken as a whole. There are several methods for “grading” the quality of a body of evidence, as we discuss in Chapter 27. Clearly, you would need to put most weight on the most rigorous studies. Preappraised evidence is already screened and evaluated, but you may still need to judge its integrity.
Magnitude of Effects You also need to assess what the results actually are and whether they are clinically important. This criterion considers not whether the results are valid but what they are and how powerful are the effects. For example, consider clinical scenario number 3 cited earlier, which corresponds to the following clinical question: Does the use of compression stockings (I) lower the risk of flight-related deep vein thrombosis (O) for high-risk patients (P)? In our search, we find a relevant systematic review in the nursing literature—a meta-analysis of nine randomized controlled trials (Hsieh & Lee, 2005)—and another in the Cochrane database (Clarke et al., 2006). The conclusion of these reviews, based on reliable evidence, is that
compression stockings are effective and the magnitude of the effect, in terms of risk reduction, is fairly substantial. Thus, advice about the use of compression stockings may be appropriate, pending an appraisal of other factors.
Determining the magnitude of the effect for quantitative findings is especially important when an intervention is costly or when there are potentially negative side effects. If, for example, there is good evidence that an intervention is only marginally effective in improving a health problem, it is important to consider other factors (e.g., evidence of effects on quality of life).
There are various ways to quantify the magnitude of effects, many of which are described later in this book. An index known as the effect size, for example, can provide estimates of the magnitude of effects for outcomes for which average values can be computed (e.g., average body temperature). When outcomes can be dichotomized (e.g., occurrence versus nonoccurrence of a health problem), estimates of magnitude of the effect can be calculated as absolute risk reduction (ARR) or relative risk reduction (RRR). For example, if the RRR for the use of compression stockings was 50%, this would mean that this intervention reduced the risk of deep vein thrombosis by 50%, relative to what would occur in its absence. We describe methods of calculating these and other related indexes in Chapter 16.
The magnitude of effects also has a bearing on clinical significance. We discuss how to assess whether the findings from a study are clinically significant in Chapter 20.
Precision of Estimates Another consideration, relevant with quantitative evidence, is how precise the estimate of effect is. This level of appraisal requires some statistical sophistication and so we postpone our discussion of confidence intervals to Chapter 17. Suffice it to say that research results provide only an estimate of effects and it is useful to understand not only the exact estimate but also the range within which the actual effect probably lies.
If the evidence is judged to be valid and the magnitude of effects supports further consideration, supplementary information may still be important in guiding decisions. One issue concerns peripheral benefits and costs, evidence for which may emerge during your search. In framing your clinical question, you would have identified the key outcomes in which you were interested—for example, weight stabilization or weight gain for interventions to address cancer cachexia. Primary research on this topic, however, may have involved an examination of other outcomes that could be taken into account—for example, effects on quality of life, side effects, satisfaction, and so on.
Financial Issues Another issue concerns the financial cost of using the evidence. In some cases, costs may be small or nonexistent. For example, in clinical scenario 4, where the question concerned the experience of CPAP treatment, nursing action would be cost neutral because the evidence would be used to provide information and reassurance to the patient. Some interventions, however, are costly and so the amount of resources needed to put best evidence into practice would need to be factored into any decision. Of course, while the cost of a clinical decision needs to be considered, the cost of not taking action is equally important.
Clinical Relevance Finally, it is important to appraise the evidence in terms of its relevance for the clinical situation at hand—that is, for your patient in a specific clinical setting. Best practice evidence can most readily be applied to an individual patient in your care if he or she is similar to people in the study or studies under review. Would your patient have qualified for participation in the study—or is there some factor such as age, illness severity, or comorbidity that would have excluded him or her? DiCenso and colleagues (2005), who advised clinicians to ask whether there is some compelling reason to conclude that the results may not be applicable in their clinical situation, have written some useful tips on applying research results to individual patients.
Actions Based on Evidence Appraisals Appraisals of the evidence may lead you to different courses of action. You may reach this point and conclude that the evidence is not sufficiently sound, or that the likely effect is too small, or that the cost of applying the evidence is too high. The integration of appraisal information may suggest that “usual care” is the best strategy—or it may suggest the need for a new EBP inquiry. For instance, in the example about cachexia, you likely would have learned that recent best evidence suggests that fish oil– enhanced nutritional supplements may be an ineffective treatment (Ries et al., 2012). However, during your search you may have come across a Cochrane review that concluded that megestrol acetate improves appetite and weight gain in patients with cancer (Ruiz Garcia et al., 2013). This may lead to a new evidence inquiry and to discussions with other members of your health care team about nutrition protocols for your clinical setting. If, however, the initial appraisal of evidence suggests a promising clinical action, then you can proceed to the next step.
Integrating Evidence As the definition for EBP implies, research evidence needs to be integrated with other types of information, including your own clinical expertise and knowledge of your clinical setting. You may be aware of factors that would make implementation of the evidence, no matter how sound or promising, inadvisable.
Patient preferences and values are also important. A discussion with the patient may reveal negative attitudes toward a potentially beneficial course of action, contraindications (e.g., comorbidities), or possible impediments (e.g., lack of health insurance).
One final issue is the importance of integrating evidence from qualitative research, which can provide rich insights about how patients experience a problem, or about barriers to complying with a treatment. A new intervention with strong potential benefits may fail to achieve desired outcomes if it is not implemented with sensitivity and understanding of the patients’ perspectives. As Morse (2005) has so aptly noted, evidence from a clinical trial may tell you whether a pill is effective, but qualitative research can help you understand why patients may not swallow the pill.
Implementing the Evidence and Evaluating Outcomes After the first four steps of the EBP process have been completed, you can use the resulting information to make an evidence-based decision or provide research-informed advice. Although the steps in the process, as just described, may seem complicated, in reality, the process can be efficient—if there is an adequate evidence base and especially if it has been skillfully preprocessed. EBP is most challenging when findings from research are contradictory, inconclusive, or “thin”—that is to say, when better quality evidence is needed.
One last step in an individual EBP effort concerns evaluation. Part of the evaluation involves assessing whether your action achieved the desired outcome. Another part concerns an evaluation of how well you are performing EBP. Sackett and colleagues (2000) offer self-evaluation questions that relate to the five EBP steps, such as asking answerable questions (Am I asking any clinical questions at all? Am I asking well- formulated question?) and finding external evidence (Do I know the best sources of current evidence? Am I becoming more efficient in my searching?). A self-appraisal may lead you to conclude that at least some of the clinical questions in which you are interested are best addressed as a group effort.
EVIDENCE-BASED PRACTICE IN AN ORGANIZATIONAL CONTEXT Most nurses practice in organizations, such as hospitals or long-term care settings. For some clinical scenarios that trigger an EBP effort, individual nurses may have sufficient autonomy that they can implement research- informed actions on their own (e.g., answering patients’ questions about experiences with CPAP). In many situations, however, decisions are best made among a team of nurses working together to solve a common clinical problem. This section describes some additional considerations that are relevant to institutional efforts at EBP—efforts designed to result in a formal policy or protocol affecting the practice of many nurses.
Many of the steps in organizational EBP projects are similar to the ones described in the previous section. For example, asking questions and gathering and appraising evidence are key activities in both. However,
there are additional issues of relevance at the organizational level.
Selecting a Problem for an Organizational Evidence-Based Practice Project An institutional EBP effort can emerge in response to clinical scenarios such as those presented earlier but can also arise in other contexts such as quality improvement efforts. Some EBP projects are “bottoms-up” efforts that originate in discussions among clinicians who develop ideas for problem-solving innovations. Others are “top-down” efforts in which administrators take steps to stimulate creative thought and the use of research evidence. This latter approach often occurs as part of the Magnet recognition process.
Several EBP models distinguish two types of “triggers” for an EBP project—(1) problem-focused triggers—a clinical practice problem in need of solution, or (2) knowledge-focused triggers—readings in the research literature. Problem-focused triggers may arise in the normal course of clinical practice, as in the clinical scenarios described earlier. A problem- focused approach is likely to have staff support if the problem is widespread.
A second catalyst, knowledge-focused triggers, is research evidence itself. Sometimes this catalyst is a new clinical guideline, and in other cases, the impetus emerges from discussions in a journal club. For EBP projects with knowledge-focused triggers, an assessment of clinical relevance might be needed—that is, will a problem of significance to nurses in that setting be solved by introducing an innovation? Titler and Everett (2001) offered suggestions for selecting interventions, using concepts from Rogers’s (1995) influential Diffusion of Innovations Theory.
With both types of triggers, consensus about the problem’s importance and the need for improving practice is crucial. In the Iowa Model (Figure 2.2), the first decision point involves determining whether the topic is a priority for the organization considering practice changes. Titler and colleagues (2001) advised that, when finalizing a topic, the following issues be taken into account: the topic’s fit with the organization’s strategic plan, the magnitude of the problem, the number of people
invested in the problem, support of nurse leaders and of those in other disciplines, costs, and possible barriers to change.
Addressing Practical Issues in Organizational Evidence- Based Practice Efforts The most pervasive barriers to EBP are organizational, and so one upfront issue is that nurse administrators need to create structures and processes that facilitate research translation. Nursing leaders can support EBP as an approach to clinical decision making in many ways, including providing nurses with sufficient time away from their daily clinical responsibilities to undertake EBP activities, making available financial and material resources, and developing collaborations with mentors who can provide guidance and direction in the search for and appraisal of evidence.
In an organizational EBP project, some practical matters should be resolved even before a search for evidence begins. One issue concerns the team itself. A motivated and inspiring team leader is essential. The recruitment and development of EBP team members often requires an interdisciplinary perspective. Identifying tasks to be undertaken, developing a realistic timetable and budget, assigning members to tasks, and scheduling meetings are necessary to ensure that the effort will progress. Finally, it is wise for the team to solicit the support of stakeholders who might influence project activities and the eventual implementation of EBP changes.
Finding and Appraising Evidence for Organizational Evidence-Based Practice For an organizational EBP effort, the best possible scenario involves identifying an appropriate clinical practice guideline, care bundle, or other decision support tool that has been based on rigorous research evidence. For some problem areas, however, clinical guidelines will need to be developed based on the evidence and not just implemented or adapted for use.
If a relevant guideline is identified, it should be carefully appraised. Several guideline appraisal instruments are available, but the one that has gained the broadest support is the Appraisal of Guidelines Research and
Evaluation (AGREE) Instrument, now in its second version (Brouwers et al., 2010). This tool has been translated into many languages and has been endorsed by the World Health Organization. The AGREE II instrument consists of ratings of quality on a 7-point scale (from strongly agree to strongly disagree) for 23 quality dimensions organized in six domains: scope and purpose, stakeholder involvement, rigor of development, clarity and presentation, applicability, and editorial independence (plus two global assessment ratings). As examples, one of the statements in the Scope and Purpose domain is: “The population (patients, public, etc.) to whom the guideline is meant to apply is specifically described”; one of the statements in the Rigor of Development domain is: “The guideline has been externally reviewed by experts prior to its publication.” The AGREE instrument should be applied to the guideline under consideration by a team of two to four appraisers. The Supplement to this chapter briefly discusses aspects of the AGREE II instrument. Another shorter and simpler tool for evaluating guideline quality is called the iCAHE Guideline Quality Check List (Grimmer et al., 2014).
One final issue is that guidelines change more slowly than the evidence. If a high-quality guideline is not recent, it is advisable to determine whether more up-to-date evidence would alter (or strengthen) the guideline’s recommendations. It has been recommended that, to avoid obsolescence, guidelines should be reassessed every 3 years.
Making Decisions Based on Evidence Appraisals In the Iowa Model, the synthesis and appraisal of research evidence provides the basis for a second major decision. The crux of the decision concerns whether the evidence is sufficient to justify an EBP change—for example, whether an existing clinical practice guideline is of sufficiently high quality that it can be used or adapted locally or whether (in the absence of a guideline) research evidence is sufficiently rigorous to recommend a practice innovation.
Coming to conclusions about the adequacy of research evidence can result in several possible outcomes leading to different paths. If the research base is weak, the team could either abandon the EBP project, or they could assemble other types of evidence (e.g., through consultation
with experts or surveys of clients) and assess whether these sources suggests a practice change. Another possibility is to pursue an original clinical study to address the question directly. This course of action may be impractical and would result in years of delay before conclusions could be drawn. If, on the other hand, there is a solid evidence base or a high- quality clinical practice guideline, then the team could develop plans for moving forward with implementing a practice innovation.
Assessing Implementation Potential In some EBP models, the next step is the development and testing of the innovation, followed by an assessment of organizational “fit.” Other models recommend early steps to assess the appropriateness of the innovation within the organizational context. In some cases, such an assessment may be warranted even before searching for and appraising evidence. We think an early assessment of the implementation potential (or, environmental readiness) of a clinical innovation is often sensible, although some situations have little need for a formal assessment.
In determining the implementation potential of an innovation in a particular setting, several issues should be considered, particularly the transferability of the innovation, the feasibility of implementing it, and its cost–benefit ratio.
• Transferability. Transferability concerns whether it makes sense to implement the innovation in your practice setting. If some aspects of the setting are fundamentally incongruent with the innovation—in terms of its philosophy, types of client served, staff, or administrative structure—then it might not make sense to try to adopt the innovation, even if there is evidence of clinical effectiveness in other contexts. One possibility, however, is that some organizational changes could be made to make the “fit” better.
• Feasibility. Feasibility questions address practical concerns about the availability of staff and resources, the organizational climate, the need for and accessibility of external assistance, and the potential for clinical evaluation. An important issue is whether nurses will have, or share, control over the innovation. If nurses will not have control over a new procedure, the interdependent nature of the project should be identified
early so that the EBP team will have needed interdisciplinary representatives.
• Cost–benefit ratio. A critical part of a decision to proceed with an EBP project is a careful assessment of costs and benefits of the change. The cost–benefit assessment should encompass likely costs and benefits to various groups (e.g., clients, nurses, the overall organization). If the degree of risk in introducing an innovation is high, then potential benefits must be great and the evidence must be very sound. A cost– benefit assessment should consider the opposite side of the coin as well: the costs and benefits of not instituting an innovation. The status quo bears its own risks and failure to change—especially when such change is based on firm evidence—can be costly to clients, to organizations, and to the entire nursing community.
TIP: The Toolkit for Chapter 2 in the Resource Manual has a worksheet with a series of questions for assessing the implementation potential of a potential innovation.
If the implementation assessment suggests that there might be problems in testing the innovation in that particular practice setting, then the team can either begin the process anew with a different innovation or pursue a plan to improve the implementation potential (e.g., seeking external resources if costs were the inhibiting factor).
TIP: Documentation of all steps in the EBP process, including the implementation potential of an innovation, is highly recommended. Committing ideas to writing is useful because it can help to resolve ambiguities, can serve as a problem-solving tool if problems emerge, and can be used to persuade others of the value of the project. All aspects of the EBP project should be transparent.
Developing Evidence-Based Protocols If the implementation criteria are met and the evidence is adequate, the team can prepare an action plan to move the effort forward, which includes
laying out strategies for designing and piloting the new clinical practice. In most cases, a key activity will involve developing a local evidence-based clinical practice protocol or guideline or adapting an existing one.
If a relevant clinical practice guideline has been judged to be of sufficiently high quality, the EBP team needs to decide whether to (1) adopt it in its entirety, (2) adopt only certain recommendations, while disregarding others (e.g., recommendations for which the evidence is less sound), or (3) make adaptations deemed necessary based on local circumstances. The risk in modifying guidelines is that the adaptation will not adequately incorporate the research evidence.
If there is no existing clinical practice guideline, or if existing guidelines are weak, the team will need to develop its own protocol or guideline reflecting the accumulated research evidence. Strategies for developing clinical practice guidelines are suggested in most textbooks on EBP and in several handbooks (Ansari & Rashidian, 2012; Turner et al., 2008). Whether a guideline is developed “from scratch” or adapted from an existing one, independent peer review is advisable to ensure that the guidelines are clear, comprehensive, and congruent with best existing evidence.
TIP: Guidelines should be user-friendly. Visual devices such as flowcharts and decision trees are often useful.
Implementing and Evaluating the Innovation Once an EBP protocol has been developed, the next step is to pilot test it (give it a trial run) in a clinical setting and to evaluate the outcome. Building on the Iowa Model, this phase of the project likely would involve the following activities:
1. Developing an evaluation plan (e.g., identifying outcomes to be achieved, deciding how many clients to involve, settling on when and how often to collect outcome information)
2. Collecting information on the outcomes for clients prior to implementing the innovation to develop a comparison against which the outcomes of the innovation can be assessed
3. Training staff in the use of the new protocol and, if necessary, “marketing” the innovation to users so that it is given a fair test
4. Trying the protocol out on one or more units or with a group of clients 5. Evaluating the pilot project, in terms of both process (e.g., How was the
innovation received, what implementation problems were encountered?) and outcomes (e.g., How were outcomes affected, what were the costs?)
TIP: The Registered Nurses’ Association of Ontario or RNAO (2012) has developed a toolkit to facilitate the implementation of clinical practice guidelines. The toolkit (second edition) is available at http://rnao.ca/bpg/resources/toolkit-implementation-best-practice- guidelines-second-edition.
A variety of research strategies and designs can be used to evaluate the innovation (see Chapter 11). In most cases, an informal evaluation will be adequate, for example, comparing outcome information from hospital records before and after the innovation and gathering information about patient and staff satisfaction. Qualitative information can also contribute to the evaluation: Qualitative data can uncover subtleties about the implementation process and help to explain findings.
Evaluation information should be gathered over a sufficiently long period (6 to 12 months) to allow for a true test of a “mature” innovation. An even longer time frame is useful for learning about the sustainability of an innovation. The end result is a decision about whether to adopt the innovation, to modify it for ongoing use, or to revert to prior practices. Another advisable step is to disseminate the results so that other nurses and nursing administrators can benefit. Finally, the EBP team should develop a plan for when the new protocol will be reviewed and, if necessary, updated based on new research evidence or ongoing feedback about outcomes.
TIP: Every nurse can play a role in using research evidence. Here are some strategies:
• Read widely and critically. Professionally accountable nurses keep
abreast of important developments and read journals relating to their specialty, including research reports in them.
• Attend professional conferences. Nursing conferences include presentations of studies with clinical relevance. Conference attendees have opportunities to meet researchers and to explore practice implications.
• Insist on evidence that a procedure is effective. Every time nurses or nursing students are told about a standard nursing procedure, they have a right to ask: Why? Nurses need to develop expectations that the clinical decisions they make are based on sound, evidence-based rationales.
• Become involved in a journal club. Many organizations that employ nurses sponsor journal clubs that review studies with potential relevance to practice. The traditional approach for a journal club (nurses coming togeth er as a group to discuss and critique an article) is in some settings being replaced with online journal clubs that acknowledge time constraints and the inability of nurses from all shifts to come together at one time.
• Pursue and participate in EBP projects. Several studies have found that nurses who are involved in research activities (e.g., an EBP project or data collection activities) develop more positive attitudes toward research and better research skills.
RESEARCH EXAMPLE Thousands of EBP projects are underway in practice settings. Many that have been described in the nursing literature offer useful information about planning and implementing such an endeavor. One is described here, and another full article is included in the Resource Manual.
Study: Implementing skin-to-skin contact at birth using the Iowa Model (Haxton et al., 2012).
Purpose: An evidence-based practice implementation project was undertaken at a Midwestern academic medical center. The focus of the project was to promote early skin-to-skin contact (SSC) as a best
practice for healthy term newborns. Framework: The project used the Iowa Model as its guiding framework.
The EBP team identified early SSC as having a knowledge-focused trigger: New guidelines supporting SSC had recently been published, and yet many nurses in the medical center did not routinely engage in the practice.
Early Iowa Model Steps: After conducting a small study in which 30 mothers of health newborns were interviewed, the project leaders decided to focus on “improving staff knowledge and philosophy of care related to SSC” (p. 224). They addressed the issue of organizational priorities by making a presentation of evidence on the benefits of SSC to the hospital’s Best Practice Committee. The Committee and the leadership within the hospital’s Birth Center gave their enthusiastic support for the project. Next, a project team was organized. The team consisted of nurses, physicians, nursing assistants, and lactation consultants, under the leadership of the Birth Center’s Clinical Nurse Specialist. Nurses from all shifts joined the team. Members of the team participated in a group approach to assembling, reviewing, and critiquing the evidence.
Protocol Development: The team discovered that implementing early SSC was in conflict with standard practices at the Birth Center, which involved having a second nurse in the patient’s room to accept the newborn from the physician or midwife, complete a series of common measurements and interventions, wrap the newborn in a blanket, and then hand the infant to the mother. After reviewing evidence that delaying common newborn interventions would not put the baby at risk, the team concluded that if a newborn was stable and the mother expressed a desire to perform SSC, then the second nurse could be dismissed earlier. Subsequently, the team “negotiated a detailed step-by- step protocol for early SSC intervention after vaginal birth” (p. 226). Educational material for mothers was developed. The electronic medical record also had to be modified.
Evaluation: The protocol was then pilot tested, beginning with the development and implementation of four training sessions for the labor and delivery nursing staff. Once training was complete, several
evaluation activities were undertaken. For example, the team first examined whether SSC was actually being offered and delivered. In the pilot study, several outcomes were assessed such as the rate of breastfeeding initiation and maternal satisfaction. The evaluation involved examination of charts before and after implementing the protocol and interviews with staff and mothers.
Findings and Conclusions: Based on chart review in the four quarters before and after initiating the pilot, the researchers found that the rate of breastfeeding initiation increased from around 74% to 84%. Most mothers who were interviewed acknowledged that their nurse had explained SSC and its benefits to them, and most made positive comments about the SSC experience. Nurses anecdotally reported that the protocol did not increase their workload and did not result in delays in transferring mothers to the postpartum unit. The authors of the article acknowledged that the project was challenging and many barriers were encountered, but they emphasized that they were able to overcome the challenges and were encouraged by the preliminary results from the pilot.
• Evidence-based practice (EBP) is the conscientious integration of current best evidence with clinical expertise and patient preferences in making clinical decisions; it is a clinical problem-solving strategy that de-emphasizes decision making based on custom.
• Research utilization (RU) and EBP are overlapping concepts that concern efforts to use research as a basis for clinical decisions, but RU starts with a research-based innovation that gets evaluated for possible use in practice.
• Knowledge translation (KT) is a term used primarily about system- wide efforts to enhance systematic change in clinical practice or policies.
• Two underpinnings of the EBP movement are the Cochrane Collaboration (which is based on the work of British epidemiologist
Archie Cochrane) and the clinical learning strategy called evidence- based medicine developed at the McMaster Medical School.
• EBP typically involves weighing various types of evidence in an effort to determine best evidence. Often, an evidence hierarchy is used to grade study findings according to the strength of evidence provided, but different hierarchies are appropriate for different types of questions. In all evidence hierarchies, however, systematic reviews are at the pinnacle.
• Resources to support EBP are growing at a phenomenal pace. Among the resources are systematic reviews (and electronic databases that make them easy to locate); evidence-based clinical practice guidelines, care bundles, and other decision support tools; a wealth of other preappraised evidence that makes it possible to practice EBP efficiently; and models of EBP that provide a framework for planning and undertaking EBP projects.
• Systematic reviews are rigorous integrations of research evidence from multiple studies on a topic. Systematic reviews can involve either qualitative, narrative approaches to integration (including metasynthesis of qualitative studies), or quantitative methods (meta- analysis) that integrate findings statistically.
• Evidence-based clinical practice guidelines combine a synthesis and appraisal of research evidence with specific recommendations for clinical decision making. Clinical practice guidelines should be carefully and systematically appraised, for example, using the Appraisal of Guidelines Research and Evaluation (AGREE II) Instrument.
• Many models of EBP have been developed, including models that provide a framework for individual clinicians (e.g., the Stetler Model) and others for organizations or teams of clinicians (e.g., the Iowa Model of Evidence-Based Practice to Promote Quality Care, Promoting Action on Research Implementation in Health Services or PARIHS Model).
• Individual nurses can put research into practice, using five basic steps: (1) framing an answerable clinical question, (2) searching for relevant research evidence, (3) appraising and synthesizing the evidence, (4)
integrating evidence with other factors, and (5) assessing effectiveness.
• One scheme for asking well-worded clinical questions involves four primary components, an acronym for which is PICO: Population (P), Intervention or influence (I), Comparison (C), and Outcome (O). When there is no explicit comparison, the acronym is PIO.
• An appraisal of the evidence involves such considerations as the validity of study findings, their clinical importance, the precision of estimates of effects, associated costs and risks, and utility in a particular clinical situation.
• EBP in an organizational context involves many of the same steps as an individual EBP effort but tends to be more formalized and must take organizational and interpersonal factors into account. “Triggers” for an organizational project include both pressing clinical problems and existing knowledge.
• Team-based or organizational EBP projects typically involve the development or adaptation of clinical protocols. Before these products can be tested, there should be an assessment of the implementation potential of the innovation, which includes the dimensions of transferability of findings, feasibility of using the findings in the new setting, and the cost–benefit ratio of a new practice.
• Once an evidence-based protocol or guideline has been developed and deemed worthy of implementation, the team can move forward with a pilot test of the innovation and an assessment of the outcomes prior to widespread adoption.
Chapter 2 of the Resource Manual for Nursing Research: Generating and Assessing Evidence for Nursing Practice, 10th edition, offers study suggestions for reinforcing concepts presented in this chapter. In addition, the following questions can be addressed in classroom or online discussions:
1. Think about your own clinical situation and identify a problem area. Now pose a well-worded clinical question using the templates in Table 2.1. Identify the various components of the question—that is, population, intervention or issue, comparison, and outcome.
2. Discuss the overall approach used in the example featured at the end of this chapter (Haxton et al., 2012).
STUDIES CITED IN CHAPTER 2
*Ansari, S., & Rashidian, A., (2012). Guidelines for guidelines: Are they up to the task? A comparative assessment of clinical practice guideline development handbooks. PloS One, 7, e49864.
Bates, O. L., O’Connor, N., Dunn, D., & Hasenau, S. (2014). Applying SRAAT interventions in incremental bundles: Improving post-CABG surgical patient care. Worldviews on Evidence-Based Nursing, 11, 89–97.
Beck, C. (2009). Metasynthesis: A goldmine for evidence-based practice. AORN Journal, 90, 701–702.
*Brouwers, M., Kho, M., Browman, G., Burgers, J., Cluzeau, F., Feder, G., . . . Zitzelsberger, L. (2010). AGREE II: Advancing guideline development, reporting and evaluation in health care. Canadian Medical Association Journal, 182, E839–E842.
Brown, C. G. (2014). The Iowa model of evidence-based practice to promote quality care: An illustrated example in oncology nursing. Clinical Journal of Oncology Nursing, 18, 157–159.
Brown, S. J. (2013). Evidence-based nursing: The research-practice connection (3rd ed.). Boston: Jones and Bartlett.
Canadian Institutes of Health Research. (2004). Knowledge translation strategy 2004–2009: Innovation in action. Ottawa, Canada: Author.
Clarke, M., Hopewell, S., Juszczak, E., Eisinga, A., & Kjeldstrom, M. (2006). Compression stockings for preventing deep vein thrombosis in airline passengers. Cochrane Database of Systematic Reviews, (2), CD004002.
Craig, J., & Smyth, R. (2012). The evidence-based practice manual for nurses (3rd ed.). Edinburgh, Scotland: Churchill Livingstone Elsevier.
DiCenso, A., Guyatt, G., & Ciliska, D. (2005). Evidence-based nursing: A guide to clinical practice. St. Louis, MO: Elsevier Mosby.
DiPietro, T., Nhuyen, H., & Doran, D. (2012). Usability evaluation: Results from “evaluation of mobile information technology to improve nurses’ access to and use of research evidence.” Computers, Informatics, Nursing, 30, 440–448.
Doran, D. (2009). The emerging role of PDAs in information use and clinical decision making. Evidence-Based Nursing, 12, 35–38.
Doran, D., Haynes, R., Kushniruk, A., Straus, S., Grimshaw, J., Hall, L., . . . Jedras, D. (2010). Supporting evidence-based practice for nurses through information technologies. Worldviews on Evidence-Based Nursing, 7, 4–15.
Du, S., Dong, J., Zhang, H., Jin, S., Xu, G., Liu, Z., . . . Sun, Z. (2015). Tai chi exercise for self-rated sleep quality in older people: A systematic review and meta-analysis. International Journal of Nursing Studies, 52, 368–379.
Gawlinski, A., & Rutledge, D. (2008). Selecting a model for evidence-based practice changes. AACN Advanced Critical Care, 19, 291–300.
*Gilbert, R., Salanti, G., Harden, M., & See, S. (2005). Infant sleeping position and the sudden infant death syndrome: Systematic review of observational studies and historical review of recommendations from 1940 to 2002. International Journal of Epidemiology, 34, 874–887.
Gillespie, B., Chaboyer, W., McInnes, E., Kent, B., Whitty, J., & Thalib, L. (2014). Repositioning for pressure ulcer prevention in adults. Cochrane Database of Systematic Reviews, (4), CD009958.
Glasziou, P. (2005). Evidence-based medicine: Does it make a difference? Make it evidence informed with a little wisdom. British Medical Journal, 330(7482), 92.
*Grimmer, K., Dizon, J., Milanese, S., King, E., Beaton, K., Thorpe, O., . . . Kumar, S. (2014). Efficient clinical evaluation of guideline quality: Development and testing of a new tool. BMC Medical Research Methodology, 14, 63.
Haxton, D., Doering, J., Gingras, L., & Kelly, L. (2012). Implementing skin-to- skin contact at birth using the Iowa Model. Nursing for Women’s Health, 16, 221–230.
Horsley, J. A., Crane, J., & Bingle, J. D. (1978). Research utilization as an organizational process. Journal of Nursing Administration, 8, 4–6.
Hsieh, H. F., & Lee, F. P. (2005). Graduated compression stockings as prophylaxis for flight-related venous thrombosis: Systematic literature review. Journal of Advanced Nursing, 51, 83–98.
Kring, D. L. (2008). Clinical nurse specialist practice domains and evidence-based practice competencies: A matrix of influence. Clinical Nurse Specialist, 22, 179–183.
Logan, J., & Graham, I. (1998). Toward a comprehensive interdisciplinary model of health care research use. Science Communication, 20, 227–246.
Melnyk, B. M. (2014). Evidence-based practice versus evidence-informed practice: A debate that could stall forward momentum in improving healthcare quality, safety, patient outcomes, and costs. Worldviews on Evidence-Based
Nursing, 11, 347–349. Melnyk, B. M., & Fineout-Overholt, E. (2015). Evidence-based practice in
nursing and healthcare (3rd ed.). Philadelphia: Lippincott Williams & Wilkins. Morse, J. M. (2005). Beyond the clinical trial: Expanding criteria for evidence.
Qualitative Health Research, 15, 3–4. Newhouse, R., Dearholt, S., Poe, S., Pugh, L. C., & White, K. M. (2005).
Evidence-based practice: A practical approach to implementation. Journal of Nursing Administration, 35, 35–40.
Registered Nurses’ Association of Ontario. (2012). Toolkit: Implementation of best practice guidelines (2nd ed.). Retrieved from http://rnao.ca/bpg/resources/toolkit-implementation-best-practice-guidelines- second-edition
Registered Nurses’ Association of Ontario. (2013). Assessment and management of foot ulcers for people with diabetes (2nd ed.). Retrieved from http://rnao.ca/bpg/guidelines/assessment-and-management-foot-ulcers-people- diabetes-second-edition
Ries, A., Trottenberg, P., Elsner, F., Stiel, S., Haugen, D., Kaasa, S., & Radbruch, L. (2012). A systematic review on the role of fish oil for the treatment of cachexia in advanced cancer. Palliative Medicine, 26, 294–304.
Rogers, E. M. (1995). Diffusion of innovations (4th ed.). New York: Free Press. Rosswurm, M. A., & Larrabee, J. H. (1999). A model for change to evidence-
based practice. Image: The Journal of Nursing Scholarship, 31, 317–322. Ruiz Garcia, V., López-Briz, E., Carbonell Sanchis, R., Gonzalvez Perales, J., &
Bort-Marti, S. (2013). Megestrol acetate for treatment of anorexia-cachexia syndrome. Cochrane Database of Systematic Reviews, (3), CD004310.
Rycroft-Malone, J. (2010). Promoting Action on Research Implementation in Health Services(PARIHS). In J. Rycroft-Malone & T. Bucknall (Eds.), Models and frameworks for implementing evidence-based practice: Linking evidence to action (pp. 109–133). Malden, MA: Wiley-Blackwell.
*Rycroft-Malone, J., Kitson, A., Harvey, G., McCormack, B., Seers, K., Titchen, A., & Estabrooks, C. (2002). Ingredients for change: Revisiting a conceptual framework. Quality and Safety in Health Care, 11, 174–180.
Sackett, D. L., Straus, S. E., Richardson, W. S., Rosenberg, W., & Haynes, R. B. (2000). Evidence-based medicine: How to practice and teach EBM (2nd ed.). Edinburgh, Scotland: Churchill Livingstone.
Sandelowski, M., Voils, C. I., Crandell, J. L., & Leeman, J. (2013). Synthesizing qualitative and quantitative research findings. In C. T. Beck (Ed.), Routledge international handbook of qualitative nursing research (pp. 347–356). New York: Routledge.
Schaffer, M. A., Sandau, K., & Diedrick, L. (2013). Evidence-based practice
models for organizational change: Overview and practical applications. Journal of Advanced Nursing, 69, 1197–1209.
Schmidt, N. A., & Brown, J. M. (2011). Evidence-based practice for nurses: Appraisal and application of research (2nd ed.). Sudbury, MA: Jones & Bartlett.
Scott, K., & McSherry, R. (2009). Evidence-based nursing: Clarifying the concepts for nurses in practice. Journal of Clinical Nursing, 18, 1085–1095.
Shapiro, S. E. (2005). Evaluating clinical decision rules. Western Journal of Nursing Research, 27, 655–664.
Sigma Theta Tau International. (2008). Sigma Theta Tau International position statement on evidence-based practice, February 2007 summary. Worldviews of Evidence-Based Nursing, 5, 57–59.
Stetler, C. B. (2001). Updating the Stetler model of research utilization to facilitate evidence-based practice. Nursing Outlook, 49, 272–279.
Stetler, C. B. (2010). Stetler model. In J. Rycroft-Malone & T. Bucknall (Eds.), Models and frameworks for implementing evidence-based practice: Linking evidence to action (pp. 51–77). Malden, MA: Wiley-Blackwell.
Tao, H., Songwathana, P., Isaralali, S., & Zhang, Y. (2014). Personal awareness and behavioural choices on having a stoma: A qualitative metasynthesis. Journal of Clinical Nursing, 23, 1186–1200.
Thorne, S. (2009). The role of qualitative research within an evidence-based context: Can metasynthesis be the answer? International Journal of Nursing Studies, 46, 569–575.
Titler, M. (2010). Iowa model of evidence-based practice. In J. Rycroft-Malone & T. Bucknall (Eds.), Models and frameworks for implementing evidence-based practice: Linking evidence to action (pp. 169–182). Malden, MA: Wiley- Blackwell.
Titler, M. G., & Everett, L. Q. (2001). Translating research into practice. Considerations for critical care investigators. Critical Care Nursing Clinics of North America, 13, 587–604.
Titler, M. G., Kleiber, C., Steelman, V., Rakel, B., Budreau, G., Everett, L., . . . Goode, C. (2001). The Iowa model of evidence-based practice to promote quality care. Critical Care Nursing Clinics of North America, 13, 497–509.
*Turner, T., Misso, M., Harris, C., & Green, S. (2008). Development of evidence- based clinical practice guidelines (CPGs): Comparing approaches. Implementation Science, 3, 45.
*World Health Organization. (2005). Bridging the “Know–Do” gap: Meeting on knowledge translation in global health. Retrieved from http://www.who.int/kms/WHO_EIP_KMS_2006_2.pdf
*A link to this open-access journal article is provided in the Toolkit for this chapter in the accompanying Resource Manual.
3 Key Concepts and Steps in Qualitative and Quantitative Research
his chapter covers a lot of ground—but, for many of you, it is familiar ground. For those who have taken an earlier research course,
this chapter provides a review of key terms and steps in the research process. For those without previous exposure to research methods, this chapter offers basic grounding in research terminology.
Research, like any discipline, has its own language—its own jargon. Some terms are used by both qualitative and quantitative researchers, but others are used mainly by one or the other group. To make matters more complex, some nursing research jargon has its roots in the social sciences, but sometimes different terms for the same concepts are used in medical research; we cover both.
FUNDAMENTAL RESEARCH TERMS AND CONCEPTS When researchers address a problem—regardless of the underlying paradigm—they undertake a study (or an investigation). Studies involve people working together in different roles.
The Faces and Places of Research Studies with humans involve two sets of people: those who do the research and those who provide the information. In a quantitative study, the people being studied are called subjects or study participants (Table 3.1). In a qualitative study, those cooperating in the study are called study participants, informants, or key informants. Collectively, study
participants comprise the sample.
The person who conducts a study is the researcher or investigator. When a study is done by a team, the person directing the study is the principal investigator (PI). In large-scale projects, dozens of individuals may be involved in planning, managing, and conducting the study. The following examples of staffing configurations span the continuum from an extremely large project to a more modest one.
Examples of Staffing on a Quantitative Study: The first author of this book was involved in a multicomponent, interdisciplinary study of poor women living in four major U.S. cities. As part of the study, she and two colleagues prepared a report documenting the health problems of 4,000 welfare mothers who were interviewed twice over a 3-year period (Polit et al., 2001). The project was staffed by over 100 people, including lead investigators of six project components (Polit was one), over 50 interviewers, and dozens of research assistants, computer programmers, and other support staff. Several health consultants, including a prominent nurse researcher (Linda Aiken), served as reviewers.
Examples of Staffing on a Qualitative Study: Beck (2009) conducted a qualitative study focusing on the experiences of mothers caring for their children with a brachial plexus injury. The team consisted Beck as the PI (who gathered and analyzed all the data), members of the United Brachial Plexus Executive Board (who helped to recruit mothers for the study), a transcriber (who listened to the tape-recorded interviews and typed them up verbatim), and an undergraduate nursing student (who checked the accuracy of the interview transcripts against the tape-recorded interviews).
Research can be undertaken in a variety of settings (the specific places where information is gathered) and in one or more sites. Some studies take place in naturalistic settings in the field, such as in people’s homes, but some studies are done in controlled laboratory or clinical settings. Qualitative researchers are especially likely to engage in fieldwork in natural settings because they are interested in the contexts of people’s experiences. The site is the overall location for the research—it could be an entire community (e.g., a Haitian neighborhood in Miami) or an institution (e.g., a hospital in Toronto). Researchers sometimes engage in multisite studies because the use of multiple sites offers a larger or more diverse sample of participants.
The Building Blocks of Research Phenomena, Concepts, and Constructs Research involves abstractions. For example, pain, quality of life, and resilience are abstractions of particular aspects of human behavior and characteristics. These abstractions are called concepts or, in qualitative studies, phenomena.
Researchers also use the term construct. Like a concept, a construct is an abstraction inferred from situations or behaviors. Kerlinger and Lee (2000) distinguish concepts from constructs by noting that constructs are abstractions that are deliberately invented (constructed) by researchers. For example, self-care in Orem’s model of health maintenance is a construct. The terms construct and concept are sometimes used interchangeably but, by convention, a construct typically refers to a more complex abstraction than a concept.
Theories and Conceptual Models A theory is a systematic, abstract explanation of some aspect of reality. Theories, which knit concepts together into a coherent system, play a role in both qualitative and quantitative research.
Quantitative researchers may start with a theory, framework, or conceptual model (distinctions are discussed in Chapter 6). Based on theory, researchers predict how phenomena will behave in the real world if the theory is true—researchers use deductive reasoning to go from a theory to specific hypotheses. Predictions deduced from theory are tested through research, and results are used to support, reject, or modify the theory.
In qualitative research, theories may be used in various ways. Sometimes conceptual or sensitizing frameworks, derived from qualitative research traditions we describe later in this chapter, offer an orienting worldview. In such studies, the framework helps to guide the inquiry and to interpret research evidence. In other qualitative studies, theory is the product of the research: The investigators use information from participants inductively to develop a theory rooted in the participants’ experiences.
Deductive and inductive logical reasoning processes are described more fully on the Supplement to this chapter on .
Variables In quantitative studies, concepts often are called variables. A variable, as the name implies, is something that varies. Weight, fatigue, and anxiety are variables—each varies from one person to another. In fact, most aspects of humans are variables. If everyone weighed 150 pounds, weight would not be a variable, it would be a constant. It is precisely because people and conditions do vary that most research is conducted. Quantitative researchers seek to understand how or why things vary and to learn if differences in one variable are related to differences in another. For example, lung cancer research is concerned with the variable of lung cancer, which is a variable because not everyone has this disease. Researchers have studied factors that might be linked to lung cancer, such as cigarette smoking. Smoking is also a variable because not everyone
smokes. A variable, then, is any quality of a person, group, or situation that varies or takes on different values. Variables are the building blocks of quantitative studies.
When an attribute is highly varied in the group under study, the group is heterogeneous with respect to that variable. If the amount of variability is limited, the group is homogeneous. For example, for the variable height, a sample of 2-year-old children would be more homogeneous than a sample of 21-year-olds.
Variables may be inherent characteristics of people, such as their age, blood type, or weight. Sometimes, however, researchers create a variable. For example, if a researcher tests the effectiveness of patient-controlled analgesia as opposed to intramuscular analgesia in relieving pain after surgery, some patients would be given patient-controlled analgesia and others would receive intramuscular analgesia. In the context of this study, method of pain management is a variable because different patients get different analgesic methods.
Continuous, Discrete, and Categorical Variables. Some variables take on a wide range of values. A person’s age, for instance, can take on values from zero to more than 100, and the values are not restricted to whole numbers. Continuous variables have values along a continuum and, in theory, can assume an infinite number of values between two points. Consider the continuous variable weight: between 1 and 2 pounds, the number of values is limitless: 1.05, 1.8, 1.333, and so on.
By contrast, a discrete variable has a finite number of values between any two points, representing discrete quantities. For example, if people were asked how many children they had, they might answer 0, 1, 2, 3, or more. The value for number of children is discrete, because a number such as 1.5 is not meaningful. Between 1 and 3, the only possible value is 2.
Other variables take on a small range of values that do not represent a quantity. Blood type, for example, has four values—A, B, AB, and O. Variables that take on a handful of discrete nonquantitative values are categorical variables. When categorical variables take on only two values, they are dichotomous variables. Gender, for example, is dichotomous: male and female.
Dependent and Independent Variables. Many studies seek to unravel and understand causes of phenomena. Does a nursing intervention cause improvements in patient outcomes? Does smoking cause lung cancer? The presumed cause is the independent variable, and the presumed effect is the dependent variable (or, the outcome variable). In terms of the PICO scheme discussed in Chapter 2, the dependent variable corresponds to the “O” (outcome). The independent variable corresponds to the “I” (the intervention, influence, or exposure) plus the “C” (the comparison). In searching for existing evidence, you might want to learn about the effects of an intervention or influence (I), compared to any alternative, on an outcome (O) of interest. In a study, however, researchers must always specify the comparator (the “C”).
Variation in the dependent variable is presumed to depend on variation in the independent variable. For example, researchers study the extent to which lung cancer (the dependent variable) depends on smoking (the independent variable). Or, investigators might study the extent to which patients’ pain (the dependent variable) depends on different nursing actions (the independent variable). The dependent variable is the outcome that researchers want to understand, explain, or predict.
The terms independent variable and dependent variable can also be used to indicate direction of influence rather than a cause and effect. For example, suppose a researcher studied the mental health of the spousal caregivers of patients with Alzheimer’s disease and found lower depression for wives than for husbands. We could not conclude that depression was caused by gender. Yet the direction of influence clearly runs from gender to depression: A patient’s level of depression does not influence their gender. Although it may not make sense to infer a cause- and-effect connection, it is appropriate to consider depression as the outcome variable and gender as an independent variable.
Most outcomes have multiple causes or influences. If we were studying factors that influence people’s body mass index (the dependent variable), we might consider height, physical activity, and diet as independent variables. Two or more dependent variables also may be of interest. For example, a researcher may compare the effects of two methods of nursing care for children with cystic fibrosis. Several dependent variables could be
used to assess treatment effectiveness, such as length of hospital stay, number of recurrent respiratory infections, and so on. It is common to design studies with multiple independent and dependent variables.
Variables are not inherently dependent or independent. A dependent variable in one study could be an independent variable in another. For example, a study might examine the effect of an exercise intervention (the independent variable) on osteoporosis (the dependent variable) to answer a Therapy question. Another study might investigate the effect of osteoporosis (the independent variable) on bone fracture incidence (the dependent variable) to address a Prognosis question. In short, whether a variable is independent or dependent is a function of the role that it plays in a particular study.
Example of Independent and Dependent Variables: Research question (Etiology/Harm question): Are interruptions during patient medication rounds in a mental health hospital associated with higher rates of nurses’ medication- administration errors? (Cottney & Innes, 2015) Independent variable: Interruptions during medication rounds Dependent variable: Medication administration errors
Conceptual and Operational Definitions Concepts are abstractions of observable phenomena, and researchers’ worldviews shape how those concepts are defined. A conceptual definition presents the abstract or theoretical meaning of concepts under study. Even seemingly straightforward terms need to be conceptually defined. The classic example is the concept of caring. Morse and colleagues (1990) examined how researchers and theorists defined caring and identified five classes of conceptual definition: as a human trait, a moral imperative, an affect, an interpersonal relationship, and a therapeutic intervention. Researchers undertaking studies of caring need to clarify which conceptual definition they have adopted.
In qualitative studies, conceptual definitions of key phenomena may be a major end product, reflecting an intent to have the meaning of concepts defined by those being studied. In quantitative studies, however, researchers must define concepts at the outset because they must decide
how the variables will be observed and measured. An operational definition of a concept specifies what the researchers must do to measure the concept and collect needed information.
Variables differ in the ease with which they can be operationalized. The variable weight, for example, is easy to define and measure. We might operationally define weight as the amount that an object weighs, to the nearest half pound. This definition designates that weight will be measured using one system (pounds) rather than another (grams). We could also specify that weight will be measured using a digital scale with participants fully undressed after 10 hours of fasting. This operational definition clarifies what we mean by the variable weight.
Few variables are operationalized as easily as weight. Most variables can be measured in different ways, and researchers must choose the one that best captures the variables as they conceptualize them. Take, for example, anxiety, which can be defined in terms of both physiologic and psychological functioning. For researchers choosing to emphasize physiologic aspects, the operational definition might involve a measure such as pulse rate. If researchers conceptualize anxiety as a psychological state, the operational definition might be scores on a paper-and-pencil test such as the State Anxiety Scale. Readers of research articles may not agree with how variables were conceptualized and measured, but definitional precision is important for communicating exactly what concepts mean within the study.
TIP: Operationalizing a concept is often a two-part process that involves deciding (1) how to accurately measure the variable and (2) how to represent it in an analysis. For example, a person’s age might be obtained by asking them to report their birthdate but operationalized in an analysis in relation to a threshold (e.g., under 65 versus 65 or older).
Example of Conceptual and Operational Definitions: Fogg and colleagues (2011) developed a scale to measure people’s beliefs and intentions about HIV screening. The scale relied on constructs from a theory called the Theory of Planned Behavior (see Chapter 6). The researchers provided examples of both conceptual and operational definitions of key constructs. For example,
“subjective norm” was conceptually defined as “the overall perception of social pressure to perform or not perform the behavior” and a scale item used to measure this construct in the context of HIV screening was “The people in my life whose opinions I value are regularly tested for HIV” (p. 76).
Data Research data (singular, datum) are the pieces of information obtained in a study. In quantitative studies, researchers identify and define their variables and then collect relevant data from study participants. Quantitative researchers collect primarily quantitative data—data in numeric form. For example, suppose we conducted a quantitative study in which a key variable was depression. We might ask, “Thinking about the past week, how depressed would you say you have been on a scale from 0 to 10, where 0 means ‘not at all’ and 10 means ‘the most possible’?” Box 3.1 presents quantitative data for three fictitious people. Subjects provided a number along the 0 to 10 continuum representing their degree of depression—9 for subject 1 (a high level of depression), 0 for subject 2 (no depression), and 4 for subject 3 (little depression). The numeric values for all people, collectively, would comprise the data on depression.
BOX 3.1 Example of Quantitative Data
Question: Thinking about the past week, how depressed would you say you have been on a scale from 0 to 10, where 0 means “not at all” and 10 means “the most possible”?
Data: 9 (Subject 1)
0 (Subject 2)
4 (Subject 3)
In qualitative studies, researchers collect qualitative data, that is, narrative descriptions. Narrative information can be obtained by having conversations with participants, by making detailed notes about how people behave in naturalistic settings, or by obtaining narrative records, such as diaries. Suppose we were studying depression qualitatively. Box
3.2 presents qualitative data for three people responding conversationally to the question, “Tell me about how you’ve been feeling lately—have you felt sad or depressed at all, or have you generally been in good spirits?” The data consist of rich narrative descriptions of participant’s emotional state.
BOX 3.2 Example of Qualitative Data
Question: Tell me about how you’ve been feeling lately—have you felt sad or depressed at all, or have you generally been in good spirits?
Data: “Well, actually, I’ve been pretty depressed lately, to tell you the truth. I wake up each morning and I can’t seem to think of anything to look forward to. I mope around the house all day, kind of in despair. I just can’t seem to shake the blues, and I’ve begun to think I need to go see a shrink.” (Participant 1)
“I can’t remember ever feeling better in my life. I just got promoted to a new job that makes me feel like I can really get ahead in my company. And I’ve just gotten engaged to a really great guy who is very special.” (Participant 2)
“I’ve had a few ups and downs the past week, but basically things are on a pretty even keel. I don’t have too many complaints.” (Participant 3)
Relationships Researchers are rarely interested in isolated concepts, except in descriptive studies. For example, a researcher might describe the percentage of patients receiving intravenous (IV) therapy who experience IV infiltration. In this example, the variable is IV infiltration versus no infiltration. Usually, however, researchers study phenomena in relation to other phenomena—that is, they focus on relationships. A relationship is a bond or a connection between phenomena. For example, researchers repeatedly have found a relationship between cigarette smoking and lung cancer.
Both qualitative and quantitative studies examine relationships but in different ways.
In quantitative studies, researchers examine the relationship between the independent and dependent variables. Researchers ask whether variation in the dependent variable (the outcome) is systematically related to variation in the independent variable. Relationships are usually expressed in quantitative terms, such as more than, less than, and so on. For example, let us consider a person’s weight as our dependent variable. What variables are related to (associated with) body weight? Some possibilities are height, caloric intake, and exercise. For each independent variable, we can make a prediction about its relationship to the outcome variable:
Height: Taller people will weigh more than shorter people. Caloric intake: People with higher caloric intake will be heavier than
those with lower caloric intake. Exercise: The lower the amount of exercise, the greater will be the
Each statement expresses a predicted relationship between weight (the dependent variable) and a measurable independent variable. Terms such as more than and heavier than imply that as we observe a change in one variable, we are likely to observe a change in weight. If Alex is taller than Tom, we would predict (in the absence of any other information) that Alex is heavier than Tom.
Quantitative studies can address one or more of the following questions about relationships:
• Does a relationship between variables exist? (e.g., Is cigarette smoking related to lung cancer?)
• What is the direction of the relationship between variables? (e.g., Are people who smoke more likely or less likely to get lung cancer than those who do not?)
• How strong is the relationship between the variables? (e.g., How much higher is the risk that smokers will develop lung cancer?)
• What is the nature of the relationship between variables? (e.g., Does
smoking cause lung cancer? Does some other factor cause both smoking and lung cancer?)
As the last question suggests, variables can be related to one another in different ways. One type of relationship is called a cause-and-effect (or causal) relationship. Within the positivist paradigm, natural phenomena are assumed not to be haphazard; they have antecedent causes that are presumably discoverable. In our example about a person’s weight, we might speculate that there is a causal relationship between caloric intake and weight: Consuming more calories causes weight gain. As noted in Chapter 1, many quantitative studies are cause-probing—they seek to illuminate the causes of phenomena.
Example of a Study of Causal Relationships: Townsend-Gervis and colleagues (2014) studied whether interdisciplinary rounds and a structured communication protocol had an impact on patient satisfaction, patient readmission, and Foley catheter removal compliance.
As noted earlier, not all relationships between variables can be interpreted as causal ones. There is a relationship, for example, between a person’s pulmonary artery and tympanic temperatures: People with high readings on one tend to have high readings on the other. We cannot say, however, that pulmonary artery temperature caused tympanic temperature nor that tympanic temperature caused pulmonary artery temperature. This type of relationship is called a functional (or an associative) relationship rather than as a causal relationship.
Example of a Study of Associative Relationships: Hsieh and colleagues (2014) examined the relationship between physical activity, body mass index, and cardiorespiratory fitness among Taiwanese school children.
Qualitative researchers are not concerned with quantifying relationships nor in testing causal relationships. Qualitative researchers seek patterns of association as a way to illuminate the underlying meaning and dimensionality of phenomena. Patterns of interconnected themes and processes are identified as a means of understanding the whole.
Example of a Qualitative Study of Patterns: Martsolf and colleagues (2012) investigated patterns of dating violence in 88 young adults aged 18 to 21 who had experienced violent dating relationships as teenagers. Analysis of the in- depth interviews revealed four patterns of adolescent dating violence based on the number of violent relationships in which each teen had been involved.
MAJOR CLASSES OF QUANTITATIVE AND QUALITATIVE RESEARCH Researchers usually work within a paradigm that is consistent with their worldview and that gives rise to questions that excite their curiosity. The maturity of the focal concept also may lead to one or the other paradigm: When little is known about a topic, a qualitative approach is often more fruitful than a quantitative one. In this section, we briefly describe broad categories of quantitative and qualitative research.
Quantitative Research: Experimental and Nonexperimental Studies A basic distinction in quantitative studies is between experimental and nonexperimental research. In experimental research, researchers actively introduce an intervention or treatment—most often, to address Therapy questions. In nonexperimental research, researchers are bystanders— they collect data without intervening (most often, to address Etiology, Prognosis, or Diagnosis questions). For example, if a researcher gave bran flakes to one group of people and prune juice to another to evaluate which method facilitated elimination more effectively, the study would be experimental because the researcher intervened in the normal course of things. If, on the other hand, a researcher compared elimination patterns of two groups whose regular eating patterns differed, the study would be nonexperimental because there is no intervention. In medical research, an experimental study usually is called a clinical trial, and a nonexperimental inquiry is called an observational study. As we discuss in Chapter 9, a randomized controlled trial or RCT is a particular type of clinical trial.
TIP: On the evidence hierarchy shown in Figure 2.1, the two rungs
below systematic reviews (RCTs and quasi-experiments) involve interventions and are experimental. The four rungs below that are nonexperimental.
Experimental studies are explicitly cause-probing—they test whether an intervention caused changes in the dependent variable. Sometimes nonexperimental studies also explore causal relationships, but the resulting evidence is usually less conclusive. Experimental studies offer the possibility of greater control over confounding influences than nonexperimental studies, and so causal inferences are more plausible.
Example of Experimental Research: Williams and colleagues (2014) tested the effect of an intervention called Reasoning Exercises in Assisted Living on residents’ problem solving and reasoning. Some study participants received the cognitive training intervention, and others did not.
In this example, the researchers intervened by giving some patients the special intervention but not giving it to others. In other words, the researcher controlled the independent variable, which in this case was receipt or nonreceipt of the cognitive training intervention.
Example of Nonexperimental Research: Huang and colleagues (2014) studied factors that predicted fatigue severity in Taiwanese women with breast cancer 1 year after surgery. They found, for example, that women who were married and who had poorer functional performance at diagnosis had higher levels of fatigue.
In this nonexperimental study to address a Prognosis question, the researchers did not intervene in any way. Their intent was to explore existing relationships rather than to test a potential solution to a problem.
Qualitative Research: Disciplinary Traditions The majority of qualitative studies can best be described as qualitative descriptive research. Many qualitative studies, however, are rooted in research traditions that originated in anthropology, sociology, and
psychology. Three such traditions that are prominent in qualitative nursing research are briefly described here. Chapter 21 provides a fuller discussion of these traditions and the methods associated with them.
Grounded theory research, with roots in sociology, seeks to describe and understand the key social psychological processes that occur in social settings. Most grounded theory studies focus on a developing social experience—the social and psychological processes that characterize an event or episode. A major component of grounded theory is the discovery of not only the basic social psychological problem but also a core variable that is central in explaining what is going on in that social scene. Grounded theory researchers strive to generate explanations of phenomena that are grounded in reality. Grounded theory was developed in the 1960s by two sociologists, Glaser and Strauss (1967).
Example of a Grounded Theory Study: Ramirez and Badger (2014) conducted a grounded theory study to explore the social psychological processes of men who suffer from depression. They uncovered six stages through which men navigated in their experiences with depression.
Phenomenology, rooted in a philosophical tradition developed by Husserl and Heidegger, is concerned with the lived experiences of humans. Phenomenology is an approach to thinking about what life experiences of people are like and what they mean. The phenomenologic researcher asks the questions: What is the essence of this phenomenon as experienced by these people? Or, What is the meaning of the phenomenon to those who experience it?
Example of a Phenomenologic Study: Ekwall and co-researchers (2014) conducted in-depth interviews to explore the lived experience of having recurring ovarian cancer.
Ethnography, the primary research tradition in anthropology, provides a framework for studying the patterns, lifeways, and experiences of a defined cultural group in a holistic manner. Ethnographers typically engage in extensive fieldwork, often participating in the life of the culture
under study. Ethnographic research can be concerned with broadly defined cultures (e.g., Hmong refugee communities) but sometimes focuses on more narrowly defined cultures (e.g., the culture of an intensive care unit). Ethnographers strive to learn from members of a cultural group, to understand their worldview, and to describe their customs and norms.
Example of an Ethnographic Study: Broadbent and colleagues (2014) conducted ethnographic fieldwork to investigate the emergency department triage environment and its effect on triage practices for clients with a mental illness.
MAJOR STEPS IN A QUANTITATIVE STUDY In quantitative studies, researchers move from the beginning of a study (posing a question) to the end point (obtaining an answer) in a reasonably linear sequence of steps that is broadly similar across studies. In some studies, the steps overlap; in others, some steps are unnecessary. Still, a general flow of activities is typical in a quantitative study (see Figure 3.1). This section describes that flow, and the next section describes how qualitative studies differ.
Phase 1: The Conceptual Phase Early steps in a quantitative study typically have a strong conceptual element. These activities include reading, conceptualizing, theorizing, and reviewing ideas with colleagues or advisers. During this phase, researchers call on such skills as creativity, deductive reasoning, and a firm grounding in previous research on a topic of interest.
Step 1: Formulating and Delimiting the Problem Quantitative researchers begin by identifying an interesting, significant research problem and formulating research questions. Good research requires starting with good questions. In developing research questions, nurse researchers must attend to substantive issues (What kind of new evidence is needed?), theoretical issues (Is there a conceptual context for understanding this problem?), clinical issues (How could evidence from
this study be used in clinical practice?), methodologic issues (How can this question best be studied to yield high-quality evidence?), and ethical issues (Can this question be rigorously addressed in an ethical manner?)
TIP: A critical ingredient in developing good research questions is personal interest. Begin with topics that fascinate you or about which you have a passionate interest or curiosity.
Step 2: Reviewing the Related Literature Quantitative research is conducted in a context of previous knowledge. Quantitative researchers typically strive to understand what is already known about a topic by undertaking a literature review. A thorough literature review provides a foundation on which to base new evidence and usually is conducted before data are collected. For clinical problems, it may also be necessary to learn the “status quo” of current procedures and to review existing practice guidelines.
Step 3: Undertaking Clinical Fieldwork Unless the research problem originated in a clinical setting, researchers embarking on a clinical nursing study benefit from spending time in relevant clinical settings, discussing the problem with clinicians and administrators, and observing current practices. Clinical fieldwork can provide perspectives on recent clinical trends, current diagnostic procedures, and relevant health care delivery models; it can also help researchers better understand clients and the settings in which care is provided. Such fieldwork can also be valuable in gaining access to an appropriate site or in developing research strategies. For example, in the course of clinical fieldwork, researchers might discover the need for research assistants who are bilingual.
Step 4: Defining the Framework and Developing Conceptual Definitions Theory is the ultimate aim of science: It transcends the specifics of a particular time, place, and group and aims to identify regularities in the
relationships among variables. When quantitative research is performed within the context of a theoretical framework, the findings often have broader significance and utility. Even when the research question is not embedded in a theory, researchers should have a conceptual rationale and a clear vision of the concepts under study.
Step 5: Formulating Hypotheses Hypotheses state researcher’s expectations (predictions) about relationships between study variables. The research question identifies the study concepts and asks how the concepts might be related; a hypothesis is the predicted answer. For example, the research question might be: Is preeclamptic toxemia related to stress during pregnancy? This might be translated into the following hypothesis: Women with high levels of stress during pregnancy will be more likely than women with lower stress to experience preeclamptic toxemia. Most quantitative studies involve testing hypotheses through statistical analysis.
Phase 2: The Design and Planning Phase In the second major phase of a quantitative study, researchers decide on the methods they will use to address the research question. Researchers usually have flexibility in designing a study and make many decisions. These methodologic decisions have crucial implications for the integrity and generalizability of the resulting evidence.
Step 6: Selecting a Research Design The research design is the overall plan for obtaining answers to the research questions. Many experimental and nonexperimental research designs are available. In designing the study, researchers select a specific design and identify strategies to minimize bias. Research designs indicate how often data will be collected, what types of comparisons will be made, and where the study will take place. The research design is the architectural backbone of the study.
Step 7: Developing Protocols for the Intervention In experimental research, researchers create an intervention (the
independent variable), and so they need to develop its specifications. For example, if we were interested in testing the effect of biofeedback on hypertension, the independent variable would be exposure to biofeedback compared with either an alternative treatment (e.g., relaxation) or no treatment. An intervention protocol for the study must be developed, specifying exactly what the biofeedback treatment would entail (e.g., what type of feedback, who would administer it, how frequently and over how long a period the treatment would last, and so on) and what the alternative condition would be. The goal of well-articulated protocols is to have all people in each group treated in the same way. (In nonexperimental research, this step is not necessary.)
Step 8: Identifying the Population Quantitative researchers need to clarify the group to whom study results can be generalized—that is, they must identify the population to be studied. A population is all the individuals or objects with common, defining characteristics (the “P” component in PICO questions). For example, the population of interest might be all patients undergoing chemotherapy in Dallas.
Step 9: Designing the Sampling Plan Researchers collect data from a sample, which is a subset of the population. Using samples is more feasible than collecting data from an entire population, but the risk is that the sample might not reflect the population’s traits. In a quantitative study, a sample’s adequacy is assessed by its size and representativeness. The quality of the sample depends on how typical, or representative, the sample is of the population. The sampling plan specifies how the sample will be selected and recruited and how many subjects there will be.
Step 10: Specifying Methods to Measure Research Variables Quantitative researchers must develop or borrow methods to measure their research variables. The primary methods of data collection are self-reports (e.g., interviews), observations (e.g., observing the sleep–wake state of infants), and biophysiologic measurements. Self-reported data from
patients is the largest class of data collection methods and is often referred to as patient-reported outcomes (PROs). The task of measuring research variables and developing a data collection plan is complex and challenging.
Step 11: Developing Methods to Safeguard Human/Animal Rights Most nursing research involves humans, and so procedures need to be developed to ensure that the study adheres to ethical principles. A formal review by an ethics committee is usually required.
Step 12: Reviewing and Finalizing the Research Plan Before collecting their data, researchers often take steps to ensure that plans will work smoothly. For example, they may evaluate the readability of written materials to assess if participants with low reading skills can comprehend them, or they may pretest their measuring instruments to see if they work well. Normally, researchers also have their research plans critiqued by peers, consultants, or other reviewers before implementing it. Researchers seeking financial support submit a proposal to a funding source, and reviewers usually suggest improvements.
TIP: For major studies, researchers often undertake a small-scale pilot study to test their research plans. Strategies for designing effective pilot studies are described in Chapter 28.
Phase 3: The Empirical Phase The empirical phase of quantitative studies involves collecting data and preparing the data for analysis. Often, the empirical phase is the most time- consuming part of the investigation. Data collection typically requires months of work.
Step 13: Collecting the Data The actual collection of data in quantitative studies often proceeds according to a preestablished plan. The plan typically spells out procedures
for training data collection staff, for actually collecting data (e.g., where and when the data will be gathered), and for recording information.
Technologic advances have expanded possibilities for automating data collection.
Step 14: Preparing the Data for Analysis Data collected in a quantitative study must be prepared for analysis. One preliminary step is coding, which involves translating verbal data into numeric form (e.g., coding gender information as “1” for females and “2” for males). Another step may involve transferring the data from written documents onto computer files for analysis.
Phase 4: The Analytic Phase Quantitative data must be subjected to analysis and interpretation, which occur in the fourth major phase of a project.
Step 15: Analyzing the Data Quantitative researchers analyze their data through statistical analyses, which include simple procedures (e.g., computing an average) as well as ones that are complex. Some analytic methods are computationally formidable, but the underlying logic of statistical tests is fairly easy to grasp. Computers have eliminated the need to get bogged down with mathematic operations.
Step 16: Interpreting the Results Interpretation involves making sense of study results and examining their implications. Researchers attempt to explain the findings in light of prior evidence, theory, and their own clinical experience—and in light of the adequacy of the methods they used in the study. Interpretation also involves drawing conclusions about the clinical significance of the results, envisioning how the new evidence can be used in nursing practice, and clarifying what further research is needed.
Phase 5: The Dissemination Phase In the analytic phase, the researcher comes full circle: Questions posed at
the outset are answered. Researchers’ responsibilities are not completed, however, until study results are disseminated.
Step 17: Communicating the Findings A study cannot contribute evidence to nursing practice if the results are not shared. Another—and often final—task of a study is the preparation of a research report that summarizes the study. Research reports can take various forms: dissertations, journal articles, conference presentations, and so on. Journal articles—reports appearing in professional journals such as Nursing Research—usually are the most useful because they are available to a broad, international audience. We discuss journal articles later in this chapter.
Step 18: Utilizing the Findings in Practice Ideally, the concluding step of a high-quality study is to plan for the use of the evidence in practice settings. Although nurse researchers may not themselves be able to implement a plan for using research findings, they can contribute to the process by making recommendations for utilizing the evidence, by ensuring that adequate information has been provided for a systematic review, and by pursuing opportunities to disseminate the findings to clinicians.
ACTIVITIES IN A QUALITATIVE STUDY Quantitative research involves a fairly linear progression of tasks— researchers plan the steps to be taken to maximize study integrity and then follow those steps as faithfully as possible. In qualitative studies, by contrast, the progression is closer to a circle than to a straight line— qualitative researchers continually examine and interpret data and make decisions about how to proceed based on what has already been discovered (Figure 3.2).
Because qualitative researchers have a flexible approach, we cannot show the flow of activities precisely—the flow varies from one study to another, and researchers themselves do not know ahead of time exactly how the study will unfold. We provide a sense of how qualitative studies are conducted, however, by describing some major activities and indicating when they might be performed.
Conceptualizing and Planning a Qualitative Study Identifying the Research Problem Qualitative researchers usually begin with a broad topic area, focusing on an aspect of a topic that is poorly understood and about which little is known. Qualitative researchers often proceed with a fairly broad initial question, which may be narrowed and clarified on the basis of self- reflection and discussion with others. The specific focus and questions are usually delineated more clearly once the study is underway.
Doing a Literature Review Qualitative researchers do not all agree about the value of doing an up- front literature review. Some believe that researchers should not consult the literature before collecting data because prior studies could influence
conceptualization of the focal phenomenon. In this view, the phenomena should be explicated based on participants’ viewpoints rather than on prior knowledge. Those sharing this opinion often do a literature review at the end of the study. Other researchers conduct a brief preliminary review to get a general grounding. Still others believe that a full early literature review is appropriate. In any case, qualitative researchers typically find a fairly small body of relevant previous work because of the types of question they ask.
Selecting and Gaining Entrée into Research Sites Before going into the field, qualitative researchers must identify an appropriate site. For example, if the topic is the health beliefs of the urban poor, an inner-city neighborhood with low-income residents must be identified. Researchers may need to engage in anticipatory fieldwork to identify a suitable and information-rich environment for the study. In some cases, researchers have ready access to the study site, but in others, they need to gain entrée. A site may be well suited to the needs of the research, but if researchers cannot “get in,” the study cannot proceed. Gaining entrée typically involves negotiations with gatekeepers who have the authority to permit entry into their world.
TIP: The process of gaining entrée is usually associated with doing fieldwork in qualitative studies, but quantitative researchers often need to gain entrée into sites for collecting data as well.
Developing an Overall Approach in Qualitative Studies Quantitative researchers do not collect data before finalizing their research design. Qualitative researchers, by contrast, use an emergent design that materializes during the course of data collection. Certain design features may be guided by the qualitative research tradition within which the researcher is working, but nevertheless, few qualitative studies adopt rigidly structured designs that prohibit changes while in the field.
Although qualitative researchers do not always know in advance exactly how the study will progress, they nevertheless must have some sense of how much time is available for fieldwork and must also arrange
for and test needed equipment, such as recording equipment or laptop computers. Other planning activities include such tasks as hiring and training interviewers to assist in the collection of data; securing interpreters if the informants speak a different language; and hiring appropriate consultants, transcribers, and support staff.
Addressing Ethical Issues Qualitative researchers, like quantitative researchers, must also develop plans for addressing ethical issues—and, indeed, there are special concerns in qualitative studies because of the more intimate nature of the relationship that typically develops between researchers and study participants. Chapter 7 describes these concerns.
Conducting a Qualitative Study In qualitative studies, the tasks of sampling, data collection, data analysis, and interpretation typically take place iteratively. Qualitative researchers begin by talking with or observing a few people who have first-hand experience with the phenomenon under study. The discussions and observations are loosely structured, allowing for the expression of a full range of beliefs, feelings, and behaviors. Analysis and interpretation are ongoing, concurrent activities that guide choices about the kinds of people to sample next and the types of questions to ask or observations to make.
The process of data analysis involves clustering together related types of narrative information into a coherent scheme. As analysis and interpretation progress, researchers begin to identify themes and categories, which are used to build a rich description or theory of the phenomenon. The kinds of data obtained and the people selected as participants tend to become increasingly purposeful as the conceptualization is developed and refined. Concept development and verification shape the sampling process—as a conceptualization or theory develops, the researcher seeks participants who can confirm and enrich the theoretical understandings as well as participants who can potentially challenge them and lead to further theoretical development.
Quantitative researchers decide up-front how many people to include in a study, but qualitative researchers’ sampling decisions are guided by the
data. Qualitative researchers use the principle of data saturation, which occurs when themes and categories in the data become repetitive and redundant, such that no new information can be gleaned by further data collection.
Quantitative researchers seek to collect high-quality data by measuring their variables with instruments that have been demonstrated to be reliable and valid. Qualitative researchers, by contrast, are the main data collection instrument and must take steps to demonstrate the trustworthiness of the data. The central feature of these efforts is to confirm that the findings accurately reflect the experiences and viewpoints of participants rather than the researcher’s perceptions. One confirmatory activity, for example, involves going back to participants and sharing interpretations with them so that they can evaluate whether the researcher’s thematic analysis is consistent with their experiences.
Disseminating Qualitative Findings Qualitative nurse researchers also share their findings with others at conferences and in journal articles. Regardless of researchers’ positions about when a literature review should be conducted, a summary of prior research is usually offered in qualitative reports as a means of providing context for the study.
Quantitative reports almost never contain raw data—that is, data in the form they were collected, which are numeric values. Qualitative reports, by contrast, are usually filled with rich verbatim passages directly from participants. The excerpts are used in an evidentiary fashion to support or illustrate researchers’ interpretations and thematic construction.
Example of Raw Data in a Qualitative Report: Gitsels-van der Wal and colleagues (2015) did an in-depth study of how pregnant Muslim women living in the Netherlands make decisions about antenatal anomaly screening. The researchers found that the women were hesitant about the test uptake. Here is an illustrative quote:
I thought, “What if there is something wrong?” It would have to be something really major before you’d want a termination, but I think it would be a desperately difficult choice. . . . It seems like a very awkward choice, because you got pregnant
in the first place because you wanted a baby, and then it doesn’t matter whether it’s disabled or not. (p. e45)
Like quantitative researchers, qualitative nurse researchers want their findings used by others. Qualitative findings often are the basis for formulating hypotheses that are tested by quantitative researchers, for developing measuring instruments for both research and clinical purposes, and for designing effective nursing interventions. Qualitative studies help to shape nurses’ perceptions of a problem or situation, their conceptualizations of potential solutions, and their understanding of patients’ concerns and experiences.
RESEARCH JOURNAL ARTICLES Research journal articles, which summarize the context, design, and results of a study, are the primary method of disseminating research evidence. This section reviews the content and style of research journal articles to ensure that you will be equipped to delve into the research literature. A more detailed discussion of the structure of journal articles is presented in Chapter 30, which provides guidance on writing research reports.
Content of Journal Articles Many quantitative and qualitative journal articles follow a conventional organization called the IMRAD format. This format, which loosely follows the steps of quantitative studies, involves organizing material into four main sections—Introduction, Methods, Results, and Discussion. The text of the report is usually preceded by an abstract and followed by cited references.
The Abstract The abstract is a brief description of the study placed at the beginning of the article. The abstract answers, in about 250 words, the following: What were the research questions? What methods did the researcher use to address the questions? What did the researcher find? What are the implications for practice? Readers review abstracts to assess whether the
entire report is of interest. Some journals have moved from traditional abstracts—single paragraphs summarizing the study’s main features—to longer, structured abstracts with specific headings. For example, in Nursing Research, the abstracts are organized under the following headings: Background, Objectives, Method, Results, and Conclusions.
The Introduction The introduction communicates the research problem and its context. The introduction, which often is not be specifically labeled “Introduction,” follows immediately after the abstract. This section typically describes (1) the central phenomena, concepts, or variables under study; (2) the population of interest; (3) the current state of evidence, based on a literature review; (4) the theoretical framework; (5) the study purpose, research questions, or hypotheses to be tested; and (6) the study’s significance. Thus, the introduction sets the stage for a description of what the researcher did and what was learned. The introduction corresponds roughly to the conceptual phase (Phase 1) of a study.
The Method Section The method section describes the methods used to answer the research questions. This section lays out methodologic decisions made in the design and planning phase (Phase 2) and may offer rationales for those decisions. In a quantitative study, the method section usually describes (1) the research design, (2) the sampling plan for selecting participants from the population of interest, (3) methods of data collection and specific instruments used, (4) study procedures (including ethical safeguards), and (5) analytic procedures and methods.
Qualitative researchers discuss many of the same issues but with different emphases. For example, a qualitative study often provides more information about the research setting and the study context and less information on sampling. Also, because formal instruments are not used to collect qualitative data, there is less discussion about data collection methods, but there may be more information on data collection procedures. Increasingly, reports of qualitative studies are including descriptions of the researchers’ efforts to enhance the trustworthiness of the study.
The Results Section The results section presents the findings (results) from the data analyses. The text summarizes key findings, and (in quantitative reports) tables provide greater detail. Virtually all results sections contain a description of the participants (e.g., their average age, percent male/female).
In quantitative studies, the results section provides information about statistical tests, which are used to test hypotheses and evaluate the believability of the findings. For example, if the percentage of smokers who smoke two packs or more daily is computed to be 40%, how probable is it that the percentage is accurate? If the researcher finds that the average number of cigarettes smoked weekly is lower for those in an intervention group than for those not getting the intervention, how probable is it that the intervention effect is real? Is the effect of the intervention on smoking likely to be replicated with a new sample of smokers—or does the result reflect a peculiarity of the sample? Statistical tests help to answer such questions. Researchers typically report:
• The names of statistical tests used. Different tests are appropriate for different situations but are based on common principles. You do not have to know the names of all statistical tests—there are dozens of them—to comprehend the findings.
• The value of the calculated statistic. Computers are used to calculate a numeric value for the particular statistical test used. The value allows researchers to draw conclusions about the meaning of the results. The actual numeric value of the statistic, however, is not inherently meaningful and need not concern you.
• The significance. A critical piece of information is whether the value of the statistic was significant (not to be confused with important or clinically relevant)