WHY DO PEOPLE RESIST HEALTHCARE IT?

LITERATURE ANALYSIS, MODEL

TESTING, AND REFINEMENT

By

BAHAE SAMHAN

A dissertation submitted in partial fulfillment of

the requirements for the degree of

DOCTOR OF PHILOSOPHY

WASHINGTON STATE UNIVERSITY

Carson College of Business

MAY 2016

© Copyright by BAHAE SAMHAN, 2016

All Rights Reserved
 ProQuest Number: 10139705

All rights reserved

INFORMATION TO ALL USERS

The quality of this reproduction is dependent upon the quality of the copy submitted.

In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.

ProQuest 10139705

Published by ProQuest LLC (2016). Copyright of the Dissertation is held by the Author.

All rights reserved.

This work is protected against unauthorized copying under Title 17, United States Code
Microform Edition © ProQuest LLC.

ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, MI 48106 - 1346
© Copyright by BAHAE SAMHAN, 2016
All Rights Reserved
To the Faculty of Washington State University:

The members of the Committee appointed to examine the dissertation of

BAHAE SAMHAN find it satisfactory and recommend that it be accepted.

___________________________________
K.D. Joshi, Ph.D., Chair

___________________________________
Kenneth Butterfield, Ph.D.

___________________________________
Terence Saldanha, Ph.D.

ii
 ACKNOWLEDGMENTS

Firstly, I would like to express my sincerest gratitude to my advisor and chair of my

committee, Prof. K.D. Joshi for her continuous support of my Ph.D study, for her patience,

motivation, and immense knowledge. I am very fortunate to have had you as my advisor and as

the chair of my dissertation committee. I could not have imagined having a better mentor for my

Ph.D study. I am confident that your guidance will have a lasting positive impact on my future

career. There are no words great enough to express my sincerest appreciation and gratitude to all

of the support I received from you throughout the years.

I would also like to thank the rest of my dissertation committee members: Dr. Kenneth

Butterfield, and Dr. Terence Saldanha, for their insightful comments and encouragement. Dr.

Saldanha, you have been of tremendous help for me. You agreed to step in and join my

dissertation committee when it was most-needed. I could not imagine completing my dissertation

without your presence, perspective, and guidance. Dr. Butterfield, I really appreciate you for

always being there for me. I will be forever grateful for your continuous support for me as a

member on my dissertation committee and as well as my department chair. You are a real role

model to me. I sincerely hope that our relationship continues throughout my career.

My sincere thanks also goes to Dr. Mauricio Featherman, I honestly cherished each

moment I worked with you teaching students how to program. I have learned a lot from you and

it has been my honor to have you as a great mentor and a close friend. I trust that our relationship

will continue endlessly.

Special thanks goes to Prof. Deborah Compeau. Although we shared a relatively short

period of time at WSU, but I am forever grateful for all of your constructive feedback that you

iii
never hesitated to provide on my work. Thank you for your wisdom and continuous

encouragement.

Next, I would like to thank my family for all of their support, love, and encouragement

throughout this process. To my great father Dr. Marouf Samhan, you have always been the

source of my inspiration. If I can be half as good a scholar as you have always been, then that

will be my greatest achievement. Thank you for all of your teachings, wise advise, endless

support, and encouragement. To my dear mom Aida, you are everything to me. Thank you for

always being there for me. You always encouraged me to pursue my academic career because

you believed in me. I am very happy that I was able to make your dream come true. To my

lovely wife Sara, I am truly blessed to have you in my life. I deeply acknowledge all of what you

had to go through to make this journey possible. Thank you for believing in me and always

teaching me how to make the impossible reachable. I would have not been able to accomplish

this without you in my life. To our beautiful sons Adam and Alex, you have gave true meaning

to my life and a reason to keep going no matter what. My brother Waseem, thank you for always

being my “go-to” person when I needed help. My sweet sister Katherine, thank you for always

reminding me that when the going gets tough, the tough gets going. I am truly blessed to have all

of you in my life.

Finally, I must mention the students in my cohort who have been an important part of this
process. Majid Dadgar, Joseph Taylor, and Benyawareth (Yaa) Nithithanatchinnapat. I am truly

grateful to have had you all along. Majid, you have been more than a brother to me. I would also

like to thank my senior cohort: Tanya Beaulieu, Chris Callif, Xiaolin Lin, and Todd Martin. I am

grateful that I have had you as colleagues and friends. Special thanks goes to my dear friend

from the junior cohort, Mina Jafarijoo. Talking with you has always reenergized me and kept me

sane. I am sure you will be finishing your dissertation sooner than you can imagine. To everyone

else that I may have unintentionally forgot, please accept my apologies and I truly thank you.

iv
“One child, one teacher, one book, and one pen can change the world. Education is the only

solution. Education first”

- Malala Yousafzai, (UN, New York, 2013)

v
 WHY DO PEOPLE RESIST HEALTHCARE IT? LITERATURE ANALYSIS, MODEL

TESTING, AND REFINEMENT

Abstract

by Bahae Samhan, Ph.D.

Washington State University
May 2016

Chair: K.D. Joshi

Health Information Technology (HIT) has the potential of improving the overall

performance of healthcare organizations. However, there are worldwide evidence of HIT

resistance and avoidance behaviors. IS researchers have provided valuable insights about these

behaviors towards IT in general. Very limited work has aimed to explain these behaviors in

healthcare settings and especially towards Electronic Health Record (EHR) systems. In this

dissertation, we investigate the resistance and avoidance behaviors in the context of a hospital,

specifically having the EHR system as the focal technology for my study. Overall, the

dissertation is comprised of five chapters and subscribes to a mixed-methods approach. The first

chapter will provide an extensive review the literature to highlight the current status of research

on technology resistance and the main research gaps. In the second chapter, we empirically

validate the User Resistance Model (Kim and Kankanhalli, 2009). We aim to explain the

healthcare providers’ resistance to change from the paper-based recording system to the new

EHR system. Based on an analysis of survey data from healthcare providers, we investigate the

effects of the key determinants from URM on EHR resistance. The third chapter is a qualitative

vi
research in which we collect responses from healthcare providers using open-ended questions.

We use the Reasoned Causal Mapping (RCM) methodology to uncover the main predictors of

EHR resistance and the key concepts shaping those predictors. In the fourth chapter, we aim to

explain the EHR avoidance behaviors after the implementation of the system is complete and has

been enforced to all users. We adopt the Technology Threat Avoidance Theory (TTAT) (Liang

and Xue, 2009) and empirically test the complete proposed conceptual model. Based on an

analysis of survey data from healthcare providers, we investigate the effects of perceived threats

and perceived avoidability on avoidance motivations, and the direct effects of avoidance

motivation on avoidance behavior. In the fifth chapter, we use RCM to reveal the main

constructs impacting EHR avoidance as well as the key concepts forming these constructs. We

analyze qualitative data collected from healthcare providers using open-ended questions.

Contributions to research and practice are discussed within each chapter.

vii
 TABLE OF CONTENTS

Page

ACKNOWLEDGEMENTS............................................................................................. iii

ABSTRACT ................................................................................................................... vi

LIST OF TABLES ........................................................................................................ xiv

LISTOF FIGURES ....................................................................................................... xvi

CHAPTER ONE: RESISTANCE OF HEALTH INFORMATION TECHNOLOGIES:

LITERATURE REVIEW, ANALYSIS, AND GAPS

1. ABSTRACT ...................................................................................................... 1

2. INTRODUCTION ............................................................................................. 2

3. METHODOLOGY ............................................................................................ 4

4. EMERGING RESEARCH GAPS ...................................................................... 7

Lack of a Unified Frame Work of Resistance ............................................... 7

Patients are Ignored as Potential HIT Users.................................................. 8

Limited Work in the Healthcare Setting ....................................................... 9

The Nature of Technology Support-System Functionalities and Supported

Work Processes ............................................................................................ 9

The Resistance Construct is Multifaceted ....................................................10

The Role of Beliefs in IT Resistance Needs Greater Theorization ...............11

Individual Differences in IT Resistance .......................................................13

5. IMPLICATIONS ..............................................................................................14

Implications for Research............................................................................14

Implications for Practice .............................................................................15

viii
 6. FUTURE WORK .............................................................................................15

7. REFERENCES .................................................................................................17

CHAPTER TWO: SWITCHING TO ELECTRONIC HEALTH RECORD SYSTEMS: AN

APPLICATION OF THE USER RESISTANCE MODEL

1. ABSTRACT .....................................................................................................24

2. INTRODUCTION ............................................................................................25

3. LITERATURE REVIEW .................................................................................27

4. INVESTIGATIVE CONTEXT .........................................................................32

5. THEORATICAL BACKGROUND ..................................................................35

The User Resistance Model.........................................................................35

6. MODEL CONSTRUCTS AND HYPOTHESES..............................................38

Switching Benefits ......................................................................................39

Switching Costs ..........................................................................................39

Perceived Value ..........................................................................................41

Self-Efficacy for Change .............................................................................41

Organizational Support for Change .............................................................42

Colleague Opinion ......................................................................................42

User Resistance...........................................................................................43

Control Variables ........................................................................................44

7. METHODOLOGY ...........................................................................................46

Instrument Development .............................................................................46

Sample and Data Collection ........................................................................46

Instrument Validation .................................................................................48

8. RESULTS ........................................................................................................49

ix
 9. DISCUSSION OF FINDINGS..........................................................................53

10. CONTRIBUTIONS ..........................................................................................58

Contribution to Research.............................................................................58

Contribution to Practice ..............................................................................59

11. LIMITATIONS AND FUTURE DIRECTIONS ...............................................60

12. REFERENCES .................................................................................................62

13. APPENDIX ......................................................................................................69

Measurement Items .....................................................................................69

Instrument Validation .................................................................................71

CHAPTER THREE: UNDERSTANDING ELECTRONIC HEALTH RECORDS

RESISTANCE: A REVEALED CAUSAL MAPPING APPROACH

1. ABSTRACT .....................................................................................................73

2. INTRODUCTION ............................................................................................74

3. LITERATURE REVIEW ON TECHNOLOGY RESISTANCE .......................75

4. THEORY IN USE ............................................................................................77

5. RESEARCH METHODOLOGY ......................................................................81

Participants and Procedure ..........................................................................82

Data Analysis..............................................................................................83

6. RESULTS ........................................................................................................86

EHRS Self-Efficacy ....................................................................................88

EHRS Organizational Support ....................................................................89

EHRS Costs ................................................................................................90

EHRS Benefits ............................................................................................91

x
 EHRS Perceived Value ...............................................................................92

EHRS Perceived Threat ..............................................................................94

Facilitating Conditions ................................................................................95

EHRS Resistance ........................................................................................96

7. DISCUSSION ....................................................................................................97

8. CONTRIBUTIONS .......................................................................................... 100

For Research ............................................................................................. 100

For Practice............................................................................................... 101

9. LIMITATIONS AND FUTURE DIRECTIONS ............................................. 102

10. REFERENCES ............................................................................................... 103

11. APPENDIX ....................................................................................................107

Interview Questions .................................................................................. 107

CHAPTER FOUR: ELECTRONIC HEALTH RECORDS AVOIDANCE: AN EMPIRICAL

INVESTIGATION FROM THE TECHNOLOGY THREATS AVOIDANCE THEORY
PERSPECTIVE

1. ABSTRACT ...................................................................................................108

2. INTRODUCTION .......................................................................................... 109

3. TECHNOLOGY AVOIDANCE ..................................................................... 110

4. LITERATURE REVIEW ............................................................................... 112

5. INVESTIGATIVE CONTEXT .......................................................................113

6. THEORATICAL BACKGROUND OF TTAT ............................................... 114

7. RESEARCH MODEL AND HYPOTHESES ................................................. 117

8. METHODOLOGY ........................................................................................... 120

Instrument Development ........................................................................... 120

xi
 Sample and Data Collection ......................................................................121

Instrumental Validation............................................................................. 123

9. RESULTS ........................................................................................................ 124

Suggested Model ...................................................................................... 129

10. CONTRIBUTIONS ........................................................................................ 133

Contributions to Research ......................................................................... 133

Contributions to Practice ........................................................................... 134

11. LIMITATIONS .............................................................................................. 134

12. REFERENCES ............................................................................................... 136

13. APPENDIX ....................................................................................................140

Measurement Items ................................................................................... 140

Instrument Validation ............................................................................... 143

CHAPTER FIVE: UNDERSTANDING ELECTRONIC HEALTH RECORDS AVOIDANCE:

A REVEALED CAUSAL MAPPING APPROACH

1. ABSTRACT ...................................................................................................145

2. INTRODUCTION .......................................................................................... 146

3. THEORATICAL BACKGROUND ................................................................ 147

4. RESEARCH METHODOLOGY ....................................................................149

Data Analysis............................................................................................ 149

5. RESULTS ......................................................................................................152

Perceived Harm ........................................................................................ 154

Threat Appraisal ....................................................................................... 155

Coping Appraisal ...................................................................................... 156

Problem Focused Coping .......................................................................... 156

xii
 Emotion Focused Coping .......................................................................... 157

The Aggregated RCM ............................................................................... 158

6. DISCUSSION ................................................................................................ 160

7. CONTRIBUTIONS ........................................................................................ 162

For Research ............................................................................................. 162

For Practice............................................................................................... 163

8. LIMITATINS AND FUTURE DIRECTIONS ................................................ 163

9. REFERENCES ............................................................................................... 165

APPENDIX

A. SURVEY FORMS .......................................................................................... 167

xiii
 LIST OF TABLES

1.1 Research Gaps in HIT Resistance Literature vs IT Resistance Literature ...........14

2.1 Summary of IS Research EHR Resistance ........................................................27

2.2 User Resistance Model Constructs Composition ...............................................38

2.3 URM Hypotheses .............................................................................................45

2.4 Measurement Items...........................................................................................69

2.5 Descriptive Statistics of Respondents ................................................................48

2.6 Items Loadings and CFA Model Fit Statistics ...................................................71

2.7 Nonresponse Bias Test......................................................................................49

2.8 URM Model-Fit Indices ....................................................................................50

2.9 Standardized Path Coefficients .........................................................................51

2.10 Suggested Model Fit Indices .............................................................................57

3.1 Descriptive Statistics of Respondents ................................................................82

3.2 Interviews’ Open-Ended Questions ................................................................. 107

3.3 An Illustration of the 5-Step Procedure for Constructing Causal Map ...............84

3.4 Revealed Constructs of EHRS Resistance and Its Concepts ..............................86

3.5 EHRS Self-Efficacy Revealed Key Concepts ....................................................89

3.6 EHRS Organizational Support Revealed Key Concepts ....................................90

3.7 EHRS Costs Revealed Key Concepts ................................................................90

3.8 EHRS Benefits Revealed Key Concepts ...........................................................92

xiv
3.9 EHRS Perceived Value Revealed Key Concepts ...............................................93

3.10 EHRS Perceived Threats Revealed Key Concepts .............................................94

3.11 Facilitating Conditions Revealed Key Concepts ................................................96

3.12 EHRS Resistance Revealed Key Concepts ........................................................96

4.1 TTAT Model Constructs ................................................................................. 117

4.2 TTAT Hypotheses .......................................................................................... 119

4.3 Measurement Instrument ................................................................................ 140

4.4 Descriptive Statistics of Respondents .............................................................. 123

4.5 Items Loadings and CFA Model Fit Statistics ................................................. 143

4.6 Nonresponse Bias Test.................................................................................... 124

4.7 TTAT Model Fit Indices ................................................................................. 125

4.8 Standardized Path Coefficients .......................................................................126

4.9 Suggested Model Fit Indices TTAT Model Fit Indices ....................................131

5.1 An Illustration of the 4-Step Procedure for Constructing Causal Maps ............ 150

5.2 Revealed Constructs of EHR Avoidance and Its Concepts .............................. 153

xv
 LIST OF FIGURES

1.1 Search, Review, and Analysis Process ............................................................... 5

1.2 Literature Review Space .................................................................................... 6

2.1 Different Types of User Resistance Behavior ....................................................44

2.2 URM ................................................................................................................46

2.3 URM Hypotheses Testing Results.....................................................................52

2.4 Suggested Model Results ..................................................................................57

3.1 The Integrative Framework of URM .................................................................78

3.2 Point of Redundancy.........................................................................................86

3.3 Aggregated Revealed Causal Map of EHRS Resistance ....................................98

4.1 TTAT Conceptual Model ................................................................................ 119

4.2 TTAT Hypotheses Testing Results ................................................................. 126

4.3 Suggested Model Results ................................................................................ 131

5.1 Threat Avoidance Feedback Loop...................................................................149

5.2 Point of Redundancy....................................................................................... 152

5.3 Aggregated Revealed Causal Map of EHR Threat Avoidance ......................... 159

xvi
 Dedication

For my dad, who has been there for me with every step.

xvii
 CHAPTER ONE:

RESISTANCE OF HEALTHCARE INOFRMATION TECHNOLOGIES: LITERATURE

REVIEW, ANALYSIS, AND GAPS

ABSTRACT

The Implementing of Health Information Technology (HIT) is increasing, yet it is still

argued that HIT is being resisted by the same people who are expected to use it the most. IS

research in this area is still considered as under-researched. Prior Research has mainly focused

on the adoption aspect of the technology with limited effort towards explaining resistance

behaviors especially in healthcare organizations. HIT investments are high and expectations of

getting better performance, after its implementation are thus high. HIT, if meaningfully used, has

the potential of changing the way of applying healthcare services and highly benefit the

healthcare organizational performance. The benefits of HIT cannot be measured if the

technology was resisted. This paper provides an extensive review of the literature to uncover the

current status of the research in this area, to highlight the main research gaps that need to be

addressed, and to reveal how relatively little attention has been paid to understand HIT

resistance. Aiming to provide a research agenda that would encourage scholars to conduct

research in this area.

1
 CHAPTER ONE:

RESISTANCE OF HEALTHCARE INOFRMATION TECHNOLOGIES: LITERATURE

REVIEW, ANALYSIS, AND GAPS

INTRODUCTION

There is a dearth of prior studies investigating the phenomenon of resistance to HIT

(Bhattacherjee and Hikmet, 2007; and Poon et al, 2004), and despite the effort of IS researchers

in this area, it still remains understudied (Karahanna et al, 1999). This paper provides an

extensive review of the literature to uncover the current status of the research in this area, to

highlight the main research gaps that need to be addressed, and to reveal how relatively little

attention has been paid to understand HIT resistance. User resistance to information systems has

been identified as one of the salient reasons for the failure of new systems (Kim and

Kankanhalli, 2009). This is evident in the healthcare setting, (Spil et al, 2004) developed a ‘USE

IT’ model describing four dimensions of HIT use based on users’ needs: resistance, relevance,

requirements, and the availability of resources. The authors found that resistance was the

strongest determinant of not-using the electronic prescription systems (EPS). The technology

was not used in 72% of the cases, and that resistance was the cumulative consequence of the

remaining three determinants. Another example was reported on the Cedars-Sinai Medical

Center at Los Angeles where doctors resisted to use the newly installed Computerized Physician

Order Entry (CPOE) system, which caused the system to fail and result in a complete withdrawal

from implementation after it was already implemented in two-thirds of the 870-bed hospital

(Bhattacherjee et al, 2013). System failure caused by resistance has a potential of negative

impacts on users. This is very crucial in the healthcare setting where impacts are directly aimed

on patients. Result could be life-threatening cases, for example, the resistance of using a blood

2
sugar monitor by a type 2 diabetic patient may result in many health complications, especially if

the patient is in rural area and must rely on a mobile monitoring device to alert the doctor and the

patient when insulin is needed.

HITs are expected to increase patient safety, reduce cost of healthcare delivery, and

improve efficiency in the healthcare industry (Bhattacherjee and Hikmet, 2007), yet these

technologies, in many cases, are facing wide resistance (Lapointe and Rivard, 2005). User

resistance to new HITs needs to be understood and managed in order to mitigate HIT failures

(Poon et al, 2004).

There is an extensive body of research that has focused on technology adoption. It started

in 1989 when the Technology Acceptance Model (TAM) was first introduced (Davis, 1989).

Today, More than 345 articles focusing on technology acceptance have been published in the top

19 journals of the IS community in the last 20 years (Williams et al, 2009). However, it is argued

that technology in healthcare context has unique characteristics that makes it different from the

more general environments in which technology acceptance models have been successfully

tested in prior studies, therefor these models have less explanatory power when applied in

healthcare settings (Smith et al., 2014). Also, Practitioners usually use complex and difficult

medical equipment, so “hard to use” fails to explain the non-acceptance for them (Hare et al,

2006). Difficulty of use has a potential effect on patients’ perspectives towards using or resisting

the HIT. However, no studies, that includes patients in the examination of HIT resistance, were

found. Moreover, these theories are designed to explain voluntary IT usage and have limited

capabilities in explaining mandated IT usage which is present in the healthcare settings

(Bhattacherjee et al, 2013). Thus, it is believed that applying the acceptance models in healthcare

3
settings is not sufficient to completely explain the individuals’ behaviors towards new HIT and a

different perspective is still required to provide more insights regarding user resistance to HIT.

Some early IS research was conducted to uncover the concept of resistance, for example,

(Jha et al, 2008; Bhattacherjee and Hikmet, 2007; and Markus, 1983). However, the review

conducted by (Lapointe and Rivard, 2005), looked into 20 IS/IT journals over the past 25 years

and found 43 articles that considered resistance as an implementation issue. They reported that

only 9 of the 43 articles had defined resistance. Resistance is not only under researched but also

shapes a practice phenomenon in the healthcare sector, healthcare lags behind other industries in

adopting information technology by as much as 10–15 years (Goldschmidt, 2005).

This study provides an extensive review of IT resistance literature in IS and healthcare

disciplines to describe the current status of this literature and to analyze the literature to provide

researchers with a clear understanding of the emerging research gaps that need to be addressed in

future studies.

METHODOLOGY

To uncover the current status of the research and find the main research gaps, we

examined the extant literature. As shown in Figure 1.1, we started with conducting an extensive

search for technology resistance in different IS and healthcare journals and key IS conferences.

The keywords used were: IS, IT, adoption barriers, resistance, rejection, and healthcare.

4
 Figure 1.1: Search, review, and analysis process

This search resulted in 121 articles that covered the four areas of research as depicted in

Figure 1.2.

5
 Figure 1.2. Literature Review Space

Section 1 included studies about HIT with no resistance behavior, so these papers were

considered irrelevant to our research. Section 2 included papers about resistance to healthcare but

not to the technology used. These papers were also considered irrelevant to our research. Section

3 included papers that looked into resistance to IT, in general, not specifically in healthcare

settings. These papers are considered relevant to our research because HIT is merely a subset of

IT and understanding the general issue of resistance to IT is vital in developing a better

understanding of resistance in the healthcare settings. Section 4 included the most relevant

papers of our research, these papers looked into the resistance behavior towards technology in

healthcare settings.

Only 47 studies were considered as relevant after eliminating all irrelevant articles. The

47 relevant articles include 30 articles on IT resistance and 17 on HIT resistance. Relevant

papers were then coded, analyzed and synthesized in order to highlight the emerging gaps in this

area.

6
 EMERGING RESEARCH GAPS

This study aims to uncover the research gaps in the IT resistance literature with a focus

on HIT resistance. In order to reveal the research gaps of HIT resistance it was essential to

uncover those of IT resistance in general, and then format it in the context of healthcare. To find

those gaps, It was first necessary to characterize the current status of the IS research by

summarizing the literature that studied IT resistance in both the general context and the

healthcare context. It was clear that despite researchers’ attempts to understand IT resistance, this

area is still considered under researched (Laumer and Eckhardt, 2012). There are critical gaps in

the IT resistance literature, especially in healthcare settings, that need to be addressed (Poon et

al, 2004).

In the next section we provide a list of the research gaps revealed from the literature

review and analysis. Table 1.1 provides a visual summary of these gaps.

Lack of a Unified Framework of Resistance

Prior research that examined IT resistance, explained the resistance to IT using a sub-set

of factors that shape resistance behavior, such as loss of power (Markas and Hornik, 1996),

political concerns (Lin and Ashcraft, 1990), evaluation of the net gain and the change in the

status quo (Joshi, 1991), fear of losing jobs because of the introduction to the new technology

(Yoon et al, 1995), bad experiences perceived from the change to a newer technology and lack of

user involvement (Gill, 1996). However, (Karahanna et al, 1999) argued that prior perspectives

on IT resistance are still valuable, but the antecedents of IT resistance evolve and expanded over

time as individuals start to use IT more regularly and on different scales in different areas. The

study conducted by (Bhattacherjee and Hikmet, 2007) found that the prior research on IT

resistance is still limited and fragmented with no unified theory to explain the resistance behavior

7
towards technologies. In a recent study, (Laumer and Eckhardt, 2012) reviewed 9 IS models and

theories of resistance and concluded that IS literature still lacks a unified theory of user

resistance. The review conducted in this study indicates that the IS literature still needs to find

ways to define and measure IT resistance beyond the prior perspectives and demands a unified

framework explaining the IT resistance phenomenon and how similar or different is it from the

HIT resistance phenomenon.

Patients Are Ignored as Potential HIT Users

Patients are major stakeholders in the healthcare process Campbell et al, 2009). Today

the technology is advancing rapidly and is facilitating healthcare options to patients through a

variety of methods such as medical mobile apps that are widely spread and easily accessed by

patients (Ahmad et al, 2002). Also there is a wide spread of diagnostic tools that are available

and easily accessed over the web (Yager et al, 2006). However in most of the HIT studies such

as (Bhattacherjee and Hikmet, 2007; Joshi, 1991; and Lapointe and Rivard, 2005) the focus was

only on the practitioners, physicians, nurses, managers or Clinical laboratory personnel but not

the patients. It is important to understand who is resisting the technology and why are they

resisting it. This includes not only practitioners but also patients. Section 2 depicted in figure1 is

the area of articles that looked into patients’ resistance to healthcare not to the technology. These

patients could be resisting the HIT as a result of resisting the care itself. The indirect resistance

of patients towards healthcare occurred via the technology. While examining patients’ resistance,

it is crucial to explain what patients are actually resisting, and uncouple the resistance behavior

to understand whether the patients are rejecting the technology or the care itself.

The literature showed limited work on patients’ resistance to HIT. This is a major

research gap and researchers are encouraged to include patients in their future research designs.

8
Patients might have different reasons for resisting the technologies and uncovering these reasons

will contribute to both research and practice.

Limited Work in the Healthcare Setting

HIT resistance is very critical as it has a direct effect on human health, the resistance of

HIT may result in many medical errors that could lead to serious injuries and deaths (Poon et al,

2004). However, HIT resistance is clearly present in the U.S. according to (Jha et al., 2008) who

conducted a study comparing the level of HIT adoption in seven developed nations, and revealed

that U.S.A ranked last along with Canada with only 10 to 30 percent adoption rate. Also, within

the U.S. organizations, health technologies are being resisted more than technologies in different

disciplines (Goldschmidt, 2005).

HIT resistance is still not well understood (Laumer and Eckhardt, 2012), and a better

understanding of user resistance to HIT may help design better systems that are functional and

also adoptable by their potential users (Shah et al, 2006), it will also help in conducting

appropriate interventions to minimize the resistance behavior (Poon et al, 2004). However, the

review of the literature showed that very little attention has been given to the resistance of HIT.

In 2004 Poon et al., (Poon et al, 2004) found that this critical area is was considered under

researched in the IS literature. Ten years later, still very few studies are found on HIT resistance,

only 17 articles were conducted in the healthcare setting. Researchers are encouraged to give

more attention to this critical area and conduct research that will benefit academic research and

provide practitioners with a better understanding of the phenomenon and potentially suggest

solutions to avoid the HIT resistance behavior.

The Nature of Technology Support- System Functionalities and Supported Work Processes

9
 The literature review uncovered a number of different technologies used when examining

resistance. Understanding the functionality of the examined technology is important as the

performed task has a potential impact on individuals’ behavior towards the technology (Haymes,

2008), and could possibly change the behavior of potential users towards resistance. HIT used in

prior research confined to electronic health records (EHR), computerized patient orders entry

systems (CPOES) and other systems used for front desk clerk tasks such as admission, discharge,

and transfer tasks. Exploring the different technologies used for different work processes, such as

diagnostics, care giving, monitoring, etc., and examining resistance towards those technologies

could provide us with a more nuanced understanding of the resistance behavior.

Researchers are encouraged to expand their studies to include a wider variety of

technologies and work processes.

The Resistance Construct is Multifaceted

The prior literature defined and conceptualized resistance in a variety of ways. Resistance

behaviors were characterized by low levels of use, by lack of use, or by dysfunctional use

(Martinko et al, 1996). One study conducted by Keen (Keen, 1981) defined resistance as “social

inertia”. Some other definitions were based on the concept of resistance to change, e.g., (Zaltman

and Duncan, 1977), others defined it based on behavioral expressions and opposition to use, e.g.,

(Markus, 1983). Not only defined differently, resistance was also explained in a variety of ways;

Markus (Markus, 1983) explained the resistance behavior as a result of individuals perception on

power gain or loss associated with the change. This perspective is similar to the net gain concept

that Joshi (Joshi, 1991) proposed in his findings, he posits that users react to change by

evaluating their net gain based on change in their inputs and outcomes and comparing their

relative outcomes with that of other user groups and the employer. Bahattacherjee and Hikmet

10
(Bhattacherjee and Hikmet, 2007) is one of the few papers that explains resistance to change as

an important aspect of organizational change. They studied resistance in the healthcare setting

and also explained resistance differently building on the dual factor structure of IT usage

proposed by (Cenfetelli, 2004) that posits inhibitors as the main reason of resistance, defining

inhibitors as the negative factors that discourage IT usage when present, but have no effect when

absent.

This summary of IT resistance conceptualized in the extant literature illustrates that this

construct is multifaceted and needs to be developed further. The IS scholars should strive for a

more comprehensive yet integrated characterization of IT resistance. A lack of unified

conceptualization of the resistance behavior will impede scholarship in the research area which is

in its incipient stages. A rich and deep conceptualization of this multifaceted construct will allow

researcher to instantiate it in diverse settings including healthcare.

The Role of Beliefs in IT Resistance Needs Greater Theorization

IS research has provided limited insights on the role of behavioral and social beliefs in

shaping IT resistance. Prior research mainly focused on perceived threat as the belief that shifts

individuals’ behavior towards resisting the technology (Laumer, 2011). This is evident in studies

that focused on IT resistance as well as on HIT resistance. For example, (Markas and Hornik,

1996) proposed perceived threat as an antecedent to IT resistance whereas Bahattacherjee and

Hikmet (Bhattacherjee and Hikmet, 2007) used it to explain HIT resistant. However, perceived

threat is a broad concepts that may include a number of antecedents and causes. Prior research

has characterized threat differently, Bahattacherjee and Hikmet (Bhattacherjee and Hikmet,

2007) explained threat as physician’s loss of control over their work, which is similar to that

observed by (Lapointe and Rivard, 2005). Another attempt to explain resistance using the

11
concept of threat is the work of (Kim and Kankanhalli, 2009) who defined threat as the perceived

costs verses the perceived benefits that potential users consider before rejecting the technology.

This is similar to the findings of Joshi (Joshi, 1991). Perceived threat still has no one unified

definition or explanation. Other beliefs were considered when explaining resistance; (Cenfetelli,

2004) categorized potential users’ beliefs into inhibitors and enablers that would discourage or

encourage the usage behavior. However, a general explanation was given to show how inhibitors

discourage usage with no further investigation on what these inhibitors could be. (Markas and

Hornik, 1996) proposed that both fear and stress generated from the introduction of the

technology into the world of the potential user has impact on the resistance behavior. They

explained how we must be mindful that people “do not resist new technology, rather they resist

the effect that the new technology may have on their own lives.” This goes in hand with the

study of (Martinko et al, 1996) that also posited that resistance is affected by individuals’ beliefs.

They explained that negative outcomes result from individuals’ beliefs towards technology

usage. (Enns et al., 2003) studied how uncertainty and the loss of status can change the balance

between acceptance and resistance.

Not only behavioral believes require further theorization, also social beliefs impact must

be considered when researching resistance. Opinions and behavior of individuals are influenced

by their social networks (Laumer and Eckhardt, 2012). This is explained by the social impact

theory which shows how social influence has an important impact on individuals’ behaviors.

Individuals’ decision towards technology adoption or resistance is extensively affected by the

influence of other users (Keen, 1981). However, limited attention has been devoted to the study

of social influence on resisting technologies within organizations (Laumer, 2011). Also limited

attention has been given to the social influence stemmed from private sources outside of the

12
organization such as family and friends (Laumer and Eckhardt, 2012). Our review found that

only (Eckhardt et al, 2009) studied social influence on individual resistance. They expanded the

work of UTAUT (Venkatesh et al, 2003) to explain the impact of social influence on non-

adoption intentions, and these influences were limited to work place colleagues and superiors not

including the influence of other individuals outside of the organizations’ boundaries, such as

friends and family. This is very important to examine, especially when measuring its impact on

patients resistance to HIT. Patients, in most cases are not employees of the healthcare facility,

and in some cases might be resisting these technologies while being remote from the healthcare

facility, therefor the influence of employees at the facility has limited impact on patients’

decisions. Articles in section 2 of figure1 revealed that patients resist receiving care from the

health providers for reasons related to their personal beliefs such as religion. When examining

resistance to HIT, it is important to uncouple patients’ resistance to the technology and their

resistance to the actual care based on their beliefs and previous experiences. Moreover, patients

may be resisting care as a result of resisting the technology implemented. Both behavioral and

social beliefs impact on resistance are interesting areas that still needs more investigating.

The IS scholars should strive for a more comprehensive understanding of the role of

individuals’ believes in IT resistance. Understanding the impact of beliefs on resistance and how

these beliefs are related helps in developing a better understanding of resistance (Laumer and

Eckhardt, 2012).

Individual Differences in IT Resistance

The review of prior literature indicates that not only demographic variables of the

potential users affect IT resistance, but also personality factors such as levels of flexibility, desire

for change and growth, and tolerance for ambiguity are closely related to resistance (Majchurzak

13
et al, 1988). The case study of IT resistance at the Hmong American Partnership (Vang, 2008)

revealed that “the one antecedent that withstood time in relation to user acceptance or rejection

of information technology was individual indifferences, such as a user’s age, education, and

gender”. IS research has considered individual differences in a variety of studies. However, few

studies considered individual differences when examining IT resistance, according to (Laumer,

2011) only one study (Klaus et al, 2010) considered individual differences when examined IT

resistance.

The literature review confirms that few number of articles studied the impact of

individual differences on IT resistance and none were conducted in the healthcare setting.

Studying the impact of individual differences on IT resistance could result in interesting findings

that contributes to research as well as to practice.

Table 1.1. Research Gaps in HIT Resistance Literature vs IT Resistance Literature

Emerging Research Gap HIT Resistance IT

Resistance
(Out of
Health
Context)
Lack of a Unified Framework of Resistance X X
Patients are Ignored as Potential HIT Users X
Limited Work in The Healthcare Setting X
The Nature of Technology Support-System Functionalities X X
and Supported Work Processes
The Resistance Construct is Multifaceted X X
The Role of Beliefs in IT Resistance Needs X X
Greater Theorization
Individual Differences in IT Resistance X X

IMPLICATIONS

Implications for Research

This study has a number of implications for research. First, our analysis of the current

status of resistance in the IS literature was not only limited to HIT resistance, rather it provided a

14
general analysis on IT resistance including healthcare settings. Second, it provides research gaps

analysis and highlights a research agenda for future studies that are encouraged to build on the

findings and address the emerging research gaps that were looked over in the past. Third, it

summarized the comprehensive description of the core concepts of resistance to IT and HIT.

Finally, the study encourages researchers to start aiming their work towards finding a unified

theory of resistance that considers individual believes and differences, as well as to include

patients as a major stakeholder when theorizing on HIT resistance.

Implications for Practice

This study provide implications for practice, by establishing a better understanding of

user resistance to IT, system designers will be able to build better systems that are functional and

also adoptable by their potential users (Poon et al, 2004). Furthermore, understanding how

resistance could be treated will have a positive impact on system success (Kim and Kankanhalli,

2009). This paper seeks to find answers to IT resistance in general and focuses on the HIT

resistance behavior. Reducing the resistance to HIT by patients and practitioners will help in

avoiding catastrophic medical errors (Kohn et al, 2000) and will help in improving the overall

quality, efficiency and performance of the healthcare provider (Bhattacherjee et al, 2006). This

study aims to encourage researchers to build on the emerging gaps to provide clear guidelines

that potentially could help in building and implementing technologies with lower rates of

resistance.

FUTURE WORK

We hope that this study will spark interest in the IT resistance phenomena in general and

in healthcare settings in particular and encourage researchers to conduct studies in this

unexplored yet potentially fertile area of research. In particular, future work needs to design

15
research studies that are patient centric. Future research needs to decouple the effects of the

factors that contribute to patients’ resistance to healthcare from their resistance to IT used to

support the care they are receiving. We encourage scholars to address the research gaps

uncovered in this study by conducting empirical research in this area in an aim to discover the

significant antecedents shaping the HIT resistance behavior. We also hope that this study will

stimulate more theoretical contributions in this area to form a better understanding of the

phenomenon which will help both researchers in formulating a unified understanding of

technology resistance and designers in building more adoptable systems. The main contribution

of this paper is in shaping the future research agenda by revealing the research gaps in this

critical area. By revealing the research status this work encourages researchers to focus on the

key research gaps while conducting studies related to IT and HIT resistance.

16
 REFERENCES
Agarwal, R 2000. "Individual acceptance of information technologies," Framing the domains of
IT management: Projecting the future through the past, pp. 85-104.
Agarwal, R, 2010. "Research commentary-The digital transformation of healthcare: current
status and the road ahead," Information Systems Research, 21, 4, pp. 796-809.
Ahmad, A., Teater P., Bentley T.D., Kuhen L., Kumar R.R., Thomas A., and Mekhijan H.S.,
2002. “Key attributes of a successful physician order entry system implementation in a
multi-hospital environment,” Journal of American Medical Informatics Association, 9, 1,
pp. 16–24.
Ajzen, I. and Fishbein, M., 1980. “Understanding attitudes and predicting social behavior,”
Englewood Cliffs, NJ: Prentice-Hall.
Ajzen, I., 1991. “The theory of planned behavior.” Organizational Behavior and Human
Decision Processes”, 50, pp. 179–211.
Anderson, Catherine L., and Ritu Agarwal, 2011. "The digitization of healthcare: boundary risks,
emotion, and consumer willingness to disclose personal health information," Information
Systems Research, 22, 3, pp. 469-490.
Ang, J., and Pavri, F., 1994. "A Survey and Critique of the Impacts of Information Technology,"
International Journal of Information Management, 14, 2, pp. 122-133.
Angst, C., R. Agarwal, V. Sambamurthy, and K. Kelley, 2010. “Social contagion and
information technology diffusion: The adoption of electronic medical records in U.S.
hospitals,” Management Science, 56, 8, pp. 1219–1241.
Baker, L., T. H. Wagner, S. Singer, M. K. Bundorf, 2003. “Use of the Internet and e-mail for
health care information: Results from a national survey,” Journal of American Medical
Association, 289, 18, pp. 2400–2406.
Bates, W., Leape, L.L., Cullen, D.J., Laird, N., Peterson, L.A., and Teich J.M., 1998. “Effect of
computerized physician order entry and team intervention on prevention of serious
medication errors,” Journal of American Medical Association, 280, 15, pp. 1311–1316.
Bhattacherjee, A., and Hikmet, N., 2007. "Physicians' resistance toward healthcare information
technology: a theoretical model and empirical test," European Journal of Information
Systems, 16, 6, pp. 725-737.
Bhattacherjee, A., Hikmet, N., Menachemi, N., Kayhan, V. O., & Brooks, R. G., 2006. “The
differential performance effects of healthcare information technology adoption,”
Information Systems Management, 24, 1, pp. 5-14.

17
Bhattacherjee, A., and Hikmet, N., 2007. "Physicians' resistance toward healthcare information
technologies: a dual-factor model," 40th Annual Hawaii International Conference on
System Science, IEEE.
Bhattacherjee, A, Davis C., and Hikmet N., 2013. "Physician Reactions to Healthcare IT: An
Activity-Theoretic Analysis," 46th Hawaii International Conference on system science,
IEEE.
Bhattacherjee, A., Davis, C., Hikmet, N., and Kayhan, V., 2008. "User reactions to information
technology: Evidence from the healthcare sector," Association for Information Systems.
Borzekowski, R, 2002. “Health care finance and the early adoption of hospital information
systems,” Divisions of Research & Statistics and Monetary Affairs, Federal Reserve
Board. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=337960
Ram, B., Mala, Grocott, P., and Weir, H., 2008. "Issues and challenges of involving users in
medical device development," Health Expectations, 11, 1, pp. 63-71.
Campbell, S.M., Shield, T., Rogers, A., and Gask, L., 2004. "How do stakeholder groups vary in
a Delphi technique about primary mental health care and what factors influence their
ratings?" Quality and Safety in Health Care 13.6 (2004), pp. 428-434.
Campbell E., K. Guappone, D. Sittig, R. Dykstra, J. S. Ash., 2009. “Computerized provider order
entry adoption: Implications for clinical workflow,” Journal of General Internal
Medicine 24, 1, pp. 21–26.
Cenfetelli R.T., 2004. “Inhibitors and enablers and dual factor concepts in technology usage,”
Journal of the Association of Information Systems, 5, 11, (2004), pp. 472–492.
Chen, J.M., Suen, M.W., Chiu, S.C., and Lin, M.J., 2004. “Social dilemma studies in Taiwan,”
International Journal of Psychology, Vol. 39, No. 5-6, pp. 577-577.
Chen, P., and Hitt, L. M., 2002. “Measuring Switching Costs and the Determinants of Customer
Retention in Internet-Enabled Online Brokerage Industry,” Information Systems
Research, 13, 2, pp. 255-274.
Chiasson, M., E. Davidson, 2002. “Getting the two to dance: Examining barriers to health
information systems research in mainstream IS journals”, Proceedings of the 35th Annual
Hawaii International Conference on System Science, Big Island, HI, pp. 1966-1974.
Chismar, William G., and Wiley-Patton, S., 2003. "Does the extended technology acceptance
model apply to physicians," System Sciences Proceedings of the 36th Annual Hawaii
International Conference on System Science. IEEE.
Coetsee, L. D., 1993. "A Practical Model for the Management of Resistance to Change: An
Analysis of Political Resistance in South Africa," International Journal of Public
Administration, 16, 11, pp. 1815-1858.

18
Coetsee, L. D., 1999. "From Resistance to Commitment," Public Administration Quarterly, 23,
2, pp. 204-222.
Cutler, D. M., N. E. Feldman, J. R. Horwitz, 2005.“U.S. adoption of computerized physician
order entry systems,” Health Affairs, 24, 6, (2005), pp. 1654–1663.
Davis, F. D., 1989. “Perceived Usefulness, Perceived Ease of Use, and User Acceptance of
Information Technology,” MIS Quarterly, 13, 3, pp. 319-340.
DeSanctis, G., and Courtney, J.F., 1983. "Toward Friendly User MIS Implementation,"
Communications of the ACM, 26, 10, pp. 732-738.
Doolin, B., 2004. "Power and resistance in the implementation of a medical management
information system," Information Systems Journal, 14.4, pp. 343-362.
Eckhardt, A., Laumer S., and Weitzel, T., 2009. "Who influences whom and quest; analyzing
workplace referents' social influence on IT adoption and non-adoption," Journal of
Information Technology, 24, 1, pp. 11-24.
England, I., D. Stewart, S. Walker, 2000. “Information technology adoption in health care:
When organizations and technology collide,” Australian Health Review, 23, 3, pp. 176–
185.
Enns, H.G., Huff, S.L., and Higgins, C.A., 2003. "CIO Lateral Influence Behaviors: Gaining
Peers' Commitment to Strategic Information Systems," MIS Quarterly, 27, 1, pp. 155-
175.
Fishbein, Martin, and Ajzen, I., 1975. “Belief, attitude, intention and behavior: An introduction
to theory and research.”
Fisher, S., 2008. “Improving Physician Adoption of CPOE Systems,”
http://www.strategiestoperform.com/volume3_issue2/docs/ImprovingPhysicianAdoption.
pdf
Ford, Jeffrey D., Laurie W. Ford, and Angelo D'Amelio, 2008. "Resistance to change: The rest
of the story,” Academy of Management Review, 33, 2, pp. 362-377.
Freudenhiem, M., 2004. “Many hospitals resist computerized patient care,” New York Times.
http://www.nytimes.com/2004/04/06/technology/06errors.html.
Gill, T. Grandon, 1996. "Expert systems usage: task change and intrinsic motivation," MIS
quarterly, pp.301-329.
Goh, Jie Mein, Guodong Gao, and Ritu Agarwal, 2011. "Evolving work routines: adaptive
routinization of information technology in healthcare," Information Systems Research,
22,3, pp. 565-585.
Goldschmidt, Peter G., 2005. "HIT and MIS: implications of health information technology and
medical information systems," Communications of the ACM, 48, 10, pp. 68-74.

19
Hare, K., Whitworth, B., Deek, F., and Norris, T., 2006. “Identifying Barriers to IT Adoption in
Clinical Health Care: A Pilot Study,” Health Care and Informatics Review Online, ISSN
1176-4201, http://hcro.enigma.co.nz/website/index.cfm
Haymes, T., 2008. "The three-e strategy for overcoming resistance to technological change,"
Educause Quarterly, 31, 4, pp. 67-69.
Hirschheim, R., and Newman, M., 1988. “Information Systems and User Resistance: Theory and
Practice,” Computer Journal, 31, 5, pp. 398-408.
Holmes, Thomas J., and James Schmitz, 1995. "Resistance to new technology and trade between
areas." Federal Reserve Bank of Minneapolis Quarterly Review, 19, 1, pp. 2-17.
Jha, A. K., Doolan, D., Grandt, D., Scott, T., and Bates, D. W., 2008. “The use of health
information technology in seven nations,” International journal of medical informatics,
77, 12, pp. 848-854.
Jiang, James J., Waleed A. Muhanna, and Gary Klein, 2000. "User resistance and strategies for
promoting acceptance across system types," Information & Management, 37, 1, PP. 25-
36.
Joshi, K., 1991. “A Model of User Perspective on Change: The Case of Information Systems
Technology Implementation,” MIS Quarterly, 15, 2, pp. 229-242.
Karahanna, Elena, Detmar W. Straub, and Norman L. Chervany, 1999. "Information technology
adoption across time: A cross-sectional comparison of pre-adoption and post-adoption
beliefs,” MIS quarterly, 23, 2.
Kazley, A., Y. Ozcan, 2007. “Organizational and environmental determinants of hospital EMR
adoption: A national study,” Journal of Medical Systems, 31, 5, pp. 375–384.
Keen, P.G., 1981. "Information Systems and Organizational Change," Communications of the
ACM, 24, 1, pp. 24-33.
Kim, H.W., and Kankanhalli, A., 2009. "Investigating user resistance to information systems
implementation: a status quo bias perspective,” MIS quarterly, 33, 3.
Klaus, T., S. C. Wingreen and J. E. Blanton, 2010. "Resistant Groups in Enterprise System
Implementations: A Q-Methodology Examination," Jouransl of Information Technology,
25, 1, pp. 91-106.
Kohn, Linda T., Janet M. Corrigan, and Molla S. Donaldson, 2000. "To err is human: building a
safer health system: A report of the Committee on Quality of Health Care in America,”
Institute of Medicine.
Lapointe, L., and Rivard, S., 2005. “A Multiple Model of Resistance to Information Technology
Implementation,” MIS Quarterly, 29, 3, pp. 461-491.
Lapointe, L., and Rivard, S., 2007. “A Triple Take on Information System Implementation”,
Organization Science, 18, 1, pp.89-107.
20
Latané, B., 1981. “The psychology of social impact,” American Psychologist, 36, pp. 343-356.
Laumer, S., and Eckhardt, A., 2012. "Why do people reject technologies: a review of user
resistance theories," Information Systems Theory. Springer New York, pp. 63-86.
Laumer, S., 2011. “Why Do People Reject Technologies–A Literature-Based Discussion of the
Phenomena Resistance to Change In Information Systems and Managerial Psychology
Research”
Lee, H.G., and Clark, T.H., 1997. "Market Process Reengineering through Electronic Market
Systems: Opportunities and Challenges," Journal of Management Information Systems,
13, 3, Winter, pp. 113-137.
Lehoux, P., 2008. “Why examining the desirability of health technology matters,” Politiques de
Santé.
Leonardi, Paul M., 2009. "Why do people reject new technologies and stymie organizational
changes of which they are in favor? Exploring misalignments between social interactions
and materiality,” Human Communication Research, 35, 3, 407-441.
Lin, E.M., and Phillip Ashcraft, 1990. "A case of system-development in a hostile environment,”
Journal of Systems Management 41, 4, pp. 11-14.
Lin, C., Lin I., and Roan J., 2012. "Barriers to physicians’ adoption of healthcare information
technology: an empirical Study on multiple hospitals," Journal of medical systems, 36, 3,
pp. 1965-1977.
Lorence, D.P., and Richards, M.C., 2003. “Adoption of regulatory compliance programs across
United States healthcare organizations: a view of institutional disobedience,” Health
Services Management Research, 16, 3, pp. 67–178.
Majchrzak, A., and Cotton, J., 1988. "A longitudinal study of adjustment to technological
change: From mass to computer‐automated batch production," Journal of Occupational
Psychology, 61, 1, pp. 43-66.
Marakas, G. M., and Hornik, S., 1996. “Passive Resistance Misuse: Overt Support and Covert
Recalcitrance in IS Implementation,” European Journal of Information Systems, 5, 3, pp.
208-219.
Markus M. L., 1983. “Power, Politics, and MIS Implementation”, Communications of the ACM,
26, 6, pp. 430-444.
Martinko, M. J., Henry, J. W., and Zmud, R. W., 1996. “An Attributional Explanation of
Individual Resistance to the Introduction of Information Technologies in the Workplace,”
Behavior and Information Technology, 15, 5, pp. 313-330.
Massaro T.A., 1993. “Introducing physician order entry at a major academic medical center:
Impact on organizational culture and behavior,” Academic Medicine 68, 1, pp. 20–25.

21
Meissonier, R., and Emmanuel H., 2009. "Attitudes to information technology in health care
professions," European Conference on Information Systems.
Mekhjian, S., Kumar R., Kuehn, Bentley, Teater, P., Thomas, A., Payne, B., and Ahmad A.,
2002. "Immediate benefits realized following implementation of physician order entry at
an academic medical center," Journal of the American Medical Informatics Association,
9, 5, pp. 529-539.
Gonzalez, P. A., and Chan, Y.E., 2013. “Exploring Healthcare Professionals’ Resistance to
Using Wireless Communication Technology,” Proceedings of the 19th Americas
Conference on Information Systems, Chicago, Illinois.
Parthasarathy, M., and Bhattacherjee A., 1998. “Understanding post adoption behavior in the
context of online services,” Information Systems Research 9, 4, 362–379.
Piderit, S.K., 2000. “Rethinking resistance and recognizing ambivalence: a multidimensional
view of attitudes toward an organizational change,” Academy of Management Review 25,
4, pp. 783-794.
Poon, E., Blumenthal, D., Jaggi, T., Honour, M., Bates, D.W., and Kaushal, R., 2004.
“Overcoming barriers to adopting and implementing computerized physician order entry
systems in U.S. hospitals,” Health Affairs 23, 4, pp. 184–190.
President’s Information Technology Advisory Committee (2004), Revolutionizing Health Care
through Information Technology, National Coordination Office for Information
Technology Research and Development, Arlington, Virginia. http://www.nitrd.gov
Shah, S., and Robinson I., 2006. "User involvement in healthcare technology development and
assessment: structured literature review,” International Journal of Health Care Quality
Assurance, 19, 6, pp. 500-515.
Smith, Tim, Gerald G., and Ramirez, A., 2014. "Investigating the influence of psychological
ownership and resistance on usage intention among physicians,” 47th Hawaii
International Conference on System Science, IEEE.
Spil, T. A., Schuring, R. W., and Michel-Verkerke, M. B., 2004. “Electronic prescription system:
do the professionals use it?” International Journal of Healthcare Technology and
Management, 6, 1, pp. 32-55.
Taylor, R., Manzo, J., and Sinnett, M., 2002. "Quantifying value for physician order-entry
systems: a balance of cost and quality." Healthcare financial management: journal of the
Healthcare Financial Management Association, 56, 7, pp. 44-48.
Kate, T., Haverkamp, S., Mahmood, F., and Feldberg, F., 2010. “Social network influences on
technology acceptance: A matter of tie strength, centrality and density,” BLED
Proceedings, Paper, 40.

22
Tulu, Bengisu, Richard J. Burkhard, and Thomas A. Horan, 2006. "Continuing use of medical
information systems by medical professionals: empirical evaluation of a work system
model,” Communications of the AIS, 18, 31, pp. 641–656.
Vang, T. S., 2008. “User resistance to information technology: A case study of the Hmong
American partnership,” ProQuest.
Venkatesh, V., M.G. Morris, G.B. Davis, and F.D. Davis, 2003. “User Acceptance of
Information Technology: Toward A Unifying View,” MIS Quarterly, 27, 3, pp. 425-478.
Venkatesh, V., and Susan A. B., 2001. "A longitudinal investigation of personal computers in
homes: adoption determinants and emerging challenges." MIS quarterly, pp. 71-102.
Williams, Michael D., Yogesh K. Dwivedi, Banita Lal, and Schwarz, A., 2009. "Contemporary
trends and issues in IT adoption and diffusion research," Journal of Information
Technology, 24, no. 1, pp. 1-10.
Yager, P., Edwards, T., Fu, E., Helton, K., Nelson, K., Tam, M. R., & Weigl, B. H., 2006.
“Microfluidic diagnostic technologies for global public health,” Nature, 442, pp. 412-
418.
Yoon, Y., and Guimaraes, T., 1995. "Assessing expert systems impact on users' jobs," Journal of
Management Information Systems, 12, 1, pp. 225-249.
Zaltman, G., and Duncan R., 1977. “Strategies for Planned Change,” John Wiley & Sons, New
York, NY.

23
 CHAPTER TWO:

SWITCHING TO ELECTRONIC HEALTH RECORD SYSTEMS: AN APPLICATION

OF THE USER RESISTANCE MODEL

ABSTRACT

Despite the positive effects of electronic health records within healthcare practices, the

adoption rate of such systems is still low and faces resistance from healthcare providers. In this

paper we aim to understand why healthcare providers resist to change from the traditional

paper-based recording systems to the newly implemented electronic health record systems. Our

study tests the user resistance model using a questionnaire to collect data from a random sample

of healthcare providers of a large public hospital in Amman-Jordan which recently implemented

and enforced a new EHR system. Data was collected in 2 stages: pilot study and main study. 226

valid respondents participated in the main study survey. Results were then analyzed to describe

and evaluate resistance to change to the new EHR system. Our finding contributes to IS research

by highlighting main issues surrounding resistance to change from traditional paper-based

systems to EHR systems which is still considered as under-researched in IS. Additionally, we

contribute to practice by providing understanding of the resistance phenomenon which would

help EHR system developers in the process of updating, designing, and developing these systems

in an aim to have higher adoption outcomes.

24
 CHAPTER TWO:

SWITCHING TO ELECTRONIC HEALTH RECORD SYSTEMS: AN APPLICATION

OF THE USER RESISTANCE MODEL

INTRODUCTION

Electronic Health Records (EHR) are expected to bring a number of benefits to the health

organization. EHR can increase patient safety, reduce cost of healthcare delivery, and improve

efficiency in the healthcare industry (NITRD, 2004; Hillestad et al, 2005). However, switching

to EHR is challenging as it requires hospitals and its staff to change the way work is conducted

(Sheikh et al, 2011). For instance, in the U.S. the Meaningful-Use Mandate, enforced by the

Department of Health and Human Services, requires that all healthcare organizations must adopt

and use EHR meaningfully (Blumenthal and Tavenner, 2010). However, it has now been over

six years since this initiative was launched, and early reports show that it is still a challenge for

physicians and other healthcare providers to accomplish tasks using the new EHR systems

(Heath and Appan, 2014). In 2009 it was estimated that 73% of EHR implementations are, “not

using the system as intended 12 months after implementation” (Terry, 2012). This is evident on a

global scale. International case studies from 2013 showed that the U.S., the U.K., France, and

India all share similar struggles of EHR implementation (Stone, 2014).

One of the primary challenges of EHR usage is the labor intensive data entry that

healthcare providers have little tolerance for because they believe it impedes with their primary

goal of taking care of their patients (Cantrill, 2010). Additionally, issues such as technical

problems and design difficulties (Cantrill, 2010), and security problems and data privacy issues

(Fu, 2009) also pose usage obstacles. These challenges are contributing in the shaping of the

phenomenon of healthcare providers resisting to change to EHR systems. A better understanding

25
of the phenomenon may help design systems that are functional, adoptable by their potential

users, and have higher success rates (Bhattacherjee and Hikmet, 2007; Shea and Hripcsak, 2010;

Jha et al, 2009).

Despite that resisting to change to new EHR systems is a global phenomenon, this topic

is still considered underexplored (Samhan and Joshi, 2015). There is a lack of empirically tested

models aiming to explain “why” and “how” healthcare providers resist EHR systems (Boonstra

and Broekhuis, 2010; Weeger et al., 2011; Olaniran, 2015). Therefore, in an effort to explain

healthcare providers’ motivations to resist the switch to EHR systems, in this study we examine

the following research question - Why do healthcare providers resist to switch to EHR systems?

Because we aim to capture resistance at early stages of the EHR implementation it was

appropriate to investigate the resistance to change behaviors where healthcare providers are

switching from a Paper Based Recording System (PBRS) to a new EHR system.

To our knowledge, this work is the first attempt in the IS literature to apply a

comprehensive theoretically grounded model of resistance with a focus on the early

implementation stage of a new EHR system. Our study conducts an empirical test of the User

Resistance Model (URM) (Kim and Kankanhalli, 2009) to assess resistance to change from

PBRS to EHR. We contribute to literature by extending and enhancing this model by, first,

applying the model in a healthcare context. Second, this study captures resistance to change

right after the implementation of the new EHR system, whereas this model in the past has been

posited and tested during the pre IT system implementation stage. This could serve as a

comparative between resistance at different stages (i.e. pre and post implementation). Third,

while URM provided a conceptual argument only on the effect of loss aversion on the status quo

bias, we empirically validate the principle of loss aversion in our model. Finally, while URM

26
limited social influence to that from peers and coworkers only, we include peers, colleagues, and

superiors as individuals with potential influence on an employee.

LITERATURE REVIEW

Prior research defined and conceptualized resistance to change differently. The work of

(Keen, 1981) viewed resistance to change from the costs-benefits perspective, he argued that

resistance to change occurs if the costs of change are perceived as greater than the benefits. This

is similar to the conceptualization used in the Equity Implementation Model (EIM) (Joshi, 1991)

which suggests that the greater the perceived inequity the greater the resulting distress would be,

and people with greater distress are more likely to resist it by attempting to minimize their inputs

and others’ outcomes as well as attempting to increase others’ inputs. In a more recent study

(Bahattacherjee and Hikmet, 2007) incorporated resistance to change with the Technology

Acceptance Model (TAM) (Davis, 1989). They explained user resistance to change by

explaining physician resistance to HIT that was a result of the perceived threats. This is similar

to the work of (Walter and Lopez, 2008) who posited that perceived threat influences intentions

to use which results in resistance to change behaviors. The User Resistance Theory (Kim and

Kankanhalli, 2009) posited the user resistance model which explained resistance to change from

a costs-benefits perspective and defined resistance as the opposition of a user to change

associated with a new IS implementation. However, it is found that there is still a need for a

unified theory of resistance in IS (Samhan and Joshi, 2015). Further, our literature review

revealed that prior IS research has captured resistance to EHR systems mainly by capturing

lower levels of adoption. (Table 2.1) summarizes prior studies on EHR resistance.

Table 2.1. Summary of IS Research on EHR Resistance

Article Definition of resistance Stage of Resistance Method
Angst and Agarwal, 2009 Opposing to use the system Post Implementation Survey

27
Boonstra and Broekhuis, 2010 Lower Levels of EHR adoption Review of both Pre and Conceptual
Post Implementation Literature
Review
Burns et al, 2015 Work-Around Behaviors Post Implementation Survey

Burt et al, 2005 Lower Levels of EHR Post Implementation Survey

Adoption
Davidson et al 2007 Lower Levels of EHR Post Implementation Interviews
Adoption
DesRoches et al, 2008 Lower Levels of EHR Post Implementation Survey
Adoption
Dinev et al, 2016 Lower Levels of EHR Post Implementation Survey
Adoption
Earnest et al, 2004 Lower Levels of EHR Post Implementation Survey &
Adoption Interviews
Gan and Cao, 2014 Lower Levels of EHR Post Implementation Survey
Adoption
Hewitt and McLeod, 2011 Lower Levels of EHR Post Implementation Survey
Adoption
Hung, 2013 Lower Levels of EHR Post Implementation Survey
Adoption
Jha et al, 2008 Lower Levels of HER Post Implementation Survey
Adoption
Kemper et al, 2006 Lower Levels of HER Post Implementation Survey
Adoption
Lapointe and Rivard, 2005 A range of behaviors from Longitudinal: early and Case Studies
late stages of
Passive Resistance to
implementation
Aggressive Resistance.
Laerum et al, 2001 Lower Levels of EHR Post Implementation Survey
Adoption
Loomis et al, 2002 Lower Levels of EHR Post Implementation Survey
Adoption
Ludwick et al, 2009 Lower Levels of EHR Post Implementation Interviews
Adoption
Najaftorkaman et al, 2015 Lower Levels of EHR Review of both pre and Conceptual
post implementation
Adoption Literature
Review

28
Ngafeeson and Midha, 2014 A covert or overt intention Post Implementation Survey
that opposes change towards
the use of an information
system
Nöltner and Krönung, 2015 Lower Levels of EHR Post Implementation Literature
Adoption Review
Meade et al, 2009 Lower Levels of EHR adoption Post Implementation Survey
Menachemi et al, 2007 Lower Levels of EHR adoption Post Implementation Survey
Miller et al, 2004 Lower Levels of EHR adoption Post Implementation Interviews
Olaniran, 2015 Lower Levels of EHR Post Implementation Conceptual
Adoption
Pizziferri et al, 2005 Lower Levels of EHR Post Implementation Qualitative
Adoption Observations
Poon et al, 2006 Opposing adoption of HIT Post Implementation Interviews

Randeree, 2007 Lower Levels of EHR Post Implementation Interviews

Adoption
Reardon and Davidson, 2007 Lower Levels of EHR Post Implementation Survey
Adoption
Shachak et al, 2009 Lower Levels of EHR Post Implementation Interviews
Adoption
Simon et al, 2007b Lower Levels of EHR Post Implementation Survey
Adoption
Simon et al, 2007c Lower Levels of EHR Post Implementation Survey
Adoption
Terry et al, 2008 Lower Levels of EHR Post Implementation Interviews
Adoption
Valdes et al, 2004 Lower Levels of EHR Post Implementation Survey
Adoption
Vishwanath et al, 2007 Lower Levels of EHR Post Implementation Conceptual
Adoption
Walter et al, 2008 Lower Levels of EHR Post Implementation Survey
Adoption
Weeger et al., 2011 Resistance is not a behavior but Post Implementation Interviews
a cognitive force preluding
potential behavior

29
 One of the most insightful reviews of EHR adoption (Najaftorkaman et al, 2015)

categorized prior findings into 8 main factors and argued that these factors has the potential to

impact EHR adoption levels. And that lower levels of adoptions would be interpreted as

resistance. These factors are: Individual factors such as age and gender, physiological factors

such as satisfaction and attitudes, behavioral factors such as behavioral change and automatic

behavior, environmental factors such as competition and reputation of practice, organizational

factors such as size and age of the organization, financial factors such as start-up costs and on-

going costs, legal factors such as privacy and security issues, and finally technical factors such as

technical training and technical support. The work of (Desroches et al., 2008) investigated EHR

resistance in the same direction and found that the most common hurdles for EHR adoption are

financial cost related issues, not finding a system that meets healthcare providers’ needs, and

concerns that the EHR system would become “obsolete”. Other research emphasized on the

concept of low self-efficacy as the main contributor towards having lower EHR adoption rates

(Jha et al, 2008; Olaniran, 2015). The work of (Hung et al, 2013) also posited that computer self-

efficacy influences EHR adoption but also included perceived service levels (reliability and

responsiveness of the EHR system) along with perceived risks as contributors that impact

healthcare providers’ decision of EHR adoption. Additionally, the work of (Alghamdi, 2015)

posited that the main hurdles of EHR adoption are self-efficacy, privacy and security concerns

cost of implementation, and EHR maintenance “down time”. The task technology fit theory was

used to explain EHR resistance by lower levels of adoption (Gan and Cao, 2014). They argued

that adoption of EHR may be explained by the intention of the organization to improve

organizational performance through the fit between task and technology. Furthermore, the

30
concept of strength of attitude was used to explain EHR resistance (Nöltner and Krönung, 2015).

They suggested that resistance may be explained by “weaker” attitudes of adoption.

Prior empirical research on EHR resistance focused on finding solutions to the resistance

behavior rather than testing the resistance variable and conceptualize it based on well-defined

antecedents. For example, (Hewitt and McLeod, 2011; Dinev et al, 2016; Burns et al, 2015)

argued that security and privacy concerns may be a main hurdle in the process of EHR diffusion.

(Angst and Agarwal, 2009) investigated whether individuals can be persuaded to change their

attitudes and “opt-in” behavioral intentions toward EHRs, and allow their medical information to

be digitized even in the presence of significant privacy concerns. They found that an individual’s

concerns for information privacy (CFIP) interact with argument framing and issue involvement

to affect attitudes toward the use of EHR, and that attitude toward EHR use and CFIP directly

influence opt-in behavioral intentions. Their work shows that even when people have high

concerns for privacy, their attitudes can still be positively altered with appropriate message

framing. Another important study was conducted by (Weeger et al, 2011) who focused on the

perceived risks that could lead to EHR resistance. They explained how physicians associate

several risks with adopting EHR, these risks are: performance risks, social risks, psychological

risks, and privacy risks. They argue that these risks will need to be mitigated to enable proper

user acceptance of EHR. However, the study had no exploration to other antecedents to EHR

resistance and mainly focused on the risks associated with the introduction to EHR systems.

Other studies explored the phenomenon using qualitative methods. The study conducted

by (Poon et al, 2006) argued that the limited incentives given for EHR adoption to care providers

is effecting the diffusion of the technology. Furthermore, they argue that the high cost associated

with the implementation of EHR is shaping a burden on small physician practices and in turn is

31
limiting its diffusion. (Lapointe and Rivard, 2005) conducted case studies on EHR resistance.

They studied the phenomenon on group levels over time. They found that group resistance

behaviors vary during implementation. When a system is introduced, users in a group will first

assess it in terms of the interplay between its features and individual and/or organizational-level

initial conditions. They then make projections about the consequences of its use. If expected

consequences are threatening, resistance behaviors will result.

In summary, our review uncovered interesting research gaps that need to be explored

through future research. There are limited IS research studies on EHR resistance that considered

theoretical lenses other than adoption theories. Additionally, we find that prior studies aimed to

capture resistance at later stages of the implementation of the EHR systems.

In this study we aim not only to understand resistance towards EHR systems but also to

address the research gaps we found in the literature. This study is the first to apply the

comprehensive theoretically grounded model of the URM with a focus on the early

implementation stage of a new EHR system.

INVESTIGATIVE CONTEXT

The target hospital for this study is a public governmental hospital located in the capital

city of the Jordanian kingdom, Amman. The hospital consists of multiple health departments and

33 specialists’ clinics. The total inpatient capacity is 450 beds, with a plan of adding 100 more

beds by early 2016. The hospital has recently implemented a new EHR system that is mandated

to all employees. It is one of two health organizations in the country that have completely

implemented the system. The project started in 2009, implemented in 2014, and was enforced

early 2015. No more paper records are being used at the hospital. The new EHR system is named

“Hakeem”, the Arabic word for “Doctor”. Hakeem was implemented through a governmental

32
program which is the first national e-health initiative in Jordan. The system was designed and

implemented by Electronic Health Solutions (EHS) which is a non-profit, innovative,

technology-driven company established in early 2009. EHS is a partnership between the main

healthcare stakeholders: Ministry of Health, Ministry of Information and Communications

Technology, Royal Medical Health Awareness Society and Private Hospitals Association.

According to EHS, 4.5 million JDs (around $6.4 million) were invested in specialized

resources to develop and implement Hakeem. Hakeem is a combination of an informational

portal as well as a transactional system. It serves several key functions, such as storage, retrieval

and updating of the EHR of patients cared for by all of the participating healthcare facilities in

Jordan. Implementation of Hakeem brought substantial changes to the organization in terms of

how employees perform their tasks. Employees had to switch from using the previous paper-

based system to the new EHR system Hakeem. When entering data into the system, they are

asked to complete the data entry of each form before being able to move to the next page. This is

taking some of the freedom they had with the paper based system. Additionally, the system is

only accessible from specific computers that are located within the hospital which challenges

mobile and remote working.

Understanding resistance to change to the new way of working with the EHR system

requires special attention to the unique characteristics of the healthcare context. Healthcare

organizations differ from other environments in many ways. First, healthcare providers leverage

a different level of power which makes them capable of resisting the change to a newly

implemented technology even if it was mandated by the organization. Second, physicians are

seldom employed by the hospital (Halamka, 2011) which allows them more freedom of choice

and less managerial control over their decision of using or resisting a specific technology. Third,

33
healthcare providers usually perceive the use of some technologies as unnecessary tasks that

hurdles them from performing their original job of providing care to patients, and some consider

these technologies inappropriate for the physicians to use because it requires certain tasks that

represents threat to their professional status (Lapointe and Rivard, 2005). Some healthcare

providers have also expressed concerns about EHR systems affecting their communication with

patients such as loss of eye contact, which is against the social norm for physicians to avoid

using the computer while with the patient (Cotea, 2010). Additionally it was found that

healthcare providers fear that the EHR systems will take time away from actually caring for the

patient and make the art of their job impersonal (Sassen, 2009). Finally, healthcare providers do

not believe EHR systems are useful if compared to the PBRS (Stream, 2009).

Given the uniqueness of the healthcare context, it becomes essential not to investigate

resistance to change as simple as less EHR adoption behaviors. The healthcare context makes it

different from the more general “corporate settings” in which adoption theories have been

successfully tested in prior studies, therefor when applying these theories in the healthcare

context they will have less explanatory power (Smith et al, 2014). Also, Practitioners usually use

complex and difficult medical equipment, so “hard to use” fails to explain the non-acceptance for

them (Hare et al, 2006). Moreover, these theories are designed to explain voluntary IT usage and

have limited capabilities in explaining mandated IT usage which is present in the healthcare

settings (Bhattacherjee and Hikmet, 2013). It is believed that only applying the acceptance

models, with no extensions, in the healthcare settings is not sufficient to completely explain the

individuals’ resistance to change behaviors towards new EHR systems. Thus a different

perspective is required to provide more insights regarding resistance to change to EHR systems

34
while still leveraging the technology acceptance literature in examining resistance to change, and

not define resistance as the opposite of usage or adoption (Bhattacherjee and Hikmet, 2007).

THEORETICAL BACKGROUND

The User Resistance Model

The URM was for this study for many reasons. First, it aims to explain resistance to

change associated with the implementation of a new IS system. Our theoretical development

focuses on the early implementation stage of the EHR system. In this stage the hospital is

keeping both the old paper-based record system (PBRS) and the new EHR system in place, and

employees are gradually changing from the PBRS to the EHR. In a later stage the EHR system

will be mandated to all employees and the PBRS will no longer be available for them to use. In

the context of our study, we aim to capture the resistance behaviors reported by employees when

changing from the PBRS to the EHR system. Second, it integrates different perspectives of user

resistance including the Status Quo Bias Theory (SQBT) (Samuelson and Zekhauser, 1988)

which will help explaining employees’ resistance to change due to the preference to stay with the

current status of working with the PBRS. Additionally, the URM includes the Equity

Implementation Model (EIM) (Joshi, 1991) perspective which will help in explaining

employees’ resistance to change to the EHR system based on cost-benefit analysis of the change

or the “net equity” associated with the change to the new EHR system. Third, the URM is a

comprehensive model which draws from previous literature various antecedents for technology

acceptance and resistance. Which allows us to integrate concepts from the technology acceptance

literature including Attitude, Subjective Norms, and Behavioral Control from the Theory of

Planned Behavior (TPB) (Ajzen, 1991). Fourth, prior research that tested the URM on EHR

systems were not found in the IS literature. Finally, replicating the URM in a different context

35
for the first time will help in making findings more generalizable and will contribute to the

progress of research in this area. (Nosek et al., 2015) reported that research progress relies on

both innovations which points out possible paths and replication which points out likely paths.

Thus, this work will increase certainty when findings are reproduced and promote innovation

when they are not.

The URM is a comprehensive framework which integrates multiple theories in an aim to

explain user resistance to change. Namely it integrates the Equity Implementation Model (EIM)

(Joshi, 1991), the Status Quo Bias Theory (SQBT) (Samuelson and Zekhauser, 1988), and the

Theory of Planned Behavior (TPB) (Ajzen, 1991). In this section we briefly explain each of

these theories and how do these concepts correspond to the constructs of the URM.

The Equity Implementation Model posits that individuals evaluate the change associated

with the new IS implementation based on the “net equity”. The net equity is perceived as the

difference between changes in outcomes (advantages) and changes in inputs (disadvantages)

with the new IS. The outcomes are measured as (increase in outcomes – decrease in outcomes)

and inputs are measured as (increase in inputs – decrease in inputs). Additionally, net equity is

determined after individuals compare the change in their relative outcomes with that of the

employer and that of other individuals in the reference group. If Inequity (decline in net gain) is

perceived, then individuals would be resistant to the change; the greater the inequity the greater

the resistance to change would be.

The SQBT provides a theoretically driven explanations of resistance associated with the

new IS implementation based on the evaluation of the current status of the individual and the

perceived future status of the individual after accepting the change. It posits that resistance can

be due to the preference to stay with the current situation. The SQBT is explained using three

36
categories: (1) Rational Decision Making; this includes the evaluation of costs and benefits

associated with the change. If costs were perceived to be larger than benefits then this would lead

to the status quo bias. SQBT identifies two types of costs in this category: First, Transition Costs,

and these are the costs incurred in the process of adapting to the new situation, it includes costs

that happen during the change (transient costs) and costs that result from the change (permanent

costs). Second, Uncertainty Costs, and these are explained as the perception of risk associated

with the change to the new situation. Switching to a new IS can trigger uncertainty costs on

users because they may be unsure and anxious about the resulting changes. (2) Cognitive

Misperception; this is mainly explained by the psychological concept of loss aversion. Loss

aversion is manifested when individuals weigh losses more than gains while making decisions on

value, so relatively small losses associated with the change may be perceived as larger as they

actually are (Kahneman and Tversky 1979). (3) Psychological Commitment; Samuelson and

Zeckhauser explained this category using three factors: First, sunk cost, and these refer to

previous commitments, which may cause resistance to change. For example, if the current status

requires certain skills that a person had put time and effort into mastering those skills, and the

future alternative would require complete different set of skills, then the individual would

perceive his skills as sunk costs that may affect negatively on his decision to switch to the new

status. Second, Social norms, and these refer to the perceptions of people at work about the

change. These social norms may influence the individual’s status quo bias. Third, Efforts to Feel

in Control, and these are individuals’ desires to determine and control their own situation. If an

individual feels that s/he will lose control by changing to an unknown or unfamiliar way of

working, then status quo bias would be observed.

37
 The Technology Acceptance (TA) literature was used in building the URM. The TA

theories have enriched the URM by including concepts such as social norms, control beliefs and

attitudes from the TPB. According to TPB, behavioral beliefs produce a favorable or unfavorable

attitude toward the behavior, normative beliefs result in subjective norms, and control beliefs

increases perceived behavioral controls. Because colleagues are usually the important referents

for individuals in work-related issues (Kim and Kankanhalli, 2009), the URM has incorporated

the concepts of colleague opinion which corresponds to the social norms from the SQBT.

Additionally, the perceived behavioral controls are accounted for by both external and internal

controls in URM. Self-efficacy for change and organizational support for change respectively

represent the internal and external means of achieving control of the changed situation (Ajzen

2002). Moreover, attitude is represented by perceived value in URM, which refers to the overall

evaluation of change related to a new IS implementation based on the comparison between

benefits and costs (Kahneman and Tversky 1979).

MODEL CONSTRUCTS AND HYPOTHESES

The integrative framework of URM consists 7 main constructs. In this section we provide

a definition for each of these constructs in terms of the context of our study, and how does the

EIM, SQBT and TA theories correspond to each of the model constructs (see Table 2.2).

Table 2.2: User Resistance Model Constructs Composition

URM Corresponding
Source Definition
Construct Construct
Increase in
Switching EIM The perceived utility an individual would enjoy in
Outcomes
Benefits switching from the PBRS to the new EHR system.
Decrease in Inputs. EIM
Decreased in
EIM
Outcomes
Switching Increased in Inputs EIM The perceived disutility an individual would incur in
Costs Transition Costs SQBT switching from the PBRS to the new EHR system.
Uncertainty Costs SQBT
Sunk Costs SQBT
Net Equity EIM

38
 Net Benefits SQBT
Perceived The net benefits of switching from the PBRS to the
Loss Aversion SQBT
Value new EHR system.
Attitude TA
Control SQBT Individual’s confidence in own ability to change and
Self-Efficacy
adapt with the new EHR system.
for Change Behavioral Control TA

Control SQBT
Organizational The perceived facilitation provided by the hospital to
Support for make individual’s adaptation to new EHR system
Change easier.
Behavioral Control TA

Colleague Social Norms SQBT The perception that colleagues favor the change to the
Opinion Subjective Norms TA new EHR system.
User Any act of opposition to change from the PBRS to the
User Resistance URM
Resistance EHR system.

Switching Benefits

Switching Benefits in our study refers to the benefits perceived after changing from the

PBRS to the new way of working with the EHR system. Switching benefits correspond to the

increase in outcomes (e.g., improved job productivity) and the decrease in inputs (e.g.,

performing job faster) from the EIM. Based on URM, we theorize that the switch to the new

EHR would result in benefits as enhancements in the job productivity and performances of

employees, as a result these enhancements would be associated with certain rewards, therefor it

would increase the perceived value of the change. Thus, we adapt the URM hypothesis that

higher switching benefits would increase the perceived value of change to the new system.

Switching Costs in our study refers to the costs perceived after changing from the PBRS

to the new way of working with the EHR system. Switching costs correspond to the decrease in

outcomes (e.g., less job productivity) and the increase in inputs (e.g., performing a task slower)

from the EIM. Additionally, switching costs correspond to the transition costs, uncertainty costs,

and sunk costs from the SQBT.

Switching Costs

39
 First, Transition costs. In our study these are the costs associated with the change from

the PBRS to the new EHR system. Any switch from the current status quo to a new status incurs

transition costs (Kim, 2011). When undertaking the transition to the new system, users must

acquire new skills and knowledge in order to use the new system and to perform their tasks

(Burnham et al, 2003). Users would strongly oppose the change when it requires transition costs

and have a tendency to keep their current status quo (Samuelson and Zekhauser, 1988).

Second, Uncertainty costs. These are the perceived levels of risk on users’ performances

associated with the change (Burnham et al, 2003). If users have limited knowledge about the

new system, they will be uncertain about their job performance with this system, and would

perceive this as a risk (Whitten and Wakefield, 2006), this would produce unpleasant

psychological reactions such as anxiety, which in turn would influence their status quo bias

(Inder and O’Brien, 2003).Thus, users will prefer to maintain the status quo (i.e. PBRS) and

resist switching to the new system (i.e. EHR).

Third, Sunk costs. These are the investments and costs already put into mastering the

current way of working (Kim, 2011). In our study, these are the costs in time and effort that have

already been invested by users in mastering the PBRS. The larger is the past investment, the

harder will be the change (Samuelson and Zekhauser, 1988).

Based on URM, we theorized that switching costs would negatively affect the net

benefits of the switch because net benefits are evaluated by weighing benefits against costs.

Additionally, the corresponding costs from the SQB (i.e. transition costs, uncertainty costs, and

sunk costs) would strongly influence the individual’s decision on change. Thus, we adapt the

URM hypothesis that switching costs have negative effects on perceived value and positive

effects on resistance.

40
 Perceived Value

Perceived value in our study refers to the perceived net benefits (benefits vs costs) of

switching from the PBRS to the new way of working with the EHR system. Perceived value

corresponds to the net equity and net benefits concepts from EIM and SQBT respectively.

Additionally, perceived value corresponds to the concept of loss aversion from SQBT because it

contributes explanations of how the perceived value of change is assessed (i.e. losses are viewed

larger than they actually are). Following Kim and Kankanhalli’s definition of the construct, we

define perceived value in our study as the net benefits of switching from the PBRS to the new

EHR system. Based on URM, we theorize that people tend to like maximizing value when

making a decision (Sirdeshmukh et al, 2002), if the perceived value of the change is low, then

users are likely to have greater resistance to change (i.e. switching to the new EHR system). On

the other hand, if the perceived value is high, then users are likely to have lower resistance to

switching to the new EHR system. Thus, we adapt the URM hypothesis that perceived value has

a negative effect on resistance.

Self-Efficacy for Change

Self-Efficacy for Change in or study refers to the internal means that can enhance

achieving control of the changed situation (i.e. switching to the new EHR system) (Ajzen 2002).

Self-efficacy corresponds to the control concept from the SQBT. If individuals feels that there

will be a loss in control associated with the change to the new unfamiliar way of working, then

status quo bias would be observed. This loss of control may be manifested as low self-efficacy.

Following Kim and Kankanhalli’s definition of the construct, we define self-efficacy for change

as an individual’s confidence in his or her own ability to change and adapt with the new EHR

system. Based on URM, we theorized that difficulties employees face during the change to the

41
EHR system will be viewed as threats to be avoided or skills to be mastered depending on the

level of self-efficacy the employee has (Bandura 1995). So individuals with high levels of self-

efficacy would face the change confidently and, on the other hand, individuals with low levels of

self-efficacy would be more inclined to resist the change. Additionally, individuals with high

levels of self-efficacy for change would be less likely to experience anxiety and uncertainty

regarding the change (Bandura 1995; Compeau et al. 1999), which implies less switching costs

(i.e. uncertainty and transition costs). Thus, we adapt both of URM hypotheses: First, that self-

efficacy for change has negative effect on user resistance, and second, that self-efficacy for

change has negative effect on switching costs.

Organizational Support for Change

Organizational Support for Change, like the Self-Efficacy for Change, it also corresponds

to the control concept from the SQBT. Following Kim and Kankanhalli’s definition of the

construct, we define organizational support for change as the perceived facilitation provided by

the hospital to make individual’s change to the new EHR system easier. Based on URM we

theorize that organizational support mechanisms such as training and providing resources could

affect individual’s reactions towards change to the new EHR system (Hirschheim and Newman

1988). Additionally, these mechanisms would minimized the perceived difficulty of adapting to

the new EHR system (Lewis et al. 2003), and in turn would reduce costs of time and effort

required to learn the new way of working. Thus, we adapt both of URM hypotheses: First, that

organizational support for change has a negative effect on user resistance, and second

organizational support for change has a negative effect on switching costs.

Colleague Opinion

42
 This construct corresponds to social norms from SQBT. Kim and Kankanhalli’s

definition of the construct limited social influence to only that coming from colleagues and

peers. In our study we included peers, colleagues and superiors as important others. Thus, we

used named this construct as “Social Influence” rather than “Colleague Opinion”. We define

Social Influence as the degree to which an individual perceives that important others believe he

or she should use the new EHR system (Venkatesh et al, 2003). We are including both types of

influence: (1) informational influence which comes mostly from colleagues, and (2) normative

influence which comes mostly from superiors (Burnkrant and Cousineau 1975). Building on

URM, we theorize that individuals have the tendency to conform to their important others’

opinions because of the need for social companionship and the fear of sanction for

noncompliance (Ajzen 2002; Lewis et al. 2003). So individual’s resistance behaviors are directly

affected by what other people at work think about the change to the new EHR system.

Additionally important others’ opinion about the new EHR system may alter the original

perceptions an individual has about switching costs and benefits (e.g. informational influence -

Bunkrant and Cousineau, 1975). Positive opinions toward the new change can serve to reduce

users’ uncertainty and therefore lower their perceptions of switching costs. Also, positive

opinions about the change to the new EHR would lead to a greater perception of switching

benefits among users. Thus, we adapt both of URM hypotheses that positive opinions (positive

social influence) about the change to the EHR has a negative effect on both user resistance and

switching costs, and has a positive effect on switching benefits.

User Resistance

The outcome variable of the URM, User Resistance, was constructed based on the frame

work of resistance behaviors (Bovey and Hede, 2001) which distinguishes between overt

43
resistance (physical behavior) and covert resistance (psychological mid-set), and between active

and passive resistance. Resistance can occur on a spectrum from covert passive (e.g., ignoring) to

overt active (e.g., sabotage) behaviors (see Figure 2.1). Based on Kim and Kankanhalli’s

definition of the construct, we define user resistance as any act of opposition to change from the

PBRS to the EHR system.

Figure 2.1: Different Types of User Resistance Behavior

Control Variables

In addition to assessing key variables of URM, we included several demographic control

variables. First, because it is argued that employees with different positions would perceive the

change differently. While physicians consider data entry into EHR is a “clerical task” (Lapointe

and Rivard, 2005), other employees may enjoy and appreciate the change. Thus, we included

position at hospital as a control variable. Second, because social influence may have effect on

user resistance it was appropriate to control for the number of employees in a unit. Third, we

controlled for all of age, gender, and level of education to assess how individual differences may

affect individuals’ resistance behaviors.

44
 Table 2.3 summarizes the research model hypothesis and Figure 2.2 illustrates the posited

structural model of URM (Kim and Kankanhalli, 2009) that will be tested in the context of this

study in aim to explain resistance towards EHR.

Table 2.3: URM Hypotheses

No. Hypothesis

H1 Switching Benefits has a positive effect on Perceived Value.

H2 Switching Costs has a negative effect on Perceived Value.

H3 Switching Costs has a positive effect on User Resistance.

H4 Perceived Value has a negative effect on User Resistance.

H5 Self-Efficacy for Change has a negative effect on Switching Costs.

H6 Self-Efficacy for Change has a negative effect on User Resistance.

H7 Organizational Support has a negative effect on Switching Costs.

H8 Organizational Support has a negative effect on User Resistance.

H9 Social Influence has a negative effect on Switching Costs.

H10 Social Influence has appositive effect on Switching Benefits.

H11 Social Influence has a negative effect on User Resistance.

45
 Figure 2.2: URM (Kim and Kankanhalli, 2009)

METHODOLOGY

Instrument Development

Existing validated scales were adopted for this study. We mainly adopted the instrument

items used for the URM’s original study. However all items were modified to fit the context of

our study. Social Influence included an additional item (SOI4) to capture superiors’ influence.

Also unlike URM, Perceived Value included Loss Aversion items in the survey (LAV1-6) these

were adopted from (Gächter et al, 2007). Measurement items were anchored on five-point Likert

scales (1 = strongly disagree, 5 = strongly agree). The instrument was reviewed by IS researchers

before the study. The final version of the questionnaire was developed for the main study as

shown in the Appendix (Table 2.4).

Sample and Data Collection

46
 The total estimated number of employees who work at the hospital is 5,500. Majority of

these employees are potential users of the new system. However administrative staff are less

likely to use the system as they do not participate in the care giving process, however they still

have access to the system and would be able to print out reports as needed. The sample is a

combination of all stakeholders of the system (i.e. physicians, nurses, lab specialists, radiologists,

pharmacists, admin staff, and resident medical students). Data was collected in 2 phases. In the

first phase, we collected data for the pilot study from employees who works only for the

Dermatology unit at the hospital, we collected a total of (126) responses. After conducting

Explanatory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) of the data

collected for the pilot test, it was necessary to alter some of the items to address a number of

issues related to the psychometric properties of factors. Additionally, some new items were

added to better capture the concepts of some constructs. First, SWC1 had a low loading of

(0.451). The item is developed to capture sunk costs which refers to the investments and costs

already put into mastering the current way of working (Kim, 2011). Thus, it was necessary to

reword the item to reflect the costs healthcare providers have invested in the previous PBRS, so

we added “I already have” at the beginning of the survey question. The loading of the item went

up to (0.702) after collecting data for the main study. Second, SFC1 also had a low loading of

(0.595). The item was reworded to match the same fashion as other factors of the same construct.

This resulted in an increase in its loading in the main study to become (0.761). Additionally, we

added item SFC4 into the main study survey to capture the levels of confidence healthcare

providers have in their ability to change to the new EHR system. The new item had a loading of

(0.740). Third, we included the fourth item (OGS4) to capture healthcare providers’ opinions on

the levels of assistants provided by the hospital. The new item had a loading of (0.833). Fourth,

47
because our study aims to capture social influence sourcing from all possible employees at the

hospital, it was necessary to include item SOI4 which captures social influence of superiors on

the healthcare providers. This new item had a loading of (0.846). Finally, we included the

concepts of loss aversion within the perceived value construct. We also added a fourth item

PVL4 which captured healthcare providers’ perceptions of value after considering the limitations

of Hakeem. Loss aversion had a negative loading, in the main study, and no statistical

significance (p = 0.11 > 0.05), so Loss Aversion item was dropped from the scale. However the

new item PVL4 had a loading of (0.976). After refining the survey items based on the pilot

data’s psychometric properties, we collected data for the main study from the entire hospital, but

employees who filled the first survey (i.e. pilot study) were specifically asked not to fill in the

survey for the second time. We collected a total of (226) responses, descriptive statistics are

reported in (Table 2.5).

Table 2.5: Descriptive Statistics of Respondents

Demographic Variables Data
Gender Male 133 (58.80%)
Female 93 (41.20%)
Age <30 56 (24.80%)
(Mean = 33.02, S.D. = 5.19) 31 - 40 87 (38.50%)
41 - 50 61 (27.00%)
51 - 60 22 (9.70%)
>60 0 (0.00%)
Position Physician 57 (25.20%)
Nurse 70 (31.00%)
Radiologist 23 (10.20%)
Lab Specialist 32 (14.20%)
Pharmacist 15 (6.60%)
Admin Staff 20 (8.80%)
Medical Student 0 (0.00%)
Total 226 (100%)

Instrument Validation

To validate the survey instrument, the psychometric properties of the survey were

assessed by conducting Explanatory Factor Analysis (EFA) and Confirmatory Factor Analysis

48
(CFA) using MPlus 7.1 (Muthén and Muthén, 1998-2006). Cronbach’s α was performed using

SPSS 20 (IBM Corp. 2013). After the pilot data was collected, we refined the items to keep those

with statistical significant loading larger of (0.7). Cronbach’s α reliability tests for all constructs

exceeded (0.8). After the main data was collected, we again conducted CFA analysis and again

all items had a significant loading greater than (0.7) except for the dichotomous factor indicator

Loss Aversion within Perceived Value. It had a negative loading with no statistical significance

(p = 0.11 > 0.05), so Loss Aversion items was dropped from the scale. All constructs had

Cronbach’s α values exceeding (0.8). The CFA analysis provided strong support for our

measurement model, which suggested that the items under each of the constructs were

adequately measuring the constructs. Table 2.6 in the Appendix lists the items’ loadings and the

model fit statistics for the CFA.

The survey responses came in (2) waves: (W1: N=164; W2: N=62). We checked for

nonresponse bias by testing the difference in means between waves (see Table 2.7). We found no

significant difference between the two respondent groups based on the sample attributes (gender,

age, and position).

Table 2.7: Nonresponse Bias Test

Sig
First Wave (N=164) Second Wave (N=62) t-Value (p<0.05)
df = 224
Age Mean = 38.0 Mean = 36.7 0.923 Ns
Gender Male = 97 (59.1%) Male = 36 (58.1%)
0.146 Ns
Female = 67 (40.9%) Female = 26 (41.9)
Position Physician = 40 (24.39%) Physician = 17 (27.41%)
Nurse = 50 (30.48%) Nurse = 20 (32.25%)
Pharmacist = 17 (10.36%) Pharmacist = 6 (9.67%)
Admin Staff = 26 (15.83%) Admin Staff = 6 (9.67%) 0.239 Ns
Radiologist = 11 (6.70%) Radiologist = 4 (6.45%)
Lab Specialist = 14 (8.53%) Lab Specialist = 6 (9.67%)
Medical Student = 6 (3.65%) Medical Student = 3 (4.83%)

RESULTS

49
 The original research model (URM) was tested by applying Structural Equation

Modeling (SEM) using Mplus version 7.1 (Muthén and Muthén, 1998-2006). We applied the

maximum likelihood estimator with robust standard errors (MLR). Because the URM is not

saturated (i.e., not all possible regression paths were included) we evaluated the model fit

indicators depicted in Table 2.8.

Table 2.8: URM Model Fit Indices

Indicator Value
X2 86.149
df 22
P-Value 0.000
Calculated NC 3.915
CFI 0.882
RMSEA 0.114
90 CI 0.089 – 0.139
SRMR 0.069

Following (Bollen, 1998) suggestions on evaluating Chi-Sqaure (X2), we calculated the

Normed X2 (NC=3.91). It is suggested that NC value between (2.0 and 6.0) indicates reasonable

fit. However, we notice that (CFI = 0.882), this indicates illness in the model fit. The rule of

thumb for the CFI and other incremental indices is that values greater than roughly (.90) may

indicate reasonably good fit of the researcher’s model (Hu and Bentler, 1999). Additionally, the

Root Mean Square Error of Approximation (RMSEA) is (0.114). According to the rules of

thumb by (Browne and Cudeck, 1993) RMSEA value between (0.05) and (0.08) suggest

reasonable error of approximation and values greater than (0.1) suggests poor fit. In order to

improve the goodness-of-fit of the model we conducted modification indices test using MPlus.

This suggests possible paths to be added to the model to improve its goodness-of-fit.

50
 The suggested model and its new improved model fit indices are included in the

discussion section of this paper. The standardized path coefficients, its level of significance, and

the standard errors (S.E) are provided in (Table 2.9), and results are depicted in (Figure 2.3).

Table 2.9: Standardized Path Coefficients

Hypothesized Relationships STD β (S.E.)
SWB – PVL 0.858 (0.029)***
SWC – PVL -0.069 (0.040)
SFC – SWC -0.170 (0.067)*
OGS – SWC 0.245 (0.067)***
SOI – SWC -0.172 (0.072)*
SOI – SWB 0.439 (0.050)***
SWC – RES 0.519 (0.063)***
PVL – RES -0.348 (0.056)***
SFC – RES -0.033 (0.053)
OGS – RES 0.052 (0.067)
SOI – RES -0.056 (0.060)
*p<0.05; **p<0.01; ***p<0.001

51
 (*p<0.05; **p<0.01; ***p<0.001)

Figure 2.3: URM Hypotheses Testing Results

After comparing our findings with the hypotheses of the original theory we posit mixed

results (i.e. some hypothesis were confirmed and others were not). Our study found that the

positive effect switching benefits has on perceived value is confirmed. On the other hand

switching costs effect on perceived value was not confirmed but switching cost direct effect on

resistance was confirmed. The positive effect of Social Influence on switching benefits was

confirmed, as well as its negative affect on switching costs. However social influence had no

significant direct effect on resistance, so the original hypothesis was not confirmed. Similarly,

52
self-efficacy for change and organizational support had no direct effect on resistance, and both

hypotheses were not confirmed. The negative effect of Self-efficacy for change on switching cost

was confirmed. Finally, organization support had a positive direct effect on switching costs. This

contradicts the original hypothesis that suggested a negative relation between both constructs.

None of the control variables had significant effect on resistance. Perceived value and switching

costs effects on user resistance explained (48.6%) of its variance. Switching benefits explained

(75.1%) of the variance in perceived value. Social influence explained (19.2%) of the variance in

switching benefits. Social influence, organizational support, and self-efficacy for change effects

on switching costs explained (9.2%) of its variance.

DISCUSSION OF FINDINGS

Our findings show that most of posited hypotheses were supported. This indicates that the

URM has the potential of being generalized and could successfully explain resistance to change

in different contexts. However, we still found that few hypotheses were not supported in addition

to a number of interesting issues surrounding the resistance to change to a new EHR system. In

this secession we provide a discussion of these findings. First, organizational support had a

positive direct effect on switching costs which in turn has a positive direct effect on resistance.

This means the more healthcare providers perceive the hospital being supportive to the change to

Hakeem the more they perceive the switch to Hakeem as being costly in terms of their transition,

uncertainty, and sunk costs. This opposes the original theory hypothesis that suggested

organizational support would have negative influence on switching costs. Healthcare

organizations are aware of a number of compelling issues surrounding EHR systems that keeps

them motivated to provide support to their EHR users: that EHR systems consist of complex

software and hardware that requires certain skills in order to be applied in the workplace

53
(Boonstra and Broekhuis, 2010), that one of the most important barriers to EMR adoption is a

lack of computer proficiency (Najaftorkaman et al, 2015), and that some healthcare providers are

still interested in documenting their work using PBRS instead of using EHR systems. If these

providers cannot embrace EHR in their workplaces, the resistance of EHR systems can rapidly

increase in healthcare organizations (Pinaire, 2009). Therefore, healthcare organizations tend to

offer training sessions to their healthcare providers who are expected to interact with the EHR

system, as well as provide technical support to users as needed. However, healthcare providers at

the hospital may be perceiving increased support from the hospital as an indication that the

transition is difficult and requires special support and this may result in an increase in the

perceived transition costs. Furthermore, hospital support may be in the form of training sessions

which means there will be new skills to be learned and new knowledge to be mastered, and this

may increase levels of sunk costs. Moreover, hospital support for the change may be unclear on

how would this effect healthcare providers’ productivities and performances, which may lead to

an increase in their uncertainty costs. Thus, we found that hospital support had a positive effect

on switching costs. This positive effect of hospital support on switching costs is an interesting

finding especially that switching costs is a composite construct and the effect may be directed to

one or more of the types of costs. We encourage future research to further explore this finding.

Second, switching costs had no effect on perceived value. However, our findings confirm

the direct effect of switching costs on resistance. In other words, healthcare providers are more

influenced to resist the change to Hakeem when higher switching costs are perceived, however

their perception of costs has no influence on their perception of the value of the switch. So one

with high perceptions of switching costs may be influenced to resist the switch without affecting

his perceptions of the switching value. This may be different if loss aversion items were included

54
in the perceived value scale items. Loss aversion suggests that when evaluating a situation

people will value losses more than gains. This is especially true in the healthcare context where

losses could be directly aimed towards the health of patients. When healthcare providers are

delivering care with a suggested success rate of the treatment they are expected to value the

failure of the procedure heavily and direct their focus on avoiding it. We believe that, as a matter

of automatic behavior, healthcare providers will weigh costs much more than they weigh benefits

in any given evaluation case. Therefore, the loss aversion concept can be used to help explain

this finding. Additionally, our finding could be explained by the concept of perceived

consequences which refers to the cognitive evaluation of the probable consequences of the

behavior (Gagnon et al., 2010). The work of (Najaftorkaman et al, 2015) suggested that

healthcare providers’ perceived consequences of using EHR systems have a significant impact

on their adoption behaviors of the EHR. Thus, when healthcare providers know about the various

costs and consequences of using EHR systems they begin to resist EHR systems. This perception

of consequences is found to be stronger and independent from their beliefs of the value

associated with the EHR system.

Third, self-efficacy for change has no direct impact on user resistance. Rather, there is an

indirect effect mediated through switching costs. This result indicates that self-efficacy for

change decreases resistance by reducing the perception of switching costs. We believe that this

finding could be explained partly by the fact that we have captured self-efficacy to change to the

new way of working with Hakeem rather than self-efficacy of actually using Hakeem. Healthcare

providers may be confident in their ability to adapt to the transition process of the hospitals in

terms of redirecting the use of resources, work processes, budget allocations, or any other modes

of operation that will reshape the hospital, but not confident in their ability to use the new

55
technology on their own. It has been shown in the literature that computer literacy has a negative

impact on EHR resistance and healthcare providers with less computer skills are less likely to

adopt the systems (Stream, 2009; Terry et al, 2009). Using EHR systems could be a completely

new experience for some healthcare providers in their workplace. Some EHR users are not

confident about their ability of using it (Simon et al, 2007a; Terry et al, 2009). Therefore, we call

on future research to investigate the impact of self-efficacy of using the EHR technology on

resistance behaviors.

Fourth, social influence has no direct effect on resistance, but there is an indirect effect

mediated through switching costs. This indicates that social influence decreases resistance by

reducing the perception of switching costs. This could be partly explained by the fact that we

included items to capture influence from superiors. Some healthcare providers may be influenced

by the fear of punishment if they resist the enforced change while their superiors are supportive

of it. Further, different opinions may be perceived from different sources. Thus, the superiors

influence, when included with other type of influences (i.e. from colleagues and peers), may

have contributed to the result of having no direct relation between social influence and

resistance.

Fifth, organizational support has no direct effects on resistance. This is related to our

previous argument on hospital support. Healthcare organizations need to be mindful of users’

needs and must choose the proper method of support carefully. Our findings indicates that the

type of support chosen by the hospital is increasing healthcare providers’ perceptions of

Hakeem’s costs and is not impacting on their resistance behaviors. We encourage future research

to further explore what type of support was used by the hospital in this study and why did

healthcare providers respond negatively to hospital support. Additionally, future research may

56
aim to provide a comparative analysis between different types of support provided by healthcare

organizations and how do healthcare providers respond to each.

Finally, our findings indicated problems with the goodness-of-fit of the tested model. We

conducted a model indices test in MPlus to find alternative models that are aligned with the

URM theory and has an improved fit. We found the best model fit achieved by adding the

following three paths: (1) direct effect of switching benefits on resistance, (2) self-efficacy on

switching benefits, and (3) organizational support on switching benefits. Additionally, the model

dropped all 3 direct effects of social influence, organizational support, and self-efficacy for

change on user resistance. The new suggested model is depicted in (Figure 2.4) and the new

model indices are presented in (Table 2.10).

(*p<0.05; **p<0.01; ***p<0.001)

Figure 2.4: Suggested Model Results

Table 2.10: Suggested Model Fit Indices

Indicator Value
X2 52.330
df 22
P-Value 0.0003

57
 Calculated NC 2.378
CFI 0.944
RMSEA 0.078
90 CI 0.051 – 0.106
SRMR 0.046

The suggested model had no notable change in the amount of variance explained, except

for switching benefits (28.9%) which increased from (19.2%). However the suggested model had

a better fit. NC was at (2.3), CFI increased to (0.944) which is in the recommended range (>0.9),

and RMSEA became lower (0.078) also in the reasonable range (0.05 – 0.10). The only new

significant path is the positive effect of self-efficacy for change on switching benefits. The more

healthcare providers feel confident in their skills and abilities to master the new way of working

with the EHR system the more they will perceive the change as beneficial. This falls along the

URM theory which hypothesized the negative effect self-efficacy for change has on switching

costs.

CONTRIBUTIONS

Contribution to Research

Our study makes several contributions to IS research. First, we explained resistance to

change to the new way of working using EHR systems based on net-benefit evaluation which

influences how healthcare providers would perceive the value of the switch to the new EHR

system. Resistance to change to the new EHR system is effected directly by healthcare provider’

perceived value of the EHR system and their perceived costs of switching to the new EHR

system. These findings contributes to the IS research which still requires advancements in

understanding resistance behaviors in healthcare settings. Second, we found that the URM, if

appropriately modified, could be adapted to explain resistance to change to new IS in different

contexts. For this study we tested the model in the healthcare context for the first time and found

58
that URM’s basic solid foundations are also applicable in such stetting. Third, we found a direct

effect, which was not reported in the original theory, between self-efficacy for change and

switching benefits. Fourth, our study reveals a counterintuitive finding; the positive direct effect

organizational support has on switching costs. This interesting finding should motivate future

work to better investigate different possibilities surrounding this phenomenon. Fifth, we included

superiors’ influence in our scale when capturing social influence. This addition confirms the

effect of normative influence as well as informative influence on healthcare providers when

evaluating the net-befit of switching to the new EHR system. Sixth, our study was conducted at

early stages of implementation, this allowed us to capture resistance to change as it is happening.

Finally, our findings show that there is no difference in the way healthcare providers are

influenced to resist the change to the new EHR system based on different demographics such as

their age, gender, position at the hospital, or levels of education.

Contributions to Practice

Our study provides a number of implications to practice. First, we provide possible

solutions to hospitals and other health organizations when trying to reduce resistance behaviors

towards the change to new EHR systems. Switching cost has both a direct and indirect effect on

resistance. Thus, reducing it is critical when aiming to eliminate resistance behaviors. This could

be achieved by enhancing positive opinions about the change both from colleagues and

superiors. Also, it is very crucial to understand what type of support is more suitable for the

healthcare providers. Our findings show that hospital support has a positive effect on switching

costs. This may be due to inappropriate delivery of the support (e.g. long training sessions for

healthcare providers after hours). Additionally, investing in healthcare providers’ development to

increase their self-efficacy would have direct impacts on their perception of the switch in terms

59
of its costs vs benefits. Also, it is important to emphasize the advantages of the EHR system and

to explain the core benefits of the switch at early stages before the implementation of the EHR

system, this would help in increasing switching benefits and perceived value which in turn would

reduce resistance behaviors. Second, because this study is conducted in a developing country,

which is still at the first stages of digitizing healthcare organizations, it becomes very important

to understand what predicts resistance to change to new EHR systems and what do new users

consider as benefits or costs to the switch. Answers to these questions would help system

developers and system designers build improved EHR systems that overcomes sources of costs.

Third, understanding EHR resistance would help healthcare providers get the appropriate support

to reduce resistance motivations. Governments in many countries have largely invested in the

transition to EHR systems (Khalifa, 2013). Thus, being able to explain causes of HIT resistance

would help in finding solutions to make these systems more adopted and with higher success

rates. Finally, our work shows that URM could be applied successfully in different contexts.

Thus, the theory could be replicated using different technologies across industries. Findings

would help practitioners better understand resistance in the context of which it was tested in, and

that would serve as the first step towards solving problems associated with resistance behaviors.

LIMITATIONS AND FUTURE DIRECTIONS

This study has its limitations. First, in order to fully understand the positive effect

between organizational support and switching costs, further investigations are required. It is

interesting to study how different type of support provided by healthcare organizations may

influence cost perceptions of EHR systems differently. Second, because our survey data is cross-

sectional and collected from a single source, common method bias may be a concern (Xue et al,

2014). Future research can apply a longitudinal study or collect data from multiple sources to

60
validate the findings. Finally, although we have collected loss aversion in perceived value’s

scale, the items had poor loadings and thus were dropped from the survey. Future studies could

attempt to empirically validate these items.

61
 REFERENCES

Ajzen, I. 2002. “Constructing a TPB questionnaire: Conceptual and methodological

considerations.”
Ajzen, I., 1991. “The theory of planned behavior,” Organizational Behavior and Human
Decision Processes, 50, pp. 179–211.
Alghamdi, A. S., 2015. “Factors associated with the implementation and adoption of electronic
health records (EHRs) in Saudi Arabia,” (Doctoral dissertation, Rutgers University-
School of Health Related Professions).
Angst, C. M., & Agarwal, R. 2009. “Adoption of electronic health records in the presence of
privacy concerns: The elaboration likelihood model and individual persuasion,” MIS
quarterly, 33(2), 339-370.
Bandura, A. 1995. “Exercise of Personal and Collective Efficacy in Changing Societies,” Self-
Efficacy in Changing Societies, New York: Cambridge University Press, pp. 1-45.
Bhattacherjee, A, Davis C., and Hikmet N., 2013. "Physician Reactions to Healthcare IT: An
Activity-Theoretic Analysis," 46th Hawaii International Conference on system science,
IEEE.
Bhattacherjee, A., & Hikmet, N., 2007. “Physicians' resistance toward healthcare information
technology: a theoretical model and empirical test,” European Journal of Information
Systems, 16(6), 725-737.
Blumenthal, D., & Tavenner, M. 2010. “The Meaningful Use regulation for electronic health
records,” New England Journal of Medicine, 363(6), 501-504.
Bollen, K. A., 1998. "Structural equation models," Encyclopedia of biostatistics.
Boonstra, A., & Broekhuis, M. 2010. “Barriers to the acceptance of electronic medical records
by physicians from systematic review to taxonomy and interventions,” BMC health
services research, 10(1), 1.
Bovey, W. H., and Hede, A. 2001. “Resistance to Organizational Change: The Role of Defense
Mechanisms,” Journal of Managerial Psychology, (16:7), pp. 534-548.
Browne, M. W., and Cudeck, R. "Alternative ways of assessing model fit," Sage Focus Editions,
154 (1993): 136-136.
Burnham, T., Frels J., and Mahajan V., 2003. “Consumer switching costs: A typology,
antecedents, and consequences,” Journal of Academic Marketing Science, vol. 31, no. 2,
pp. 109–203.
Burnkrant, R. E., and Cousineau, A. 1975. “Informational and Normative Social Influence in
Buyer Behavior,” Journal of Consumer Research, (2:3), pp. 206-214.

62
Burns, A. J., Young, J., Roberts, T. L., Courtney, J. F., & Ellis, T. S., 2015. “Exploring the Role
of Contextual Integrity in Electronic Medical Record (EMR) System Workaround
Decisions: An Information Security and Privacy Perspective,” AIS Transactions on
Human-Computer Interaction, 7(3), 142-165.
Burt C.W., Sisk J.E., 2005. “Which Physicians and Practices Are Using Electronic Medical
Records?” Health Affairs, 24(5):1334-1343.
Cantrill, M., and Stephen, V., 2010. “Computers in Patient Care: The Promise and the
Challenge,” ACM Queue, (8:8).
Compeau, D., Higgins, C. A., and Huff, S. 1999. “Social Cognitive Theory and Individual
Reactions to Computing Technology: A Longitudinal Study,” MIS Quarterly (23:2), pp.
145-158.
Cotea, C., 2010. “Electronic health record adoption: Perceived barriers and facilitators,” Centre
for Military and Veterans' Health, The University of Queensland.
Davis, F. D., 1989. “Perceived Usefulness, Perceived Ease of Use, and User Acceptance of
Information Technology,” MIS Quarterly, 13, 3, pp. 319-340.
DesRoches, C. M., Campbell, E. G., Rao, S. R., Donelan, K., Ferris, T. G., Jha, A., &
Blumenthal, D., 2008. “Electronic health records in ambulatory care—a national survey
of physicians,” New England Journal of Medicine, 359(1), 50-60.
Dinev, T., Albano, V., Xu, H., D’Atri, A., & Hart, P., 2016. “Individuals’ Attitudes Towards
Electronic Health Records: A Privacy Calculus Perspective,” Advances in Healthcare
Informatics and Analytics, (pp. 19-50), Springer International Publishing.
Earnest M.A., Ross S.E., Wittevrongel L., Moore L.A., Lin C.T., 2004. “Use of a Patient-
Accessible Electronic Medical Record in a Practice for Congestive Heart Failure: Patient
and Physician Experiences,” Journal of the American Medical Informatics Association,
11(5):410 417.
Electronic Health Solutions (EHS) http://www.ehs.com.jo/en
Fu, K. 2009. “Viewpoints: Inside Risks Reducing Risks of Implantable Medical Devices,”
Communications of the ACM (52:6), pp. 25-27.
Gächter, S., Johnson, E. J., & Herrmann, A., 2007. “Individual-level loss aversion in riskless and
risky choices.”
Gagnon, M.-P., Ouimet, M., Godin, G., Rousseau, M., Labrecque, M., Leduc, Y., 2010. “Study
protocol multi-level analysis of electronic health record adoption by health care
professionals: A study protocol,” Implementation Science, 30(5), 1–10.
Gan, Q., & Cao, Q., 2014. “Adoption of Electronic Health Record System: Multiple Theoretical
Perspectives,” System Sciences (HICSS), 47th Hawaii International Conference on
System Science, (pp. 2716-2724), IEEE.

63
Halamka, J., 2011 “Healthcare is different”, http://geekdoctor.blogspot.com/2011/08/healthcare-
is-different.html
Hare, K., Whitworth, B., Deek, F., and Norris, T., 2006. “Identifying Barriers to IT Adoption in
Clinical Health Care: A Pilot Study”, Health Care and Informatics, Review Online, ISSN
1176-4201, http://hcro.enigma.co.nz/website/index.cfm
Heath, M., & Appan, R., 2014. “Impact of Information Technology Culture Conflict on
Meaningful Use of Electronic Health Records System: A Conceptual Framework.”
Hewitt, B., & McLeod, A., 2011. “Modeling Security in Acceptance of Electronic Health Record
Systems,” Journal of Information Privacy and Security, 7(3), 23-45.
Hillestad, R., Bigelow, J., Bower, A., Girosi, F., Meili, R., Scoville, R., & Taylor, R. 2005. “Can
electronic medical record systems transform health care? Potential health benefits,
savings, and costs,” Health Affairs, 24(5), 1103-1117.
Hirschheim, R., and Newman, M., 1988. “Information Systems and User Resistance: Theory and
Practice,” Computer Journal, 31, 5, pp. 398-408.
Hu, L.T. and Bentler, P.M., 1999. "Cutoff Criteria for Fit Indexes in Covariance Structure
Analysis: Conventional Criteria versus New Alternatives," Structural Equation
Modeling, 6 (1), 1-55.
Hung, S. Y., Yu, W. J., Tsai, M. F., & Yen, D. C., 2013. “What Drives Physicians' Intention to
Use Electronic Medical Record System: The Roles of Perceived Service Level, Computer
Self-Efficacy, and Perceived Risk,” PACIS, (p. 194).
IBM Corp. Released 2013. IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM
Corp.
Inder, B., and O’Brien, T., 2003. “The endowment effect and the role of uncertainty,” Bull.
Econ. Res., vol. 55, no. 3, pp. 289–301.
IT-toolbox ERP Implementation Survey, 2004. www.vvsgidas.lt/get.php?id=403
Jha, A. K., DesRoches, C. M., Campbell, E. G., Donelan, K., Rao, S. R., Ferris, T. G., &
Blumenthal, D., 2009. “Use of electronic health records in US hospitals,” New England
Journal of Medicine, 360(16), 1628-1638.
Joshi, K., 1991. “A Model of User Perspective on Change: The Case of Information Systems
Technology Implementation,” MIS Quarterly, 15, 2, (1991), pp. 229-242.
Kahneman, D., & Tversky, A., 1979. “Prospect theory: An analysis of decision under risk,”
Journal of the Econometric Society, 263-291.
Keen, P.G., 1981. "Information Systems and Organizational Change," Communications of the
ACM, 24, 1, pp. 24-33.

64
Kemper A.R., Uren R.L., Clark S.J., 2006. “Adoption of Electronic Health Records in Primary
Care Pediatric Practices,” Pediatrics, 118(1):20-24.
Khalifa, M., 2013. “Barriers to health information systems and electronic medical records
implementation. A field study of Saudi Arabian hospitals,” Procedia Computer Science,
21, 335-342.
Kim, H. W., 2011. “The effects of switching costs on user resistance to enterprise systems
implementation,” Engineering Management, IEEE Transactions on, 58(3), 471-482.
Kim, Hee-Woong, and Atreyi Kankanhalli, 2009. "Investigating user resistance to information
systems implementation: a status quo bias perspective,” MIS quarterly, 33, 3.
Laerum H, Ellingsen G, Faxvaag A., 2001. “Doctors’ Use of Electronic Medical Records
Systems in Hospitals: Cross Sectional Survey,” British Medical Journal,
323(7325):1344-1348.
Lapointe, L., and Rivard, S. 2005. .“A Multiple Model of Resistance to Information Technology
Implementation,” MIS Quarterly, (29:3), pp. 461-491.
Lewis, W., Agarwal, R., and Sambamurthy, V., 2003. “Sources of Influence on Beliefs about
Information Technology Use: An Empirical Study of Knowledge Workers,” MIS
Quarterly, (27:4), pp. 657-678.
Loomis GA, Ries S, Saywell RM, Thakker NR., 2002. “If Electronic Medical Records Are So
Great, Why Aren’t Family Physicians Using them?” Journal of Family Practice,
51(7):636-641.
Ludwick DA, Doucette J., 2009. “Primary Care Physicians’ Experience with Electronic Medical
Records: Barriers to Implementation in a Fee-for-Service Environment,” International
Journal of Telemedicine and Applications, 853524.
Meade B, Buckley D, Boland M., 2009. “What Factors Affect the Use of Electronic Patient
Records by Irish GPs?” International Journal of Medical Informatics, 78(8):551-558.
Menachemi N, Langley A, Brooks RG., 2007. “The Use of Information Technologies Among
Rural and Urban Physicians in Florida,” Journal of Medical Systems, 31(6):483–488.
Miller RH, Sim I., 2004. “Physicians’ Use of Electronic Medical Records: Barriers and
Solutions,” Health Affairs, 23(2):116-126.
MPLUS (Version 6.11). [Computer Software]. Los Angeles, CA: Muthén & Muthén
Najaftorkaman, M., Ghapanchi, A. H., Talaei‐Khoei, A., & Ray, P., 2015. “A taxonomy of
antecedents to user adoption of health information systems: A synthesis of thirty years of
research,” Journal of the Association for Information Science and Technology, 66(3),
576-598.
Ngafeeson, M. N., and Midha, V., 2014. "An exploratory study of user resistance in healthcare
IT," International Journal of Electronic Finance, 8, no. 1, 74-94.
65
Nöltner, M., & Krönung, J. 2015, “Bolstering Attitudes: An Introduction to the Foundations of
Attitude Strength and Their Consideration in IS Adoption Research.”
Nosek, B. A., Alter, G., Banks, G. C., Borsboom, D., Bowman, S. D., Breckler, S. J., ... &
Contestabile, M., 2015. “Promoting an open research culture: Author guidelines for
journals could help to promote transparency, openness, and reproducibility,” Science
(New York, NY), 348(6242), 1422.
Olaniran, B. A., 2015. “Contextual factors and challenges to e-health literacy,” Knowledge
Management & E-Learning: An International Journal (KM&EL), 7(4), 616-628.
Pinaire, K. (2009). “An overview of barriers to the adoption of electronic medical records,”
MWAIS Proceedings, 14.
Pizziferri L, Kittler AF, Volk LA, Honour MM, Gupta S, Wang S, Wang T, Lippincott M, Li Q,
Bates DW., 2005. “Primary Care Physician Time Utilization Before and After
Implementation of an Electronic Health Record: A time motion Study,” Journal of
Biomedical Informatics, 38(3):176-188.
Poon, E., Blumenthal, D., Jaggi, T., Honour, M., Bates, D.W., and Kaushal, R., 2004.
“Overcoming barriers to adopting and implementing computerized physician order entry
systems in U.S. hospitals,” Health Affairs, 23, 4, pp. 184–190.
Poon, Eric G., Jennifer L. Cina, William Churchill, Nirali Patel, Erica Featherstone, Jeffrey M.
Rothschild, Carol A. Keohane, Anthony D. Whittemore, David W. Bates, and Tejal K.
Gandhi, 2006. "Medication dispensing errors and potential adverse drug events before
and after implementing bar code technology in the pharmacy," Annals of Internal
Medicine, 145, no. 6, 426-434.
President’s Information Technology Advisory Committee, 2004. Revolutionizing Health Care
through Information Technology, National Coordination Office for Information
Technology Research and Development, Arlington, Virginia, http://www.nitrd.gov
Randeree E., 2007. “Exploring Physician Adoption of EMRs: A Multi-Case Analysis,” Journal
of Medical System, 31(6):489-496.
Reardon JL, & Davidson E., 2007. “An Organizational Learning Perspective on the Assimilation
of Electronic Medical Records among Small Physician Practices,” European Journal of
Information Systems, 16:681-694.
Samhan, B., & Joshi, K. D., 2015. “Resistance of Healthcare Information Technologies;
Literature Review, Analysis, and Gaps,” System Sciences (HICSS), 2015 48th Hawaii
International Conference on (pp. 2992-3001). IEEE.
Samuelson, W., & Zeckhauser, R., 1988. “Status quo bias in decision making,” Journal of risk
and uncertainty, 1(1), 7-59. Chicago.
Sassen, E.J., 2009. “Love, hate, or indifference: How nurses really feel about the electronic
health record system,” Computers, Informatics, Nursing, 27(5), 281–287.

66
Shea, S., & Hripcsak, G., 2010. “Accelerating the use of electronic health records in physician
practices,” New England Journal of Medicine, 362(3), 192-195.
Sheikh, A., Cornford, T., Barber, N., Avery, A., Takian, A., Lichtner, V. & Morrison, Z., 2011.
“Implementation and adoption of nationwide electronic health records in secondary care
in England: final qualitative results from prospective national evaluation in “early
adopter” hospitals,” BMJ, 343, d6054.
Simon SR, Kalshal R, Cleary PD, Jenter CA, Volk LA, Oray EJ, Burdick E, Poon EG, Bates
DW., 2007b. “Physicians and Electronic Health Records: A Statewide Survey,” Archives
of Internal Medicine, 167(5):507-512.
Sirdeshmukh, D., Singh, J., and Sabol, B., 2002. “Consumer Trust, Value and Loyalty in
Relational Exchange,” Journal of Marketing, (66:1), pp. 15-37.
Smith, Tim, Gerald Grant, and Alejandro Ramirez, 2014. "Investigating the influence of
psychological ownership and resistance on usage intention among physicians,” System
Sciences (HICSS), 47th Hawaii International Conference on. IEEE.
Stone, C., 2014. “A Glimpse at EHR Implementation Around the World: The Lessons the US
Can Learn,” The Health Institute for E-Health Policy, http://www.e-
healthpolicy.org/docs/A_Glimpse_at_EHR_Implementation_Around_the_World1_Chris
Stone.pdf
Stream, G.R, 2009. “Trends in adoption of electronic health records by family physicians in
Washington State,” Informatics in Primary Care, 17(3), 145–152.
Terry AL, Thorpe CF, Giles GG, Brown JB, Harris SB, Reid GJ, Thind A, Stewart M, 2008.
“Implementing Electronic Health Records,” Canadian Family Physician, 54:730-736.
Terry, A.L., Giles, G., Brown, J.B., Thind, A., & Stewart, M., 2009. “Adoption of electronic
medical records in family practice: The providers’ perspective,” Family Medicine, 41(7),
508.
Terry, K., 2012. “EHR Adoption: U.S. Remains The Slow Poke,” Information Week. Retrieved
from http://www.informationweek.com/healthcare/electronic-healthrecords/ehr-adoption-
us-remains-the-slow-poke/d/d-id/1107410
Valdes I, Kibbe DC, Tolleson G, Kunik ME, Petersen LA., 2004. “Barriers to Proliferation of
Electronic Medical Records,” Informatics in Primary Care, 12(1):3-9.
Venkatesh, V., M.G. Morris, G.B. Davis, and F.D. Davis, 2003. “User Acceptance of
Information Technology: Toward A Unifying View,” MIS Quarterly, 27, 3, pp. 425-478.
Vishwanath A, Scamurra SD., 2007. “Barriers to the Adoption of Electronic Health Records:
Using Concept Mapping to Develop a Comprehensive Empirical Model,” Health
Informatics Journal, 13(2):119-134.

67
Walter, Z., and Lopez, M. S., 2008. “Physician Acceptance of Information Technologies: Role of
Perceived Threat to Professional Autonomy,” Decision Support Systems (46:1), pp. 206-
215.
Weeger, A., Gewald, H., & Vriesman, L. J., 2011. “Do Risk Perceptions Influence Physician's
Resistance to Use Electronic Medical Records? An Exploratory Research in German
Hospitals,” AMCIS.
Whitten, D., and Wakefield R., 2006. “Measuring switching costs in IT outsourcing services,”
Journal of Strategic Information Systems, vol. 15, pp. 219–248.
Xue, Z., Gebremichael, M., Ahmad, R., Weldu, M. L., & Bagtzoglou, A. C., 2011. “Impact of
temperature and precipitation on propagation of intestinal schistosomiasis in an irrigated
region in Ethiopia: suitability of satellite datasets,” Tropical Medicine & International
Health, 16(9), 1104-1111.

68
 APPENDIX

Measurement Items

Table 2.4: Measurement Items

Construct Item Wording
Changing to the new way of working with Hakeem enhances my effectiveness
SWB1
on the job more than working using the previous paper-based system.
Changing to the new way of working with Hakeem enables me to accomplish
SWB2 relevant tasks more quickly than working using the previous paper-based
Switching system.
Benefits Changing to the new way of working with Hakeem increases my productivity
SWB3 more than working with the previous paper-based system.
Changing to the new way of working with Hakeem improves the quality of the
SWB4
work I do more than working with the previous Paper based system.
Considering the time and effort that I have to spend, the change to the new
PVL1
way of working with Hakeem is worthwhile.
Considering the loss that I incur, the change to the new way of working with
PVL2
Hakeem is of good value.
Considering the hassle that I have to experience, the change to the new way of
PVL3
working with Hakeem is beneficial to me.
Considering Hakeem’s limitations, the change to the new way of working with
PVL4
Hakeem is advantageous.
Consider you are asked to participate in a game with the following rules:
You start with 100 points. When the game ends, the higher the points you have
the bigger the prize you get. You are asked to fill in the following before you
LAV
flip a coin.
Perceived You can gain or lose points as described in the table. Please indicate for each
Value choice which would you accept and which would you reject:
If the coin turns up heads, then you lose 2 points; if it turns up tails, you win 6
LAV1
points
If the coin turns up heads, then you lose 3 points; if it turns up tails, you win 6
LAV2
points
If the coin turns up heads, then you lose 4 points; if it turns up tails, you win 6
LAV3
points
If the coin turns up heads, then you lose 5 points; if it turns up tails, you win 6
LAV4
points
If the coin turns up heads, then you lose 6 points; if it turns up tails, you win 6
LAV5
points
If the coin turns up heads, then you lose 7 points; if it turns up tails, you win 6
LAV6
points
I already have put a lot of time and effort into mastering the previous paper-
SWC1
based system.
It took a lot of time and effort to switch to the new way of working with
SWC2
Switching Hakeem.
Costs Switching to the new way of working with Hakeem resulted in unexpected
SWC3
hassles.
I lost a lot in my work after switching to the new way of working with
SWC4
Hakeem.
Social Most of my colleagues think the change to the new way of working with
SOI1
Influence Hakeem is a good idea.

69
 My peers are supportive of the change to the new way of working with
SOI2
Hakeem.
Most people whom I deal with in my job encourage my change to the new
SOI3
way of working with Hakeem.
Most of my superiors think that the change to the new way of working with
SOI4
Hakeem is a good idea.
I was able to change to the new way of working with Hakeem easily based on
SFC1
my own knowledge, skills and abilities.
I was able to change to the new way of working with Hakeem without the help
SFC2
of others.
Self-Efficacy
I was able to change to the new way of working with Hakeem reasonably well
for Change SFC3 on my own.
I had confidence in my ability to change to the new way of working with
SFC4 Hakeem without any difficulties.
My hospital provided me with guidance on how to change to the new way of
OGS1
working with Hakeem.
The management at my hospital provided the necessary help and resources to
OGS2
Organizational enable me to change to the new way of working with Hakeem.
Support for I was given the necessary support and assistance to change to the new way of
Change OGS3 working with Hakeem by my hospital.
The assistance provided by my hospital made the change to the new way of
OGS4 working with Hakeem easier.
I would rather not comply with the change to the new way of working with
RES1
Hakeem.
User I would rather not cooperate with the change to the new way of working with
RES2
Resistance Hakeem.
RES3 I oppose the change to the new way of working with Hakeem.
RES4 I do not agree with the change to the new way of working with Hakeem.

70
 Instrument Validation

Table 2.6: Items Loadings and CFA Model Fit Statistics

Construct Item Std. Mean STD Cronbach’s α CFA Model Fit
Loading Statistics
SWB1 0.935 X2: 6.605
df: 2
SWB2 0.946 P < 0.05
SWB 3.341 1.201 0.976 CFI: 0.983
RMSEA: 0.101
SWB3 0.987 CI: 0.022 - 0.191
SRMR: 0.007
SWB4 0.947

SWC1 0.702 X2: 11.513

df: 2
SWC2 0.859 P < 0.01
SWC 3.301 0.868 0.896 CFI: 0.961
SWC3 0.848 RMSEA: 0.145
CI: 0.072 - 0.231
SWC4 0.898 SRMR: 0.029

PVL1 0.953 X2: 8.881

df: 2
PVL2 0.969 P < 0.05
PVL 3.239 1.090 0.981 CFI: 0.971
PVL3 0.956 RMSEA: 0.059
CI: 0.000 - 0.120
PVL4 0.976 SRMR: 0.181

SFC1 0.761 X2: 36.673

df: 2
SFC2 0.888 P < 0.001
SFC 3.447 0.889 0.910 CFI: 0.904
SFC3 0.977 RMSEA: 0.227
CI: 0.203 - 0.359
SFC4 0.740 SRMR: 0.048

SOI1 0.925 X2: 15.969

df: 2
SOI2 0.927 P < 0.001
SOI 3.642 0.848 0.945 CFI: 0.957
SOI3 0.902 RMSEA: 0.176
CI: 0.103 – 0.206
SOI4 0.846 SRMR: 0.023

OGS1 0.851 X2: 9.820

df: 2
OGS2 0.956 P < 0.05
OGS 3.602 0.755 0.943 CFI: 1.00
OGS3 0.958 RMSEA: 0.000
CI: 0.000 - 0.107
OGS4 0.833 SRMR: 0.004

71
 RES1 0.968 X2: 8.420
df: 2
RES2 0.992 P < 0.05
RES 2.938 1.129 0.982 CFI: 1.00
RES3 0.985 RMSEA: 0.000
CI: 0.000 - 0.102
RES4 0.919 SRMR: 0.001

72
 CHAPTER THREE:

UNDERSTANDING ELECTRONIC HEALTH RECORDS RESISTANCE: A

REVEALED CAUSAL MAPPING APPROACH

ABSTRACT

The implementation of Electronic Health Records (EHR) has potential benefits to

healthcare organizations. However, evidence show that EHR systems are being resisted by

healthcare providers. Little attention in IS research has been paid to understand this

phenomenon. In this study we aim to reveal the main antecedents of EHR resistance and the

concepts shaping these constructs. We use Revealed Causal Mapping (RCM) methodology to

analyze data gathered from interviewing healthcare providers who are introduced to a newly

implemented EHR system in a large international hospital. We argue that better understanding

of EHR resistance will contribute to both practice and research. Therefore, in this study, we

explore the key determinants that drive users’ resistance to EHR system.

73
 CHAPTER THREE:

UNDERSTANDING ELECTRONIC HEALTH RECORDS RESISTANCE: A

REVEALED CAUSAL MAPPING APPROACH

INTRODUCTION

Healthcare organizations are in the process of changing the way healthcare providers

document, monitor, and share information about healthcare delivery from using Paper-Based

Recording Systems (PBRS) to using Electronic Health Records (EHR) (DesRoches et al, 2013).

Using EHR is expected to reduce medical errors, improve continuity and quality of care, and

facilitate availability of patients’ medical history (Hillestad et al, 2005). However, despite the

potential benefits EHR could provide, there are evidence that suggests EHR is being resisted by

the same people who would benefit from its use. In 2013, only 44 percent of hospitals in the U.S.

reported having and using EHR systems (DesRoches et al, 2013). This phenomenon is also

taking place around the globe. Case studies conducted in all of the U.S., U.K., France, and India

show that healthcare organizations in all of those countries shared similar EHR implementation

challenges (Stone, 2014). Resistance behaviors towards technology has been identified as one of

the salient reasons for the failure of new systems (Kim and Kankanhalli, 2009). This includes

EHR and other types of Health Information Technology (HIT). One example comes from the

Cedars-Sinai Medical Center at Los Angeles, where doctors resisted to use the newly

implemented Computerized Physician Order Entry (CPOE) system, which caused the system to

fail and result in a complete withdrawal after it was already implemented in two-thirds of the

870-bed hospital (Bhattacherjee et al., 2008).

Despite the importance of understanding EHR resistance phenomenon to be able to find

solutions, IS research in this area is found to be limited (Samhan and Joshi, 2015; Lapointe and

74
Rivard, 2005). In this paper, motivated by the urgency of understanding HER resistance coupled

by the limited prior efforts at explaining EHR resistance, we explore the following research

question: What are the key determinants that drive users’ resistance to EHR system. Our goal is

to go beyond the typical findings available in the literature and provide more insights by

conducting a qualitative analysis. We believe that while quantitative analysis could explain and

predicts EHR resistance behaviors, a qualitative analysis could provide deeper and richer insights

about individuals’ experiences with the new EHR system that help shape their resistance

behavior. We asked healthcare providers, working a hospital located in Amman-Jordan, open-

ended questions to capture their perceptions of a newly implemented EHR system. To analyze

their responses, we used Revealed Causal Mapping (RCM), a qualitative methodology

commonly used to identify constructs and linkages revealed from respondents’ statements

(Nelson et al., 2000). RCM has proven to be useful in studying emerging phenomena that require

rich and contextualized understanding (Narayanan and Armstrong, 2005). Given that our goal is

to understand the perceptions of individuals interacting directly with the newly implemented

EHR system, we believe that the RCM method is appropriate for our investigation of these

perceptions.

LITERATURE REVIEW ON TECHNOLOGY RESISTANCE

Some early IS research was conducted to uncover the concept of technology resistance

(e.g., Hirschheim and Newman, 1988; Keen, 1981; Markus, 1983). However, recent reviews of

the literature (Samhan, and Joshi, 2015; Lapointe and Rivard, 2005) show that limited work has

been conducted to explain resistance behaviors especially within the healthcare context. Prior

work in IS aimed to explain the resistance phenomenon based on the concept of perceived

threats. Perceived threats are identified by expressions such as "overwhelming emotional pain"

75
(Freud 1919) or "the perception of a dangerous situation" (Marakas and Hornik 1996). IS

research suggested that new technology will be resisted when people believe it will result in a

threat situation such as loss in individual’s status, revenue, or power. Perceived threat was

explained in (Lapointe and Rivard, 2005) case study of physician resistance. They posited that

when a system is introduced, users will first assess it and then make predictions about the

consequences of its use. If results were threatening, resistance behaviors will occur. Similarly,

(Bhattacherjee and Hikmet, 2007) tested healthcare providers’ behaviors towards a new

Computerized Physician Order Entry system (COPE) and considered perceived threats to be an

important agent that helps determine individuals’ attitude towards the new technology. They

suggested that higher levels of perceived threat would result in greater resistance towards the

new system. The work of (Joshi, 1991) viewed perceived threats as the inequity which leads

people to resist the implementation of a new system. The work of (Dent and Goldberg, 1999)

suggests that when people resist change they are in fact reacting to the threats associated with

that change. Another stream of research explained technology resistance by applying resistance

to change theories (e.g. Kim and Kankanhalli, 2009; Joshi, 1991; Samuelson and Zekhauser,

1988). It is suggested that resistance behaviors could be predicted by the evaluation of the net

benefits of using the new technology. Antecedents of this theoretical perspective deal with the

outcomes of changing from the current way of working to the new way after implementing the

new technology (e.g. gaining more benefits), while others focus on the costs associated with the

change. Some other work explained EHR resistance by lower levels of adoption (e.g. Boonstra

and Broekhuis, 2010; Gan and Cao, 2014; Najaftorkaman et al, 2015; Jha et al, 2008). However,

in our work we aim to conceptualize resistance distinctively from the use construct for a number

of reasons. First, the unique characteristics of the healthcare context makes it different from the

76
general corporate environment in which the techchnology use theories were applied (Smith et al.,

2014). Thus, we suggest that applied those theories will have limited explanation power when

aiming to explain EHR resistance. Second, prior technology use theories explained voluntarily

use behaviors, whereas in our study the EHR system is mandated to all healthcare providers

which requires a different perspective to carefully explain resistance behaviors (Bhattacherjee

and Hikmet, 2013). Finally, it is found that resistance and use behaviors may coexist (Lapointe

and Rivard, 2005). For example, resistance may be manifested in the form of indirect usage or

delayed usage. This suggests that both use and resistance constructs are distinctive. In summary,

the review of prior IS literature revealed that prior findings on technology resistance are mixed

and that research on EHR resistance are limited. Additionally, we found that prior studies had

limited explanation of the concepts shaping the antecedents of resistance. For example, prior

findings suggested that perceived threat is a main antecedent to technology resistance, however

limited explanation was given to what users may have considered to be a threat. In this paper we

aim to provide a qualitative exploration to the concepts and constructs serving as determinants

that drive healthcare providers to resist new EHR systems.

THEORY IN USE

Our study is broadly informed by the theoretical foundations of the User Resistance

Model (URM) (Kim and Kankanhalli, 2009). The URM explains user resistance to the

technology by using the resistance to change concepts. It provides an integrative framework

combing three main theories: (1) the Status Quo Bias Theory (SQBT) (Samuelson and

Zeckhauser, 1988), (2) the Equity Implementation Model (EIM) (Joshi, 1991), and (3) the

Theory of Planned Behavior (TPB) (Ajzen, 1991, 2002). (Figure 3.1) shows how constructs from

the 3 theories correspond to each of the URM constructs. We draw upon the semantics of the key

77
constructs introduced in URM to help us reveal the main concepts that influences resistance

behaviors towards the EHR system. The URM suggests that user resistance is predicted by

perceived value which is determined by evaluating the costs and benefits associated with the

change to the newly implemented technology. It also includes concepts from the technology

acceptance literature including Self-Efficacy for change, Organizational Support, and Colleague

Opinion.

Figure 3.1: The Integrative Framework of URM (Kim and Kankanhalli, 2009)

According to URM, individuals begin to evaluate the new technology based on the

concept of net-benefits (Kim and Kankanhalli, 2009; Joshi, 1991; Samuelson and Zeckhauser,

1988). This evaluation would be mainly based on the concepts of switching benefits and

switching costs. Switching benefits correspond to the increase in outcomes (e.g., improved job

productivity) and the decrease in inputs (e.g., performing job faster) from the EIM. Based on
78
URM, it is suggested that the switch to the new EHR may result in benefits as enhancements in

the job productivity and performances of employees, as a result these enhancements would be

associated with certain rewards, therefor it would increase the perceived value of the change and

decrease the perceived threat of the change. The second agent in the evaluation phase is

switching costs. Switching costs correspond to the decrease in outcomes (e.g., less job

productivity) and the increase in inputs (e.g., performing a task slower) from the EIM.

Additionally, switching costs correspond to the transition costs, uncertainty costs, and sunk costs

from the SQBT. Transition costs in our study are the costs associated with the change from the

PBRS to the new EHR system. Any switch from the current status quo to a new status incurs

transition costs (Kim, 2011). When undertaking the transition to the new system, users must

acquire new skills and knowledge in order to use the new system and to perform their tasks

(Burnham et al, 2003). Users would strongly oppose the change when it requires transition costs

and have a tendency to keep their current status quo (Samuelson and Zekhauser, 1988).

Uncertainty costs are the perceived levels of risk on users’ performances associated with the

change (Burnham et al, 2003). If users have limited knowledge about the new system, they will

be uncertain about their job performance with this system, and would perceive this as a risk

(Whitten and Wakefield, 2006), this would produce unpleasant psychological reactions such as

anxiety, which in turn would influence their status quo bias (Inder and O’Brien, 2003). Sunk

costs are the investments and costs already put into mastering the current way of working (Kim,

2011). In our study, these are the costs in time and effort that have already been invested by users

in mastering the PBRS. URM suggests that switching costs would negatively affect the net

benefits of the switch because net benefits are evaluated by weighing benefits against costs.

Additionally, the corresponding costs from the SQBT (i.e. transition costs, uncertainty costs, and

79
sunk costs) would strongly influence the individual’s decision on change. Thus, switching costs

have negative effects on perceived value and positive effects on perceived threats. Which implies

influencing individuals’ attitudes towards the change.

Perceived value corresponds to the net equity and net benefits concepts from EIM and

SQBT respectively. Additionally, perceived value corresponds to the concept of loss aversion

from SQBT because it contributes explanations of how the perceived value of change is assessed

(i.e. losses are viewed larger than they actually are). URM suggests that people tend to like

maximizing value when making a decision (Sirdeshmukh et al, 2002), if the perceived value of

the change is low, then users are likely to have greater resistance to change (i.e. switching to the

new EHR system). On the other hand, if the perceived value is high, then users are likely to have

lower resistance to switching to the new EHR system. Thus, perceived value would have

negative effects on resistance behaviors.

Self-efficacy for change refers to the internal means that can enhance achieving control

of the changed situation (i.e. switching to the new EHR system) (Ajzen 2002). Self-efficacy

corresponds to the control concept from the psychological commitment category of the SQBT. If

individuals feels that there will be a loss in control associated with the change to the new

unfamiliar way of working, then status quo bias would be observed. This loss of control may be

manifested as low self-efficacy. URM suggests that different levels of self-efficacy for change

would result in different levels of anxiety and uncertainty regarding the change (Bandura 1995;

Compeau et al. 1999), which implies differences in their evaluation of the change. Individuals

with greater self-efficacy would evaluate the change as beneficial, on the other hand, individuals

with lower self-efficacy would evaluate the change costly.

80
 The organizational support for change also corresponds to the control concept of the

psychological commitment category from the SQBT. URM suggests that organizational support

mechanisms such as training and providing resources would minimized the perceived difficulty

of adapting to the new EHR system (Lewis et al. 2003), and in turn would reduce costs of time

and effort required to learn the new way of working. Thus, organizational support for change

would influence individuals’ evaluations of the change. The more support provided by the

hospital, the more healthcare providers would perceive the change as beneficial, and vice-versa.

Social influence corresponds to the social norms concept from the psychological

commitment category of the SQBT. It is the degree to which an individual perceives that

important others believe he or she should use the new EHR system (Venkatesh et al, 2003).

URM has limited social influence sources to colleagues and peers but not superiors. URM,

suggests that individuals have the tendency to conform to their important others’ opinions

because of the need for social companionship and the fear of sanction for noncompliance (Ajzen

2002; Lewis et al. 2003). Important others’ opinion about the new EHR system may alter the

original perceptions an individual has about switching costs and benefits. Thus, social influence

would impact individuals’ evaluation of the change.

RESEARCH METHODOLOGY

For this study we used the Revealed Causal Mapping (RCM) method, a sub-category of

cognitive mapping that involves an inductive process of revealing constructs and linkages

between constructs, all from the respondents’ statements (Narayanan and Armstrong, 2005).

RCM is used commonly in the management and organization science field, as reflected in the

special issues on the RCM method and its application in the Journal of Management Studies

(1987) and Organization Science (1996). In IS research, (Nelson et al., 2000) applied RCM to

81
identify the skills and expertise requirements in IS maintenance, while (Deng and Chi, 2012)

used this method to reveal system usage problems and causes in organizational use of business

intelligence systems. We intend to investigate what makes people oppose the change from the

PBRS and resist the use of the new EHR system, rather than to hypothesize or test cause-and-

effect relationships. Thus, we believe that the RCM method is appropriate for identifying key

constructs and linkages in this complex and emerging phenomenon. Informed by prior

theorizations, we seek to develop an analytical generalization of the constructs influencing

resistance and the main properties of these constructs.

Participants and Procedure

The sample used for this study is a combination of all stakeholders of the system (i.e.

physicians, nurses, lab specialists, radiologists, pharmacists, admin staff, and resident medical

students). We collected responses through open ended questions. Participants received a link to

the questions on their email, and their answers were retrieved automatically after submission via

a research software. This was repeated twice, one for a pilot study and one for the main study.

For the pilot study we targeted only employees who work at the dermatology unit at the hospital.

We collected data for the main study from the entire hospital, but employees who participated in

the pilot study were specifically asked not to answer the questions for the second time (i.e. the

main study). We collected a total of (59) responses. Descriptive statistics of all (59) Interviewees

are reported in (Table 3.1).

Table 3.1: Descriptive Statistics of Respondents

Demographic Variables Data
Gender Male 34
Female 25
Age <30 17
(Mean = 41.06, S.D. = 7.16) 31 – 40 23
41 – 50 12
51 – 60 7

82
 >60 0
Position Physician 20
Nurse 18
Radiologist 4
Lab Specialist 5
Pharmacist 4
Admin Staff 7
Medical Student 1
Total 59 (100%)

Our sample size is similar to that in prior IS studies using the RCM method (Deng and

Chi, 2012; Nelson et al., 2000). Studies based on the RCM method use the point-of-redundancy

concept to calculate the adequacy of sampling (Narayanan and Armstrong, 2005). As the point of

redundancy is not known until RCMs are constructed, the number of required respondents is a

judgment call. We adopted a sample size consistent with (Nelson et al., 2000), in which 50

experts in the IS maintenance context were interviewed. The point of redundancy in our data

analysis (as explained below) suggested that our sample was sufficient for exhausting the

concepts influencing resistance behavior.

Data Analysis

Our qualitative analysis method aims at extracting causal relationships from a

respondent’s explicit statements. We used a four-step process: (1) data elicitation, (2)

construction of revealed causal maps, (3) validation of the maps, and (4) interpretation of the

maps.

Step 1: Data Elicitation; we conducted the open-ended questions using Qualtrics. Table

3.2 in the Appendix includes the questions used for this study. Fifty-nine respondents typed their

responses online to the open-ended questions. In applying the RCM method, researchers often

rely on two major types of data: text-based data and interview data (Narayanan and Armstrong,

2005). As our research objective was to reveal the cognitive structure pertaining to the resistance

83
behavior domain, it was appropriate to use open-ended questions, similar to an interview

approach (Nelson et al., 2000), and to analyze the narratives of individual participants for their

experiences, perceptions, and beliefs with regard to the new EHR system.

Step 2: Construction of Revealed Causal Maps; following (Nelson et al., 2000), we

derived revealed causal maps of perceptions step by step. First, we identified causal statements

from an informant’s narrative by looking for key words such as “because,” “therefore,” “so,”

“while,” and “if–then.” Then, we coded each causal statement into cause and effect and replaced

the “cause” and “effect” in a statement with the concepts (the key words or phrases) and linked

“cause” and “effect” with an arrowed line to obtain a causal map at the concept level. We

highlighted repeated words and phrases and grouped them into constructs. Finally, we

aggregated all concept-level maps across all informants to obtain a causal map at the construct

level, Table 3.3 shows an example of the described process. At the end a total of 23 concepts and

375 linkages were identified.

Table 3.3: An Illustration of the Five–Step Procedure for Constructing Causal Maps
Step 1: Identifying causal statement
Step 2: Constructing raw causal maps
Step 3: Coding
Example of causal statement:
“I find Hakeem threating to my job because it takes too much
time to fill data so I cannot serve more patients…. We need more
training and practice on how to use Hakeem faster.”
Cause:
1. It takes too much time to fill in data.
2. We need more training on how to use it faster.
Effect:
I find Hakeem threating to my job.
Raw Phase (Coded Concept)
1. Takes much time to fill in data (Wastage of Time)
2. Needs more training and practice (lack of training)
3. Threating to my job (Perceived Threat)

84
 Step 4: Recasting raw causal maps
into concept-level revealed causal
maps

Step 5: Creating a construct-level

revealed causal map

Step 3: Validation; we reviewed the resistance literature to validate the concepts and

constructs identified in the RCMs. Similar to (Nelson et al., 2000), we used the point of

redundancy to evaluate the convergence of concepts elicited from the responses (Axelrod, 1976).

In our analysis, the 28 concepts converged at the 26th respondent--meaning that no new concept

was revealed in the map of the 27th respondent or beyond-- indicating the sufficiency of our

sample size. The additional relations yielded by adding the revealed causal map of each

individual respondent is measured by plotting a curve as shown in Figure 3.2.

85
 Figure 3.2: Point of Redundancy

Step 4: Interpretation of the Maps; we examined the key constructs and linkages

uncovered in the map (see Figure 3.3). The constructs and their associations demonstrate what

influences people to resist the new EHR system.

RESULTS

Our data analysis revealed nine constructs that are essential to understanding resistance

towards the EHR System (EHRS). Four are causes: EHRS Benefits, Organizational Support,

EHRS Self-Efficacy, and Facilitating Conditions; and four are major outcomes: EHRS Costs,

EHRS Perceived Value, EHRS Perceived Threats, and EHRS Resistance. Our data analysis

revealed new constructs that were not considered in prior research on technology resistance (e.g.,

facilitating conditions). It also revealed the main concepts associated with each revealed

construct. (Table 3.4) lists the constructs and concepts identified in the study with the percentage

of each of these concepts occurrence over all of the other concepts within the same construct.

Table 3.4: Revealed Constructs of EHRS Resistance and Its Concepts

Construct Definition Concepts Definition %
EHRS Self- Individuals’ Computer Basic, nontechnical 0.22
Efficacy perceptions of their Literacy knowledge about EHRS
ability to use EHRS to and how to use it.

86
 accomplish their tasks. Familiarity The feeling of comfort 0.39
(Compeau and with knowledge of the
Higgins, 1995) EHRS.
Required The need for resources 0.39
Training and services dedicated
to helping educate
healthcare providers on
EHRS-related topics.
Organizational The perceived Technical Services provided by 0.11
Support facilitation provided Support the hospital to help
by the hospital to make healthcare providers
individuals’ adaptation with and advice about
to EHRS easier. (Kim the EHRS.
and Kankanhalli, Training Any course of 0.89
2009) Sessions instruction on EHRS.
This may be delivered
in class, via a software
product installed on a
single computer,
through hospital’s
intranet, or over the
Internet as Web-based
training.
EHRS Costs The perceived Time Perceived waste in time 0.25
disutility an individual Consuming when using the EHRS.
would incur with the Technical Any technical problem 0.21
use of EHRS. Difficulties healthcare providers
(Derived from the encounter when using
study) the EHRS
Design Perceived costs 0.31
Difficulties associated with the
interface, design, and/or
the enforced process of
using the EHRS forms.
Access The limited possibility 0.23
Limitations of accessing the EHRS
remotely, from any
device, or at any time.
EHRS Benefits The perceived utility Design Includes the look and 0.34
an individual would Advantages feel of the interface, the
enjoy with the use of level of data
EHRS. (Derived from organization, and ease
the study) of use.
Data Data continues to be 0.39
Availability available at a required
level of performance.
Fast The response rate of the 0.20
Processing EHRS in terms of
loading forms and
storing data.
Accuracy Retrieving medical 0.07
information that is free
of mistakes and errors.

87
EHRS The overall evaluation Valuable The EHRS is of great (+)
Perceived of EHRS based on the (Worthy) worth to healthcare 0.49
Value comparison between providers. (-)
benefits and costs 0.51
(Kahneman and
Tversky 1979).
EHRS The extent to which an Lower Low effectiveness of 0.38
Perceived individual perceives Productivity productive effort.
Threat EHRS as dangerous. Evaluated in terms of
(Liang and Xue, the rate of
2009) accomplishing tasks per
unit of input (i.e.
patients).
Lower Low accomplishments 0.41
Performances of tasks or functions.
Potential Loss The probability of 0.21
of Data losing medical data
because of technical
errors, cyber-attacks, or
any other difficulty
when interacting with
the EHRS.
Facilitating The objective factors Alternative The ability to
Conditions that individuals User accomplish tasks by 0.17
perceive as facilitating having someone else
their ability to avoid interact with the EHRS.
using EHRS. Alternative The ability to 0.83
(Thompson et al., System accomplish tasks by
1991) using a method or
system other than the
EHRS.
EHRS Any act of opposition No Usage The absence of any 0.12
Resistance intended to prevent the EHRS usage behavior.
usage of EHRS Partial Usage Using part of the EHRS 0.53
(Lapointe and functionality and
Rivard, 2005; resisting other parts.
Markus, 1983) For example: using the
EHRS to retrieve data
and resist to enter the
data in the EHRS
forms.
Delayed Using the EHRS after a 0.18
Usage period of time by which
tasks become late or
postponed.
Indirect Usage Using the EHRS via an 0.17
alternative system or
person.

EHRS Self-Efficacy

88
 When employees are introduced to a newly implemented system they normally manifest

different levels of confidence in their own ability to utilize the system in doing their jobs

(Venkatesh et al., 2003). Our study revealed that employees at the hospital first create judgments

of their abilities to interact with Hakeem to accomplish their tasks. Then use these judgements as

key factors that influences their perception of Hakeem’s value. Three main dimensions of self-

efficacy were revealed: Computer Literacy, Familiarity, and Required Training. Table 3.5

presents examples from respondents for each concept.

Table 3.5: EHRS Self-Efficacy Revealed Key Concepts

Concept Quote Respondent

Computer Literacy “…what I really do not like about Hakeem (Female, 36, Nurse, Dermatology)

is that it needs a lot of skills in computing

that most of us do not have”

Familiarity “…at the beginning I thought it [Hakeem] (Male, 43, Doctor, Anesthesiology)

was from a different planet…it took me

awhile to get used to it”

Required Training “The system needs a lot of training to know (Male, 28, Lab Specialist, Human

everything about it…I practiced a lot to be Medical Lab Testing)

able to cope with it.”

Organizational Support

The organizational support theory suggests that organizations provide support to

employees as an exchange of their positive actions (Eisenberger et al., 1986). Thus, employees

tend to perform better to reciprocate received rewards and favorable treatment (Eisenberger et

89
al., 1986). However, our data analysis shows that employees refer to organizational support as

the facilitations provided by the hospital to make employees utilize Hakeem in an easy and more

comfortable fashion, regardless of employees’ performance. Our data analysis also revealed that

employees’ perception on Hakeem’s value is influenced by the levels of support provided by the

hospital. Two main dimensions of organizational support were revealed: Technical Support and

Training Sessions. Table 3.6 provides examples from respondents on each concept.

Table 3.6: EHRS Organizational Support Revealed Key Concepts

Concept Quote Respondent

Technical Support “We have an on-call facility that provides (Male, 41, Office Manager,

help whenever we need it. They can Dermatology Unit)

connect us to the main office if they do not

know how to fix the problem. I usually use

it when I cannot find a form outside of my

daily routine use”

Training Sessions “At the beginning we were taught by (Female, 33, Nurse, General

Hakeem academy which give us training Medicine)

sessions to learn how to use the system”

EHRS Costs

Prior research has conceptualized costs as those being associated with the switch or the

change to the new way of working with the newly implemented technology (Kim and

Kankanhalli, 2009; and Kim, 2011). However, in this study, the responces revealed that costs are

associated with the technology itself rather than the costs associated with the change. Four types

of costs were revealed from responses: Time Consuming, Technical Difficulties, Design

90
Difficulties, and Access Limitations. Table 3.7 provides examples from respondents on each

concept.

Table 3.7: EHRS Costs Revealed Key Concepts

Concept Quote Respondent

Time Consuming “…it takes a lot of time to fill-in all data (Female, 27, Nurse, Dermatology)

fields, some of it is not necessary… I spend

more time in front of the computer now”

Technical “I do not like it when it is not working (Male, 33, Lab Specialist, Lab)

Difficulties properly. The page [Form] freezes

sometimes and I want to get the

information quickly to process the case

[taking samples from patients]…If the

system goes down I cannot read the orders

nor write the results”

Design Difficulties “The system’s design is not very (Male, 38, Nurse, Dermatology)

nice…sometimes I find difficulties finding a

selection from a menu so I must call

technical support and they will get back to

me within 24 hours”

Access Limitation “…I cannot access the system from my (Male, 33, Lab Specialist, Lab)

phone or iPad or any other device, if I

want to see what are tomorrows orders I

have to go to the hospital and login from

my computer”

EHRS Benefits

91
 The URM (Kim and Kankanhalli, 2009) was the first to utilize the concept of net-benefits

when theorizing technology resistance behaviors. This was adapted from the EIM (Joshi, 1991)

and the SQBT (Samuelson and Zekhauser, 1988). URM suggested that users determine how

valuable the new system would be based on evaluating the switching benefits against the

switching costs. However in this study we revealed benefits being associated with the technology

itself rather than the changing benefits. These benefits are: Design Advantages, Data

Availability, Processing Speed, and Accuracy of completing the task. Table 3.8 provides an

example from respondents on each concept.

Table 3.8: EHRS Benefits Revealed Key Concepts

Concept Quote Respondent

Design Advantages “What I really like about Hakeem is that (Female, 43, Material Management

information input is done in simple easy Director, Environmental Services)

steps and it takes only few clicks to finalize

any task”

Data Availability “I really like it because it gives (Male, 32, Nurse, ER)

information about everything the patients

did at the hospital or at any other hospital

that uses Hakeem. If an unconscious

patients come to the ER we can track all of

his data without talking to him”

Processing Speed “Hakeem changed the way of (Male, 42, Doctor, ENT)

working…now I like it because it increase

the effectiveness of time…everything could

be accomplished faster”

Accuracy “...And even patients’ labs test results now (Female, 34, Nurse, Dermatology)

with Hakeem are being obtained when

92
 needed accurately with no manual errors

like we had in the past”

EHRS Perceived Value

Perceived value in our study was determined by integrating responses indicating

employees’ positive feelings about Hakeem as well as their negative feelings. Positive perceived

value were coded from respondents who used words such as: “valuable”, “worthy”, and “like”

whereas negative perceived value were coded from respondents who used words such as: “not

valuable” and “dislike”. We found costs being associated directly with negative perceived

values. On the other hand benefits were associated directly with positive perceived value. This is

closely related to the conceptualization of perceived value used in the URM which suggests that

perceived value is a result of evaluating benefits against costs. Table 3.9 provides examples from

respondents on their negative and positive perceptions of Hakeem.

Table 3.9: EHRS Perceived Value Revealed Key Concepts

Concept Quote Respondent

Negative Perception “I do not like it and see that it did not add (Male, 46, Nurse, Dermatology)

much value to my job…the system is

sometimes down which means to have 10

to 15 reports that needs to be filled before

I can leave”

“ …Hakeem had no value added to my (Female, 39, Doctor, Dermatology)

work I only spend more time in front of

the computer now”

Positive Perception “If a medicine contradicts patient’s (Male, 27, Pharmacist, Internal

allergies the system will give me an alert. Pharmacy)

93
 For example if the patient is pregnant

certain medicines cannot be given to her

and the system know that”

“Electronic Care is the future of medical (Male, 48, Doctor, Dermatology)

care and starting to implement Hakeem in

Amman at this time is a huge

advancement and an early step towards

improving the way we deliver care…. I

like using Hakeem…it is better and more

efficient”

EHRS Perceived Threat

Prior IS theories conceptualized perceived threat as the main antecedent to resistance

(e.g, Bhattacherjee and Hikmet, 2007; Bhattacherjee et al, 2008; Lapointe and Rivard, 2005).

However very little was revealed on what is considered to be a threat. Users’ perceptions of

threat may vary on basis of technology, task, or context. The study conducted by (Liang and

Xue, 2009) suggested that threat would be determined by perceived susceptibility and perceived

severity. Perceived susceptibility is defined as the employees’ subjective probability that the new

system will negatively affect them. Perceived severity is defined as the extent to which

employees perceive the negative consequences caused by the system are severe. When users

believe that system makes them vulnerable to dangerous situations and that the consequence of

that is serious, only then they will perceive a threat (Liang and Xue, 2009). In this study we

revealed three dimensions of threat: Lower Productivity, Lower Performances, and Potential

Loss of Data. Table 3.10 provides examples from respondents on each of these concepts.

Table 3.10: EHRS Perceived Threat Revealed Key Concepts

94
 Concept Quote Respondent

Lower Productivity “…the threat of Hakeem is that it (Female, 30, Nurse, Dermatology)

kills time so we cannot serve more

patients”

Lower “…we are not working as fast as (Female, 24, Nurse, Dermatology)

Performances we used to do in the past…may be if

the forms are locally installed on

my computer so the data is always

found on the computer even after

the system down phase… this would

make me move a lot faster like we

used to work with the paper file in

folder for each patient…”

Potential Loss of “…our increased dependence on (Female, 35, Admin Assistant,

Data software material which needs General Medicine)

more care in saving and sharing

information increases the risk of

information being lost as well as

being attacked [hacked]”

Facilitating Conditions

Facilitating conditions are objective factors in the environment of which individuals

agree that it is making an act easy to do (Thompson et al, 1991). Venkatesh et al., (2003) gave an

example of how returning items purchased online would be facilitated if no fee is charged to

return the item. For this study we define facilitating conditions as the factors that individuals

95
perceive as facilitating their ability to avoid using the new system. This is similar to the concept

of perceived avoidability by (Liang and Xue, 2009) which suggests that people will assess their

likelihood of avoiding a system by using specific safeguarding measures. We revealed two main

factors: Alternative User and Alternative System. Table 3.11 provides examples from

respondents on each of these concepts.

Table 3.11: Facilitating Conditions Revealed Key Concepts

Concept Quote Respondent

Alternative User “…I ask my nurse during the (Male, 51, Doctor, Primary Care Unit)

check up to input data of my

evaluation of the case”

Alternative System “sometimes we have to use (Male, 29, Pharmacist, Pharmacy)

email system to get messages to

pharmacy from ER and floors

[the clinics] if the system is

down”

EHRS Resistance

In this study we argue that resistance may coexist with or after the usage of the system,

and that resistance does not necessarily mean there is no user acceptance (Nah et al, 2004). Thus

there will be a variety of manifestations of the resistance behavior rather than only the non-usage

behavior. Our study revealed four main types of EHRS resistance behaviors: No Usage, Partial

Usage, Delayed Usage, and Indirect Usage. Table 3.12 provides examples from respondents on

each of these concepts.

Table 3.12: EHRS Resistance Revealed Key Concepts

Concept Quote Respondent

96
 No Usage “…to cope with the threats on (Male, 39, Doctor, Emergency Room)

my job from Hakeem I do not

use the system”

Partial Usage “…avoiding Hakeem is a good (Female, 57, Doctor, General

idea…I would like to use a Medicine)

paper based system when

needed”

Delayed Usage “…Digitizing the data could be (Female, 32, Nurse, ER)

done after we finish the patient

visit”

Indirect Usage “…I understand the importance (Male, 39, Doctor, ENT)

of having records digitized, but

someone else needs to do the

job”

DISCUSSION

Our analysis of responses using the RCM methodology revealed the aggregated map

presented in Figure 3.3. It shows the key constructs and the associations between constructs. In

the map, an influencing factor (cause) is depicted by an oval shape and an outcome (effect) by a

square shape. The arrow indicates the link between a cause and an effect. Resistance behavior

towards EHRS was shaped by the confluence of constructs revealed in this study. The map is

based on 375 linkages identified from 59 respondents. The number associated with a linkage

reflects the percentage of that linkage over all the linkages identified.

97
 Figure 3.3: Aggregated Revealed Causal Map of EHRS Resistance

Our findings show that EHRS Perceived Value is affected by EHRS Benefits, EHRS

Costs, Organizational Support, and EHRS Self-Efficacy. EHRS Benefits has a positive effect on

how employees perceive the value of Hakeem. As more benefits of Hakeem are considered, the

more employees perceive Hakeem as valuable to them.

“…what makes it valuable is being able to view all patient information on the computer without requesting
info from different departments.” (Female, 30, Nurse, Dermatology).

On the other hand, EHRS Costs has a negative effect on how employees perceive the

value of Hakeem. The more costs of Hakeem are considered the less employees perceive

Hakeem as valuable to them.

“I do not think of Hakeem as an added value…We can only access the system from certain machines inside
the hospital, we cannot use the system from outside the hospital…so sometimes I stay late to complete my
data entry into the system.” (Male, 35, Office Manager, Dermatology).

Organizational support was revealed in terms of technical support and training sessions.

It was found that the more support provided by the hospital the more the employees perceive

Hakeem as valuable to them.

98
 “Hakeem added value to my work…at the beginning I did not know how to use all of its features but after
we were offered some technical courses at an IT academy I really enjoyed all of options of Hakeem.”
(Male, 43, Nurse, Primary Care Unit).

However, it was reported by only one that she did not find Hakeem of high value and

later she complained about how the training sessions were provided outside of normal working

hours and it was unpaid. Thus, she found the training session as an impediment rather than a

support.

“…the main challenge is for me is to go attend Hakeem academy sessions after long days of work…these
sessions are outside of my working hours and they do not pay us extra to attend those classes.” (Female,
31, Nurse, ER).

This may result in a negative relation between organizational support and perceived

value. However, we do not have enough evidence in this study to support this argument, and we

encourage further research examine all possible attitudes towards organizational support.

EHRS Self-Efficacy had a positive effect on EHRS Perceived Value. The more

employees believe in their ability to use Hakeem, more they perceive it as valuable to them.

However, all of the responses were associated with the negative aspect of self-efficacy (i.e. only

low self-efficacy was reported) but this is posited as a positive effect because it was associated

with lower perceptions of Hakeem’s value.

“I do not see the difference [in value] when doctors use Hakeem…some doctors are not used to work with
computers so they are not gaining any value when asked to use a computer to do their work.” (Male, 49,
Doctor, Dermatology).

We also found that EHRS Perceived Threat was affected by ERHS costs. The more costs

were considered by employees the more the perceived Hakeem as a threat to their job.

“…when it is slow or it is hanging [not responding] some data could be lost and we will have to enter it
again, this causes a lot of time to be wasted and patients’ appointments will be delayed.” (Female, 32,
Nurse, Dermatology).

Finally, EHRS Resistance was found to be affected by EHRS Perceived Value, EHRS

Perceived Threat, and Facilitating Conditions. EHRS Perceived Value had a negative effect on

EHRS resistance. This means the less Hakeem was perceived as valuable to the employees the

99
more the employees would resist Hakeem. We did not find responses on how higher value

perceptions may result in less resistance. However, we consider this implicitly explained within

this linkage.

“…I think of Hakeem as a second choice with the old system. Today we cannot finish the patient visit
without finishing all forms and reports, this takes time and most of this information at my clinic is not
applicable because I care for kids. So I write all inputs on the patient’s folder and by the end of the day I
ask someone [nurse] to enter the data…this makes the flow of patients faster [Indirect Usage].” (Female,
41, Doctor, Pediatric Unit).

EHRS Perceived Threats was found to have a positive effect on EHRS Resistance. The

more employees consider Hakeem as a threat the more they will manifest shapes of resistance.

“Hakeem is a threat to my job because I spend too much time with the system and less with patients…this
effects my productivity…as a safeguard I like to keep the paper records in place to be used when needed.
Digitizing the data could be done after we finish the patient visit [Delayed Usage]” (Male, 32, Nurse, ER)

Facilitating Conditions was also found to have a positive effect on EHRS Resistance. The

higher the assessment results of employees’ likelihood of avoiding Hakeem by using specific

alternatives, the more likely employees would use the alternate rather than using Hakeem.

“…to avoid the threats I use the paper filing system when needed [Alternative System].”
(Male, 49, Doctor, ENT).

“…and have someone with us during the check up to input data during evaluations [Alternative User].”

(Male, 43, Doctor, Primary Care).

CONTRIBUTIONS

For Research

Our study has a number of contributions to the IS research. The study provides

qualitative insights on the EHR resistance phenomenon which is not well understood in IS

(Samhan and Joshi, 2015). We revealed a new construct, facilitating conditions, which emerged

from this study and was not tested in any of the prior IS resistance theories. Furthermore, we

provided a broader explanation of the resistance phenomenon by combining concepts from the

URM theory, perceived threats, and the new construct: facilitating conditions. We revealed how

100
perceived threat and perceived value may both effect resistance simultaneously. This has not

been tested in any of the prior IS theories. Moreover, our results uncovered the underlying

concepts of each construct which provides a richer understanding of these constructs. Prior IS

theorizations use these constructs abstractly with very limited insights on the dimensions shaping

each construct. Our findings is the first in IS to account for all possible shapes of resistance.

Prior theories relied merely on the conceptualization of resistance as “no usage” (e.g. Kim and

Kankanhalli, 2009; Bhattacherjee and Hikmet, 2007). Finally, our findings provided a revealed

causal map which can be the basis for empirical testing in future research.

For Practice

We have a number of contributions to practice. Our findings provide a better

understanding of the resistance phenomenon to decision makers in the healthcare sector. We

revealed that perceived threats, perceived value, and facilitating conditions affect resistance

directly. Thus, we argue eliminating threats and facilitating conditions as well as increasing the

perceived value, which could be achieved by enhancing benefits and reducing costs, is critical

when aiming to eliminate resistance behaviors. Also, it is very crucial to understand how the

provided support may influence healthcare providers’ decisions. Our findings shows that

organizational support has a positive effect on perceived value. However, there was one response

which considered the support as an overhead which leads to negative feelings about the EHR

system. This may be due to inappropriate delivery of the support (e.g. long training sessions after

hours). Additionally, investing in healthcare providers’ development to increase their self-

efficacy would have direct impacts on their perception of the value of the EHR system. It is also

important to explain and emphasize the advantages of the system, this would help in increasing

perceived value and reducing costs which in turn would reduce resistance behaviors.

101
Our findings could be tested on different technologies and in different contexts. This would help

better understand resistance in the context of which it was tested in, and that would serve as the

first step towards solving problems associated with resistance behaviors. Our study provides

insights about where EHRS designer efforts and resources should be directed in order to develop

EHRS with higher adoption rates.

LIMITATIONS AND FUTURE DIRECTIONS

Our study has a number of limitations. First, the sample size may be a concern. We

recommend conducting empirical studies on a larger scale to validate and modify the RCM

developed in this study. Second, because our data was collected from a single source, validation

concerns may arise. Future research can apply a longitudinal study or collect data from multiple

sources to validate the findings. Finally, we did not find responses on the possible effects of

social influence on the revealed constructs in this study. We encourage future research to

consider different types of social influence when testing the revealed causal map.

102
 REFERENCES
Ajzen, I, 2002. Constructing a TPB questionnaire: Conceptual and methodological
considerations.
Armenakis, A. A., and Harris, S. G., 2002. Crafting a change message to create transformational
readiness. Journal of organizational change management, 15(2), 169-183.
Axelrod, R. 1976 Structure of Decision: The Cognitive Maps of Political Elites, Princeton
University Press, Princeton, NJ.
Bandura, A. (1995). Self-efficacy in changing societies. Cambridge university press.
Bates W, Leape LL, Cullen DJ, Laird N, Peterson LA and Teich JM (1998) Effect of
computerized physician order entry and team intervention on prevention of serious
medication errors. Journal of American Medical Association 280 (15), 1311–1316.
Bhattacherjee, A., and Hikmet, N. 2007. "Physicians' resistance toward healthcare information
technologies: a dual-factor model." System Sciences, 2007. HICSS 2007. 40th Annual
Hawaii International Conference on, IEEE, 2007.
Bhattacherjee, Anol, Christopher J. Davis, Neset Hikmet, and Varol Kayhan, 2008. "User
reactions to information technology: Evidence from the healthcare sector." Association
for Information Systems.
Bovey, W. H., and Hede, A. (2001). Resistance to organisational change: the role of defence
mechanisms. Journal of Managerial Psychology, 16(7), 534-548.
Burnham, T., Frels J., and Mahajan V., 2003 “Consumer switching costs: A typology,
antecedents, and consequences,” J. Acad. Marketing Sci., vol. 31, no. 2, pp. 109–203.
Burnkrant, R. E., and Cousineau, A. (1975). Informational and normative social influence in
buyer behavior. Journal of Consumer research, 206-215.
Compeau, D., Higgins, C. A., and Huff, S. (1999). Social cognitive theory and individual
reactions to computing technology: A longitudinal study. MIS quarterly, 145-158.
Davis, F. D., 1989 “Perceived Usefulness, Perceived Ease of Use, and User Acceptance of
Information Technology”, MIS Quarterly, 13, 3, pp. 319-340.
Deng, X., and Chi, L. (2012). Understanding Post-adoptive Behaviors in IS Use: A Longitudinal
Analysis of System Use Problems in the Business Intelligence Context. Journal of
Management Information System, 29(3), 305-340.
Dent, E. B., and Goldberg, S. G. (1999). Challenging “resistance to change”. The Journal of
Applied Behavioral Science, 35(1), 25-41.
DesRoches, C. M., Charles, D., Furukawa, M. F., Joshi, M. S., Kralovec, P., Mostashari, F. &
Jha, A. K. (2013). Adoption of electronic health records grows rapidly, but fewer than
half of US hospitals had at least a basic system in 2012. Health Affairs, 10-1377.

103
Deutsch, M., and Gerard, H. B. (1955). A study of normative and informational social influences
upon individual judgment. The journal of abnormal and social psychology, 51(3), 629.
Eisenberger, R., Jones, J. R., Aselage, J., and Sucharski, I. L. (2004). Perceived organizational
support. The employment relationship: Examining psychological and contextual
perspectives, 206-225.
Electronic Health Solutions (EHS) http://www.ehs.com.jo/en
Fine, M., Weis, L., Weseen, S., and Wong, L. (2000). For whom. Qualitative research,
representations, and social responsibilities. In NK Denzin and YS Lincoln (Eds.),
Handbook of qualitative research, 2, 107-131.
Freud, S. (1919). Turnings in the ways of psychoanalytic therapy. Collected papers, 2, 392-402.
FREUDENHEIM M (2004). Many hospitals resist computerized patient care. New York Times,
http://www.nytimes.com/2004/04/06/technology/06errors.html.
Halamka, John. “Healthcare is different”, (2011)
http://geekdoctor.blogspot.com/2011/08/healthcare-is-different.html
Hare, K., Whitworth, B., Deek, F., and Norris, T., 2006 “Identifying Barriers to IT Adoption in
Clinical Health Care: A Pilot Study”, Health Care and Informatics
Review Online, (2006), ISSN 1176-4201, http://hcro.enigma.co.nz/website/index.cfm
Hillestad, R., Bigelow, J., Bower, A., Girosi, F., Meili, R. and Scoville, R. 2005. Can Electronic
Medical Record Systems Transform Health Care? Potential Health Benefits, Savings, and
Costs. Health Affairs, 24, 1103-1117.
Hirschheim, R., and Newman, M. (1988). Information systems and user resistance: theory and
practice. The Computer Journal, 31(5), 398-408.
Inder, B., and O’Brien, T., “The endowment effect and the role of uncertainty,” Bull. Econ. Res.,
vol. 55, no. 3, pp. 289–301, 2003.
Joshi, K. 1991. “A Model of User Perspective on Change: The Case of Information Systems
Technology Implementation”, MIS Quarterly, 15, 2, pp. 229-242.
Keen, P.G., 1981. "Information Systems and Organizational Change," Communications of the
ACM, 24, 1, pp. 24-33.
Kelman, H. C. (1958). Compliance, identification, and internalization: Three processes of
attitude change. Journal of conflict resolution, 51-60.
Kim, H. W. (2011). The effects of switching costs on user resistance to enterprise systems
implementation. Engineering Management, IEEE Transactions on, 58(3), 471-482.
Kim, Hee-Woong, and Atreyi Kankanhalli. 2009. "Investigating user resistance to information
systems implementation: a status quo bias perspective”, MIS quarterly 33, 3.

104
Kotter, J. P., and Schlesinger, L. A. (2008). Choosing strategies for change. Harvard business
review, 86(7/8), 130.
Lapointe, L., and Rivard, S. 2005. .“A Multiple Model of Resistance to Information Technology
Implementation.” MIS Quarterly (29:3), pp. 461-491.
Lapointe, L., and Rivard, S., 2007. “A Triple Take on Information System Implementation”,
Organization Science, 18, 1, (2007), pp.89-107
Lewis, W., Agarwal, R., and Sambamurthy, V. 2003. .“Sources of Influence on Beliefs about
Information Technology Use: An Empirical Study of Knowledge Workers.” MIS
Quarterly (27:4), pp. 657-678.
Liang, H., and Xue, Y. (2009). Avoidance of information technology threats: a theoretical
perspective. MIS quarterly, 71-90.
Marakas, G. M., and Hornik, S. 1996. .“Passive Resistance Misuse: Overt Support and Covert
Recalcitrance in IS Implementation” European Journal of Information Systems (5:3), pp.
208-219.
Markus M. L. 1983. .“Power, Politics, and MIS Implementation” Communications of the ACM
(26:6), pp. 430-444.
Martinko, M. J., Henry, J. W., and Zmud, R. W., “An Attributional Explanation of Individual
Resistance to the Introduction of Information Technologies in the Workplace”, Behavior
and Information Technology, 15, 5, (1996), pp. 313-330.
Mekhijan HS, Kumar RR, Kuehn L, Bentley TD, Teater P, Thomas A, Payne B, and Ahmad A
(2002). Immediate benefits realized following implementation of physician order
entry at an Academic Medical Center. Journal of American Medical Informatics
Association 9(5), 529–539.
Miller, V. D., Johnson, J. R., and Grau, J. (1994). Antecedents to willingness to participate in a
planned organizational change.
Nah, F. F., Tan, X., and Teh, S. H. 2004. .“An Empirical Investigation on End-Users.’
Acceptance of Enterprise Systems.” Information Resources Management Journal (17:3),
pp. 32-53.
Narayanan, V.K.., and Armstrong, D. J. (2005). Causal Mapping for Research in Information
Technology. Hershey, PA: Idea Group Publishing.
Nelson, K.M., Nadkarni, S., Narayanan, V.K., and Ghods, M. (2000). Understanding Software
Operations Support Expertise: A Revealed Causal Mapping Approach. MIS Quarterly,
24(3), 475-507.
Samhan, B., and Joshi, K. D. (2015, January). Resistance of Healthcare Information
Technologies; Literature Review, Analysis, and Gaps. In System Sciences (HICSS), 2015
48th Hawaii International Conference on (pp. 2992-3001). IEEE.

105
Samuelson, W., and Zeckhauser, R. (1988). Status quo bias in decision making. Journal of risk
and uncertainty, 1(1), 7-59. Chicago
Sirdeshmukh, D., Singh, J., and Sabol, B. (2002). Consumer trust, value, and loyalty in relational
exchanges. Journal of marketing, 66(1), 15-37.
Smith, Tim, Gerald Grant, and Alejandro Ramirez. "Investigating the influence of psychological
ownership and resistance on usage intention among physicians”, System Sciences
(HICSS), 47th Hawaii International Conference on. IEEE, (2014).
SPIL TAM, SCHURING RW and MICHEL-VERKERKE MB (2004) Electronic prescription
system: do the professional USE IT. International Journal of Healthcare Technology
Management 6(1), 32–55.
Stone, C. 2014 “A Glimpse at EHR Implementation around the World: The Lessons the US Can
Lern,” The Health Institute for E-Health Policy (available at http://www.e-
healthpolicy.org/docs/A_Glimpse_at_EHR_Implementation_Around_the_World1_Chris
Stone.pdf).
TAYLOR R, MANZO J and SINNET M (2002) Quantifying value for physician order-entry
systems: a balance of cost and quality. Healthcare Financial Management 7, 44–48.
Thompson, R. L., Higgins, C. A., and Howell, J. M. “Personal Computing: Toward a Conceptual
Model of Utilization,” MIS Quarterly (15:1), 1991, pp. 124-143.
Tichy, N. (1983). The essentials of strategic change management. Journal of Business Strategy,
3(4), 55-67.
Venkatesh, V., and Davis, F. D. (2000). A theoretical extension of the technology acceptance
model: Four longitudinal field studies. Management science, 46(2), 186-204.
Venkatesh, V., Davis, F. D., and Morris, M. G. (2007). Dead or alive? The development,
trajectory and future of technology adoption research. Journal of the association for
information systems, 8(4), 267.
Venkatesh, V., M.G. Morris, G.B. Davis, and F.D. Davis, 2003. “User Acceptance of
Information Technology: Toward A Unifying View,” MIS Quarterly, 27, 3, pp. 425-478.
Wanberg, C. R., and Banas, J. T. (2000). Predictors and outcomes of openness to changes in a
reorganizing workplace. Journal of Applied Psychology, 85(1), 132.
Watson, G. (1971). Resistance to change. The American Behavioral Scientist, 14(5), 745.
Whitten, D., and Wakefield R., 2006. “Measuring switching costs in IT outsourcing services,” J.
Strategic Inf. Sys., vol. 15, pp. 219–248.
Williams, Michael D., Yogesh K. Dwivedi, Banita Lal, and Andrew Schwarz. 2009.
"Contemporary trends and issues in IT adoption and diffusion research", Journal of
Information Technology, 24, no. 1, (2009), pp. 1-10.

106
 APPENDIX

Interview Questions

Table 3.2: Interviews’ Open-ended Questions

No. Question
1 What is your job title at the hospital?
2 Briefly describe your job duties at the hospital.
3 Describe how you use Hakeem to do your job. Please provide some examples.
4 Has the use of Hakeem changed the way you do your job? Please explain
5 What do you like about Hakeem? Why? Please explain using examples
6 What do you dislike about Hakeem? Why? Please explain using examples.
7 Describe the challenges you encountered while switching to the new way of working with
Hakeem.
8 What do your colleagues and coworkers like or dislike about Hakeem? Please explain using
examples.
9 What type of organizational support did you receive when switching from the paper based system
to Hakeem? Was it helpful? Why or why not?
10 What would you like to change about Hakeem? Why?
11 What are the threats, risks, and perils of you using Hakeem to do your job? Please explain using
examples
12 What do you think are appropriate safeguards and protections if employed will help in reducing
threats, risks, and perils of you using Hakeem to do your job?
13 How do you cope with the threats, risks, and perils of you using Hakeem to do your job?

107
 CHAPTER FOUR:

ELECTRONIC HEALTH RECORDS AVOIDANCE: AN EMPIRICAL

INVESTIGATION FROM THE TECHNOLOGY THREATS AVOIDANCE THEORY

PERSPECTIVE

ABSTRACT

This study empirically tests the conceptual model of the Technology Threats Avoidance

Theory (TTAT) (Liang and Xue, 2009) in an international healthcare context to investigate

electronic healthcare avoidance behaviors. We tested the complete conceptual model and based

on the modification indices test we suggested a theoretical model with an aim to explain

avoidance behaviors towards electronic health records. We found that the electronic health

record system avoidance is predicted by levels of perceived threat, avoidance motivation,

perceived effectiveness of the safeguarding measure, and the social influence of using the

safeguarding measure. We also controlled for a number of variables and found that levels of

education and number of healthcare providers in a medical unit has significant influence on

avoidance behavior towards the electronic health record system.

108
 CHAPTER FOUR:

ELECTRONIC HEALTH RECORDS AVOIDANCE: AN EMPIRICAL

INVESTIGATION FROM THE TECHNOLOGY THREATS AVOIDANCE THEORY

PERSPECTIVE

INTRODUCTION

Electronic Health Records (EHR) systems have the potential to provide numerous

benefits (Callen et al, 2013). It provides healthcare providers with important information about

patients’ medical history including previous diagnoses and medications, medication allergies,

treatments, surgeries, summaries of hospital admissions and discharges, lab test results, and

summaries of any previous doctor visits. All of this information could be retrieved in timely

fashion which helps healthcare providers make more appropriate decisions (Pope et al, 2013).

EHR systems can also help providers from different units of a hospital, or even from different

hospitals, to coordinate the care-giving process by knowing the medical history of the patient

they are treating (Cordell et al, 1998). Additionally, prior research suggested that the use of EHR

systems by healthcare providers can enhance quality of patient care and reduce medical errors

(Hillestad et al, 2005). These benefits cannot be achieved if providers avoid EHR systems

(Romanow et al, 2012).

However, despite the great potential EHR systems has to generate substantial value to

healthcare providers, evidence of avoidance behaviors from healthcare providers towards EHR

are still reported (Kane and Labianca, 2011). We also find a dearth of IS research in this area and

the findings are mixed and unclear (Kellermann and Jones, 2013). Prior research has paid some

attention to the consequences of resistance to information systems such as (Markus, 1983; Poon

et al, 2004; Lapointe and Rivard, 2005; Bhattacherjee and Hikmet, 2008), however very little

109
attention has been given on IS avoidance (Liang and Xue, 2009) and particularly in the

healthcare context (Kane and Labianca, 2011). As a result, we know little concerning why

Health Information Technology (HIT) is being avoided and specifically why healthcare

providers avoid EHR systems in the postadoption stages. In this study we aim to understand the

following research question - why healthcare providers avoid EHR systems?

We tested the complete conceptual research model from the Technology Threat

Avoidance Theory (TTAT) (Liang and Xue, 2009) to explain how healthcare providers develop

threat perceptions, evaluate safeguard measures, and engage in avoidance behavior towards the

EHR system. We undertake our study in the context of an international hospital that has enforced

the use of an EHR system. We used survey data to test the TTAT model. Findings and

discussions on how well these findings support the original conceptual model are provided in this

paper. This work is one of the very few attempts to provide empirical explanations about the

phenomenon of EHR avoidance. Plus, to our knowledge, this is one of the first studies to

empirically validate and test the entire TTAT model in the healthcare context. Doing so will not

only advance the theoretical understanding of EHR avoidance behaviors, but also offer practical

insights that could help with managing EHR avoidance behaviors.

TECHNOLOGY AVOIDANCE

In this paper we define EHR avoidance as the process in which individuals aim to enlarge

the distance between their current safe state and the undesired end state caused by using the EHR

system (Carver and Scheier, 1982; Edwards, 1992). In our study we distinguish avoidance

behaviors from the acceptance and the resistance behaviors when theorizing EHR avoidance.

Avoidance and acceptance behaviors may coexist. For example, behaviors such as indirect usage

and delayed usage are considered avoidance behaviors while the technology is still being

110
accepted/used. We believe that EHR acceptance and EHR avoidance constructs are different and

one cannot be explained as the opposite of the other (Liang and Xue, 2009; Carver and White,

1994; Elliot, 2006; Elliot and Covington, 2001). The distinction between adoption and avoidance

constructs is more important in this study because we aim to explain avoidance in a healthcare

setting which has unique characteristics that makes it different from the more general “corporate

environments” in which technology adoption models have been successfully tested in prior

studies. Therefore, applying adoption models within the healthcare context will have less

explanatory power (Smith et al, 2014). Moreover, the adoption theories are designed to explain

voluntary IT usage and have limited capabilities in explaining mandated IT usage; which is the

case for most EHR implementation initiatives (Bhattacherjee and Hikmet, 2013). Nevertheless,

technology acceptance theories can provide useful findings when assessing the adoption of

safeguards which individuals use to avoid threats associated with using the EHR systems, yet

this approach does not provide a complete explanation of the technology avoidance

phenomenon. For example, physicians may be influenced to avoid the EHR system if they

considered the threat of losing valuable time using the EHR system while performing a medical

task. Healthcare providers must first perceive the loss in time as a threat and then can take

several actions such as using a paper-based system or asking someone else to fill in the EHR

while continue doing their medical task. If we only study adoption of the paper-based system, we

will form partial understanding of the avoidance phenomenon because we ignored the evaluation

of threat and alternative actions. Individual’s aim to avoid threats associated with the EHR

system not to adopt safeguards. The safeguards are only tools which help healthcare providers

reach their goal of avoiding the system and its threats.

111
 Additionally, we believe that EHR avoidance behaviors differ from EHR resistance

behaviors mainly in terms of the stage in which each is manifested (Kane and Labianca, 2011).

Healthcare providers may first resist the change from working with the Paper-Based Recording

System (PBRS) to the newly implemented EHR system; and overtime, when the EHR system is

completely implemented and enforced, avoidance behaviors may become present (Kim and

Kankanhalli, 2009). Therefore, we focus our study on avoidance behaviors towards the EHR

system because the EHR system has already been implemented and mandated for all employees

to use, thus positioning our work in the post adoption stage.

LITERATURE REVIEW

Prior studies in the IS literature has focused on the resistance behaviors prior to system

implementation, it focused on the concept of resistance to change associated with the

implementation of an IS (example: Kim and Kankanhalli, 2009). However, relatively little effort

has been spent to investigated how avoidance occurs in post-implementations stages, especially

in the healthcare settings (Kane and Labianca, 2011). One of the few attempts to theorize

technology avoidance was the work of (Liang and Xue, 2009) who introduced the Technology

Threat Avoidance Theory (TTAT). It suggests that when users perceive a threat associated with

the technology they are using (threat appraisal), they become motivated to actively avoid the

threat and start to find safeguard that would help them avoid the threats associated with the

system (coping appraisal). Liang and Xue derived a measurement model from the TTAT and

empirically tested it in the context of spyware threats (Liang and Xue, 2010), another model was

derived from TTAT and tested in the context of telemedicine (Xue et al, 2014; and Xue et al,

2015). Additionally, (Arachchilage and Love, 2013) tested the TTAT in the context of game

design framework for avoiding phishing attacks. Similar to the TTAT, the work of (Rho and

112
Ryu, 2011) examined the cognitive process of avoidance through coping behavior, they

explained how users first perceive a threat such as the exposure of personal information, which

then influences emotion-focused coping that in turn influences technology avoidance. (Hwang,

2005) conceptualized Uncertainty Avoidance as an informal control in the ERP implementation

and tested its influence on the technology acceptance variables. Another study on HIT avoidance

was conducted by (Kane and Labianca, 2011), they examined HIT avoidance as a predictor of

patient care outcomes across different levels of analysis in a healthcare organization.

In summary, the review of prior IS literature revealed a number of research gaps. First,

very few studies aimed to investigate avoidance in general and specifically avoidance towards

EHR systems. Second, there is an evident lack of empirically tested unified model of HIT

avoidance. Third, majority of the prior research was conducted in the North-American context

with very limited insights from global perspectives. In this paper we will empirically test the

complete conceptual model of TTAT, including all of its constructs, with a focus on the post

implementation stage of an EHR system.

INVESTIGATIVE CONTEXT

Our study was conducted in the same hospital used for chapter two. It is a large public

hospital located in the capital city of Jordan, Amman. The hospital consists of multiple health

departments and 33 specialists’ clinics. The total inpatient capacity is 450 beds, with a plan of

adding 100 more beds by early 2016. The hospital has recently implemented an EHR system that

is mandated to all employees. It is one of two healthcare organizations in the country that have

completely implemented the EHR system. The project started in 2009, implemented in 2014, and

was enforced early 2015. No more paper records are being used at the hospital. The new EHR

system is named “Hakeem”, the Arabic word for “Doctor”. Hakeem was implemented through a

113
governmental program which is the first national e-health initiative in Jordan. The system was

designed and implemented by Electronic Health Solutions (EHS) which is a non-profit,

innovative, technology-driven company established in early 2009. EHS is a partnership between

the main healthcare stakeholders: Ministry of Health, Ministry of Information and

Communications Technology, Royal Medical Health Awareness Society and Private Hospitals

Association.

Understanding avoidance behaviors towards the EHR system requires special attention to

the unique characteristics of the healthcare context. Healthcare organizations differ from other

corporate environments in many ways. First, healthcare providers, in most cases, are not

employees of the hospital but rather work as private contractors delivering care through different

hospitals (Halamka, 2011). This allows them to decide whether to use or avoid the EHR system

even if it was mandated by the healthcare organization. Second, healthcare providers usually

perceive entering data into the EHR system as an unnecessary task that hurdles them from

performing their original job of providing care to patients, and some consider this task

inappropriate for the physicians because it requires doing tasks that represents threat to their

professional status (Lapointe and Rivard, 2005). Third, some healthcare providers have concerns

about how EHR systems affects their communication with patients such as loss of eye contact

while using the computer (Cotea, 2010). Finally, healthcare providers fear that the EHR systems

will take time away from actually caring for the patient and make the art of their job impersonal

(Sassen, 2009).

THEORETICAL BACKGROUND OF TTAT

The TTAT (Liang and Xue, 2009) explains individuals’ behavior of avoiding threats

associated with the use of technology. It suggests that threats are influenced by the levels of its

114
severity and individual’s believes of their susceptibility to these threats. If a threat is realized,

individuals begin to aim to avoid the threats by using a chosen safeguarding measure. The

safeguarding measure does not necessarily have to be an IT source but also could be an

alternative way to complete the task (e.g. using the paper-based record system (PBRS) as a

safeguard to avoid using the EHR system). TTAT explains that the safeguarding measure would

be evaluated by three constructs: safeguard effectiveness, safeguard cost, and self-efficacy.

Safeguard effectiveness refers to individuals’ evaluation of a safeguarding measure in regards to

how effective could it be in avoiding the threat. For example, the individual assessment

regarding how effectively the PBRS can be to avoid a threats associated with using the EHR

system. Safeguard cost refers to the physical and cognitive costs associated with performing the

safeguarding measure. It could be costs in time, money, or other inconvenient requirement of

using the safeguard measure. Self-efficacy is individuals’ confidence in being able to take the

safeguard measure. However in the context of our study, safeguarding measures are reported as

using PBRS when avoiding the ERR system. Because the PBRS requires no special skill and has

been the original way of performing tasks at the hospital before the implementation of the EHR

system, we did not include self-efficacy of using paper records in this study, and only measured

for the effect of safeguard costs and safeguard effectiveness on perceived avoidability.

According to TTAT, after perceiving a threat users would react by performing some type

of coping. Users would be motivated to avoid the system (problem-focused coping) if they knew

how to avoid the system and had the ability to do so, otherwise if users believe that the threat

cannot be fully avoided by taking safeguarding measures, they would passively avoid the system

by engaging in emotion-focused coping in which they create a false reality about the threat

environment (Liang and Xue, 2009). However, (Liang and Xue, 2010) explained how emotion-

115
focused coping reduces users’ awareness of the threat which affects their perceptions of reliable

safeguard mechanisms. This would result in less impact of the safeguarding measure on

avoidance motivation. For example, a common emotion-focused coping mechanism is to not

think about the threat to escape from the situation. Although users know that the threat is present,

they would block it from their rational thinking. Therefore, the relationship between safeguard

effectiveness and avoidance motivation will be weaker as threat levels get higher. Thus, in this

study we only included the problem-focused coping mechanism. This did not affect the

theoretical model because the emotion-focused coping construct is an outcome of the model

rather than a predictor. TTAT also included two influential variables: Risk Tolerance and Social

Influence. Risk tolerance was defined as the “minimum discrepancy between the undesired end

state and the current state that users are able to tolerate” (Liang and Xue, 2009). This means that

users with different levels of risk tolerance would endure different levels of threats associated

with using the EHR system. Thus, it is suggested that risk tolerance has direct influence on

perceived threat. Social influence in this study is suggested to affect how employees perceive

threats associated with the EHR system, their perceptions on safeguarding measures (i.e. costs

and effectiveness), and their motivation to avoid IT threats. Social influence on perceptions of

threat associated with the EHR system and social influence on the safeguarding measures (i.e.

cost and effectiveness of using the PBRs) were tested distinctively

In addition to assessing key variables of TTAT, we included several demographic control

variables: age, gender, and level of education. It is argued that employees with different positions

would perceive different threats and would tolerate risk differently. Thus, we included position at

hospital as a control variable. Addition, because social influence may have indirect effect on

avoidance behaviors and a direct effect on perceived threats, safeguarding measures, and

116
avoidance motivation, it was appropriate to control for the number of employees in a unit. We

also controlled for the size of units at the hospital based on the number of beds in a unit.

RESEARCH MODEL AND HYPOTHESES

The main constructs of the conceptual model of TTAT (Liang and Xue, 2009) were

adopted in this study. Table 4.1 presents a summary of all the constructs in the model, their

acronyms, and definitions in the context of our study.

Table 4.1: TTAT Model Constructs

Construct Acronym Definition
Individuals’ subjective probability that the use of the
Perceived
PSUS new EHR system will negatively affect their job
Susceptibility
productivity and performances.

The extent to which individuals perceive negative

Perceived Severity PSEV consequences caused by using the EHR system as
severe.

The maximum amount of risk a person is willing to

Risk Tolerance RIST
accept before deciding to avoid the EHR system.
Subjective Norms: Individuals’ perception that most
people who are important to them think they should or
should not use the EHR system.
Social Factors: The internalization of others subjective
Social Influence
SOIH culture, and specific agreements individuals have made
EHR
with others within the hospital.
Image: The degree to which the use of the EHR is
perceived to enhance the image or status of individuals
in the social system.
The extent to which individuals perceive the EHR
Perceived Threat PTHR system as danger to their job productivity and
performance
Subjective Norms: Individuals’ perception that most
people who are important to them think they should or
should not use the PBRS.
Social Factors: The internalization of others subjective
Social Influence
SOIP culture, and specific agreements individuals have made
PBRS
with others within the hospital.
Image: The degree to which the use of the PBRS is
perceived to enhance the image or status of individuals
in the social system.
Individuals’ physical and cognitive efforts that are
Perceived Cost PCOS needed to go to the PBRS such as time, money,
inconvenience, and comprehension.

117
 Individuals’ subjective assessment that having the
Perceived
PEFF PBRS can effectively avoid the threats associated with
Effectiveness
using the EHR system.
The individuals’ assessment of the levels of control
Perceived
PAVO they believe they have over avoiding the EHR system
Avoidability
by using the PBRS.
The degree to which users of the EHR system are
Avoidance
AVMO motivated to avoid associated threats by using the
Motivation
PBRS.
The process of aiming to avoid the use of the EHR
Avoidance system to enlarge the distance between a secure state
AVBH
Behavior and the unsafe state associated with the use of the EHR
system.

TTAT explored avoidance behavior from a threat perspective. It suggests that once a

threat is perceived users would evaluate safeguarding measures to be taken in an aim to avoid the

threat associated with the use of the system. Levels of susceptibility, severity, risk tolerance, and

social influence impact users’ threat perceptions. Once a threat is perceived, users evaluate

safeguarding measures by how effective these measures are in terms of avoiding the threat

associated with the use of the system, and by how costly is it to take these measures as an

alternative. Both effectiveness and cost of the safeguard along with social influence predict

users’ perceived avoidability. In turn, perceived threat and perceived avoidability influence

avoidance motivations which has a direct effect on the avoidance behavior. The conceptual

model is illustrated in Figure 4.1 and the model hypothesis are listed in Table 4.2.

118
 Figure 4.1: TTAT Conceptual Model (Liang and Xue, 2009)

Table 4.2: TTAT Hypotheses

No. Hypothesis
H1 Social Influence of using EHR has a negative direct effect on Perceived Threat.
H2a Perceived Susceptibility has a positive direct effect on Perceived Threat.
H2b Perceived Severity has a positive direct effect on Perceived Threat.
H2c Perceived Susceptibility and Perceived Severity have a positive interaction effect on Perceived
Threat.
H3 Risk Tolerance has a negative direct effect on Perceived Threat.
H4a Social Influence of using PBRS has a positive direct effect on Avoidance Motivation.
H4b Social Influence of using PBRS has a positive direct effect on Perceived Effectiveness.
H4c Social Influence of using PBRS has a negative direct effect on Perceived Cost.
H5 Perceived Effectiveness has a positive direct effect on Perceived Avoidability.
H6 Perceived Cost has a negative direct effect on Perceived Avoidability.
H7a Perceived Threat has a positive direct effect on Avoidance Motivation.
H7b Perceived Avoidability has a positive direct effect on Avoidance Motivation.
H7c Perceived Threat and Perceived Avoidability have a positive interaction effect on Avoidance
Motivation.
H8 Avoidance Motivation has a positive direct effect on Avoidance Behavior.

119
 METHODOLOGY

Instrument Development

Existing validated scales were adopted for this study. We adopted the instrument items

used for the model derived from TTAT (Linag and Xue, 2010). We also adopted items from

different sources to capture all of the constructs of TTAT that were not previously used in

empirical models, these include: First, Perceived Avoidability which is defined as the

individuals’ assessment of the levels of control they believe they have over avoiding the EHR

system by using the PBRS. Perceived Avoidability is formulated by considering how the

safeguarding measure effectively avoids the associated threat and what costs are associated with

the safeguarding measure. The more a user believes the safeguard is effective and with low costs

the more the likelihood of feeling they have control over avoiding the threat associated with

Hakeem. In the context of our study, the suggested safeguarding measure is using the PBRS to

avoid Hakeem. Items for Perceived Avoidability were adopted from the perceived behavioral

control scale (Ajzen, 2006). Second, Social Influence items were adopted from the scale used by

(Venkatesh et al, 2003) which includes social factors, subjective norms, and image items. First,

subjective norms: these are based on the concept of what do the important others think about

using the system. The source of these norms could be coming from peers and colleagues in the

form of knowledge transfer by sharing past experiences and opinions (informational influence)

(Burnkrant and Cousineau 1975; Deutsch and Gerard 1955), or could be from superiors

associated with feelings of compliance, internalization, and identification (normative influence)

(Kelman, 1974). Compliance is affected by the need for getting approval, rewards, or the fear of

punishment. Internalization is realized by harmonizing user’s goals and values about the system

with others’. Identification is shaped by users self-define their relationship with others. Second,

120
social factors: just like the internalization of the normative influence, these are individuals’

internalization of others subjective culture and the interpersonal agreements made with others in

specific social situations (Thompson et al, 1991). Third, image: this is the degree to which the

use of the system is perceived as an enhancement to individuals’ status in the social system

(Moore and Benbasat, 1991). We captured two types of social influence for this study: Social

Influence on using the system (Hakeem), and Social Influence on using the safeguarding

measure (PBRS). Third, Risk Tolerance items were adopted from (Cable and Judge, 1994; and

Judge et al, 1999). Finally, we added 2 items (AVBH3 & AVBH4) to capture Avoidance

Behavior, (Liang and Xue, 2010) had only 2 items for the construct and we needed at least 3 to

run Confirmatory Factor Analysis (CFA) successfully. All items were modified to fit the context

of our study. Measurement items were anchored on five-point Likert scales (1 = strongly

disagree, 5 = strongly agree). The instrument was reviewed by IS researchers before the study.

The final version of the questionnaire was developed for the main study as shown in Table 4.3 in

the Appendix.

Sample and Data Collection

The total estimated number of employees who work at the hospital is 5,500. Majority of

these employees are potential users of the new system. However administrative staff are less

likely to use the system as they do not participate in the care giving process, however they still

have access to the system and would be able to print out reports as needed. The sample is a

combination of all stakeholders of the system (i.e. physicians, nurses, lab specialists, radiologists,

pharmacists, admin staff, and resident medical students). Data was collected in 2 phases. First,

we collected data for the pilot study from employees who works only for the Dermatology unit at

the hospital, we collected a total of (126) responses.

121
 After conducting Explanatory Factor Analysis (EFA) and Confirmatory Factor Analysis

(CFA) of the data collected for the pilot test, it was necessary to alter some of the items to

address a number of issues related to the psychometric properties of the study’s constructs.

Additionally, some new items were added to better capture the concepts of some constructs.

First, we dropped 5 items from the Perceived Severity construct because they had low loadings

range (0.40 – 0.60) and kept the other 5 items which had acceptable loadings range of (0.80 –

0.99), we also dropped one item from the Perceived Effectiveness construct that had a low

loading of (0.69). Second, Perceived Threats items were reworded to include the word “using”

instead of “switching”. At this stage the EHR system has been completely implemented and the

switch from the PBRS has already happened. Thus, it was appropriate to change the items to

better reflect reality. Third. We reworded the items of Perceived Costs to include the word

“previous” when referring to the PBRS to be clear that we refer to the old way of working. We

aim to capture the costs associated with going back to the old way of working with the PBRS as

a safeguarding measure to avoid threats associated with Hakeem. Fourth, for both types of Social

Influence (PBRS and Hakeem) we included “people in my hospital” in the items to eliminate

healthcare providers mistaking the word “people” for their friends or relatives. Fifth, we dropped

2 items from the Risk Tolerance construct because they were conflicting with the cultural context

of this study. This first item was “I like (or would like) to play the lottery”, and this was dropped

for our survey because gambling is considered as sinful action within the Jordanian culture. The

second items was “I generally hold out for the best price on something, even if it means waiting

a long time”, and this was dropped from our survey because consumer behaviors and market

strategies in Jordan are different and sale prices does not normally change with time, so this item

122
did not make sense to the Jordanian population. Finally, we found that item PAVO2 from

Perceived Avoidability was reverse codded so we fixed it for the main study survey.

After refining the survey items based on the pilot data’s psychometric properties, we

collected data for the main study from the entire hospital, but employees who filled the first

survey (i.e. pilot study) were specifically asked not to fill in the survey for the second time. We

collected a total of (220) responses, descriptive statistics are reported in Table 4.4.

Table 4.4: Descriptive Statistics of Respondents

Demographic Variables Data
Gender Male 130 (59.09%)
Female 90 (40.91%)
Age <30 55 (25.00%)
(Mean = 33.06, S.D. = 5.10) 31 – 40 84 (38.18%)
41 – 50 60 (27.27%)
51 – 60 21 (9.55%)
>60 0 (0.00%)
Position Physician 56 (25.45%)
Nurse 69 (31.36%)
Radiologist 22 (10.00%)
Lab Specialist 31 (14.09%)
Pharmacist 14 (6.36%)
Admin Staff 19 (8.64%)
Medical Student 9 (4.09%)
Total 220 (100%)

Instrument Validation

To validate the survey instrument, the psychometric properties of the survey were

assessed by conducting Explanatory Factor Analysis (EFA) and Confirmatory Factor Analysis

(CFA) using MPlus 7.1 (Muthén and Muthén, 1998-2006). Cronbach’s α was performed using

SPSS 20 (IBM Corp. 2013). After the pilot data was collected, we refined the items to keep those

with statistical significant loading larger of (0.7). Cronbach’s α reliability tests for all constructs

exceeded (0.8). After the main data was collected, we again conducted CFA analysis and again

all items had a significant loading greater than (0.7). All constructs had Cronbach’s α values

123
exceeding (0.8). The CFA analysis provided strong support for our measurement model, which

suggested that the items under each of the constructs were adequately measuring the constructs.

Table 4.5 in the Appendix lists the items’ loadings and the model fit statistics for the

CFA.

The survey responses came in (2) waves: (W1: N=161; W2: N=59). We checked for

nonresponse bias by testing the difference in means between waves (See Table 4. 6). We found

no significant difference between the two respondent groups based on the sample attributes

(gender, age, and position).

Table 4.6: Nonresponse Bias Test

First Wave Second Wave
t-Value Sig (p<0.05) df = 218
(N=161) (N=59)
Age Mean = 38.0 Mean = 36.7 0.923 Not Sig.
Gender Male = 95 Male = 35
(59.00%) (59.03%)
0.138 Not Sig.
Female = 66 Female = 24
(40.99%) (40.67)
Position Physician = 40 Physician = 17
(24.39%) (27.41%)
Nurse = 50 Nurse = 20
(30.48%) (32.25%)
Pharmacist = 17 Pharmacist = 6
(10.36%) (9.67%)
Admin Staff = 26 Admin Staff = 6
0.115 Not Sig.
(15.83%) (9.67%)
Radiologist = 11 Radiologist = 4
(6.70%) (6.45%)
Lab Specialist = 14 Lab Specialist = 6
(8.53%) (9.67%)
Medical Student = Medical Student = 3
6 (3.65%) (4.83%)

RESULTS

We tested the TTAT theoretical model by applying Structural Equation Modeling (SEM)

using Mplus version 7.1 (Muthén and Muthén, 1998-2006). We applied the maximum likelihood

estimator with robust standard errors (MLR). Because our model is not saturated (i.e., not all

124
possible regression paths were included) we evaluated the model fit indicators depicted in Table

4.7.

Table 4.7: TTAT Model Fit Indices

Indicator Value
X2 642.671
df 41
P-Value 0.000
Calculated NC 15.67
CFI 0.593
RMSEA 0.258
90 CI 0.241 – 0.276
SRMR 0.07

The model fit indices showed ill fit of the model. First, according to (Bollen, 1998)

suggestions on evaluating Chi-Sqaure (X2), we calculated the Normed X2 (NC=15.67). It is

suggested that NC value outside the range (2.0 and 6.0) indicates poor fit. Second, the model had

a very low (CFI = 0.593), this indicates illness in the model fit. The rule of thumb for the CFI

and other incremental indices is that values greater than roughly (.90) may indicate reasonably

good fit of the researcher’s model (Hu & Beltler, 1999). Third, the Root Mean Square Error of

Approximation (RMSEA) is (0.258). According to the rules of thumb by (Browne and Cudeck,

1993) RMSEA value between (0.05) and (0.08) suggest reasonable error of approximation and

values greater than (0.1) suggests poor fit. We conclude that the model’s goodness of fit is

inadequate. In order to improve the goodness-of-fit of the model we conducted modification

indices test using MPlus. This suggests possible paths to be added to or dropped from the model

to improve its goodness-of-fit. Also it would suggest alternative theoretical models that would

best fit our data. The suggested model and its new improved model fit indices are included in the

next section of this paper. The standardized path coefficients, its level of significance, and the

standard errors (S.E) are provided in Table 4.8, and results are depicted in Figure 4.2.

125
 *p<0.05; **p<0.01; ***p<0.001

Figure 4.2: TTAT Hypotheses Testing Results

Table 4.8: Standardized Path Coefficients

Hypothesis Hypothesized STD β (S.E.)
Relationships
H1 SOIH - PTHR 0.106 (0.035)**
H2a PSUS - PTHR 0.337 (0.089)***
H2b PSEV - PTHR 0.518 (0.106)***
H2c PSUS X PSEV - PTHR 0.101 (0.134)
H3 RIST - PTHR -0.012 (0.028)
H4a SOIP - AVMO 0.170 (0.048)***
H4b SOIP - PEFF 0.520 (0.051)***
H4c SOIP - PCOS 0.002 (0.092)
H5 PEFF - PAVO 0.415 (0.056)***
H6 PCOS - PAVO 0.006 (0.066)
H7a PTHR - AVMO 0.777 (0.087)***
H7b PAVO - AVMO 0.016 (0.830)
H7c PTHR X PAVO - AVMO -0.062 (0.136)
H8 AVMO - AVBH 0.797 (0.022)***

126
 Our study found a number of hypotheses from TTAT that were not confirmed. First, the

interaction effect of perceived susceptibility and perceived severity on perceived threat was not

significant. This is an interesting finding that implies that if either variable takes a value of zero,

the other variable would still have an impact on perceived threat. For example, if healthcare

providers know that the EHR system will cause severe harm to their job, they will perceive the

EHR system as a threat even if they see no possibility for them to be harmed. Also, if healthcare

providers know that they are likely to be harmed from using the EHR system, they will perceive

it as a harm even if they do not view that harm as a serious problem. Second, Social influence on

using Hakeem had a positive direct effect on perceived threat. This means the more positive

influence about Hakeem the more users perceive it as a threat. This is a counter-intuitive finding

that contradicts the original theory which hypothesizes that social influence would have a

negative effect on perceived threat. This could be explained by the psychological reactance

theory (Brehm, 1966) which explains how psychological reactance occurs in response to threats

to freedoms and is used to deliberately taunt the authority who enforces these threats regardless

of the utility or disutility that the activity confers. Given that social influence includes the

opinions of healthcare providers’ superiors, then it is possible that healthcare providers manifest

there reactance behaviors in the shape of disliking and resenting the EHR system which was

reflected in their responses to perceiving Hakeem as a threat. We believe that this would require

further investigation. We suggest conducting interviews with users from our sample to better

explain this finding. Third, perceived cost of using the PBRS as a safeguarding measure had no

significant influence on perceived avoidability. When healthcare providers realize a threat

associated with the EHR system and begin to evaluate the possibility of using the PBRS as an

alternative way to accomplish their tasks, they would consider going back to the PBRS

127
regardless of how much trouble would be involved as long as the PBRS is perceived to be

effective in eliminating threats associated with using the EHR system. This is related to the loss

aversion concept from economics (Kahneman and Tversky, 1984) which suggests that people

tend to strongly prefer avoiding losses to acquire gains. When healthcare providers perceive the

EHR system to be a source of threat that will result in losses in their jobs, they would strongly

prefer avoiding it to eliminate these threats. However, healthcare providers find that enduring

costs related to using the PBRS is acceptable because it results in gaining a threat-relief situation.

This means that they valued the gains, in this situation, more than their losses. However, in

reality they are weighing the losses of using the PBRS much less than the losses stemming from

the threats associated with the EHR system. Forth, Perceived avoidability had no significant

impact on avoidance motivation. It is sufficient for healthcare providers to perceive the EHR

system as a threat to be motivated to avoid the system regardless of there was a known

safeguarding measure available for them to use. This implies having a non-usage behavior or a

resistance behavior where healthcare providers are motivated to simply not use the EHR system

before evaluating any other safeguarding measure such as using the PBRS. Our findings

indicates the limited explanation perceived avoidability has in this model. It serves as

endogenous independent variable rather than an exogenous predictor. Probably this explains why

studies that derived theoretical models from the TTAT have dropped perceived avoidability from

their model (Liang and Xue, 2010; Arachchilage and Love, 2013) and only tested the direct

effect of perceived avoidability antecedents on avoidance motivation. Fifth, the interaction effect

of perceived threat and perceived avoidability had no significant impact on avoidance

motivation. This confirms the former argument that healthcare providers are motivated by threats

alone to avoid the EHR system. Sixth, risk tolerance had no significant influence on perceived

128
threat. Barsky et al. (1997) suggest that individuals tend to show similar tolerant responses to

risky situations across settings. This implies that the no matter how much risk healthcare

providers are able to tolerate, it will not affect their perceptions of threats associated with the

EHR system. Evidence from the literature demonstrates that risk tolerance is a personal trait

related to demographic variables including age, gender, marital status, race, religion, education,

and income (Filbeck et al. 2005; Grable 2000; Hallahan et al. 2004). There is the possibility of

having some response bias in our sample in terms of the variables we did not control for (i.e.

marital status, race, religion, and income). Finally, we found that 3 of the control variables had

positive influence on avoidance behavior: number of employees in the unit, number of beds in

the unit, and age. However, gender, levels of education and position at the hospital had no effect

on the outcome variable. All other hypothesis in the original theory were confirmed. The model

explained (83%) of variance in perceived threat, (27%) in perceived effectiveness, (66%) in

avoidance motivation and (17%) in perceived avoidability. However, the main concern from

these findings is the lack of the model fit. In the following section we provide a suggested model

with the best fit indices.

Suggested model

In an aim to find an alternative model with better fit indices. We had to make appropriate

changes to the original model while still being aligned with the theory. We first dropped

perceived avoidability form the new model because it has no contribution in mediating the

effects of its antecedent on perceived avoidance. Next, we ran a modification indices test using

MPlus to evaluate what additional paths to the model would provide better fit. The new model

included the paths between perceived avoidability antecedents (perceived cost, perceived

effectiveness, and social influence) and avoidance motivation. We believe this is an appropriate

129
addition to the model since the mediator (perceived avoidability) no longer exists. Additionally

modification indices test suggested including the direct effect of social influence on avoidance

behavior. This is also an appropriate addition to the model since the control variable (number of

employees in unit) has a direct positive effect in avoidance motivation. The more employees in

the unit the more avoidance behaviors are reported. Thus, we believe social influence, including

influence from employees at the unit, would have a direct impact on avoidance behavior. Finally,

the modification indices test suggested adding the paths of the direct effects of avoidance

motivations’ antecedents (i.e. perceived effectiveness, perceived cost, and perceived threat) on

avoidance behavior. This also fall in line with the originally theory, it only suggests that there

would be a direct effect on avoidance behavior as well as an indirect effect through the mediator

perceived avoidance. The new suggested model results are shown in Figure 4.3 and the new

model fit indices are listed in Table 4.9.

130
 (*p<0.05; **p<0.01; ***p<0.001)

Figure 4.3: Suggested Model Results

Table 4.9: Suggested Model Fit Indices

Indicator Value
X2 121.932
df 22
P-Value 0.000
Calculated NC 5.5
CFI 0.903
RMSEA 0.104
90 CI 0.119 – 0.169
SRMR 0.021
The suggested model has improved model fit indices. NC within the recommended range,

CFI became (0.903) larger than the minimum cutoff (0.9), and SRMR is less than 0.05.

However, although RMSEA become much closer to the recommended range (0.0 – 0.1), it is still

131
a little bit higher than the maximum recommended value of. However, this would be due to a

number of reasons. First, the model has relatively low degrees of freedom, (Kenny et al, 2014)

argue to not even compute the RMSEA for low df models. Second, RMSEA is a parsimony-

adjusted index and its formula includes a built-in correction for model complexity (i.e. It favors

simpler models) (Hooper et al, 2008). Thus, the RMSEA would improve if the model had less

parameters. However, the suggested model is the most parsimonious while still being aligned

with the original theory. Finally, RMSEA is sensitive to sample size, (Hu and Bentler, 1999)

suggested that RMSEA does not perform well and is less preferable when the sample size is less

than 250. In conclusion, the overall model fit indices of the suggested model are considerably

better than the original model and shows a more reasonable fit.

The suggested model resulted in new significant relations between the constructs. First,

perceived threat has a positive direct effect on the avoidance behavior. This implies that the more

healthcare providers perceive the EHR system as a threat to their job the more they would avoid

the system. This is related to our argument on resistance behaviors. Healthcare providers are

willing to avoid the system without them knowing of any other safeguarding measure. This

would result in simply not using the EHR system. Second, perceived effectiveness had a direct

effect on both of avoidance motivation and avoidance behavior. Whereas perceived cost did not.

This confirms our former argument that healthcare providers are willing to use the PBRS if they

perceive it to be effective in eliminating the threats associated with the EHR system regardless of

the costs they may endure in the process. Finally, the suggested model shows that social

influence of using the PBRS had a positive effect on the avoidance behavior. However, there was

no significant direct effect from social influence about Hakeem on the avoidance behavior. This

implies that positive opinions about using the PBRS would influence healthcare providers to

132
avoid the EHR system. While positive opinions about using the EHR system does not influence

their decision of avoiding the EHR system. We can conclude that healthcare providers are

leaning more towards avoiding the EHR system and using the PBRS. Healthcare providers’

avoidance behaviors are not effected by what other people think of the EHR system nor how

much trouble it will cause to use the PBRS. On the other hand, in order increase healthcare

providers’ avoidance behaviors it is sufficient to perceive the EHR system as a threat, to perceive

the PBRS as an effective alternative, or to have positive opinions from others on using the PBRS.

The suggested model explained (81%) of the variance in avoidance behavior, (83%) in

perceived threat, and (76%) in perceived avoidability.

CONTRIBUTIONS

Contributions to Research

Our study makes several contributions to IS research. First, it tested the full conceptual

model of TTAT for the first time. This includes the effects of risk tolerance and social influence.

Our findings explains EHR avoidance phenomenon based on threat perceptions which are

effected by levels of susceptibility, severity, and social influence. Additionally, EHR avoidance

behaviors are determined by avoidance motivations which are effected by the effectiveness of

using the PBRS and levels of threats associated with the EHR system. We also found direct

effects of social influence of using the PBRS and the effectiveness of its use on the EHR

avoidance behavior. These findings contributes to the IS research which still requires

advancements in understanding EHR avoidance behaviors. Second, our model explained 81% of

the variance in avoidance behavior, this is much better than previous attempts (Liang and Xue,

2010; Arachchilage and Love, 2014) which explained at most 21% of the variance in the

outcome variable. Third, our study shows that TTAT could be adapted to explain technology

133
avoidance behaviors in different contexts. The TTAT first empirical test was conducted by

(Liang and Xue, 2010) to explain users’ avoidance of IT security threats such as malware and

cybercrimes, we find that its basic solid foundations are also applicable in the healthcare context

involving threats induced by using EHR systems. Fourth, we confirmed that perceived threat,

and social influence and effectiveness of the safeguarding measure have significant direct effect

on avoidance behaviors. Fifth, our study reveals a counterintuitive finding that positive social

influence about Hakeem has a positive direct effect on perceived threat. This interesting finding

should motivate future work to better investigate different possibilities surrounding this

phenomenon. Finally, the addition of control variables resulted in interesting findings.

Contributions to Practice

Our study provides a number of contributions to practice. First, because this study is

conducted in a developing country, which is still at the first stages of digitizing organizations, it

becomes very important to understand what predicts technology avoidance behavior and what do

new users consider as avoidable threats. Answers to these questions would help system

developers and system designers to build improved systems that overcomes sources of threat.

Second, understanding HIT avoidance and especially towards EHRs would help practitioners get

the appropriate support to reduce avoidance motivations. Governments in many countries have

largely invested in HIT systems. Thus, being able to explain causes of HIT avoidance would help

in finding solutions to make these systems more adopted and with higher success rates. Finally,

our work shows that TTAT could be applied successfully in different contexts. Thus, the theory

could be replicated using different technologies across industries. Findings would help

practitioners better understand avoidance in the context of which it was tested in, and that would

serve as the first step towards solving problems associated with avoidance behaviors.

134
 LIMITATIONS

This study has its limitations. First, in order to empirically test the model, we selected the

implications on job productivity and performances as the threats and the PBRS as a safeguarding

measure. This does not necessarily mean that threats are only directed on the productivity and

performances of employees, and that the only way to cope with a threat associated with an EHR

system is resorting to the PBRS. Threat could be related to different sources such as the design of

the system and the safeguarding measure could be manifested in different behaviors such as

delaying the use of the system. Research can be conducted with different threat sources and

safeguards to examine whether the findings of this study will change. Second, to fully understand

the positive effect between social influence on the use of the EHR system and perceived threat,

further investigations are required. It is interesting to study how social norms, social factors, and

image effect ones’ perception of threats associated with the EHR system. Third, although

TTAT’s conceptual model suggested the mediation effect of perceived avoidability, there was no

empirical evidence for its mediation and therefore it was dropped from our suggested model.

Future research may be directed towards testing perceived avoidability using antecedents other

than the ones related to the safeguarding measures (i.e. cost and effectiveness). Finally, because

our survey data is cross-sectional and collected from a single source, common method bias may

be a concern (Xue et al, 2014). Future research can apply a longitudinal study or collect data

from multiple sources to validate the findings.

135
 REFERENCES
Ajzen, I., 2006. "Constructing a theory of planned behavior questionnaire." Available at people.
umass. edu/aizen/pdf/tpb. measurement.pdf.

Arachchilage, Gamagedara, N., and Love, S., 2013. "A game design framework for avoiding
phishing attacks." Computers in Human Behavior 29, no. 3, 706-714.

Bhattacherjee, A., Davis C., Hikmet, N., and Kayhan, V., 2008. "User reactions to information
technology: Evidence from the healthcare sector." Association for Information Systems.

Bhattacherjee, A., Davis, C., and Hikmet N., 2013. "Physician Reactions to Healthcare IT: An
Activity-Theoretic Analysis." System Sciences (HICSS), 46th Hawaii International
Conference on. IEEE.

Bollen, K., 1998 "Structural equation models." Encyclopedia of biostatistics.

Brehm, J. W. (1966). A theory of psychological reactance.

Browne, M., and Cudeck R., 1993. "Alternative ways of assessing model fit." Sage Focus
Editions 154, 136-136.

Burnkrant, Robert E., and Cousineau, A., 1975. "Informational and normative social influence in
buyer behavior." Journal of Consumer research, 206-215.

Callen, J., Paoloni, R., Li, J., Stewart, M., Gibson, K., Georgiou, A., Braithwaite, J. &
Westbrook, J. 2013. Perceptions of the effect of information and communication
technology on the quality of care delivered in emergency departments: A cross-site
qualitative study. Annals of Emergency Medicine, 61, 131-144.

Carver, C. S. 2006. "Approach, Avoidance, and the Self-Regulation of Affect and Action,"
Motivation and Emotion (30), pp. 105-110.

Carver, C. S., and White, T. L. 1994. "Behavioral Inhibition, Behavioral Activation, and
Affective Responses to Impending Reward and Punishment: The BIS/BAS Scales,"
Journal of Personality and Social Psychology (67), pp. 319-333.

Clark, J. K., Wegener, D. T. & Fabrigar, L. R. 2008. Attitudinal ambivalence and message-based
persuasion: motivated processing of pro-attitudinal information and avoidance of
counter-attitudinal information. Personality and Social Psychology Bulletin, 34, 565-577.

Cordell, W. H., Overhage, J. M. & Waeckerle, J. F. 1998. Strategies for improving information
management in emergency medicine to meet clinical, research, and administrative needs.
Annals of Emergency Medicine, 31, 172-178.

136
Cotea, C. (2010). Electronic health record adoption: Perceived barriers and facilitators. Centre
for Military and Veterans' Health: The University of Queensland.

Deutsch, Morton, and Gerard H., 1955. "A study of normative and informational social
influences upon individual judgment." The journal of abnormal and social psychology 51,
no. 3, 629.

Edwards, J. (1992) "A cybernetic theory of stress, coping, and well-being in organizations,"
Academy of Management Review (17) 2, pp. 238-274.

Electronic Health Solutions (EHS) http://www.ehs.com.jo/en

Elliot, A. J. 2006. "The Hierarchical Model of Approach-Avoidance Motivation," Motivation

and Emotion (30), pp. 111-116.

Elliot, A. J., and Covington, M. V. 2001. "Approach and Avoidance Motivation," Educational
Psychology Review (13:2), pp. 73 92.

Gefen, D., D. Straub, and M. Boudreau (2000) "Structural equation modeling and regression:
guidelines for research practice," Communications of the AIS (4) 7.

Halamka, John. “Healthcare is different”, (2011)

http://geekdoctor.blogspot.com/2011/08/healthcare-is-different.html

Hare, K., Whitworth, B., Deek, F., and Norris, T., 2006. “Identifying Barriers to IT Adoption in
Clinical Health Care: A Pilot Study”, Health Care and Informatics Review Online, ISSN
1176-4201, http://hcro.enigma.co.nz/website/index.cfm

Hillestad, R., Bigelow, J., Bower, A., Girosi, F., Meili, R. & Scoville, R. 2005. Can Electronic
Medical Record Systems Transform Health Care? Potential Health Benefits, Savings, and
Costs. Health Affairs, 24, 1103-1117.

Hooper, Daire, Coughlan J., and Mullen M., 2008. "Structural equation modelling: Guidelines
for determining model fit." Articles, 2.

Hu, L.T. and Bentler, P.M. (1999), "Cutoff Criteria for Fit Indexes in Covariance Structure
Analysis: Conventional Criteria versus New Alternatives," Structural Equation Modeling,
6 (1), 1-55.

Hwang, Y., 2005. "Investigating enterprise systems adoption: uncertainty avoidance, intrinsic
motivation, and the technology acceptance model". European Journal of Information
Systems 14, no. 2, 150-161.

Kahneman, D. and Tversky, A. (1984). "Choices, Values, and Frames"(PDF). American

Psychologist 39 (4): 341–350. doi:10.1037/0003-066x.39.4.341.

137
Kane, G. C. & Labianca, G. 2011. IS Avoidance in Health-Care Groups: A Multilevel
Investigation. Information Systems Research, 22, 504-522.

Kellermann, A. L. & Jones, S. S. 2013. What it will take to achieve the as-yet-unfulfilled
promises of health information technology. Health Affairs, 32, 63-68.

Kelman, H. C. 1974. "Further Thoughts on the Processes of Compliance, Identification, and

Internalization, in Perspectives on Social Power,” J. T. Tedeschi (ed.), Chicago: Aldine,
pp. 126-171.

Kenny, David A., Kaniskan B., and McCoach D., 2014. "The performance of RMSEA in models
with small degrees of freedom." Sociological Methods & Research, 0049124114543236.

Kim, H.W., and Kankanhalli, A., 2009. "Investigating user resistance to information systems
implementation: a status quo bias perspective”, MIS quarterly 33, 3.

Lapointe, L., and Rivard, S., 2005. “A Multiple Model of Resistance to Information Technology
Implementation”, MIS Quarterly, 29, 3, pp. 461-491.

Liang, H., and Xue, Y., 2009. "Avoidance of information technology threats: a theoretical
perspective." MIS quarterly, 71-90.

Liang, H., and Xue, Y., 2010. "Understanding security behaviors in personal computer usage: A
threat avoidance perspective." Journal of the Association for Information Systems 11, no.
7, 394-413.

Markus M. L., 1983. “Power, Politics, and MIS Implementation”, Communications of the ACM,
26, 6, pp. 430-444.

Moore, G. C., and Benbasat, I. 1991. “Development of an Instrument to Measure the Perceptions
of Adopting an Information Technology Innovation,” Information Systems Research
(2:3), pp. 192-222.

Muthén, L. K., & Muthén, B. O. (1998-2011). Mplus User's Guide. Sixth Edition. Los Angeles,
CA: Muthén & Muthén.

Poon, E., Blumenthal, D., Jaggi, T., Honour, M., Bates, D.W., and Kaushal, R., 2004.
“Overcoming barriers to adopting and implementing computerized physician order entry
systems in U.S. hospitals”, Health Affairs 23, 4, pp. 184–190.

Pope, C., Halford, S., Turnbull, J., Prichard, J., Calestani, M. & May, C. 2013. “Using computer
decision support systems in NHS emergency and urgent care: ethnographic study using
normalization process theory,” BMC Health Services Research, 13, 1-13.

Rho, Ock, H., and Ryu I., 2011. "The impact of information technology threat on IT
appropriation and IT avoidance,” In Computational Intelligence, Communication

138
 Systems and Networks (CICSyN), Third International Conference on, pp. 195-200. IEEE,
2011.

Romanow, D., Cho, S. & Straub, D. 2012. “Riding the wave: Past trends and future directions for
health IT research,” MIS Quarterly, 36, iii-x.

Sassen, E.J. (2009). Love, hate, or indifference: How nurses really feel about the electronic
health record system. Computers, Informatics, Nursing, 27(5), 281–287.

Smith, Tim, Grant, G., and Ramirez, A., 2014. "Investigating the influence of psychological
ownership and resistance on usage intention among physicians”, System Sciences
(HICSS), 47th Hawaii International Conference on. IEEE.

Stream, G.R. (2009). Trends in adoption of electronic health records by family physicians in
Washington State. Informatics in Primary Care, 17(3), 145–152.

Thompson, R. L., Higgins, C. A., and Howell, J. M., 1991. “Personal Computing: Toward a
Conceptual Model of Utilization,” MIS Quarterly (15:1), pp. 124-143.

Venkatesh, V., M.G. Morris, G.B. Davis, and F.D. Davis, 2003. “User Acceptance of
Information Technology: Toward A Unifying View,” MIS Quarterly, 27, 3, pp. 425-478.

Xue, Y., Liang, H., Mbarika, V., Hauser, R., & Schwager, P. (2014). Understanding Healthcare
Professionals' Resistance of Telemedicine: an Empirical Study in Ethiopia. In PACIS (p.
225).

139
 APPENDIX
Measurement Items

Table 4.3: Measurement Instrument

Construct Item Wording Source
There is a good possibility that my job productivity and
PSUS1
performances are extremely less after using Hakeem.
There is a good possibility that the level of drop in my job
PSUS2
productivity and performances are great after using Hakeem.
Perceived There is a good possibility that my job productivity and
PSUS3 Liang and
Susceptibility performances have become less after using Hakeem.
I feel using Hakeem negatively affected my job productivity Xue, 2010
PSUS4
and performances.
It is extremely likely that my job productivity and
PSUS5
performances are negatively affected by using Hakeem.
Because I am using Hakeem, my job productivity and
PSEV1
performances are lower.
My job productivity and performances have dropped, as a Liang and
PSEV2
consequence of using Hakeem. Xue, 2010
Perceived
My job productivity and performance is lower as a result of
Severity PSEV3
using Hakeem.
Using Hakeem has negatively affected my job performance
PSEV4
and productivity.
I am not willing to take risks when choosing a job or a hospital
RIST1
to work for.
I prefer a low risk/high security job with a steady salary over a
RIST2
job that offers high risks and high rewards.
Risk I prefer to remain on a job that has problems that I know about Cable and
Tolerance RIST3 rather than take the risks of working at a new job that has Judge, 1994
unknown problems even if the new job offers greater rewards.
RIST4 I view risk of a job as a situation to be avoided at all costs. Judge, and
I am a cautious person who generally avoids risks. Thoresen,
RIST5 1999
People in my hospital who influence my behavior think that I
SOIH1
should use Hakeem.
People in my hospital who are important to me think that I
SOIH2
should use Hakeem.
I use Hakeem because of the proportion of coworkers who use
SOIH3
Social Hakeem.
Influence SOIH4 My managers have been helpful in the use of Hakeem.
(Hakeem) SOIH5 My supervisor is very supportive of the use of Hakeem. Venkatesh
SOIH6 In general, the hospital has supported the use of Hakeem. et al, 2003
People in my hospital who use Hakeem have more prestige
SOIH7
than those who do not.
SOIH8 People in my hospital who use Hakeem have a high profile.
SOIH9 Having Hakeem is a status symbol in my hospital.
Using Hakeem poses a threat to my job productivity and
PTHR1
performances.
Perceived The trouble caused by using Hakeem threatens my job
PTHR2
Threat productivity and performances.
Using Hakeem is a danger to my job productivity and Liang and
PTHR3 Xue, 2010
performances.

140
 Using Hakeem is dreadful to my job productivity and
PTHR4
performances.
Using Hakeem is a risk to my job productivity and
PTHR5
performances.
The paper based system would be useful for removing threats
PEFF1
associated with Hakeem.
The paper based system would increase my productivity and
PEFF2
performances by removing threats associated with Hakeem.
Perceived The paper based system would enable me to perform my job
Effectiveness PEFF3
faster. Liang and
PEFF4 The paper based system would make it easier to do my job. Xue, 2010
The paper based system would enhance my effectiveness in
PEFF5
removing threats associated with Hakeem.
PCOS1 Going back to the paper based system is difficult.
PCOS2 Going back to the paper based system is costly.
Perceived Going back to the paper based system is time consuming. Liang and
PCOS3
Cost Xue, 2010
I am not going back to the previous paper based system
PCOS4
because it is too much trouble.
People in my hospital who influence my behavior think that I
SOIP1
should use the paper based system.
People in my hospital who are important to me think that I
SOIP2
should use the paper based system.
I use the paper-based system because of the proportion of
SOIP3
coworkers who use the paper based system.
My managers have been helpful in the use of the paper-based
SOIP4
system.
Social
My supervisor is very supportive of the use of the paper-based
Influence SOIP5 Venkatesh
system.
(PBRS) et al, 2003
In general, the hospital has supported the use of the paper-
SOIP6
based system.
People in my hospital who use the paper-based system have
SOIP7
more prestige than those who do not.
People in my hospital who use the paper-based system have a
SOIP8
high profile.
Having the paper-based system is a status symbol in my
SOIP9
hospital.
For me to avoid Hakeem in the future would be:
PAVO1
(Very Difficult, Difficult, Neutral, Easy, Very Easy)
If I wanted to I could avoid Hakeem in the future
PAVO2 (Definitely false, Probably false, Neither True nor False,
Perceived Probably true, Definitely true) Ajzen, 2006
Avoidability How much control do you believe you have over avoiding
Hakeem?
PAVO3
(Far too Little, Too Little, About Right, Too Much, Far too
Much)
PAVO4 It is mostly up to me whether or not I avoid using Hakeem.
If I have access to the paper-based system, I intend to go back
AVMO1 to the paper based system to avoid threats resulting from using
Avoidance Hakeem.
Motivation If I have access to the paper based system, I predict I would go
AVMO2 back to the paper based system to avoid threats resulting from
using Hakeem.

141
 If I have access to the paper based system, I plan to go back to Liang and
AVMO3 the paper based system to avoid threats resulting from using Xue, 2010
Hakeem.
I avoid using Hakeem to mitigate my job threats associated
AVBH1
with Hakeem.
Avoidance I prefer to improve the previous paper based system rather
AVBH2
Behavior than switch to Hakeem. Liang and
AVBH3 I avoid using Hakeem in order to do my job well. Xue, 2010
AVBH4 I avoid Hakeem.

142
 Instrument Validation

Table 4.5: Items Loadings and CFA Model Fit Statistics

Std. Cronbach’s CFA Model Fit
Construct Item Mean STD
Loading α Statistics
PSUS1 0.947 X2:56.715
PSUS2 0.963 df: 5
P < 0.001
Perceived PSUS3 0.996 3.03 1.12 0.978 CFI: 0.877
Susceptibility
PSUS4 0.978 RMSEA: 0.217
CI: 0.168 - 0.269
PSUS5 0.861
SRMR: 0.015
PSEV1 0.991 X2: 1.238
PSEV2 0.988 df: 2
PSEV3 0.973 P > 0.05
Perceived
2.90 1.14 0.982 CFI: 1.00
Severity
RMSEA: 0.000
PSEV4 0.912 CI: 0.00 - 0.116
SRMR: 0.005
RIST1 0.949 X2: 26.074
RIST2 0.921 df: 6
RIST3 0.885 P < 0.001
Risk
RIST4 0.880 3.15 1.00 0.955 CFI: 0.961
Tolerance
RMSEA: 0.123
RIST5 0.900 CI: 0.077 - 0.174
SRMR: 0.081
SOIH1 0.841 X2: 829.697
SOIH2 0.760 df: 28
SOIH3 0.730 P < 0.001
Social SOIH4 0.915 CFI: 0.904
Influence SOIH5 0.929 3.68 0.70 0.836 RMSEA: 0.361
(Hakeem) SOIH6 0.922 CI: 0.340 - 0.382
SOIH7 0.710 SRMR: 0.443
SOIH8 0.854
SOIH9 0.874
PTHR1 0.970 X2: 429
PTHR2 0.971 df: 5
PTHR3 0.984 P < 0.001
Perceived
PTHR4 0.960 2.78 1.16 0.985 CFI: 1.00
Threat
RMSEA: 0.000
PTHR5 0.938 CI: 0.000 - 0.214
SRMR: 0.023
PEFF1 0.967 X2: 49.780
PEFF2 0.988 df: 5
PEFF3 0.884 P < 0.001
Perceived
PEFF4 0.824 3.07 1.03 0.963 CFI: 0.885
Effectiveness
RMSEA: 0.202
PEFF5 0.876 CI: 0.153 - 0.254
SRMR: 0.039
PCOS1 0.731 X2: 4.733
PCOS2 0.787 df: 2
Perceived
PCOS3 0.839 3.38 0.82 0.871 P > 0.05
Cost
CFI: 0.988
PCOS4 0.817 RMSEA: 0.079

143
 CI: 0.000 - 0.174
SRMR: 0.019
SOIP1 0.777 X2: 928.862
SOIP2 0.789 df: 27
SOIP3 0.775 P < 0.001
Social SOIP4 0.857 CFI: 0.985
Influence SOIP5 0.881 2.47 1.00 0.952 RMSEA: 0.390
(PBRS) SOIP6 0.867 CI: 0.368 - 0.411
SOIP7 0.829 SRMR: 0.102
SOIP8 0.846
SOIP9 0.847
PAVO1 0.815 X2: 16.441
PAVO2 0.840 df: 3
PAVO3 0.876 P < 0.001
Perceived
3.06 0.96 0.877 CFI: 0.930
Avoidability
RMSEA: 0.143
PAVO4 0.856 CI: 0.080 - 0.214
SRMR: 0.233
AVMO1 0.994 X2: 9.546
AVMO2 0.973 df: 2
P < 0.05
Avoidance
3.03 1.06 0.977 CFI: 0.966
Motivation
AVMO3 0.974 RMSEA: 0.131
CI: 0.056 - 0.219
SRMR: 0.011
AVBH1 0.934 X2: 7.107
AVBH2 0.929 df: 2
AVBH3 0.977 P < 0.05
Avoidance
2.81 1.29 0.971 CFI: 0.993
Behavior
RMSEA: 0.108
AVBH4 0.945 CI: 0.030 - 0.198
SRMR: 0.006

144
 CHAPTER FIVE:

UNDERSTANDING ELECTRONIC HEALTH RECORDS AVOIDANCE: A

REVEALED CAUSAL MAPPING APPROACH

ABSTRACT

We use the Revealed Causal Mapping (RCM) methodology to better understand the phenomenon

of Electronic Health Records (EHR) avoidance. Little research has aimed to explain avoidance

behavior towards technology in general and towards EHR in particular. We use the threat

avoidance feedback loop as a theoretical lens for this study. We revealed the main constructs

and the concepts shaping each of these constructs by analyzing data collected from interviewing

healthcare providers working at a large international hospital. The hospital has completely

implemented the EHR system and the system has been enforced to all healthcare providers. Our

findings show that healthcare providers decide to avoid the EHR system when they consider the

system to be a threat to their job and when there is an available option other than using the

system. Healthcare providers can avoid threats associated with the EHR system by conducting

problem-focused coping or emotion-focused coping. Contributes of our finding to both practice

and research are discussed.

145
 CHAPTER FIVE:

UNDERSTANDING ELECTRONIC HEALTH RECORDS AVOIDANCE: A

REVEALED CAUSAL MAPPING APPROACH

INTRODUCTION

The implementation of Electronic Health Records (EHR) can bring number of benefits to

healthcare providers. It helps them access patients’ medical history which helps them assess and

diagnose patients faster and more accurately (Pope et al, 2013). Additionally, EHR systems has

the potential of reducing medical errors (Hillestad et al, 2005). However, evidence show that

healthcare providers tend to avoid using the EHR systems when possible (Kane and Labianca,

2011). It is important to understand this phenomenon to help healthcare providers take better

advantage of all of the benefits EHR systems bring to the healthcare organization if used

meaningfully. IS research has very limited work in this area and findings are still mixed and

unclear (Kellermann and Jones, 2013). Prior research has mainly focused on EHR adoption (e.g.

Gan and Cao, 2014; Hewitt and McLeod, 2011; Hung, 2013; Jha et al, 2008; and Kemper et al,

2006), and some work focused on resistance (e.g., Markus, 1983; Poon et al, 2004; Lapointe and

Rivard, 2005; Bhattacherjee and Hikmet, 2008). However, limited attention has been given to IS

avoidance and specially EHR avoidance. Healthcare context has unique characteristics which

makes it difficult to generalize findings that explain avoidance towards technologies within

different industries to those technologies used in the healthcare industries. Furthermore, the

complexity of the avoidance behavior and ways of its manifestations does not allow us to simply

explain avoidance by lower levels of adoption. Thus, we argue for the need of a distinctive

theoretical perspective that aims to explain avoidance behaviors towards EHR systems. In this

study we aim to reveal the main concepts shaping the antecedents of the avoidance construct. We

146
base our study on the underpinning concepts of the threat avoidance feedback loop (Liang and

Xue, 2009. We conducted qualitative interviews with healthcare providers from a large public

hospital, located in Jordan, to capture their perceptions of the implemented EHR system. To

analyze their responses, we used Revealed Causal Mapping (RCM), a qualitative methodology

commonly used to identify constructs and linkages revealed from respondents’ statements

(Nelson et al., 2000). RCM has proven to be useful in studying emerging phenomena that require

rich and contextualized understanding (Narayanan and Armstrong, 2005). Given that our goal is

to understand the perceptions of individuals interacting directly with a newly implemented EHR

system, we believe that the RCM method is appropriate for our investigation of these

perceptions. We base our study on the conceptual process of threat avoidance presented by

(Liang and Xue, 2009).

THEORETICAL BACKGROUND

The process of threat avoidance explains individuals’ avoidance behaviors as a feedback

loop (see Figure 5.1). It posits three main stages. First, the threat appraisal; this is triggered when

users perceive a potential harm that has negative consequences in their environment associated

with the technology they are using. Once the harm is confirmed, users develop a perception of

their current state. They set being harmed as the undesired end state and compare it with their

current state. The assessment of threat depends mainly on how close both states are. The distance

between the current state and the undesired end state is inversely proportional to the strength of

the potential negative consequences (Liang and Xue, 2009). A threat is perceived when the

distance decreases to a certain value. Second, the coping appraisal; only when a threat is

perceived and confirmed users become motivated to actively avoid the threat and start to find

safeguarding measures that would help them avoid the threats associated with the system. The

147
threat must occur before the coping appraisal starts (Lazarus and Folkman, 1984). Users assess

available action options and decide what safeguarding measures to be taken to cope with the

threat. The assessment may be based on a number of factors such as the cost of the safeguarding

measure, the perceived effectiveness of the safeguarding measure, and the levels of self-efficacy

users believe they have in terms of adapting the safeguarding measure (Liang and Xue, 2009).

Third, the coping behavior; individuals can manifest two types of coping: problem-focused

coping and emotion-focused coping (Lazarus and Folkman 1984). Problem-focused coping is

performed by handling the source of the threat directly. It can be achieved by taking

safeguarding measures such as using the Paper Based Recording System (PBRS) instead of the

EHR system to avoid the threats associated with the EHR system. After taking the safeguarding

measure users perceive the distance between the current state and the undesired end state to be

larger, thus they perceive reduced threat. On the other hand, emotion-focused coping is related to

the psychological mindsets that users create to handle threats. It could be achieved by passively

creating false perceptions of the environment but without trying to actively change it. For

example, a common emotion-focused coping mechanism is to not think about the threat to

escape from the situation. Although users know that the threat is present, they would block it

from their rational thinking (Liang and Xue, 2009). If users fail to find a safeguarding measure

that can help them avoid the threat, they need to practice emotion-focused coping so that their

psychological well-being is maintained (Beaudry and Pinsonneault, 2005). When both appraisals

are complete users decide on a certain behavior (problem-focused coping or emotion-focused

coping) to be performed to avoid threats associated with the technology. These behaviors will

continue until the difference between the current state and the undesired end state are sufficiently

large so that the threat disappears (Liang and Xue, 2009).

148
 Figure 5.1: Threat Avoidance Feedback Loop

RESEARCH METHODOLOGY

We used RCM for this study, the same methodology as the one we used for the study in

chapter three. We also collected data from the same healthcare providers at the same hospital.

For information about the RCM methodology, the hospital and the participants please refer to the

methodology section of chapter three; (see Table 3.1).

Data Analysis

Our qualitative analysis method aims at extracting causal relationships from a

respondent’s explicit statements. We used a four-step process: (1) data elicitation, (2)

149
construction of revealed causal maps, (3) validation of the maps, and (4) interpretation of the

maps.

Step 1: Data Elicitation; we conducted the open-ended questions using Qualtrics. Table

3.2 in the Appendix of chapter three includes the questions used for this study Fifty-nine

respondents typed their responses online to the open-ended questions. In applying the RCM

method, researchers often rely on two major types of data: text-based data and interview data

(Narayanan and Armstrong, 2005). As our research objective was to reveal the cognitive

structure pertaining to the resistance behavior domain, it was appropriate to use open-ended

questions, similar to an interview approach (Nelson et al., 2000), and to analyze the narratives of

individual participants for their experiences, perceptions, and beliefs with regard to the new EHR

system.

Step 2: Construction of Revealed Causal Maps: following (Nelson et al., 2000), we

derived revealed causal maps of perceptions step by step. First, we identified causal statements

from an informant’s narrative by looking for key words such as “because,” “therefore,” “so,”

“while,” and “if–then.” Then, we coded each causal statement into cause and effect and replaced

the “cause” and “effect” in a statement with the concepts (the key words or phrases) and linked

“cause” and “effect” with an arrowed line to obtain a causal map at the concept level. We

highlighted repeated words and phrases and grouped them into constructs. Finally, we

aggregated all concept-level maps across all informants to obtain a causal map at the construct

level, Table 5.1 shows an example of the described process. In total, 11 concepts and 227

linkages were identified.

Table 5.1: An Illustration of the Four–Step Procedure for Constructing Causal Maps
Step 1: Identifying causal statement Example of causal statement:
“I think that data manipulation by hackers, it is serious
problem that I can think of it as a threat”

150
 Step 2: Constructing raw causal maps Cause:
1. Data manipulation by hackers.
Effect:
I can think of it as a threat
Step 3: Coding Raw Phase (Coded Concept)
1. Cyberattacks
2. Considered a threat
Step 4: Recasting raw causal maps into
concept-level revealed causal maps

Step 5: Creating a construct-level revealed

causal map

Step 3: Validation; we reviewed the avoidance literature to validate the concepts and

constructs identified in the RCMs. Similar to (Nelson et al., 2000), we used the point of

redundancy to evaluate the convergence of concepts elicited from the responses (Axelrod, 1976).

In our analysis, the 17 concepts converged at the 18th respondent--meaning that no new concept

was revealed in the map of the 19th respondent or beyond-- indicating the sufficiency of our

151
sample size. The additional relations yielded by adding the revealed causal map of each

individual respondent is measured by plotting a curve as shown in Figure 5.2.

Figure 5.2: Point of Redundancy

Step 4: Interpretation of the Maps; we examined the key constructs and linkages

uncovered in the map (see Figure 5.3). The constructs and their associations demonstrate what

influences people to avoid the new EHR system.

RESULTS

Our data analysis revealed six constructs that are essential to understanding avoidance

towards the EHR System (EHRS). Five are initially posited in the avoidance feedback loop

process: Perceived Harm, Threat Appraisal, Coping Appraisal, Problem-Focused Coping, and

Emotion-Focused Coping; Our analysis also revealed the underlying concepts associated with

each revealed construct. The list of constructs and concepts identified in the study with the

percentage of each of these concepts occurrence in relation to all of the other concepts within the

same construct are summarized in Table 5.2.

152
 Table 5.2: Revealed Constructs of EHR Avoidance and Its Concepts
Construct Definition Concepts/Dimensions Definition %
Perceived The awareness of the Technical Problems Any technical 0.14
Harm emergence of a problem healthcare
harmful event in the providers encounter
environment that when using the
results in becoming EHR.
closer to the undesired Potential Loss of Data The probability of 0.24
end state (anti-goal) losing medical data
(Liang and Xue, because of technical
2009) errors or any other
difficulty when
interacting with the
EHR.
Unnecessary Efforts Perceptions that 0.33
efforts required to
complete tasks using
the HER is not
needed and that the
task may be
completed without
these extra efforts.
Threat The sense of threat Lower Productivity Low effectiveness 0.67
Appraisal developed from of productive effort.
finding the current Evaluated in terms
state too close to the of the rate of
undesired end state accomplishing tasks
(anti-goal) per unit of input (i.e.
(Liang and Xue, patients).
2009) Cyber Attacks Attempts by hackers 0.33
to damage or
destroy the HER
system, the
computer it runs on,
the hospital’s
network, or the data
stored on the
system.
Coping The level of Ability to take appropriate The levels of
Appraisal motivation to engage safeguarding measures confidence in the 1.00
in a coping behavior to ability of using
avoid the threat by some other option to
increasing the accomplish the task
discrepancy between without the need to
the current state and use the EHR system.
the undesired end state
(anti-goal)
(Liang and Xue,
2009)
Emotion Cognitively regulating Accepting that the
Focused negative emotions Acceptance problem had 0.73
Coping arising from threat occurred but that
(Lazarus and nothing could be
Folkman1984). done about it (Tsai
et al, 2007).

153

Problem- Behaviors that directly
Focused address and/or change
Coping the problem causing
threat (Lazarus and
Folkman 1984).
Distraction Diverting attention 0.27
away from the
problem by thinking
about other things or
engaging in some
activity (Tsai et al,
2007).
Indirect Usage Using the EHR via 0.16
an alternative
system or person.
Delayed Usage Using the EHR after 0.37
a period of time by
which tasks become
late or postponed.
Partial Usage Using part of the 00.47
EHRS functionality
and resisting other
parts. For example:
using the EHRS to
retrieve data and
resist to enter the
data in the EHRS
forms.

Perceived Harm

When healthcare providers realize any harm associated with using Hakeem, they develop

a perception of their current state. They then set being harmed by using Hakeem as the anti-goal

(undesired end state) and compare it with their current state (Liang and Xue, 2009). Our study

shows that perceived harm influences healthcare providers’ perceptions of the threat associated

with Hakeem. The more harm was realized from using Hakeem the higher perceptions of threat

associated with Hakeem will be realized. We found three main dimensions of what healthcare

providers perceive as harm: Technical Problems, Potential Loss of Data, and Unnecessary

Efforts.

Technical Problems: responses revealed that one of the perceived harm definitions is

having to deal with difficulties related to technical issues such as frequent power outages. The

hospital has backup power generators that automatically regenerates power to medical equipment

and patients’ rooms in a case of a power outage. However, Hakeem is not connected to these

154
generators. Thus, the system will be ideal and healthcare providers will not be able to interact

with it until the general power source is recovered.

“Only when I face technical difficulties in the case I mentioned [power outage] it could take some time to
complete the forms after that” (Male, 33, Nurse, Dermatology).

Potential Loss of Data: the analysis revealed that healthcare providers consider the

potential of having the digital data lost is detrimental to their jobs. Although this applies to the

paper files as well, but none of the respondents revealed their concerns of having a paper file

being lost.

“…our increased dependency on software material…which causes the risk of information being lost…”
(Male, 24, Pre-Med, General Surgery).

Unnecessary Efforts: most of the responses identified the process of redundantly entering

patients’ data, and being asked unnecessary questions as impediments to their job. They

considered doing the same task repeatedly and being forced to answer questions that are

unrelated to the cases they are evaluating as a waste of their time.

“I wish you can be able to skip through unnecessary data… It keeps asking us the same questions every
time the patients is following up, sometimes things do not change and I wish we can skip through that, also
in some cases it asks us unnecessary questions that are not relevant to the case of the patient… we are
doing more unnecessary work now” (Female, 42, Nurse, Dermatology).

Threat Appraisal

Healthcare providers develop a sense of threat when they find that their current state is

too close to the anti-goal (i.e. harms associated with Hakeem) (Liang and Xue, 2009). The

evaluation of the threat based on the discrepancy between the safe current-state and the

dangerous end-state takes place within the threat appraisal stage. If a threat is confirmed,

healthcare providers will begin evaluating safeguarding measures within the coping appraisal.

We found 2 main threats identified by respondents: Lower Productivity, and Cyber Attacks.

155
 Lower Productivity: Healthcare providers’ responses revealed that the drop in their

productivity is a major contributor to their perception of Hakeem as a threat. Responses

explained productivity in terms of the number of files being completed per shift and the number

of scheduled appointments that were actually seen by the doctor.

‘’I find Hakeem a threat to my job productivity…I do not accomplish as many reports as I did using the
paper system’’ (Female, 29, Lab Specialist, Medical Lab).

“…It takes so long to fill in reports which is valuable time I could spend with other patients… I usually end
up seeing less patients” (Male, 38, Doctor, ER).

Cyber Attacks: healthcare providers revealed that a major threat of having patients’

records online via Hakeem is their fear of potential cyber-attacks. They explained how patient

data should be kept private and no unauthorized access should exist, and by having the records

online the privacy of patients is becoming compromised.

“What I consider a threat is related to having our data online which makes it easy for attackers [Cyber
Attacks/Hackers] to view and delete or do anything with the data” (Male, 24, Pre-Med, General Surgery).

Coping Appraisal

When healthcare providers confirm threats associated with Hakeem they begin to

evaluate appropriate safeguarding measures that would help in increasing the discrepancy

between their pre-identified safe status and the threat occurring from the use of Hakeem.

Healthcare providers would be motivated to engage in a coping behavior if the options they

considered are suitable, effective, feasible, and they are capable of performing related activities

associated with the safeguarding measure (Liang and Xue, 2009).

“I want a data entry specialist to interact with the system, doctors and nurses would only [implying should
be only required to] submit verbal or written requests…” (Male, 55, Doctor, Dermatology).

Problem Focused Coping

If healthcare providers believe that the threats associated with Hakeem could be avoided

be taking a safeguarding measure that they are capable of mastering, is effective, and is feasible,

156
then they would take the problem-solving approach to change their reality (Liang and Xue,

2009). Healthcare providers would deal directly with the source of the threat to eliminate it. We

revealed three main problem-focused approaches taken by healthcare providers to eliminate

threats associated with Hakeem: Indirect Usage, Delayed Usage, and Partial Usage.

Indirect Usage: Our analysis shows that healthcare providers may avoid Hakeem by

having someone else interact with the system on their behalf. For example, a doctor may ask his

appointed nurse to enter the data of the patient being checked by orally dictating the required

data to be entered.

“I usually ask our nurse to enter data into the system while I am checking on the patient… by doing this I
have my main focus on the patient not the computer.” (Male, 43, Doctor, ENT).

Delayed Usage: Our analysis revealed that some healthcare providers may aim to avoid

threats associated with Hakeem by procrastinating the usage of Hakeem. By doing so, they are

avoiding the threats temporarily.

“…I have Hakeem as a second choice with the old system [PBRS]. Today we cannot finish the patient visit
without finishing all forms and reports, this takes time and most of this information at my clinic is not
applicable because I care for kids. So I try to write all inputs on the patients’ folder and by the end of the
day we enter the data, which will make the flow of patients faster.” (Female, 46, Doctor, Pediatric Unit).

“I usually complete all the forms at the end of my shift. In this way I guarantee to have my time focused on
the patient during each visit…sometimes this means to have leftover work after the end of my shift” (Male,
Nurse, 31, ENT).

Partial Usage: we found healthcare providers who reported that they are using the system

partially by performing tasks they like and avoiding the ones they dislike. For example, some

healthcare providers revealed that they enjoyed using Hakeem when retrieving patients’ medical

history, whereas they considered entering the data into the system a hassle that needs to be

avoided.

“I like Hakeem because it helps in viewing all of the history of the patient quickly, I do not need to go and
search for files in a cabinet any more…entering data into the system takes so much time…I would like to
use the paper system when I need it…” (Male, 56, Doctor, ER).

Emotion Focused Coping

157
 If healthcare providers believe that the threat associated with Hakeem cannot be entirely

avoided by taking any safeguarding measure available to them, they would passively avoid the

threat by performing emotion-focused coping (Liang and Xue, 2009). Emotion-focused coping is

achieved by creating false reality of the environment without actually changing it. Our analysis

revealed two main dimensions of emotional focused coping performed by healthcare providers at

the hospital: Acceptance and Distraction.

Acceptance: this is identified when healthcare providers are aware of the threats but they

believe that nothing could be done about it (Tsai, et al., 2007).

“… I only fill in the forms as expected…we are following the hospitals policy and only using Hakeem”
(Female, 44, Nurse, Pediatric Unit).

Distraction: this is identified when healthcare providers divert their attention from the

threats by thinking about other things (Tsai, et al., 2007).

“I do not do anything different. I use the system without thinking of threats” (Male, 39, Pharmacist,
Outpatient Pharmacy).

The Aggregated RCM

158
 Figure 5.3: Aggregated Revealed Causal Map of EHR Threat Avoidance

Our findings revealed that the more healthcare providers perceive harms associated with

Hakeem, the more they would consider Hakeem as being a threat to their job.

“When we have technical problems with the system…this means my work will be delayed and I will be less
productive.” (Male, 40, Doctor, Urology).

The more Hakeem is considered as a threat the more healthcare providers will start to

evaluate safeguarding measures to seek elimination of the threats associated with Hakeem.

“…may be using a paper based system, then entering the data into Hakeem later… or may be make some of
the fields optional so we do not need to completely fill in the forms before submitting them…” (Female, 27,
Nurse, Dermatology).

“I would like to be able to send hard copies of the imaging when the system is down” (Male, 30,
Radiology, Imaging).

The result of the safeguarding evaluation influences healthcare providers’ decision of

which type of coping to engage in. If an appropriate safeguarding measure was selected then they

159
are more likely to engage in a problem-focused coping mechanism by objectively applying an

actual solution to eliminate the threats of Hakeem.

“I use both electronic and paper systems. If I have no time to fill in the data in the system or the patient
needs immediate care I only use the paper system” (Male, 33, Doctor, ER).

However, if no safeguarding measure was selected, or if healthcare providers believed

that threats will not be eliminated completely by any of the safeguarding measures, they would

engage in an emotion-focused coping mechanism.

“To cope with the threats…I follow instructions and complete the task as required” (Female, 24, Clinical
Assistant, Dermatology)

“I try not to see any threats” (Female, 27, Receptionist, Dermatology)

DISCUSSION

Our findings revealed a number of interesting insights about EHRS avoidance. First, we

found that healthcare providers aim to avoid the threat associated with the EHRS rather than

avoiding the EHRS itself. The emotion-focused coping mechanism indicates that healthcare

providers adapt to threats without actually avoiding the EHRS. Further, the avoidance behavior

is only triggered when a threat is realized. If no threat was confirmed then the likelihood of

avoiding the EHRS becomes too low. Second, we found that Hakeem’s design related issues is a

major contributor to healthcare providers perceptions of the EHRS. EHRS design was not

considered as a harm neither a threat, but rather as a source of negative feelings which healthcare

providers shape about using the EHRS. It is reported that Hakeem has a poor interface that is not

user-friendly. Also healthcare providers complained about how each form must be fully

completed before they can move to the next step, and in some cases the fields on the form are

perceived as irrelevant to the situation which causes irritation to the person required to fill in all

of the data into the forms. Another common issue reported was the access limitation to the

EHRS. The way Hakeem was designed made it only accessible from pre-identified computers

160
within the hospital. Healthcare providers are not able to access the EHRS remotely from outside

of the hospital or by using mobile devices even within the hospital. We believe that it is essential

to plan to eliminate these revealed design problems in order to mitigate avoidance behaviors.

Third, our analysis highlighted an interesting phenomenon related to avoidance; that is the

concept of “work arounds”. Prior work in healthcare defined work-arounds as "clever methods

for getting done what the system does not let you do easily" (Ash et al, 2003; Ash et al, 2004).

Similarly, (Kobayashi et al, 2005) defined work-arounds as "informal temporary practices for

handling exceptions to normal workflow". Additionally, (Morath and Turnbull, 2005) define

work-arounds as "work patterns an individual or a group of individuals create to accomplish a

crucial work goal within a system of dysfunctional work processes that prohibits the

accomplishment of that goal or makes it difficult". Healthcare providers aim to balance

technology and regulatory demands with the need to provide adequate care to patients, through

this process healthcare provider may see a greater need to improvise or work around intended

work practices (Halbesleben et al, 2008). Our study revealed that healthcare providers have in

common that they desire completing their tasks in the best possible way. They thrive to perform

their job using best practices. However, when using the EHRS becomes a barrier to them doing

their job as they believe it must be done, then EHRS would be avoided (work-around it). Our

study revealed that healthcare providers in the ER unit are the most to avoid the EHRS. Mainly

because the healthcare providers in the ER deal with very time-critical medical emergencies

more than all other providers at different units of the hospital. An example was reported by an

ER doctor who explained why he considers using PBRS when dealing with critical cases:

“…If we admit a person with a gun-shot wound or a stab wound which requires immediate care, and I sit
down and start typing information into Hakeem, people who are accompanying the patient will perceive
that as me being careless and only wasting time “playing” with a device rather than immediately devote
my attention to the patient” (Male, 38, Doctor, ER).

161
 In cases similar to the one reported here, it seems that healthcare providers agree that the

best practice would be to avoid the EHRS. Many times, the solutions devised to address a block

involve bypassing safety features in medical technology (Agency for Healthcare Research and

Quality (AHRQ, 2006). Finally, our analysis of responses did not reveal the linkage between

coping behaviors and threat appraisal. When healthcare providers engage in a coping behavior

they implicitly believe that the behavior is sufficient to reduce threats associated with Hakeem.

We also found that healthcare providers tend to engage in the same coping behavior for longer

periods of time after the threat has been realized. This allows them to worry less about the status

of the threat as they continuously engage in the same coping behavior which gives them more

comfort doing the task without the need to reevaluate the threat one more time. Thus, the linkage

between the behavior and threat appraisal was not revealed.

CONTRIBUTIONS

For Research

Our study has a number of implication to the IS research. First, the study provides

qualitative insights on the EHRS avoidance phenomenon which is not well understood in IS.

Second, our study revealed that when healthcare providers perceive Hakeem as a threat to their

job they begin to aim to avoid the threats rather than avoiding the EHRS itself. In the majority of

the cases this means to actually avoid using the EHRS itself by performing one of the three

revealed dimensions of problem focused coping behaviors. However, other cases shows that the

avoidance was achieved by performing emotional focused coping which implies the usage of the

EHRS rather than avoiding it. This gives a broader and more enriched understanding of how the

EHRS is being avoided and what is actually being avoided. Third, the constructs of the RCM has

some consistency with the avoidance feedback loop, however the interactions of these constructs

162
were not previously tested or confirmed. Our finding is the first study to explore the linkages

between constructs of the avoidance feedback loop. We did not find any evidence from the

responses that confirms the loopback from the coping stage to the threat appraisal stage. We

rather revealed that healthcare providers engage in a coping behavior and believe it is sufficient

to eliminate threats associated with Hakeem. Fifth, we provided the dimensions within all

construct influencing the EHRS threat avoidance behavior. This gives us a richer understanding

of each construct. Prior theorizations use these constructs abstractly with very limited insights

on the dimensions shaping each construct. Seventh, our findings is the first in IS to account for

all possible shapes of avoidance. Prior theories relied merely on the conceptualization of

avoidance as “not using”. Finally, our findings provided a model that can be the basis for

empirical testing in future research.

For Practice

Our study provides a number of implications to practice. First, we provide one of the few

qualitatively detailed explanations to the EHRS avoidance phenomenon. Our rich findings would

help system developers and system designers understand what users actually perceive when

interacting with the EHRS. Second, we revealed that one of the main contributors to avoidance

are the perceived threats associated with the EHRS. We also identified these threats. Therefore,

healthcare providers would be able to take advantage of the EHRS benefits by eliminating these

threats. Third, revealing main dimensions of harm and threats would help healthcare managers

identify these problems and eliminate them at early stages. Finally, revealing the main problem-

focused coping mechanisms would help healthcare managers better understand what healthcare

providers actually need to better accomplish their tasks.

LIMITATIONS AND FUTURE DIRECTIONS

163
Our study has a number of limitations. First, the sample size used for this study may be a

concern. We recommend that empirical studies on a larger scale be conducted to validate and

modify the RCM developed in this study. Second, because our data was collected from a single

source, validation concerns may arise. Future research can apply a longitudinal study or collect

data from multiple sources to validate the findings. Third, the dimensions of the constructs from

our study are limited to our sample. Further research is encouraged to explore other possible

diminutions that may emerge from different samples. Finally, responses analyzed in this study

did not show evidence of the linkage between the coping behaviors and the threat appraisal. We

encourage future research to attempt to capture this either by conducting a more extensive

qualitative interview or by developing appropriate survey items that can be used in an empirical

study of the avoidance phenomenon.

164
 REFERENCES
Beaudry, A., & Pinsonneault, A. (2005). Understanding user responses to information
technology: A coping model of user adaptation. MIS quarterly, 493-524.
Bhattacherjee, A., and Hikmet, N. 2007. "Physicians' resistance toward healthcare information
technologies: a dual-factor model." System Sciences, 2007. HICSS 2007. 40th Annual
Hawaii International Conference on, IEEE, 2007.
Bhattacherjee, Anol, Christopher J. Davis, Neset Hikmet, and Varol Kayhan, 2008. "User
reactions to information technology: Evidence from the healthcare sector." Association
for Information Systems.
Deng, X., and Chi, L. (2012). Understanding Post-adoptive Behaviors in IS Use: A Longitudinal
Analysis of System Use Problems in the Business Intelligence Context. Journal of
Management Information System, 29(3), 305-340.
Gan, Q., & Cao, Q., 2014. “Adoption of Electronic Health Record System: Multiple Theoretical
Perspectives,” System Sciences (HICSS), 47th Hawaii International Conference on
System Science, (pp. 2716-2724), IEEE.
Hewitt, B., & McLeod, A., 2011. “Modeling Security in Acceptance of Electronic Health Record
Systems,” Journal of Information Privacy and Security, 7(3), 23-45.
Hillestad, R., Bigelow, J., Bower, A., Girosi, F., Meili, R., Scoville, R., & Taylor, R. 2005. “Can
electronic medical record systems transform health care? Potential health benefits,
savings, and costs,” Health Affairs, 24(5), 1103-1117.
Hung, S. Y., Yu, W. J., Tsai, M. F., & Yen, D. C., 2013. “What Drives Physicians' Intention to
Use Electronic Medical Record System: The Roles of Perceived Service Level, Computer
Self-Efficacy, and Perceived Risk,” PACIS, (p. 194).
Jha, A. K., Doolan, D., Grandt, D., Scott, T., and Bates, D. W., 2008. “The use of health
information technology in seven nations,” International journal of medical informatics,
77, 12, pp. 848-854.
Kane, G. C. & Labianca, G. 2011. IS Avoidance in Health-Care Groups: A Multilevel
Investigation. Information Systems Research, 22, 504-522.

Kellermann, A. L. & Jones, S. S. 2013. What it will take to achieve the as-yet-unfulfilled
promises of health information technology. Health Affairs, 32, 63-68.

Kemper A.R., Uren R.L., Clark S.J., 2006. “Adoption of Electronic Health Records in Primary
Care Pediatric Practices,” Pediatrics, 118(1):20-24.
Lapointe, L., and Rivard, S., 2005. “A Multiple Model of Resistance to Information Technology
Implementation”, MIS Quarterly, 29, 3, pp. 461-491.

165
Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. Springer publishing
company.
Liang, H., and Xue, Y., 2009. "Avoidance of information technology threats: a theoretical
perspective." MIS quarterly, 71-90.

Liang, H., and Xue, Y., 2010. "Understanding security behaviors in personal computer usage: A
threat avoidance perspective." Journal of the Association for Information Systems, 11, no.
7, 394-413.

Markus M. L., 1983. “Power, Politics, and MIS Implementation”, Communications of the ACM,
26, 6, pp. 430-444.

Muthén, L. K., & Muthén, B. O. (1998-2011). Mplus User's Guide. Sixth Edition. Los Angeles,
CA: Muthén & Muthén
Narayanan, V.K.., and Armstrong, D. J. (2005). Causal Mapping for Research in Information
Technology. Hershey, PA: Idea Group Publishing.
Nelson, K.M., Nadkarni, S., Narayanan, V.K., and Ghods, M. (2000). Understanding Software
Operations Support Expertise: A Revealed Causal Mapping Approach. MIS Quarterly,
24(3), 475-507.
Poon, E., Blumenthal, D., Jaggi, T., Honour, M., Bates, D.W., and Kaushal, R., 2004.
“Overcoming barriers to adopting and implementing computerized physician order entry
systems in U.S. hospitals”, Health Affairs 23, 4, pp. 184–190.
Pope, C., Halford, S., Turnbull, J., Prichard, J., Calestani, M. & May, C. 2013. “Using computer
decision support systems in NHS emergency and urgent care: ethnographic study using
normalization process theory,” BMC Health Services Research, 13, 1-13.
Tsai, H. Y., Compeau, D., & Haggerty, N. (2007). Of races to run and battles to be won:
technical skill updating, stress, and coping of IT professionals. Human Resource
Management, 46(3), 395-409.

166
 APPENDIX A:

SURVEY FORMS

Cover Page

Welcome to the survey on attitudes towards the Electronic Health Record system

“Hakeem”. We thank you for participating and hope you will find this survey interesting and

useful. In this survey you will be asked questions about your beliefs and attitudes about Hakeem.

The purpose of this research is to help us understand attitudes and behaviors towards Health

Information Technology (HIT) in general and toward Electronic Health Records (EHR) in

particular.

PLEASE RESPOND TO THE SURVEY QUESTIONS BASED ON YOUR OPINIONS

AND FEELINGS AS AN EMPLOYEE AT THE HOSPITAL THAT RECENTLY DEPLOYED

HAKEEM.

On this survey, Hakeem refers to the Electronic Health Record system implemented

recently at the hospital. The survey takes approximately 30 minutes to complete. The findings

from this research will help us to develop more effective understanding about attitudes and

behaviors towards HIT.

Please understand that your answers are completely PRIVATE and CONFIDENTIAL.

No one other than the researchers will ever see your individual responses. This project was

reviewed and approved by Washington State University’s Institutional Review Boards. The

information in this consent form is provided so that you can decide whether or not you wish to

participate in this study. Your participation is considered voluntary. Even if you agree to

participate, you are free to withdraw from the study at any time. To do so, simply exit from the

Web Browser.

167
 In the event of any publication or presentation resulting from the research, no personally

identifiable information will be shared because your name is in no way linked to your responses.

Your confidentiality will be maintained to the degree permitted by the technology used.

However, no guarantees can be made regarding the interception of data sent via the Internet by

any third parties.

If you have any questions, complaints or concerns about this research, please contact

Bahae Samhan (bahae.samhan@wsu.edu). Your decision to be in this research is voluntary. You

can stop at any time. You do not have to answer any questions you do not want to answer.

You will be granted a certificate of completion for participating in taking this survey. At

the end of the survey you will be asked to enter your name as you want it to appear on your

certificate.

168
 Informed Consent

"I have read the preliminary description of this study and agree to participate. I

understand that there are no anticipated risks, and I am free to discontinue my participation at

any time without penalty.”

IMPORTANT: By clicking on the NEXT button below you are giving your consent to

participate in this study. If you do not give your consent, simply exit from the browser.

If you have already completed this survey in the past, then please STOP and do not complete this

one.

169