Rijksuniversiteit Groningen Faculteit

Student: Peter Reezigt

Studentnummer: 1656236

Titel afstudeeropdracht: Literature Study of Physicians’ Barriers towards Electronic Medical Record

Adoption

Maand en jaar: juni 2008

Literature Study of Physicians’ Barriers towards

Electronic Medical Record Adoption

By

Peter Reezigt

University of Groningen

CONTENTS

ABSTRACT

Objective: To identify all possible barriers and obstacles from physicians, medical specialists and general practitioners towards the use of EMR and synonyms, in the form of a literature study.

Data sources: Articles (2002 to 2008) which displayed studies that evaluated potential barriers from physicians, medical specialists and general practitioners towards the use of EMR and synonyms, which were searched in three databases. Articles were found in the following databases; Google Scholar, Science-Direct and EBSCO. Articles were selected based upon a search criterion which contained the introduced target group, one of the introduced Information Systems (EMR) and the following subject terms; barriers, resistance and obstacles towards EMR.

Method: Articles which were approved by the search criterion were read and analyzed with an extensive search on obstacles, barriers and other forms of resistance of physicians towards EMR use. Each barrier is presented and other references that referred to the same barrier were tallied. Each of the barriers was linked to a suitable so-called “resistance area”.

INTRODUCTION

Electronic Medical Records (EMRs) are computerized systems designed to electronically process medical records, and can contain medical images and billing information. They also work in conjunction with practice management systems, which handle schedules and billing (Moody, 2004; Miller & Sim, 2004). The eight core capabilities from an EMR, defined by the Institute of Medicine, include health information and data, result management, order management, decision support mechanisms, electronic communication and connectivity, administrative processes, reporting and patient support. These core capabilities display the opportunities which a typical EMR could bring for an organizational situation. There are various synonyms of EMRs, such as Electronic Health Records (EHR), Computerized Patients Records (CPR), Electronic Patient Files (EPF) and Personal Health Records (PHR). The advantages of EMR are well-recognized in literature. EMRs provide better support for electronic documentation and viewing, prescription and test ordering, care management reminders and efficient communication between other EMR functions. In addition, patients would benefit if those treating them had access to a comprehensive electronic record of their medical history (Boonstra, 2007; Øvretveit et al., 2007). Furthermore, EMR provide ways to faster communicate between physicians and results in fewer errors, all being stored on a digital component. Hospitals, pharmacies and general practitioners could share patients’ information and update each record when patients leave a hospital. Despite this positive overview, the adoption EMR is a constraining factor. A national survey in 2003 from the Commonwealth Fund displayed the fact that 27 percent of physicians are using these particular systems in the United States. This is supported by another survey which was taken in January 2005, among 1,061 members of the Medical Group Management Association. Twenty percent of physicians and specialists were using EMR. It has been estimated that it will take at least a decade for physician adoption to reach a general 80 percent (Versel, 2002). One of the problematic issues is the presence of barriers and resistance from physicians towards the adoption of EMR. This literature study supports a general view of main barriers by physicians, medical specialists and general practitioners towards the adoption of EMR and synonyms. The target group will be announced with the term “physicians. The term “EMR” will be used to generalize and represent EMR and the synonyms, which were introduced in this section. By analyzing articles, locating barriers from physicians towards EMR and linking them to a general area of resistance, this should support a clear overview of the current presence of barriers recognized in the field. The main purpose, thus, of this study is to:

 Review and analyze key articles on barriers towards EMR from a physician point of view;  Identify barriers and obstacles from physicians towards EMR;

 Identify general areas of resistance where barriers are related ;  Display references of researchers to each barrier;

 Provide a general overview of areas of resistance along with related barriers and references.

METHODS

The objective is to locate articles which introduce barriers, resistant behaviours and obstacles from physicians towards EMR. By searching on combinations of words, articles were found in EBSCO database, Science-direct database and Google scholar. These databases were selected because they are the only ones where I have access to. Each article was carefully selected based upon research criteria. This was accomplished by reading each article and using it for this research when the article was approved by the research criteria. When an article handled a different subject or an irrelevant perspective towards the EMR problem, the article was not used for this study. The criterion that was used for this selection included terms like the target group (physicians, medical specialist and general practitioners) who resisted EMR. If an article did not handle this subject, the article was repelled since it was not usable for the subject which is central in this literature study. In addition, to keep this study recent, only articles from the year 2002 till 2008 were selected1. Searching for articles in the search bar of Google Scholar with search terms resulted in the best hits, concerning the previously introduced research criteria. These key terms which were used to search on articles are presented in the nine search strategies displayed below. Google Scholar could not always provide the localization of an article. Therefore, in some cases EBSCO and Science-direct databases were used for localization. However, note that EBSCO and Science-Direct also provided search bars which were used to find articles. Some articles were not found by search strategies of words in databases, but by references of other articles. If these articles passed the search criterion they were used too and they were also a “hit”. Thus, note that an article is a hit when the article is approved by the search criterion. From all the performed search strategies, the following search strategies displayed articles which were approved by the search criterion, which were found by searching on key terms in databases’ search bars and therefore, could be seen as useful material for this study:

Search strategy 1: Keywords  Resistance + Electronic + Medical + Records  Google Scholar Search strategy 2: Keywords  EMR + EHR  Google Scholar

Search strategy 3: Keywords  Physicians + Electronic + Health + Records  Google Scholar Search strategy 4: Keywords  Physicians + Electronic + Medical + Records  EBSCO Search strategy 5: Keywords  Physicians + Barriers + EMR  Google Scholar

Search strategy 6: Keywords  Barriers + Electronic + Health + Records  Science-direct Search strategy 7: Keywords  Electronic + Medical + Records  Science-direct

Search strategy 8: Keywords  Physicians + Barriers + EHR  Google Scholar Search strategy 9: Keywords  EHR  Google Scholar

The following figure displays the pathway how articles were found and located by search strategies, databases and references from others. According to the figure the sample size contains 39 articles. Reasons for this size are time limitations and the last 8 articles did not introduce new barriers. The words displayed in each square contain the keywords combination. It was often the case that an article was referred to more than once. Since this is not relevant for this research, this is not displayed. The figure only presents the exact pathway on how articles were found, which were used in this literature study. See figure 1.

1

Figure 1: Pathway of selected articles

RESULTS

# _{Author(s) Year Method of findings Participants Main findings}

1 McDonald 1997 Opinion article - E1

2 Hier 2002 Opinion article - A5, E2, E4, F2, G1, F4

3 Hodge 2002 Opinion article - A1, C2, F1, F4

4 Assar 2002 Opinion article - C1, F2, G2

5 Winn 2002 Opinion article - A3, B2, D2, E1, E2, F2, G2, H3

6 Loomis et al. 2002 Mail survey 618 physicians A1, A6, D2, G1, H3

7 Waegemann 2003 Opinion article - A4, B1, E1, H4, G1

8 Wang et al. 2003 Cost-benefit analysis - H2, H3, H6

9 Blair 2004 International survey 7 countries B1, B3, E1, F3, H1, H3

10 Da’ve 2004 Opinion article - A3, B1, E2, F1, F3, G2, H3

11 Hier et al. 2004 Single case study 330 physicians A1, D2, E2, F4, G2

12 Miller & Sim 2004 90 interviews 30 clinics A3, A6, A7, C1, C2, E1, E2, F3, G2, H1, H2, H3, H4, H5

13 Moody 2004 Opinion article - A6, A7, B1, B3, G1, H2

14 Pizziferri et al 2004 Time-motion study 5 clinics A7, B1, E1, F1, G2, H3, H4, H6

15 Valdes et al. 2004 Survey 5517 AAFP

members B2, D2, F1, F2, H3, H7

16 Wald et al. 2004 Pilot study Unknown F1, F2

17 Ash & Bates 2005 Literature study - A1, A5, A4, B2, D2, F2, H2, H3, H6

18 Bar-Lev &

Harrison 2005

Single qualitative case study

B4, D1, F1, F2

19 Baron et al. 2005 Opinion article - A7, B2, H2

20 Bates 2005 Literature study - A1, A3, A5, A6, A7, B1, B2, B3, D2,

E1, F1, G2, H1, H3, H4, H5, H6

21 Clayton et al. 2005 Case study Several clinics A7, H2

22 Gans 2005 Survey Unknown A1, A3, A6, E1, H2, H3, H6

23 Gans et al. 2005 Survey 34,490 medical

groups

A1, B1, B3, D2, F2, F3, G3, H1, H2, H3

24 Meinert 2005 Mail survey 358 physicians A1, A3, A5, E4, F1, H3, G1

25 Shinn 2005 Essay - D2, E1, H3, H4

26 Stefan 2005 Literature study - A1, A3, A4, B2, B3, C1, D1, E1, E2,

F1, G2, H2, H3, H5

27 Weber 2005 Interviews Unknown A6, B1, C1, G2, H3, H4, H5

28 Anderson 2006 Literature study - A6, A7, B1, B2, D2, E1, E2, G2, H1,

H2, H3, H6

29 Boorady 2006 Opinion article - A3, D2, E4, F1, H3

30 Keshavjee 2006 Literature study - A3, A5, A7, B1, B2, C1, C2, D2, E1,

E2, E4, F1, F2, G2

31 Menachemi 2006 Mail survey 14921

physicians

A1, A6, D2, E1, E2, E4, F2, G1, G2, G3, G5, H1, H2, H3, H6

32 Pentecost 2006 Opinion article - A3, D2, E1, F2, H3, H7

33 Tang et al. 2006 Opinion article - A3, B2, D1, D2, F2

34 Mehta &

Partin 2007 Opinion article A3, B3, B4, D2, E4, F2, F3, G2, H1, H3

35 Pondrom 2007 Literature study A1, A3, C1, F3, G2, H2, H3, H6

36 Øvretveit et

al. 2007 Two case studies

Unknown and Swedish clinic

A1, A2, A3, A5, B4, C1, E2, E4, G2, G3, H3

37 Kemper et al. 2008 National mail survey 1000

physicians A1, A3, B1, B4, D2, E4, F1, F2, H1, H3

38 Smith 2008 Opinion article - F2

39 Withrow 2008 Opinion article - A1, E1, H4

There are in total eight areas found consisting of Human, Selection, Leadership, Legal, Technical, Procedural, Time and Financial. Each area has its own related barriers. Figure 2 presents each area and the amount of barriers each area has. Note that the areas Human and Financial have the most barriers consisting of seven barriers.

Human; 7

Selection; 4

Leadership; 2

Legal; 2

Technical; 4

Procedural; 4

Time ; 3

Financial; 7

Figure 2: Amount of barriers per area

Areas Barriers Authors’ references

A1) Lack of individual technical expertise  Withrow (2008, p90)  Bates (2005, p1182)  Loomis et al. (2002, p639)  Stefan (2005, p4)  Pondrom (2007, p4)  Gans et al. (2005, p1329)  Øvretveit et al. (2007, p183)  Meinert (2005, p501)  Hodge (2002, p19)  Kemper (2008, p22)  Menachemi (2006, p105)  Hier et al. (2004, p1302)  Gans (2005, p43)  Ash & Bates (2005, p10) A2) Lack of experimentation

and piloting abilities

 Øvretveit et al. (2007, p183)

A3) Individual / Cultural resistance  Øvretveit et al. (2007, p186)  Kemper (2008, p22)  Stefan (2005, p4)  Da’ve (2004, p51)  Keshavjee (2006, p5)  Mehta & Partin (2007, p828)  Winn (2002, p1)

 Boorady (2006, p318)

 Pondrom (2007, p4)  Gans (2005, p42)  Tang et al. (2006, p125)  Miller & Sim (2004, p121)  Bates (2005, p1186)  Meinert (2005, p501)  Pentecost (2006, p167)

A4) No personal benefits  Stefan (2005, p5)  Waegemann (2003,p2)

 Ash & Bates (2005, p10)

A5) No participation in selection / implementation process  Øvretveit et al. (2007, p183)  Keshavjee (2006, p5)  Bates (2005, p1182)  Hier (2002, p4)  Pondrom (2007, p4)  Ash & Bates (2005, p10)  Meinert (2005, p501) A6) Group size – adoption

rate

 Bates (2005, p1183)  Weber (2005, p7)  Moody (2004, p1)

 Miller & Sim (2004, p121)

 Loomis et al. (2002, p639)  Menachemi (2006, p105)  Anderson (2006, p481)  Gans (2005, p42) (A ) H u m a n

A7) Lack of incentives

 Miller & Sim (2004, p121)  Moody (2004, p1)  Clayton et al. (2005, p144)  Pizziferri et al. (2004, p183)  Bates (2005, p1183)  Keshavjee (2006, p10)  Baron et al. (2005, p225)  Anderson (2006, p481) B1) Difficulty to choose vendor / system  Bates (2005, p1182)  Moody (2004, p2)  Da’ve (2004, p51)  Keshavjee (2006, p5)  Blair (2004, p25)  Pizziferri et al. (2004, p183)  Weber (2005, p7),  Kemper (2008, p22)  Gans et al. (2005, p1329)  Anderson (2006, p481)  Waegemann (2003, p3) B2) Vendor’s support / product deliveries  Bates (2005, p1182)  Stefan (2005, p14)  Winn (2002, p1)  Ash & Bates (2005, p10)  Baron et al. (2005 ,p225)

 Valdes et al. (2004, p6)  Keshavjee (2006, p8)  Tang et al. (2006, p125)  Anderson (2006, p481)

B3) Lack of financial model / data from vendor

 Bates (2005, p1181)  Blair (2004, p25)

 Mehta & Partin (2007, p828)

 Moody (2004, p1)  Stefan (2005, p11)  Gans et al. (2005, p1329) (B ) S el ec ti o n

B4) Uncertain about quality improvements of EMRs

 Bar-Lev & Harrison (2005, p4)

 Øvretveit et al. (2007, p183)

 Kemper (2008, p22)  Mehta & Partin (2007,

C1) Lack of project champions / project management techniques

 Miller & Sim (2004, p120)  Weber (2005, p9)  Keshavjee (2006, p5)  Assar (2002, p2)  Øvretveit et al. (2007 p183)  Stefan (2005, p4)  Pondrom (2007, p4) (C ) L ea d er sh ip C2) Lack of management support

 Miller & Sim (2004, p121)  Keshavjee (2006, p4)

 Hodge (2002, p19)

D1) Loss of autonomy _ Bar-Lev & Harrison (2005,p4)_{Tang et al. (2006, p125)}  Stefan (2005, p4)

(D ) L eg a l

D2) Privacy & security issues  Bates (2005, p1182),  Valdes et al. (2004, p6)  Kemper (2008, p22)  Keshavjee (2006, p9)  Hier et al. (2004, p1302)  Gans et al. (2005, p1329)  Loomis et al. (2002, p639)  Mehta & Partin (2007, p829)

 Menachemi (2006, p105)  Winn (2002, p4)

 Tang et al. (2006, p125)  Anderson (2006, p481)  Ash & Bates (2005, p10)  Pentecost (2006, p167)  Boorady (2006, p318)  Shin (2005, p2)

E1) Interoperability issues / Lack of standards

 Withrow (2008, p90)  Miller & Sim (2004, p121)  Keshavjee (2006, p6)  Stefan (2005, p14)  Waegemann (2003, p2)  Menachemi (2006, p105)  Pizziferri et al. (2004, p183)  Gans (2005, p43)  Bates (2005, p1182)  McDonald (1997, p217)  Shin (2005, p2)  Winn (2002, p3)  Gans et al. (2005, p1329)  Anderson (2006, p481)  Pentecost (2006, p167)

E2) Complexity of system

 Miller & Sim (2004, p120)  Stefan (2005, p14)  Keshavjee (2006, p5)  Hier et al. (2004, p1302)  Øvretveit et al. (2007, p186)  Hier (2002, p4)  Menachemi (2006, p105)  Da’ve (2004, p51)  Winn (2002, p1)  Anderson (2006, p481) E3) Slow system speed  Hier et al. (2004, p1302)

(E ) T ec h n ic a l

E4) Unreliable technology

 Meinert (2005, p501)  Keshavjee (2006, p10)  Mehta & Partin (2007, p829)  Øvretveit et al. (2007, p183)  Menachemi (2006, p105)  Hier (2002, p4)  Boorady (2006, p318)  Kemper (2008, p22) F1) Decrease in accuracy, unreliability and productivity of work

 Bar-Lev & Harrison (2005,p4)  Meinert (2005, p501)  Stefan (2005, p5)  Keshavjee (2006, p5)  Boorady (2006, p318)  Wald et al. (2004, p1169)  Hodge (2002, p19)  Valdes et al. (2004, p6)  Kemper (2008, p22)  Bates (2005, p1182)  Pizziferri et al. (2004, p182)  Da’ve (2004, p51)

F2) Misfit with current working routines

 Bar-Lev & Harrison (2005, p4)

 Kemper (2008, p22)  Keshavjee (2006, p7)  Mehta & Partin (2007, p828)  Winn (2002, p1)

 Gans et al. (2005, p1329)  Tang et al. (2006, p125)

 Ash & Bates (2005, p10)  Menachemi (2006, p105)  Valdes et al. (2004, p6)  Smith (2008, p2)  Hier (2002, p4)  Pentecost (2006, p167)  Assar (2002, p2)  Wald et al. (2004, p1169) F3) Complex to migrate from paper-based to EMR  Blair (2004, p25)

 Mehta & Partin (2007, p828)  Miller & Sim (2004, p121)

G1) Time to enter data  Loomis et al. (2002, p639)  Menachemi (2006, p105)  Moody (2004, p2)  Meinert (2005, p501)  Hier (2002, p5)  Waegemann (2003, p2)

G2) Time to learn program / trainings

 Miller & Sim (2004, p120)  Bates (2005, p1182)  Menachemi (2006, p105)  Da’ve (2004, p51)

 Mehta & Partin (2007, p828)  Pondrom (2007, p4)  Assar (2002, p2)  Weber (2005, p7)  Øvretveit et al. (2007, p183)  Stefan (2005, p4)  Keshavjee (2006, p7)  Hier et al. (2004, p1302)  Winn (2002, p1)  Pizziferri et al. (2004, p182)  Anderson (2006, p481) (G ) T im e

G3) Time to select, contract & implement the system

 Øvretveit et al. (2007, p186)  Gans et al. (2005, p1329)

 Menachemi (2006, p105)

H1) Doubts about financial return of system

 Miller & Sim (2004, p120)  Kemper (2008, p22)  Gans et al. (2005, p1329)  Anderson (2006, p481)  Bates (2205, p1181)  Blair (2004, p25)  Menachemi (2006, p105)  Mehta & Partin (2007,

p828)

H2) High initial medical specialist time costs

 Miller & Sim (2004, p120)  Moody (2004, p2)  Menachemi (2006, p105)  Gans (2005, p43)  Clayton et al. (2005, p144)  Baron et al. (2005 ,p225)  Stefan (2005, p5)  Pondrom (2007, p4)  Gans et al. (2005, p1329)  Ash & Bates (2005, p10)  Anderson (2006, p481)  Wang et al. (2003, p398)

H3) High initial costs

 Miller & Sim (2004, p120)  Bates (2005, p1181)  Weber (2005, p7)  Moody (2004, p1)  Kemper (2008, p22)  Menachemi (2006, p105)  Da’ve (2004, p52)  Winn (2002, p1)  Øvretveit et al. (2007, p186)  Pizziferri et al. (2004, p184)  Wang et al. (2003, p398)  Mehta & Partin (2007, p828)

 Loomis et al. (2002, p639)  Meinert (2005, p501)  Valdes et al. (2004, p6)  Blair (2004, p25)  Stefan (2005, p5)  Shin (2005, p3)  Pondrom (2007, p4)  Ash & Bates (2005, p10)  Pentecost (2006, p167)  Anderson (2006, p481)  Boorady (2006, p317)  Gans (2005, p42) H4) Most of suspected

benefit goes to payers / consumers

 Miller & Sim (2004, p120)  Bates (2005, p1181)  Withrow (2008, p90)  Shin (2005, p3)  Weber (2005, p9)  Waegemann (2003, p2)  Pizziferri et al. (2004, p183) H5) Long timescale system’s

suspected payoff

 Weber (2005, p7)

 Miller & Sim (2004, p119)

 Bates (2005, p1181)  Stefan (2005, p5) H6) Costs to maintain the

system

 Bates (2005, p1183)  Menachemi (2006, p105)  Ash & Bates (2005, p10)  Pizziferri et al. (2004, p183)  Anderson (2006, p481)  Wang et al. (2003, p398)  Pondrom (2007, p4)  Gans (2005, p43) (H ) F in a n ci a l H7) Responsibility for purchase costs

According to figure 2, the areas Human and Technical had the most barriers. Figure 3 displays the number of references each area (and its related barriers) have. The leader is the area Financial with 65 references, closely followed by the area Human with 56 references. Note that it is quite interesting that the area Legal has 9 more references than the area Leadership since they both have equally divided barriers (2). This is related to the highly referred barrier “Privacy and Security”. The least addressed area according to this study is Leadership. See figure 3 for a graphical overview which contains the number of references to each area.

56 30 10 19 34 36 24 65 0 10 20 30 40 50 60 70 A. H um an B. S elec tion C. L eade rshi p D. L egal E. T echn ical F. P roce dura l G. T ime H. F inan cial

Figure 3: Amount of references per category

The last figure displays the amount of references per barrier. The barrier “High initial costs” with in total 24 references, is the most referred barrier according to this study. Barrier D2 is following with 16 references and barriers A3, E1, G2 and F2 closely follow with 15 references. See figure 4 for a graphical overview.

24 16 ₁₅ 14 12 ₁₁ 10 ₉ 8 ₇ 6 4 ₃ 2 ₁ 0 5 10 15 20 25 30 H3 D2 A3 E1 F2 G2 A1 F1 H2 B1 E2 B2 A6 A7 E4 H1 H6 A5 C1 H4 B3 F3 G1 B4 H5 A4 C2 D1 F4 G3 H7 A2 E3

Definition of barriers

To make each barrier clearer, a precise definition is required. In this section, each area along with its related barriers will be introduced and discussed according to several researchers and their findings.

Area A: Human

The “Human” area is primarily based on the human context towards EMR which is found in the literature. It contains factors such as human fears, physicians’ interpretations and cultural contexts. Seven related barriers were found according to the literature.

A1) Lack of individual technical expertise

Researchers argue that physicians have insufficient technical expertise to operate EMRs, which results in the rejection of these systems (Withrow, 2008; Bates, 2005; Loomis et al., 2002; Meinert, 2005; Stefan, 2005; Pondrom, 2007; Gans et al., 2005; Øvretveit et al., 2007; Hodge, 2002; Menachemi, 2006; Hier et al., 2004; Gans, 2005; Ash & Bates, 2005). Technical expertise can be translated to the experience and skill of a physician towards the use and knowledge of a computer and programs. In addition, Kemper addresses that most physicians have never used a computer and a mouse (Kemper, 2008). Since many researchers support this argument, Hier questions whether the average technical skill of physicians is indeed that limited to operate EMRs. In a survey among his employees who are users of an EMR system, he found that 16.8 percent of 191 house staff and 24.8 percent of 139 faculty staff agreed that a lack of computer skills was a barrier to use of the EMR. According to these contrary findings, physicians could have misinterpretations towards the system about their own abilities to operate one.

A2) Lack of experimentation and piloting abilities

EMRs are complex systems developed by several vendors. This diversity of vendors resulted in different products which all have the same functionality as a general EMR but come with different product names. Øvretveit et al. argue that physicians are unable to test such a system in the form of an experiment or a pilot. Since these options are not supported by vendors, physicians do not have the option to test these systems and to see if they are truly the solutions which the physicians had hoped for. The only option which remains is the limited information on the vendors’ website, and the ability for physicians to visit other clinics where EMRs have been implemented (Øvretveit et al., 2007).

A3) Individual / Cultural resistance

information systems and working practices. They find it often difficult to see the advantages of the introduction of a new system which they need to learn from the start (Smith, 2008). They do not like change, they do not see the benefit of it and they assume it will affect their informal working routines. Employees feel that change could result in a heavier workload and the need to train to learn the system, which could be a barrier to EMR adoption. Even when an EMR is being implemented, it means that physicians will have a parallel electronic and paper-based system during the transition period. Physicians are forced to switch between systems, thereby slowing workflow, requiring more time to manually enter data from external systems, which will lead to physicians' resistance to EMR use (Miller & Sim, 2004; Da’ve, 2004). According to Miller & Sim, it is even possible that practitioners still resist the actual use of the system when it is fully implemented. In their article, they argue that physicians still used paper processes which resulted in an extra cost of $20.000 per physician per year in the few practices that eliminated most paper processes (Miller & Sim, 2004). Thus, even if an EMR is implemented, it could be that it would not be used as it should be. In other situations, physicians may resist EMR since they have bad experience with previous ones which were implemented (Kemper, 2008; Meindert, 2005). The biggest obstacle is not the individual on its own, but the organizational culture in general. According to Stefan, an organizational culture that is entrenched in old habits, inflexible and autocratic will not support the environment needed for a successful EMR implementation. This mechanic culture is less able to change compared with the flexible and young organic culture (Stefan, 2005). Cultures can also be too flexible and academic. Since physicians all have a highly educated level, executives suspect that physicians will not need direct management support during the implementation process and the learning phase of the new system. Øvretveit et al argue that clinical leaders built programmes of excellence and expected a large degree of independence from their employees (Øvretveit et al., 2007).

A4) No personal benefits

Physicians feel there is a lack of personal benefits gained from moving to a computerized condition (Stefan, 2005; Waegemann, 2003; Ash & Bates, 2005). Waegemann argues that EMR systems require physicians to do more computer input and less hand writing. Many perceive writing short notes as easier and, in the short term at least, more cost-effective (Waegemann, 2003). This barrier is complementing resistance to change, since when physicians feel there is a lack of benefits gained, motivational behaviour is minimal which could result in resistance to change. Note that personal benefits consist of benefit which increases the working process and conditions for physicians. This barrier excludes financial rewarding (Barrier A7) which also can be seen as a personal benefit. Since both issues are often discussed separately in the literature, these aspects are therefore separate barriers.

A5) No participation in selection / implementation process

A6) Group size – adoption rate

The bigger the group of physicians that are working in a clinic, the bigger the chance this group will adopt the system. This is the group size – adoption rate. Many researchers argue that this gap is profound between large and small practices (Bates, 2005; Weber, 2005; Moody, 2004; Miller & Sim, 2004; Loomis et al., 2002; Menachemi, 2006; Anderson, 2006; Gans, 2005). Bates argues that the use of EMR is higher when more than fifty physicians will use the system. He found that only 13 percent of general practitioners use it which is far less than the 57 percent of physicians and medical specialists working in hospitals. Miller & Sim complement this finding by arguing that small groups of physicians lack technical expertise which they cannot share compared with a bigger group where more experts can share more information. In addition, The Commonweath Fund took a survey in 2005 with related outcomes compared with the previously described ones. They found that almost 60 percent of groups of 50 or more physicians were using EMR. That compared to only 25 percent of general practitioners who operates solo (Weber, 2005). Loomis et al. found that EMR nonusers were more often in a solo or small practice (two – six physicians) (Loomis et al., 2002). Moody argues that the financial sum is overwhelming small practices (Moody, 2004). These facts acknowledge that there is a group size – adoption rate present. Even vendors advise physicians to use the system with more people, since it then can become a more wise financial investment. Practices practiced below the 50 practitioners mostly resist EMR.

A7) Lack of incentives

When there is a lack of a financial reward system for physicians during the acceptation and implementation of an EMR, the use and quality of the system will decrease (Miller & Sim, 2004; Moody, 2004; Clayton et al., 2005; Pizziferri et al., 2004; Bates, 2005; Keshavjee, 2006; Baron et al., 2005; Anderson, 2006). When physicians will not benefit financially from the new system, they do not see the advantage of accepting and using it. Although this barrier is related to the barrier of no personal benefits, this barrier is being highlighted separately since it is often recognized by several researchers.

Area B: Selection

This area concerns uncertainty about who and what to select. Nowadays, there are many vendors who sell EMRs. People are uncertain about who to choose, what to choose and why to choose it. Along with the lack of data to make comparisons between vendors, the eventual support a vendor will give and the doubts whether an EMR should improve quality, people are uncertain when they are in the selection process.

B1) Difficulty to choose vendor / system

B2) Vendor’s support / product deliveries

A well-recognized barrier by several researchers is the support a vendor will give along with its product deliveries. Tang et al. argue that many vendors have not been financially successful in the market which has resulted in bankruptcy. In addition, this created a business climate that undervalues the potential for the future. New vendors have their fears about entering the market and more physicians find it harder to select a vendor since many are bankrupt and out of business (Tang et al., 2006). Besides this negative business climate, it is also possible that an installed EMR can become outdated within months (Winn, 2002) and vendors will not provide any technical support or updates (Ash & Bates, 2005). Other researchers found that physicians have fears whether a vendor should provide support, even when they say they will (Bates, 2005; Stefan, 2005; Valdes et al., 2004; Keshavjee, 2006; Tang et al., 2006; Anderson, 2006). Anderson found that 86 percent of physicians feel that vendor’s inability to deliver acceptable EMRs could be seen as a barrier towards the acceptance of an EMR. In addition, most vendors only built EMRs which can be used by larger firms. The reason for excluding smaller firms, is that larger firms have a higher potential of actually buying an EMR (Baron et al., 2005).

B3) Lack of financial model / data from vendor

There is a lack of financial data available which disables physicians to make comparisons between vendors’ offers of EMRs. This makes it difficult to choose the right vendor for the right situation, since each situation has is own characteristics (Bates, 2005; Blair, 2004; Mehta & Partin, 2007; Moody, 2004; Stefan, 2005; Gans et al., 2005). There is, however, data available which shows positive financial outcomes that physicians may find useful in their decision-making process. The real uncertainty comes from the lack of detailed data, which physicians can use to make comparisons between different vendor’s products. Physicians do not have proof whether this data is displaying the real situation. With this lack of proof, physicians found it hard to use this data to make important decisions (Bates, 2005). Even vendors acknowledge that they do not have proof or practical examples to make this data more reliable.

B4) Uncertain about quality improvements of EMRs

Since a majority of physicians are very interested to afford an ERM, there are some researchers who introduce the opposite. In some studies researchers found that physicians have their doubts about whether EMR should improve quality and productivity (Bar-Lev & Harrison, 2005; Øvretveit et al., 2007; Kemper, 2008; Mehta & Partin, 2007).

Area C: Leadership

Leadership issues are of great concern during the implementation of a new EMR project. It consists of product champions, project techniques and management support in order to have the highest positive outcomes. These issues are discussed in the next two barriers.

C1) Lack of project champions / project management techniques

al., 2007; Stefan, 2005; Pondrom, 2007). Project champions are the participants who truly believe that the introduced system will work. They will support it during the entire implementation process, because they see the advantages of the system and believe in what the system could bring. Their objective is to motivate participants and support them when needed. According to several studies, practices without the presence of physician EMR champions, may flounder in their efforts to generate quality or financial benefits from EMRs.

C2) Lack of management support

Besides leaders and champions, management should provide support during and after the implementation process. This is recognized in many other business fields. Practitioners should be supported when needed by top managers (Miller & Sim, 2004; Keshavjee, 2006; Hodge, 2002). In addition, it is also important to give a clear direction and to motivate less enthusiastic physicians. These factors are missing when managers are in the background during the implementation process of an EMR.

Area D: Legal

Physicians are obsessed with power of status and process speed. In addition, they find privacy and security issues important aspects. Two barriers are discussed next.

D1) Loss of autonomy

Physicians feel they will lose a significant part of their current autonomy by implementing an EMR (Bar-Lev & Harrison, 2005; Tang et al., 2006; Stefan, 2006). Reasons for this include easier accessible data, more physicians who share the same data and they cannot use their own paper-based methods anymore. Through these changes, physicians fear they will use their status of autonomy in their work. One reason for this is their work can be accessed by others.

D2) Privacy & security issues

Physicians have doubts whether EMR could provide security regulations and could protect privacy issues, regarding the patient information which is stored in EMR. Loomis et al. introduced an example of empirical research which supports the contrary and which shows that EMR are more secure and private than paper and notes. However, even with the fact that many physicians are aware of this proof, they still have serious concerns about the security and privacy an EMR should provide (Bates, 2005; Valdes et al., 2004; Kemper, 2008; Keshavjee, 2006; Hier et al., 2004; Gans et al., 2005; Loomis et al., 2002; Mehta & Partin, 2007; Menachemi, 2006; Winn, 2002; Tang et al., 2006; Anderson, 2006; Ash & Bates, 2005; Pentecost, 2006; Boorady, 2006; Shin, 2005). Bates suspects that electronic records can be better protected than working on paper. However, he notes that electronic breaches in the system could be more catastrophic with electronic records. Therefore, the selected EMR should provide good security regulations. Besides these disadvantages, it is recognized by the researchers in this section that working with EMRs provides more security to patient’s privacy.

Area E: Technical

E1) Interoperability issues / Lack of standards

Many argue that EMRs are technical incompatible with current information systems which are operative in hospitals (Withrow, 2008; Miller & Sim, 2004; Bates, 2005; McDonald, 1997; Keshavjee, 2006; Stefan, 2005; Waegemann, 2003; Menachemi, 2006; Pizziferri et al., 2004; Shin, 2005; Winn, 2002; Gans et al., 2005; Anderson, 2006; Pentecost, 2006). The problem is that the existing electronic data sources reside on many isolated islands with different structures and different code systems. This makes connectivity between the different systems a difficult option to realize. For example, Gans notes that 16.9 percent of his participants in his research addresses that EMR is not technical compatible with their practice billing system (Gans, 2005). In addition, there was a lack of data exchange between the EMR and the other clinical data systems (such as lab, radiology, and referral systems). Since the data was not always valid, physicians needed more time to check and correct the data. This point affected their working time which increased physicians' resistance to EMR use. Despite the difficulty, it is very attractive to achieve a technical fit between current information systems and the selected EMR, since benefits of this interoperability between Information Systems are profitable. This statement is introduced by Bates, where he argues according to a national analysis in the United States in 2005, that fully standardized interoperability in the U.S. could save the United States $77.8 billion annually.

E2) Complexity of system

Miller & Sim argue that EMR is too complex and has many capabilities and options which make it difficult to learn. They conducted between mid-2000 and the end of 2002, research concerning 90 interviews of physicians in 30 clinics that implemented an EMR. The research results showed that respondents found it difficult to use an EMR system, because it is very complex in the use of screens, navigation and options. This complexity issue is supported by other researchers (Stefan, 2005; Keshavjee, 2006; Hier et al., 2004; Øvretveit et al., 2007; Hier, 2002; Menachemi, 2006; Da’ve, 2006; Winn, 2002; Anderson, 2006).

E3) Slow system speed

It is important to note that physicians have different opinions based on the successfulness of the system compared with chiefs and directors. According to Hier, both institutions and physicians want superior patient safety, employees’ satisfaction and positive outcomes of the system for the organization. The physicians, however, are more interested in decreasing their time of current working practices, rather than decreasing operating and initial costs. Hier also mentions that physicians are obsessed with fast working procedures, so an EMR should enable this option (Hier, 2002). Hier et al. found that physicians feel that EMRs are too slow to use, so that this point becomes a barrier towards adoption (Hier et al., 2004).

E4) Unreliable technology

towards the adoption of EMRs by the suspected and unreliable conditions that physicians may have (Meinert, 2005; Keshavjee, 2006; Mehta & Partin, 2007; Øvretveit et al., 2007). Since the usability of EMRs are not always suspected to be that positive by physicians, vendors are slowly trying to improve this aspect. The problem is that vendors have doubts whether improvements and add-ons such as voice recognition, tablet computers or mobile computing will improve this resistance issue (Miller & Sim, 2004). Hier (2002) reports that voice recognition is not reliable, since it does not work like it should. Thus, even if there are doubts about whether add-ons would remove resistance, there are doubts whether these add-ons would work. These facts give rise to a high degree of uncertainty about the stability and reliability of the EMR systems.

Area F: Procedural

Physicians have concerns whether EMRs fit within their current working routines. Besides this suspected misfit, they also fear that EMRs could slow down their work and decrease their accuracy. In addition, they feel it is hard to translate the paper-based method to an electronic-based one. Since the new method requires more computers, physicians are not sure if there are enough computers available during work. These conditions address procedural barriers.

F1) Decrease in accuracy, unreliability and productivity of work

Physicians feel that EMR might have impact on their work process by slowing it down. Despite this suspected productivity decrease, physicians feel that working with EMR could result in less accuracy of their work, which results in less reliability of their working methods (Meinert, 2005; Keshavjee, 2006; Boorady, 2006; Wald et al., 2004; Hodge, 2002; Valdes et al., 2004; Pizziferri et al., 2004). Physicians feel that productivity will decrease, because they suspect that they will have a higher workload, since the new method requires more time compared with their old paper-based method (Kemper, 2008; Stefan, 2005; Da’ve, 2004; Bates, 2005). In addition, physicians feel that they are being required to “mindlessly” follow a predefined subset of standards related to the introduced EMR. Because of this, they feel they have less power in working methods and it will reduce their capacity and accuracy. These standardized methods enable physicians also to diagnose each patient as the same patient. Physicians often address that each patient needs different diagnosis and treatment. Therefore, a flexible, not-standardized system is a system that will better fit with their current working routines (Bar-Lev & Harrison, 2005). Instead of formal documented procedures, physicians require informal procedures since these are perceived to be less difficult, less time consuming and will not reduce creative approaches in the work process. Stefan addresses that informal procedures will achieve the same outcome as the formal procedure in health care (Stefan, 2005).

F2) Misfit with current working routines

unique workflows (Winn, 2002; Tang et al., 2006). Therefore, introducing EMRs will change current workflows since old workflows do not support the new technology (Stefan, 2005; Menachemi, 2006; Keshavjee, 2006; Asar, 2006; Wald et al., 2004, Mehta & Partin, 2007; Bar-Lev & Harrison, 2005; Gans et al., 2005; Menachemi, 2006; Smith, 2008). Finally, Hier argues that implementing an EMR requires high-quality well-maintained computer equipment. Since this is not often the case in hospitals and clinics, physicians should be aware of funding new equipment in order to let the EMR run (Hier, 2005). All these arguments are related to misfits with current workflows and EMR that physicians need to modify.

F3) Complex to migrate from paper-based to EMR

The transition process is very complex for most physicians as is seen for physicians as a barrier towards the acceptance of EMR. As introduced before, EMRs are not pre-installed systems that can be used “out of the box”. It requires complex, costly and time-consuming activities to create an effective EMR which has a good fit with current workflows in the clinic or hospital (Miller & Sim, 2004; Mehta & Partin, 2007; Blair, 2004; Da’ve, 2004; Pondrom, 2007). The presence of all the current paper files makes it difficult to migrate to an electronic system. Often files are missing or not migrated correctly to the EMR. With these facts, the migration is a complex and difficult task and it is too difficult to input historic medical data into the EMR (Gans et al., 2005). These issues are hard to realize and function as an obstacle to EMR use. Therefore, it is essential to provide management support, leadership roles and project techniques to support this issue.

F4) Lack of availability to computers

Physicians argue that most of the computers are not available when needed. Of course this is situationally dependent; still some researchers address this issue according to their findings (Hier, 2002; Hier et al., 2004; Hodge, 2002).

Area G: Time

Physicians are obsessed with time speed and fast workflows. Selecting, contracting, implementing and working with an EMR takes time. Especially after the implementation phase, where physicians need extra time to enter data and to learn the system through training.

G1) Time to enter data

Most EMRs require practitioners to carry out more computer input and less handwriting. As mentioned before, physicians perceive writing short notes as easier and, in the short term at least, more cost-effective. Physicians agree on the fact that it will take twice as long to enter data in the system (Loomis et al., 2002; Menachemi, 2006; Moody, 2004; Meinert, 2005; Hier, 2002; Waegemann, 2003).

G2) Time to learn program / doing trainings

where EMR was implemented in two hospitals. Physicians noted the following: “It was difficult to be able to prepare for EMR and at the same time do the ordinary work. In addition, staff had to work overtime to be able to do all that was expected of them… We needed better information about how much time and money we should have set aside in the department for this project”. Extra time needs to learn to program and to participate in training is a primary barrier for physicians’ adoption towards EMR use (Bates, 2005; Menachemi, 2006; Da’ve, 2004; Mehta & Partin, 2007; Pondrom, 2007; Assar, 2002; Weber, 2005; Stefan, 2005; Keshavjee, 2006; Hier et al., 2004; Winn, 2002; Pizziferri et al., 2004; Anderson, 2006).

G3) Time to select, contract & implement the system

Besides getting used to EMRs, it takes time to select, contract and implement the system. Physicians need extra time for these activities besides their normal current working practices. Øvretveit et al. argue that physicians complain about the lack of information they receive about how much time and money they should set aside for this project (Øvretveit et al., 2007). Along with the introduced barriers in the area Selection, it is difficult to decide how much time and money to spend on selecting and contracting an EMR form the wide range of vendors (Gans et al., 2005; Menachemi, 2006). Again, this should be clearly communicated from managers and project leaders, so uncertainty will decrease.

Area H: Financial

The most recognized area and its related barriers is the area Finance. Many researchers argue that physicians have doubts about the financial returns of the system and which actors will benefit from it. During the transition period, a decrease of productivity is normal which will result in less revenue. Along with the initial and maintenance costs, the long timescale of suspected benefits and deciding who is responsible for the costs, the area Finance is a serious area to consider.

H1) Doubts about financial return of system

Financial benefits of EMRs are hard to express and it is difficult to say when this suspected benefit could occur. Vendors promise the most positive outcomes but physicians still have their doubts whether these systems are that profitable. Because of this uncertain situation without any hard evidence, many physicians perceive this aspect as a barrier towards implementing EMRs (Miller & Sim, 2004; Kemper, 2008; Gans et al., 2005; Anderson, 206; Bates, 2005; Blair, 2004; Menachemi, 2006; Mehta & Partin, 207).

H2) High initial medical specialist time costs

H3) High initial costs

Initial costs include EMR hardware, software, installation, interfacing, training, space, maintenance, support and upgrades (Stefan, 2005). It is practically everything which is concerning financial effort. The cost of purchasing an EMR is approximately $32,600 per physician. Cost overruns are common and actual costs tend to exceed the vendor’s estimates by 25 percent (Mehta & Partin, 2007). This barrier is a serious and well-recognized issue from the literature towards physicians’ EMR adoption (Miller & Sim, 2004; Bates, 2005; Weber, 2005; Moody, 2004; Kemper, 2008; Menachemi, 2006; Da’ve, 2004; Winn, 2002; Øvretveit et al., 2007; Pizziferri et al., 2004; Wang et al., 2003; Loomis et al., 2002; Meinert, 2005; Valdes et al., 2004; Blair, 2004; Shin, 2005; Pondrom, 2007; Ash & Bates, 2005; Pentecost, 2006; Anderson, 2006; Boorady, 2006; Gans, 2005).

H4) Most of suspected benefit goes to payers / consumers

Most of the presented financial benefits introduced in articles contain savings in office supplies and administrative handling. However, some researchers argue that the main profit will directly affect payers and consumers in stead off physicians or the health care in general (Miller & Sim, 2004; Withrow, 2008; Shin, 2005; Waegemann, 2003). Bates and Weber argue that patients would spend less money on health care issues, because they profit from the more efficient processes. For example, this increased efficiency results lower drug costs for patients (Bates, 2005; Weber, 2005). Pizziferri et al. argue that patients, institutions, and payers stand to benefit from EMRs, while physicians are paying with extra time (Pizziferi et al., 2004). These extra time issues were introduced in the area “Time”.

H5) Long timescale system’s suspected payoff

It also has been argued according by medical directors that health care systems bring besides high costs also a long timescale related to the financial pay off of the system. This timescale could present a period of 24 months (Weber, 2005; Bates, 2005). This high timescale of suspected payoff leads to a barrier for physicians. Since physicians are obsessed with speed and stability, they would see the payoff within several months (Miller & Sim, 2004).

H6) Costs to maintain the system

Maintenance issues are another barrier towards EMR adoption. Medical specialists lack the time and expertise to maintain systems, and providers charge high costs when providing this service. The most common ongoing costs include product support, hardware replacements and providing maintenance trainings (Stefan, 2005). This is another financial effort where a particular clinic or hospital is responsible for and is often cited as a main barrier (Bates, 2005; Menachemi, 2006; Ash & Bates, 2005; Pizziferri et al., 2004; Anderson, 2006; Wang et al., 2003; Pondrom, 2007; Gans, 2005).

H7) Responsibility for purchase costs

DISCUSSION

Based on the previous introduced research objectives, they key of this research was to locate area’s of resistance and related barriers from physicians towards the use of EMRs. Each barrier was located through an extensive search action in three databases. A search criterion was used in order to keep the scope of this research focussed. Seven areas along with 33 barriers were found among 39 selected articles. These consist of Human (7), Selection (4), Leadership (2), Legal (2), Technical (4), Procedural (4), Time (3) and Financial (7). The top 5 barriers with the most related references to it are “High initial costs” (24), “Privacy and security issues” (16), “Misfit with current working routines” (15), “Individual / Cultural resistance” (15) and “Time to learn program / doing training” (15). A discussion point is the estimation of the sort “value” which can be linked to the amount of references each barrier has. Mostly, an author’s reference is related to an empirical research outcome or an expert opinion. It is hard to question which is more valid, since empirical research outcomes are always linked to a particular period and outcomes can differ when measured later. In contrast, an expert opinion is subjective and mostly related to a particular situation. It is therefore quit noticeable that each barrier was found based on different (research) methods.

If we assume that the amount of references per barrier is equal to the importance of it, the barrier “High initial cost” (24) is the winner. It is reasonable to think this barrier is important and relevant to many situations, since purchase costs of EMRs are high and product outcomes are unsure. Physicians are not sure whether an EMR will be profitable and improve quality and productivity, which is supported by the barriers B4 and H1 (“Uncertain about quality improvements of EMRs” (4) & “Doubts about financial return of system” (8)). In addition, there are also many vendors who sell many different systems which can lead to more uncertainty what to choose (B1 “Difficulty to choose vendor / system” (11)). Another complementing barrier could be “Time to learn program / doing trainings” (15). Besides the actual purchase and maintenance costs, spending time on learning new technology is a costly effort too. If we also take the barriers “Most of suspected benefit goes to payers / consumers” (7) and “Responsibility for purchase costs” (2) into account, it should be clear that these factors give rise to a high degree of uncertainty. Since physicians are obsessed with speedy workflows and stable working conditions, this degree of uncertainty and insecure conditions is not reinforcing their requirements.

these advantages and had a more positive attitude. It may seem obvious since EMR users actually have experience in practice with EMRs. Despite this contrary results, the major popularity of the respondents (77,6%) expressed an interest in an digital solution like an ERM system which could provide the ability to connect al the medical locations in to one centralized computed system (Loomis et al. 2002). The Chasm Theory could be an explanation why EMR is still not fully adopted yet, since this system is still in the so-called birth phase; it is new technology and the mainstream market needs to accept it.

EMRs are expensive products with different, interpreted outcomes by others. The adoption of something new is highly related to people opinions, interpretations and beliefs. This can results in many different perspectives, opinions and beliefs towards an EMR, which is difficult to handle for change managers. This also could be an explanation why the barrier “Individual / Cultural resistance” (15) is referred many times. People often do not like a changing environment. The outcomes of an EMR are quit unsure for physicians, since there is a lack of financial models to compare vendor’s products. Even when these models are present, physicians do not see them as reliable sources. In addition, physicians question the vendor’s support. These two issues are supported by the barriers B2 and B3 (“Vendor’s support / product deliveries” (9) & “Lack of financial model / data from vendor” (6)). One reason why some barriers are more referred to others could that they are related and sometimes complement each other. This relationship is discussed in the previous section. We also need to take into account that some articles are written with a particular purpose. Some author’s only focussed on particular problems and did not take other barriers into account. This does not mean that authors do not agree with other barriers that were found. It could be that these barriers were out of their research scope. Another reason could be that the barriers which are less referred are too small and practically unimportant to discuss. It is quit noticeable that the most referred barriers are more abstract and less detailed. For example, compare “Interoperability issues / Lack of standards” (15) which is broad, with the barrier “Slow system speed” (1) which is more detailed.

This research could be used as a basic framework to research particular barriers in a wider context. Furthermore, each barrier could be empirically tested and researchers could expand this framework with even more barriers, if these are present. It could also function as a theoretical input to design a diagnose tool to evaluate whether barriers are present in different situations. Practitioners can use this research as a theoretical source, which is related to practical situations. This source could create awareness of new barriers and they can use this material as a source to create interview questions. It could also make them aware of the potential presence of barriers in their clinics and hospitals. The following lessons are learned:

 Area’s of resistance and barriers could be related

 Physicians work in an academic culture and are obsessed with speed and stable working conditions  Since many will benefit from a shared EMR it is hard to select who is responsible for initial costs  Empirical research indicates that EMRs are not well suited for small practices

 EMR does not fit technically and procedurally with current working practices  The slow acceptance of EMR could be related to the Chasm Theory

CONCLUSION

REFERENCES

Ash, J.S. and Bates, D.W. (2005) Factors and forces affecting EHR system adoption: report of a 2004 ACMI discussion. J Am Med Inform Assoc 12(1), 8-12.

Assar, K. (2002) The Need for Electronic Medical Records in Primary Care. Avaible at: http://www.medscape.com/viewarticle/446702

Anderson, J.G. (2007) Social, ethical and legal barriers to e-health. International Journal of Medical Informatics 76, 480-483.

Bar-Lev, S., Shirly, H. and Michael, I. (2005) Localization of Health IT: How Users “Repair” Electronic Medical Record Systems. Academy of Management Proceeding, 1-5.

Baron, R.J., Fabens, E.L., Schiffman, M. and Wolf, E. (2005) Electronic health records: just around the corner? Or over the cliff? Ann Int Med 143, 222-226.

Bates, D. W. (2005) Physicians and ambulatory electronic health records. Health Aff 24(5), 1180-1190.

Blair, J. (2004). Survey of Electronic Health Record Trends and Usage for 2004. EHR Summit III. Medical Records Institute. Available at:

http://www.medrecinst.com/conferences/seminar/july04/proceedings/pdfProceedings/JB.pdf. Accessed on 10/5/2004.

Boonstra, A., Boddy, D., Bell, S. (2008) Stakeholder management in IOS projects: analysis of an attempt to implement an electronic patient file. European journal of information systems 17(2), 100-111.

Boorady, J. (2006) The EHR fear factor. Journal of the American Optometric Association 77(6), 317-318. Clayton, P.D., Narus, S.P., Bowes, W.A., Madsen, T.S., Wilcox, A.B., Orsmond, G., Rocha, B., Thornton, S.N., Jones, S., Jacobsen, C.A., Udall, M.R., Rhodes, M.L., Wallace, B.E., Cannon, W., Gardner, J., Huff., S.M. and Leckman, L. (2005) Physician use of electronic medical records: issues and successes with direct data entry and physician productivity. AMIA symposium, 141-145.

Da’Ve, D. (2004) Benefits and Barriers to EMR Implementation. Caring 11, 50-51.

Gans, D. (2005) Off to a slow start. Available at: www.rpiusa.com/Con_05_Oct_slow_start%5B1%5D.pdf. Gans, D., Kralewski, J., Hammons, T. and Dowd, B. (2005) Medical groups’ adoption of electronic health records and information systems. Health Aff 24, 1323-1333.

Hier, D. B. (2002). Physician buy-in for an EMR. Healthcare informatics. Available at: http://www.healthcare-informatics.com/issues/2002/10_02/commentary.htm

Hier, D. B., Rothschild, A., LeMaistre, A., and Keeler, J. (2004) Differing faculty and housestaff acceptance of an electronic health record one year after implementation. Medinfo 11, 1300-1303.

Hodge, R. (2002). Myths and realities of electronic medical records, The Physician Executive, 14-19.

Kemper, A.R., Uren, R.L. and Clark, S.J. (2006) Adoption of Electronic Health Records in Primary Care Pediatric Practices. Pediatrics 118(7), 20-24.

Keshavjee, K., Bosomworth, J., Copen, J., Lai, J., Kucukyazici, B., Lilani, R. and Holbrook, A.M. (2006) Best Practices in EMR Implementation: A Systematic Review. Available at:

Loomis, G, A., Ries, J. S., Saywell, R. M. and Thakker, N. R. (2002). If electronic medical records are so great, why aren’t family physicians using them? The Journal of Family Practice, 51(7), 636-641.

McDonald, C.J. (1997) The barriers to electronic medical record systems and how to overcome them. J Am Med Inform Assoc 4, 213-21.

Mehta, N.B. and Partin, M.H. (2007) Electronic health records: A primer for practicing physicians. Cleveland Clinic Journal of Medicine 74(11), 826-830.

Meinert, D. B. (2005) Resistance to electronic medical records (EMRs): A barrier to improved quality of care. Issues in Informing Science and Information Technology 2, 493-504.

Menachemi, N. (2006) Barriers to ambulatory EHR: who are ‘imminent adopters’ and how do they differ from other physicians? Informatics in Primary Care 14, 101-8.

Miller, R.H. and Sim, I. (204) Physicians' Use of Sectronic Medical Records: Barriers and Solutions. Health Affairs 23(2), 116-126.

Moore, G. (1999) Crossing the Chasm: Marketing and Selling High-Tech Products to Mainstream Consumers. Rev ed. New York: Harper Business

Moody, R. J. (2004) Local doctors slow to adopt records technology. The Business Journal. Available at: http://www.bizjournals.com/portland/stories/2004/04/26/story7.html

Øvretveit, J, Scott, J.T., Rundall, T., Shortell, S. and Brommels, M. (2007) Implementation of electronic medical records in hospitals: two case studies. Health Policy 84(2) 181-190.

Pentecost, M.J. (2006) Electronic Medical Records, Chapter 2: The Obstacles. Journal of the American College of Radiology 3(3), 167-168.

Pizziferri, L., Kittler, A.F., Volk, L.A., Honour, M.M, Gupta, S., Wang, S., Wang, T., Lippincott, M., Li, Q. and Bates, D.W. (2005) Primary care physician time utilization before and after implementation of an electronic health record: A time-motion study. Journal of Biomedical Informatics 38, 176-188.

Pondrom, S. (2007) Implementing an Electronic Medical Records System - You Can Do It. ENToday 2(1), 4. Shin, P. (2005) Electronic Health Records: The Benefits of/Barriers to HER and Implications for health Librarianship. Available at: http://www.sla.org/division/dpht/Annual2005/EssayProject--PeggyShin.doc Smith, J. (2008) Electronic Health Records and Practice Management Solutions: All-in-One versus Best-of-Breed. Managed care outlook 21(4), 1-5.

Stefan, S. (2005) Barriers and Strategies for Successful EHR Implementation. Available at:

http://66.102.1.104/scholar?hl=nl&lr=&q=cache:MTEeU0XeQpYJ:www.sayers.com/Healthcare/White%2520Pa per%2520%2520Barriers%2520and%2520Strategies%2520for%2520Successful%2520EHR%2520Implementati on%252010-10-2005.pdf+physicians+barriers+ehr

Tang, P., Ash, J., Bates, D., Overhage, M. and Sands, D. (2006) “Personal Health Records: Definitions, benefits and strategies for overcoming barriers of adoption”. Journal of American Medical Informatics Association. 13(2). pp.121-126.

Valdes I., Kibbe, D.C., Tolleson, G., Kunik, M.E. and Petersen, L.A. (2004) Barriers to proliferation of electronic medical records. Informatics in Primary Care 12(1), 3-9.

Versel, N. (2002). Stalking the elusive ‘Killer App’. Modern Physician, 6(5). Available at: http://modernphysician.com/articles:cms?articleID=148

Wald, J.S., Bloom, A., Walmsey, D., Gleason, M., Nelson, E., Li, Q., Epstein, M., Volk, L. and Bates, D. W. (2004) A Patient-Controlled Journal for an Electronic Medical Record: Issues and Challenges. MEDINFO, 1166-1170.

Wang, S. J., Middleton, B., Prosser, L. A., Bardon, C. G., Spurr, C. D., Carchidi, P. J., Kittler, A. F., Goldszer, R. C., Fairchild, D. G., Sussman, A. J., Kuperman, G. J. and Bates, D. W. (2003) A costbenefit analysis of electronic medical records. The American Journal of Medicine, 114(5), 397-403.

Weber, D.O. (2005) The state of the electronic health record in 2005. The Physician Executive 51, 6-10. Withrow, S.C. (2008) Why can't physicians interoperate? Barriers to adoption of EHRs. Healthcare financial management 2, 90-96