Literature Review

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 484–494

j o u r n a l h o m e p a g e : w w w . i j m i j o u r n a l . c o m

Evidence-based management of ambulatory electronic health record system implementation: An assessment of conceptual support and qualitative evidence

Ann Scheck McAlearney a,b,c,∗, Jennifer L. Hefner a, Cynthia Sieck a, Milisa Rizer a,d, Timothy R. Huerta a,d

a Department of Family Medicine, College of Medicine, Ohio State University, United States b Division of Health Services Management and Policy, College of Public Health, Ohio State University, United States c Department of Corporate Strategy and Innovation, École Polytechnique Fédérale de Lausanne, Switzerland d Department of Biomedical Informatics, College of Medicine, Ohio State University, United States

a r t i c l e i n f o

Article history:

Received in revised form

30 January 2014

Accepted 17 April 2014

Keywords:

Electronic health records

Medical informatics

Information systems

Information manage-

ment/systems/computerization

ambulatory/physician office health

information technology

EHR/EMR implementation

Best practices

a b s t r a c t

Objectives: While electronic health record (EHR) systems have potential to drive improve-

ments in healthcare, a majority of EHR implementations fall short of expectations.

Shortcomings in implementations are often due to organizational issues around the imple-

mentation process rather than technological problems. Evidence from both the information

technology and healthcare management literature can be applied to improve the likeli-

hood of implementation success, but the translation of this evidence into practice has not

been widespread. Our objective was to comprehensively study and synthesize best prac-

tices for managing ambulatory EHR system implementation in healthcare organizations,

highlighting applicable management theories and successful strategies.

Methods: We held 45 interviews with key informants in six U.S. healthcare organizations

purposively selected based on reported success with ambulatory EHR implementation. We

also conducted six focus groups comprised of 37 physicians. Interview and focus group

transcripts were analyzed using both deductive and inductive methods to answer research

questions and explore emergent themes.

Results: We suggest that successful management of ambulatory EHR implementation can be

Quality improvement guided by the Plan-Do-Study-Act (PDSA) quality improvement (QI) model. While participants

did not acknowledge nor emphasize use of this model, we found evidence that successful

implementation practices could be framed using the PDSA model. Additionally, successful

sites had three strategies in common: 1) use of evidence from published health informa-

tion technology (HIT) literature emphasizing implementation facilitators; 2) focusing on

workflow; and 3) incorporating critical management factors that facilitate implementation.

∗ Corresponding author at: Department of Family Medicine, Ohio State University, 273 Northwood and High Building, 2231 North High Street, Columbus, OH 43201, United States. Tel.: +1 614 293 8007; fax: +1 614 293 2715.

E-mail addresses: [email protected], [email protected] (A.S. McAlearney). http://dx.doi.org/10.1016/j.ijmedinf.2014.04.002 1386-5056/© 2014 Elsevier Ireland Ltd. All rights reserved.

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 484–494 485

Conclusions: Organizations seeking to improve ambulatory EHR implementation processes

can use frameworks such as the PDSA QI model to guide efforts and provide a means to for-

mally accommodate new evidence over time. Implementing formal management strategies

and incorporating new evidence through the PDSA model is a key element of evidence-

based management and a crucial way for organizations to position themselves to proactively

address implementation and use challenges before they are exacerbated.

© 2014 Elsevier Ireland Ltd. All rights reserved.

1

I i i i p i i a t o i s U U N l a a a m a

( h T b t a b g i m l t f t b i t p i b

t o E

. Introduction

n the United States there are mounting stakeholder and leg- slative pressures to integrate electronic health records (EHRs) nto office-based physician practice, with the goal of leverag- ng health information technology to reduce costs, improve atient safety, and track quality indicators [1–4]. Specifically,

mplementation of ambulatory EHRs has been linked to an ncrease in physician productivity [5] and more complete and ccurate documentation [6]. Given the pressures and incen- ives to adopt EHRs it is surprising that in 2011 only 34% of ffice-based physicians (including primary care and special-

sts) used an EHR that fits the definition of a “complete EHR ystem” [2,7]. While this is an increase from 11% in 2006, the .S. lags well behind other developed nations such as the nited Kingdom, Germany, the Netherlands, Australia, and ew Zealand – which all had nearly universal use of ambu-

atory EHRs by 2008 [8]. The slow pace of EHR adoption in mbulatory U.S. settings has been linked to organizational nd physician level barriers [1,7,9–13]. There is promise for n exploration of contextual factors and an application of anagement theories to provide evidence-based guidance to

lleviate these barriers [14–16]. Despite the potential for health information technology

HIT) to improve the quality, productivity, and operations of eath care, the majority of HIT implementations fail [17–20]. his lack of success is often due not to technological problems ut to organizational issues surrounding the implementa- ion process [18,21]. In the broader management literature, s in the field of HIT, a gap is noted between evidence- ased science and management practice [22,23]. Despite this ap, studies of financial and technological barriers to HIT nterventions are far more common than examinations of

anagement and organizational factors [16,24–26]. A 2011 iterature review of organizational barriers to HIT implemen- ation explicitly called for further exploration of management actors – incentives, structure, work flow processes – as facili- ators and/or barriers to HIT interventions [20]. Further, in the roader healthcare quality improvement (QI) literature, key

ntervention success factors have been identified, including op-level commitment, intervention champions, multidisci- linary teams, and physician buy-in [25,27–29] but research

nto how these factors are applied in HIT implementation has een limited.

One way to conceptualize the study of management fac-

ors in ambulatory EHR implementation is through the lens f Evidence-Based Health Services Management (EBHSM). BHSM, akin to evidence-based medicine in a clinical

setting, focuses on incorporating the findings of management research into the decision-making process of managers within health services organizations [30,31]. A 2004 study of 65 U.S. health services managers found a low level of evidence-based management behaviors [30]. Given the increasing focus on evidence-based practice in the ensuing decade, a goal of our study was to explore how EBHSM was used within the context of ambulatory EHR implementation. Specifically, we exam- ined data from a study of ambulatory EHR implementation best practices to determine whether what was done had con- ceptual support. The findings of our exploratory investigation have relevance for future HIT implementations, and provide broad support for the application of evidence-based manage- ment practices in healthcare.

2. Methods

2.1. Research question and research objective

We asked the following research questions:

1. “What characterizes best practices in implementation among healthcare organizations reported to have success- ful ambulatory EHR system implementations?”

2. “What can we learn from these ‘best practice’ implemen- tations to help other organizations overcome challenges associated with EHR system implementation?”

Our overall research objective was to develop an evidence- based understanding of the facilitators of ambulatory EHR system implementation, paying particular attention to opportunities to frame best practices that can improve imple- mentation processes and maximize the use of such systems.

2.2. Research design

We held key informant interviews and focus groups in an extensive qualitative study designed to explore physicians’ and organizations’ perspectives about ambulatory EHR system implementation [32]. Our study conformed to the standards of rigorous qualitative research [33,34] including guidelines for

in-depth interviews [35] and focus group techniques [36]. Eth- ical approval for conducting this research was obtained from the Behavioral and Social Sciences Institutional Review Board of The Ohio State University.

i c a l

486 i n t e r n a t i o n a l j o u r n a l o f m e d

2.3. Study sites

We purposively selected sites for participation in this study on the basis of reputed success with EHR system implementation as indicated by several criteria. Our initial list of “best practice” sites had either received an award from the Healthcare Infor- mation Management Systems Society (HIMSS), recognition from the Hospital and Health Network’s benchmark survey, or had been recommended from a leading vendor of ambula- tory EHR systems. We reviewed this initial list with a project advisory committee comprised of both industry and academic representatives who had expertise with health IT and reduced the list to reflect committee recommendations and address considerations of organizational and geographic variability. Our final sample included six health systems, including three sites that had received the HIMSS Annual Davies Award for Ambulatory EHR use within the past five years. This sam- ple proved sufficient to enable us to observe replication of themes across study sites and draw conclusions, consistent with the standards of high-quality case study research [37], as described further below.

2.4. Key informant interviews

We held key informant interviews with physicians and orga- nizational representatives across the six sites. We conducted in-depth interviews either in person as a part of organiza- tional site visits, or over the telephone, having identified key informants based on their role and/or affiliation with the organization. All interviewers used a standard interview guide to ensure consistent data collection. The interview guide included a list of open-ended questions as well as probing questions designed to seek additional information from the interviewees. Interviews lasted 30–60 min, and were led by one study investigator with a second investigator available to assist. Interviews were all recorded, then transcribed, verified, and corrected prior to analysis.

2.5. Focus groups

Focus groups covered similar topics to those addressed in the key informant interviews with particular emphasis on per- ceptions about the implementation process. The focus groups lasted 60–90 min and included a meal. One study investigator moderated each focus group, with a co-moderator available to assist. To facilitate the focus group discussion, we used an open-ended list of questions, and incorporated several follow- up questions to probe for more detailed information. Similar to the process with key informant interviews, we recorded each focus group session, and then recordings were transcribed ver- batim to permit analysis.

2.6. Study participants

Across the six sites we interviewed 45 organizational and

clinical informants (Table 1). These key informants included executives, managers, information systems directors and pro- fessionals, and clinical providers including both physicians and nurses. Our focus groups were comprised of 37 physician

i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 484–494

providers including physician residents, attending physicians, and physicians in private practice.

2.7. Analyses

We used a combination of inductive and deductive meth- ods in our data analyses. As the study progressed, we read and discussed transcripts and preliminary findings, using a grounded theory approach [38]. Then, when we had completed all interviews and focus groups, we created a coding team that included the lead project investigator and three research assistants to code the transcripts.

Our coding team first reviewed interview guides and trans- cripts to identify broad themes and issues in a preliminary coding process that allowed us to organize our data into cat- egories of findings [39,40]. The team then produced an agreed upon a list of codes and a coding frame. We applied this cod- ing frame to three common transcripts and this enabled us to compare individual coding decisions and clarify codes and themes within the data. Coders also met periodically to per- mit regular discussions about the analytic process and ensure consistency of coding. These frequent meetings also helped us to reach agreement about final themes we saw emerging from our data. In addition, an ongoing review of the literature on the topic of EHR implementation enabled us to validate, compare, and extend our findings, when appropriate [40]. We used the qualitative data analysis software program Atlas.ti (version 6.0) to assist us with detailed coding of the data, and to help us formally explore patterns and themes within the data [41].

3. Results

Reviewing the features that these “best practice” implemen- tations had in common, we propose that structuring the process using the Plan-Do-Study-Act (PDSA) quality improve- ment (QI) model can facilitate implementation success. We suggest that this framework can then be leveraged to help address challenges throughout the EHR implementation pro- cess. In addition, we found three common strategies that appeared to be associated with success in ambulatory EHR system implementation: 1) applying evidence from published literature on health information technology (HIT) and EHR implementation in an EBHSM approach; 2) focusing on work- flow; and 3) incorporating critical management factors that facilitate implementation.

3.1. Characterizing a quality improvement (QI) framework for ambulatory EHR implementation

Considering the evidence from our interviews and focus groups, we propose that an appropriate way to facilitate successful ambulatory EHR implementation is by applying a quality improvement (QI) model. While the deliberate use of a QI model was not explicitly acknowledged or identified by

study participants, we suggest that the traditional PDSA cycle framework provides a structure by which EHR implementation steps and activities can be organized [30,42]. From our partic- ipants’ comments, we characterized how this QI framework

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 484–494 487

Table 1 – Interviewees and focus group participants, by role.

Interviewees Number Focus group participants Number

Leaders/Managers 18 Practicing physicians(Attendings, Private Practice) 20 Information technology (IT) professionals 13 Physicians in training (Interns, Residents) 17

c e t a t s s

t a E p s l t n a p w g m w p “ l p s a S w c h s c s o E

3 i

3 T c W t u m o t t

Clinical providers (Physicians and Nurses) 14 Total 45

ould be applied to facilitate EHR implementation, including xploring subthemes related to each of the four main steps of he QI cycle. We describe this characterization further below, nd demonstrate how the PDSA framework can help organize he EHR implementation process by linking the cycle steps to tudy evidence with the inclusion of representative quotes as hown in Fig. 1.

First, the plan step involved distinct activities such as cus- omer engagement, workflow analyses, and training as well s leadership commitment. As one interviewee explained of HR implementations, “They really need to have a project lan and a timeline of some sort; it’s vital.” Next, the do tep included actions associated with the implementation go- ive, especially around the need for support at the elbow of he user. An informant from a different site noted how she eeded “just a really good support staff that’s going to be ble to help them through this process and I need quality eople to be able to send out there. I can’t send somebody ho’s marginal out to a practice and expect to have successful

o-live.” Third, the study step involved reflection on the imple- entation process itself, along with efforts to incorporate hat was learned from the experience of new users, as appro- riate. As one site’s implementation coordinator reflected, One thing that I’m really proud of with my team is that we earn with every implementation and we change the game lan all the time.” Finally, the act step included QI practices uch as using feedback to support improvement, and looking head to continue the improvement process. An Information ervices director in a third site reflected how, “In some cases e allowed clinics to be a little free with some of their pro-

esses, and since then this is not the way we do it. So we ad to go back and change something, so that we are more ystematized.” Framing successful implementation as a PDSA ycle appeared to be appropriate as a way to characterize ites’ efforts to “Circle back and improve” as they focused n iterative learning opportunities in successful ambulatory HR implementations.

.2. Strategies facilitating successful ambulatory EHR mplementation

.2.1. Applying the evidence he first of three strategies that appeared to facilitate suc- essful implementation involved applying EBHSM in practice. hile study participants did not indicate that their applica-

ion of research results was deliberate, participants frequently sed terms such as “champions,” “super user,” “positive com-

unicator,” and “physician buy-in” which suggests that many

f the success factors of EBHSM have permeated the cul- ure of practice. Specifically, we found support for applying he evidence in five important areas: 1) ensuring leadership

Total 37

commitment and organizational support [28]; 2) identify- ing and leveraging physician champions and super-users [12,28]; 3) involving physicians in the process [28]; 4) provid- ing opportunities for physicians to practice with the system [43]; and 5) learning from the past and from others [21]. Each of these areas of evidence and representative com- ments from study participants that demonstrate application of EBHSM as a strategy to facilitate implementation are summarized in Table 2.

3.2.2. Focusing on workflow A second strategy associated with successful ambulatory EHR implementation involved focusing on workflow as part of the implementation process. We found three distinct themes that we synthesized as recommendations related this aspect: 1) acknowledge the importance of understanding and fixing workflow; 2) include practice managers and nurses in efforts to address workflow; and 3) communicate and educate about expected changes in workflow. Below we describe study partic- ipants’ perspectives that led to these recommendations, and in Table 3 we present additional evidence in support of these themes.

First, across informants, we were repeatedly told of the need to acknowledge the importance of workflow and under- standing workflow in the EHR implementation process. As one interviewee described, “Workflow is key. Absolutely key.” Another summarized that “It’s really a practice redesign. It’s not just an EMR installation.” An aspect of recognizing the importance of understanding workflow was also build- ing awareness that ineffective workflows would not be fixed by implementing the EHR, but that EHR implementation itself might actually bring inappropriate workflow practices to light. One IT professional noted this by explaining how it was important to build “Understanding that you’re not going to fix anything with an EMR. If it’s broken going in, it’s broken coming out. . .. So that what that really means is that you’ve got to focus on your workflow. And if you get your work- flow stabilized and clarified before you walk in to your system implementation, you’re going to be in a lot better place com- ing out because you’ve got people bought in on the workflow versus I’m trying to buy them in on the workflow and on the new system.” Another interviewee noted how an implemen- tation would ideally proceed: “You really need to go in and understand what that current state workflow is. And if it’s not documented anywhere, spend the time to document it because then you’ll be able to say, ‘Ok, here’s how the software works and here’s where we have opportunity that we need to

change workflow.”’

A second workflow-related theme emerged around the importance of including practice managers and nurses in any efforts to address workflow. This issue was highlighted

488 i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 484–494

ame

Fig. 1 – Adapting a quality improvement fr

by interviewees who contrasted the perspectives of practice managers and nurses from those of physicians and other

EHR system users. As one interviewee reflected, “So you need the practice directing the project, I’m convinced.” Another IT professional similarly explained, “The manager sort of sees

work for ambulatory EHR implementation.

the process by which you do your core visits, not the docs,” thus having this individual contribute to process changes was

critical. Similarly, the inclusion of nurses who also knew how the practice operated was critical. One physician noted how, for “nurses, that’s their job. They work there all the time.

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 484–494 489

Table 2 – Translating lessons from prior research.

Applying evidence from prior research

Representative quotations from study participants

Ensure Leadership and Organizational Support [10,21,39,41,48]

• “The administration and the medical staff leadership and with the support of the boards, I think that there is that vision and that ability, that willingness really to invest.” • “I think it’s that leadership because there’s that support and that investment and there is a vision that people just buy into that it’s better if it’s an electronic record.” • “Having that support from higher up all the way down helped make it successful.”

Identify and Leverage Physician Champions and Super-Users [10,39,42,43,48]

• “So much of it is making sure that you select champions. . .your nursing champion, your super users, and your physician champion so that you get them onboard. You start getting them educated about the application, what the application does and they become an expert so then they can go back and spread the information and how it’s going to be used.” • “I think it’s really important to have strong super users and super users that really are excited about being super users and then to have a physician champion at each clinic is huge.” • “The most you can do is find a good physician champion, find a strong practice leader, set expectations, you know, up front.”

Involve Physicians [21,39,44,47]

• “Getting enough input from the clinicians during the design phase to achieve buy-in and balancing that with their limited availability.” • “The ideal answer is if the organization’s committed enough to doing this to compensate the physicians for the time they spend on design that is not being done between patients or taking away their lunch.”

Provide Opportunities for MDs to Practice with the System [40,45]

• “We have this thing called the playground so they could go in and play. It would demonstrate various things and when they came in for training . . . it would have fake users and fake patients they could use for training. They could take it with them and go back to their sites where the PCs were set up with this playground environment. . .we used this training to get going and practice scenarios to keep their training fresh.” • “We probably spent around 8 h of classroom training per physician and then they had to practice some on their own. We give them access to a test system.”

Learn from the Past and from Others [7,16,46]

• “So we’ve taken a lot of the technologies that we’ve learned from the inpatient setting and then moved them now, into the outpatient setting and what we’ve found is that, as we’ve kind of transformed the inpatient setting, the doctors were much more interested in ‘hey, I’m in my practice now, why can’t I get this information?’ You know, in an electronic form.” • “The first time around basically we didn’t do the preload piece where you’re loading in all of the

ff. . . . er ch plem

T i n h u m r

n w a f t e i i d h t r m d w

problem lists, meds, all that stu entering data from the prior pap about a year getting ready to im

hey’re using this system all the time.” As a result, as one nterviewee reflected, “if you’re willing to teach them [the urses] the right way to do it [i.e., the new workflow], they’re appy with that. They’ll learn it and they’ll do it. . . .they nderstand what it means to be part of a workflow which eans you have to do things a certain way. And they do it

eally well.” The third theme that emerged around workflow was the

eed to communicate and educate system users about the orkflow changes expected as part of improving efficiency nd documentation in the EHR implementation. As one IT pro- essional commented, “Don’t wait until training to tell them hat there’s a new workflow.” Another interviewee further xplained how “We attempt to have people take some train- ng before we go through the workflow so that they have an dea of what we’re talking about. At the very minimum, we o a demo to show them what the system looks like and ow it works; and then we go through the workflows with

hem.” This need to communicate with and educate users was eportedly particularly important for physician users. As one

anager emphasized, “Because they chose health care, they idn’t choose technology. So come in with the physicians, and ork with them!”

So what happened is the providers spent a lot more time just art-that slowed everything down. The second time around we spent ent it and it went much smoother.”

3.2.3. Incorporating critical management factors to facilitate implementation Comments from study participants also reflected the importance of considering management issues in success- ful EHR implementations. We characterized five factors that appeared critical to facilitate successful management of ambulatory EHR implementation in the organizations we studied: 1) commitment; 2) convincing/converting; 3) communication; 4) coordination; and 5) change man- agement. Each of these factors is described in further detail next, with additional supporting evidence presented in Table 4.

3.2.3.1. Commitment. The first critical management con- sideration that emerged across comments from study participations was the need for the organization to commit to the EHR. Commitment reportedly involved leadership sup- port, as well as backing by financial resources. As described by one interviewee reflecting on the success of the ambu- latory EHR implementation, “It was well-funded. It was top

management-supported. It was one of the number one prior- ity items for the organization, and they had the will to see it through.”

490 i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 484–494

Table 3 – Focus on workflow to facilitate successful EHR implementation.

Factors associated with successful focus on workflow

Representative comments

Acknowledging the Importance of Understanding Workflow

• “Workflow–understanding workflow.” • “Every process needs to be defined how you currently do it and every step because what I would say to people is, ‘here’s the deal: you probably do steps that you don’t even know you’re doing because you’re probably thinking of your grocery list tonight and you’re doing them because you’ve done them for how many years. Well guess what, when this electronic record comes, you have to think of every click, and you know, if you didn’t predefine all the things that happen, you’re going to get lost in. . .’ because you get under that pressure with the patients right there.” • “Be able to understand that workflow so that you can build new workflow into it and in some cases it may require people moving into different types of roles because the role that they had no longer exists with the way that the software works in the future. So it’s really important to understand that current state.”

Include Practice Managers and Nurses in Efforts to Address Workflow

• “I completely think this has to be owned by the managers and the operational people and . . .nurses.” • “Having a practice manager who understood how that practice runs.” • “They [nurses] see a wider variety of patients with different issues [from physicians] and they’re inputting different types of data and so I would say that they have to have a much more advanced knowledge [than many doctors].”

Communication and Education Around Expected Changes in Workflow

• “Communication that workflow’s going to change.” • “Education in how to use an EMR; education on the process of doing everything around it; how it would affect their workflow.” • “Do a whole lot of work on the workflow and changes to workflow and then more training. Stretch it out so there can be proper behavioral changes and operation changes by the actual users.” • “So it’s about really understanding the flow, going through that, making the doctors sit down and say, ‘Ok, pretend there’s a patient in that chair with upper respiratory infection’ whatever you want to do, ‘now interview them.’ You know? You close the door and make a fool of yourself in private. But you know, play (act?) them. How would you ask the questions? Where do you realize, ‘Oh this makes me go off my

patte

be difficult and time consuming appeared to help both

pattern so how do I readjust my

3.2.3.2. Convert resistors. Another important step in the implementation process was reportedly convincing and con- verting resistors. By acknowledging and directly addressing resistance, actions could be taken to convince reluctant new users to view the system in a more positive light. An infor- mant in one organization described how their implementation process included active recruitment of resistors to commit- tees to help guide EHR system improvements. She described how this step involved “. . .the people that don’t like the sys- tem the most and so they’re the ones saying, ‘Well what about this functionality?’ or ‘What about this workflow?’ And you know the theory there is if you have the people that hate it the most actually start seeing that there are improvements being made to the things that they don’t like, then they’re great champions.” Further, any problems that arose needed to be addressed quickly and constructively to avoid negative word- of-mouth that could influence attitudes toward EHR adoption, acceptance, and use.

3.2.3.3. Communication. Across informants, communication was also repeatedly emphasized as critical. Interviewees noted that all involved parties must know their roles, the timeframe for implementation, and the expected impact of the change, and this information needed to be communicated clearly and consistently. One participant explained, “One of the biggest things that we got from feedback was ‘we like to

be a part of. . . we would just like to know.’ People would be ask- ing all the time so we had to develop ways to communicate to them.”

rn?’ ’’

3.2.3.4. Coordination. The fourth important element of suc- cessful implementations involved explicit efforts to coordi- nate activities among involved groups. As one interviewee described, “So it’s about having medical leadership and management who can kind of work with IS to keep that balance in some zone that’s correct.” Given the many com- ponents of ambulatory EHR implementation and potential problems that could arise, this coordination role was viewed as critical.

3.2.3.5. Change management. A fifth critical factor facili- tating successful ambulatory EHR implementation involved framing the implementation process using concepts from organizational change theory. Application of this approach was evident in comments from respondents who used terms such as “urgency,” “vision,” and “culture change.” One interviewee reflected, “we created a sense of urgency, which is an important part of the transformation in our view. If you want to have a successful transformation, you have to have a sense of urgency.” Another noted how, “You create a vision by first of all saying why the cur- rent status quo is unacceptable, and then it’s all about the execution.” Framing the introduction and implemen- tation of the EHR system as a necessary organizational change and acknowledging that the change process would

organizations and system users better understand and accept the required behavior changes associated with the new EHR.

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 484–494 491

Table 4 – Critical management considerations facilitating ambulatory EHR implementation.

Commit • “I think what helped folks the most in this is what we call the ‘Rip the Band Aid Off’ approach. It’s just a rapid, aggressive timetable and we

said, ‘You have to jump this high and there’s no other choice’ and we stuck with it.” • “. . .once you’re in it, you’re in it. And we never de-install. That’s not an option. You have to continue to move forward.” Convert Resistors • “We’ve taken people that are providers let’s say that are absolutely not excited about an EMR and you go have them job shadow at one of

the sites where it’s already implemented. It’s how do you take the person that’s going to drag their feet in the mud the hardest and try to get them and move it along?”

• “We had a physician that was one of our top producers in the paper world who was adamantly opposed to moving to an EMR . . . he’d be looking at computer screen instead of talking to his patient. That person is now one of the strongest advocates for an EMR because he can see the connectivity . . . and the improvement in quality and his ability to go in and know everything about that patient from home or wherever without having the chart.”

• “You can frame it that this is about the system of care, it’s how we take care of patients, not about just you and making your life miserable with this one little piece.”

Communicate • “As you’re doing your implementation, you’re updating them, ‘Hey we’re up to. . .’ you know kind of give them an update of where you’re at

with how the implementation is coming. So I think that’s important because then they’re not all the sudden, ‘Whoa! I didn’t know it was happening tomorrow!”’

• “What so much of this comes down to is constantly communicating.” Coordinate • “Where there have been challenges or concerns or pushback on workflow, we escalate to the steering committee and we lay it out.” • “Document everything, yeah and we do, we share our minutes like with the team-everybody gets a copy, everybody gets an agenda the

next time we meet and the minutes from the other meeting and really just, you know have an agenda, kind of stick to it, get done what you had on your agenda and then if you have any other business put that on the end.”

• “Warning: it’s like I’m in my little world back there [as a physician]. Everything’s happening to help me do it but it’s like little elves. Just because I don’t see it happening, doesn’t mean that there’s not somebody back there doing something.”

Change Management • “Constantly developing an environment where people are comfortable about getting change introduced to them.” • “Recognize and tell them upfront, this is a significant change in how you work and don’t expect you to do it tomorrow. Don’t get angry that

you can’t operate this way. It’ll take time.” nce a ng to

4

O m b m P f z a t t a r f c

t s a h n s n s t E

• “Really setting the stage on we’re going to have a 2-hour meeting o of that. So I think building an understanding of what it really is goi

. Discussion

ur analysis of data from a study of ambulatory EHR imple- entation best practices revealed opportunities to apply

oth conceptual frameworks and evidence to facilitate imple- entation success. Conceptually, we propose that using the

lan-Do-Study-Act (PDSA) quality improvement model to rame the implementation process can help guide organi- ations struggling to introduce and integrate EHR systems nd develop appropriate corrective responses to implemen- ation barriers they encounter. In addition, we found that hree strategies common to these best practice sites – pplying the evidence, focusing on workflow, and incorpo- ating critical management factors – appear to hold promise or organizations striving to overcome EHR implementation hallenges.

These findings have relevance for future EHR implementa- ions, and provide both conceptual guidance and qualitative upport for the deliberate application of evidence-based man- gement practices in healthcare. Specifically, the PDSA model as been proposed as a core quality improvement tech- ique and a key tool of EBHSM [30,49]. Our present study upports this proposition, revealing that PDSA cycles, while

ot mentioned by name, may be implicitly applied within uccessful ambulatory EHR implementations that take advan- age of opportunities to learn from the process and optimize HR use. As health services managers strive to incorporate

week and you need to plan to do two more hours of work outside take I think is really essential.”

EBHSM methods into heath services decision-making, the PSDA model can thus serve as an important vehicle to deliber- ately introduce and incorporate evidence and best practices. In the case of ambulatory EHR implementation specifically, the strategies we found common to these best practice sites – applying the evidence and focusing on workflow – can be intention- ally implemented within the PDSA model, thereby facilitating implementation success.

With respect to applying the evidence, we found that suc- cessful EHR implementation was apparently facilitated when the implementation process incorporated key findings from the EHR literature and from prior HIT implementation efforts. Previous research has identified a large gap between aca- demic literature and practice-oriented literature in the field of human resources management [22]. However, our results would indicate that many management factors highlighted in the academic health care literature are pervasive in the cul- ture of health care management, as evidenced by participants’ use of terms such as “champions,” “super user”, “positive communicator,” and “physician buy-in.” While we cannot determine whether application of the evidence was inten- tional or coincidental, a variety of researchers have proposed that management strategies and HIT success are inextricably

linked [16,24,25], and the results of our study corroborate this hypothesis. Specifically, identifying critical management fac- tors across EHR implementations that have been reputed to be successful fills a gap in the literature identified by Lluch in

i c a l

The Ohio State University during the study. Finally we thank our editor and two anonymous reviewers for helpful sugges- tions that led to the improvement of this manuscript. This

492 i n t e r n a t i o n a l j o u r n a l o f m e d

2011 when she highlighted the need for an elucidation of the management factors that facilitate HIT success [20].

The strategy of incorporating these critical management factors, including commitment, convincing/converting, com- munication, coordination, and change management, however, requires acknowledgment on the part of managers that implementing an EHR requires a cultural change within the organization. For a cultural shift to occur, attention must be given to the organizational vision, leadership support, good communication, and encouragement of discourse among team members [50,51]. In EHR implementation, this includes top management, IT personnel and the end users of the sys- tem. Communication among all individuals involved in the implementation is essential to learning, as scrutinizing exist- ing technologic frames is necessary for new knowledge to be assimilated, thus allowing a shift in the technologic frames that permit successful adoption of the technology [52]. Par- ticipants from all sites in our study emphasized the need to consistently communicate with clinicians and administrative staff throughout the EHR implementation process. This find- ing is consistent with previous research, which stressed that the needs and requirements of different users must be taken into account for successful EHR implementation [6].

5. Limitations of this study

This research has several limitations. First is the small num- ber of hospitals involved in this study. The time and energy requirements associated with qualitative studies create sig- nificant barriers to large-scale studies, and limits our ability to generalize results across settings with different organi- zational characteristics or market constraints. However, the results from our in-depth exploration of this topic can now be further investigated and validated in larger samples using surveys developed based on our findings. Another limitation is that we studied only organizations that had been identified on the basis of reputation for success with EHR implementa- tion. While the management factors we present are common to these “best practice” implementations, we are unable to definitively state that the presence or absence of these factors distinguishes between implementation success or failure. A third limitation is the possibility for our findings to be influ- enced by the stage of EHR implementation and familiarity with the EHR. Future qualitative and quantitative efforts could explore these variables. Finally, our study is unable to link EHR implementation strategies to either clinical or financial out- comes, thus our sense of best practices is constrained by our sample selection.

6. Suggestions for future studies

In addition to providing insight into successful EHR imple- mentation practices, this exploratory study identified several

areas that merit further investigation and study. Specific top- ics for future research into ambulatory EHR implementation include further exploration of challenges for physicians and their practices to return to full productivity, and continued

i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 484–494

evaluation of the impact of EHR implementation on quality and efficiency in ambulatory care settings.

7. Conclusions

Organizations seeking to improve ambulatory EHR implemen- tation processes can use frameworks such as the PDSA quality improvement model to guide efforts and provide a means to formally accommodate new evidence over time. By paying attention to the growing body of literature as well as height- ened scrutiny of physician EHR use, organizations can position themselves to proactively address implementation and use challenges before they are exacerbated.

Authors’ contributions

Ann Scheck McAlearney designed the study, led data acqui- sition, and led analysis and interpretation of the paper; she led the drafting of article, and approved the final version. Jen- nifer Hefner made substantial contributions to the analysis and interpretation of the data, contributed to drafting the article, and approved the final version. Cynthia Sieck made substantial contributions to the analysis and interpretation of the data, contributed to drafting the article, and approved the final version. Milisa Rizer made substantial contributions to the interpretation of the data, reviewed the article critically for important intellectual content related to physicians’ per- spectives, revised the article critically, and approved the final version. Timothy Huerta made substantial contributions to the analysis and interpretation of the data, revised the article critically for important intellectual content related to man- agement and implementation issues, and approved the final version.

Conflict of interest statement

All authors report no conflicts of interest with this research.

Acknowledgements

The authors are extremely grateful to the organizations and informants who participated in this study, and to the health system members of our Project Advisory Team. We also thank our research team members, Drs. Paula Song, Julie Robbins, Deena Chisolm, Annemarie Hirsch, and Maria Jorina, as well as our research assistants, Trevor Young and Emily Kathryn Orcutt Knecht, and clinical consultants, Dr. John Mahan, Dr. David Rich, Dr. Scott Holiday, all of whom were affiliated with

research was funded by the Center for Health Management Research, but the study sponsors had no role in study design, collection, analysis or interpretation of data, nor the decision to submit this manuscript for publication.

i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n

Summary points What was already known on the topic:

• Only one-third of U.S. office-based physicians use a fully functional EHR. The slow pace of EHR adoption in ambulatory settings has been linked to organizational and physician level barriers.

• Despite the potential for ambulatory EHRs to improve the quality and operation of healthcare, many EHR implementations fail, often due not to technological problems but to organizational issues surrounding the implementation process.

• Leadership support has been identified as a critical success factor in prior studies involving implemen- tation of hospital-based EHRs and other information technology innovations, but prior studies have not explored what this means in ambulatory care, nor assessed how this factor may be applied in practice.

What this study added to our knowledge:

• Across six health systems reporting success at man- aging ambulatory EHR implementation, the imple- mentation process appeared consistent with the Plan-Do-Study-Act (PDSA) quality improvement (QI) model, but use of this model was not recognized or acknowledged. We propose that the PSDA model can be deliberately applied to frame and facilitate EHR implementation, thereby providing a vehicle through which management techniques and best practices, such as focusing on workflow, can be introduced.

• Ambulatory EHR implementation can benefit from intentional application of research evidence as well as consideration of critical management factors including organizational commitment, efforts to convert resis- tors, communication, coordination, and use of change management principles.

• In addition to their relevance for future EHR imple- mentations, these findings provide support for the application of evidence-based management practices

r

in healthcare.

e f e r e n c e s

[1] E.W. Kokkonen, S.A. Davis, H. Lin, T.S. Dabade, S.R. Feldman, A.B. Fleischer, Use of electronic medical records differs by specialty and office settings, J. Am. Med. Inform. Assoc. 20 (e1) (2013) e33–e38.

[2] C. Hsiao, E. Hing, T.C. Socey, B. Cai, Electronic health record systems and intent to apply for meaningful use incentives among office-based physician practices: United States, 2001-2011, NCHS Data Brief. 79 (2011) 1–8.

[3] Centers for Medicare and Medicaid Services. EHR incentive programs - data and program reports [Internet].

[4] D. Blumenthal, Stimulating the adoption of health

information technology, N. Engl. J. Med. 360 (15) (2009) 1477–1479.

[5] A.D. Cheriff, A.G. Kapur, M. Qiu, C.L. Cole, Physician productivity and the ambulatory EHR in a large academic

f o r m a t i c s 8 3 ( 2 0 1 4 ) 484–494 493

multi-specialty physician group, Int. J. Med. Inf. 79 (7) (2010) 492–500.

[6] K. Häyrinen, K. Saranto, P. Nykänen, Definition, structure, content, use and impacts of electronic health records: a review of the research literature, Int. J. Med. Inf. 77 (5) (2008) 291–304.

[7] C.W. Burt, E. Hing, D. Woodwell. Electronic medical record use by office-based physicians: United States, 2005. Health E-Stats. 2006.

[8] A.K. Jha, D. Doolan, D. Grandt, T. Scott, D.W. Bates, The use of health information technology in seven nations, Int. J. Med. Inf. 77 (12) (2008) 848–854.

[9] E. Hing, M.J. Hall, J.J. Ashman. Use of electronic medical records by ambulatory care providers: United States, 2006. Citeseer; 2010.

[10] N.M. Lorenzi, A. Kouroubali, D.E. Detmer, M. Bloomrosen, How to successfully select and implement electronic health records (EHR) in small ambulatory practice settings, BMC Med. Informat. Decision Making 9 (1) (2009) 15.

[11] C.M. DesRoches, E.G. Campbell, S.R. Rao, K. Donelan, T.G. Ferris, A. Jha, R. Kaushal, D.E. Levy, S. Rosenbaum, A.E. Shields, Electronic health records in ambulatory care—a national survey of physicians, N. Engl. J. Med. 359 (1) (2008) 50–60.

[12] R.H. Miller, I. Sim, Physicians’ use of electronic medical records: barriers and solutions, Health Aff. 23 (2) (2004) 116–126.

[13] D.A. Ludwick, J. Doucette, Adopting electronic medical records in primary care: Lessons learned from health information systems implementation experience in seven countries, Int. J. Med. Inf. 78 (1) (2009) 22–31.

[14] A. Boonstra, M. Broekhuis, Barriers to the acceptance of electronic medical records by physicians from systematic review to taxonomy and interventions, BMC Health Serv. Res. 10 (2010) 231, 6963-10-231.

[15] M. Greiver, J. Barnsley, R.H. Glazier, R. Moineddin, B.J. Harvey, Implementation of electronic medical records: theory-informed qualitative study, Can. Fam Physician 57 (10) (2011 Oct) e390–e397.

[16] A.S. McAlearney, J. Robbins, N. Kowalczyk, D.J. Chisolm, P.H. Song, The role of cognitive and learning theories in supporting successful EHR system implementation training A qualitative study, Med. Care Res. Rev. 69 (3) (2012) 294–315.

[17] L. Lapointe, S. Rivard, A multilevel model of resistance to information technology implementation, Mis Quart. 46 (2005) 1–91.

[18] E. Iveroth, P. Fryk, B. Rapp, Information technology strategy and alignment issues in health care organizations, Health Care Manage. Rev. 38 (July–September (3)) (2013) 188–200.

[19] B. Kaplan, K. Harris-Salamone, Health IT success and failure: Recommendations from literature and an AMIA workshop, J. Am. Med. Inform. Assoc. 16 (2009) 291–299.

[20] M. Lluch, Healthcare professionals’ organisational barriers to health information technologies—a literature review, Int. J. Med. Inf. 80 (12) (2011) 849–862.

[21] B. Kaplan, K.D. Harris-Salamone, Health IT success and failure: Recommendations from literature and an AMIA workshop, J. Am. Med. Informat. Assoc. 16 (3) (2009) 291–299.

[22] S.L. Rynes, T.L. Giluk, K.G. Brown, The very separate worlds of academic and practitioner periodicals in human resource management: implications for evidence-based management, Acad. Manage. J. 50 (5) (2007) 987–1008.

[23] D.M. Rousseau, S. McCarthy, Educating managers from an evidence-based perspective, Acad. Manage. Learn. Educ. 6 (1)

(2007) 84–101.

[24] P. Nambisan, G.L. Kreps, S. Polit, Understanding electronic medical record adoption in the United States:

i c a l

494 i n t e r n a t i o n a l j o u r n a l o f m e d

communication and sociocultural perspectives, Interact. J. Med. Res. 2 (March (1)) (2013) e5.

[25] H.C. Kaplan, P.W. Brady, M.C. Dritz, D.K. Hooper, W.M. Linam, C.M. Froehle, P. Margolis, The influence of context on quality improvement success in health care: a systematic review of the literature, Milbank Q. 88 (December (4)) (2010) 500–559.

[26] N.M. Lorenzi, R.T. Riley, Organizational issues change, Int. J. Med. Inf. 69 (2) (2003) 197–203.

[27] M. Leonard, S. Graham, D. Bonacum, The human factor: the critical importance of effective teamwork and communication in providing safe care, Qual. Safety Health Care 13 (Suppl. 1) (2004) i85–i90.

[28] E.G. Poon, D. Blumenthal, T. Jaggi, M.M. Honour, D.W. Bates, R. Kaushal, Overcoming barriers to adopting and implementing computerized physician order entry systems in US hospitals, Health Aff. 23 (4) (2004) 184–190.

[29] S.L. Taylor, S. Dy, R. Foy, S. Hempel, K.M. McDonald, J. Øvretveit, P.J. Pronovost, L.V. Rubenstein, R.M. Wachter, P.G. Shekelle, What context features might be important determinants of the effectiveness of patient safety practice interventions? BMJ Quality & Safety 20 (7) (2011) 611–617.

[30] A.R. Kovner, T. Rundall, Evidence-based management reconsidered, Front Health Serv. Manage. 22 (3) (2005) 3–22.

[31] K. Walshe, T.G. Rundall, Evidence-based management: from theory to practice in health care, Milbank Q. 79 (3) (2001) 429–457.

[32] J.A. Maxwell, Qualitative Research Design: An Interactive Approach, Sage Publications, Incorporated, 2004.

[33] M.B. Miles, A.M. Huberman, Qualitative Data Analysis: An Expanded Sourcebook, 2nd ed., SAGE Publications, Inc, 1994.

[34] W.L. Miller, B.F. Crabtree, Doing Qualitative Research, Sage Publications, Incorporated, 1999.

[35] G. McCracken, The Long Interview, Sage, 1988. [36] D.L. Morgan, Focus Groups as Qualitative Research, Sage,

1997. [37] R.K. Yin, Case Study Research: Design and Methods, 4th ed.,

Sage, Thousand Oaks, CA, 2009. [38] A. Strauss, J.M. Corbin, Basics of Qualitative Research:

Techniques and Procedures for Developing Grounded

Theory, 2nd ed., Sage Publications, Inc, 1998.

[39] M.A. Constas, Qualitative analysis as a public event: the documentation of category development procedures, Am. Educ. Res. J. 29 (2) (1992) 253–266.

i n f o r m a t i c s 8 3 ( 2 0 1 4 ) 484–494

[40] B. Glaser, A. Strauss, The Discovery of Grounded Theory: Strategies for Qualitative Research, Aldine, Chicago, IL, 1967.

[41] Scientific Software Development. Atlas.ti. 6.0th ed. Scientific Software Development, Berlin, 2008.

[42] M. Best, D. Neuhauser, Walter A shewhart, 1924, and the Hawthorne factory, Qual. Safety Health Care 15 (2) (2006) 142–143.

[43] A.S. McAlearney, S.B. Schweikhart, M.A. Medow, Organizational and physician perspectives about facilitating handheld computer use in clinical practice: results of a cross-site qualitative study, J. Am. Med. Inform. Assoc. 12 (5) (2005) 568–575.

[44] A. Boonstra, M. Broekhuis, Barriers to the acceptance of electronic medical records by physicians from systematic review to taxonomy and interventions, BMC Health Services Res. 10 (1) (2010) 231.

[45] J.T. Scott, T.G. Rundall, T.M. Vogt, J. Hsu, Kaiser permanente’s experience of implementing an electronic medical record: a qualitative study, BMJ: Br. Med. J. 331 (7528) (2005) 1313.

[46] D.M. Schuster, S.E. Hall, C.B. Couse, D.S. Swayngim, K.Y. Kohatsu, Involving users in the implementation of an imaging order entry system, J. Am. Med. Inform. Assoc. 10 (4) (2003) 315–321.

[47] L. Lapointe, S. Rivard, Getting physicians to accept new information technology: insights from case studies, Can. Med. Assoc. J. 174 (11) (2006) 1573–1578.

[48] C.A. Banas, A.R. Erskine, S. Sun, S.M. Retchin, Phased implementation of electronic health records through an office of clinical transformation, J. Am. Med. Inform. Assoc. 18 (5) (2011) 721–725.

[49] W.A. Sollecito, J.K. Johnson, McLaughlin and Kaluzny’s Continuous Quality Improvement in Health Care, Jones & Bartlett Learning, 2011.

[50] D. Robey, M. Boudreau, G.M. Rose, Information technology and organizational learning: a review and assessment of research, Acc. Manage. Inform. Technol. 10 (2) (2000) 125–155.

[51] P.H. Cheney, R.I. Mann, D.L. Amoroso, Organizational factors affecting the success of end-user computing, in: End-user Computing, John Wiley & Sons, Inc, 1989, pp. 295–310.

[52] W.J. Orlikowski, Learning from Notes: organizational issues in groupware implementation, in: Proceedings of the 1992 ACM conference on Computer-supported cooperative work, ACM, 1992, pp. 362–369.