Page 1
RESEARCH Open Access
Towards successful coordination of electronic
health record based-referrals: a qualitative
analysis
Sylvia J Hysong1,2*, Adol Esquivel3, Dean F Sittig4, Lindsey A Paul5, Donna Espadas1,2, Simran Singh6 and
Hardeep Singh1,2
Abstract
Background: Successful subspecialty referrals require considerable coordination and interactive communication
among the primary care provider (PCP), the subspecialist, and the patient, which may be challenging in the
outpatient setting. Even when referrals are facilitated by electronic health records (EHRs) (i.e., e-referrals), lapses in
patient follow-up might occur. Although compelling reasons exist why referral coordination should be improved,
little is known about which elements of the complex referral coordination process should be targeted for
improvement. Using Okhuysen & Bechky’s coordination framework, this paper aims to understand the barriers,
facilitators, and suggestions for improving communication and coordination of EHR-based referrals in an integrated
healthcare system.
Methods: We conducted a qualitative study to understand coordination breakdowns related to e-referrals in an
integrated healthcare system and examined work-system factors that affect the timely receipt of subspecialty care.
We conducted interviews with seven subject matter experts and six focus groups with a total of 30 PCPs and
subspecialists at two tertiary care Department of Veterans Affairs (VA) medical centers. Using techniques from
grounded theory and content analysis, we identified organizational themes that affected the referral process.
Results: Four themes emerged: lack of an institutional referral policy, lack of standardization in certain referral
procedures, ambiguity in roles and responsibilities, and inadequate resources to adapt and respond to referral
requests effectively. Marked differences in PCPs’ and subspecialists’ communication styles and individual mental
models of the referral processes likely precluded the development of a shared mental model to facilitate
coordination and successful referral completion. Notably, very few barriers related to the EHR were reported.
Conclusions: Despite facilitating information transfer between PCPs and subspecialists, e-referrals remain prone to
coordination breakdowns. Clear referral policies, well-defined roles and responsibilities for key personnel,
standardized procedures and communication protocols, and adequate human resources must be in place before
implementing an EHR to facilitate referrals.
Background
Successful referrals require considerable coordination
and interactive communication among the primary care
provider (PCP), the subspecialist, and the patient, which
may be challenging in the outpatient setting [1-3]. Sev-
eral studies at the interface of primary and subspecialty
care [4-9] suggest poor referral coordination and com-
munication as an important contributor to delays in
care,[10,11] mainly due to inappropriate timing and
detail of information [12] and lost paperwork. The use
of information technology has significant potential to
improve care coordination [13]. For instance, referrals
may be more successful when transmitted through an
integrated electronic health record (EHR; i.e., e-refer-
rals), allowing the PCP and subspecialist to exchange
information electronically, and both have immediate
* Correspondence: sylvia.hysong@va.gov
1Houston VA Health Services Research & Development Center of Excellence,
Michael E. DeBakey Veterans Affairs Medical Center, Houaron, Texas, USA
Full list of author information is available at the end of the article
Hysong et al. Implementation Science 2011, 6:84
http://www.implementationscience.com/content/6/1/84
Implementation
Science
© 2011 Hysong et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons
Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
Page 2
access to the entire patient record. However, in recent
work we found failures in referral completion despite e-
referrals;[14] about 6% of e-referrals lacked timely fol-
low-up by subspecialists, whereas when subspecialists
discontinued or deferred e-referrals and returned them
to PCPs for additional actions, 7% were lost to follow-
up [15]. Incomplete prerequisite workup and subspecia-
lists’ determination that the referral was not required
were cited frequently as reasons for discontinuing e-
referrals. This suggests a better understanding of referral
coordination and communication may be needed to
maximize the benefits of an EHR to the referrals process
[16].
Despite recommendations that referral coordination
should be improved, [1,3,17] Available: http://www.bio-
medcentral.com/1472-6963/9/62, [18] the healthcare lit-
erature sheds little light on which elements of
coordination should be targeted. Although a recent
measurement framework of coordinated care is a start,
[19] it does not identify the specific tools (e.g., routines,
plans, schedules) and processes healthcare providers use
to collectively and effectively transition patient care
from primary to secondary care setting and vice versa
[20,21]. However, literature from business management
may provide guidance on operationalizing many ele-
ments of effective coordination and shed additional light
on this issue.
Elements of coordination: an integrative framework
Okhuysen & Bechky [22] propose an integrative frame-
work explaining the mechanisms of coordination and
the integrating conditions necessary to achieve it effec-
tively. According to this framework, five basic organiza-
tional arrangements (i.e., mechanisms) allow individuals
to accomplish a collective performance, that is, to coor-
dinate:
1) Plans and rules: “purposive elements of formal
organizations” [22] (p . 473); for example, who is
allowed to place a referral request?
2) Objects and representations: technologies, tools,
and any device used to “create a common referent
around which people interact, align their work, and
create shared meaning” [22] (p. 474); for example,
how to use a template to place a referral request.
3) Roles: expectations of specific individuals; for
example, which provider is supposed to follow-up
with the patient after he/she visits the subspecialist?
4) Routines: “repeated patterns of behaviour that are
bound by rules and customs” [22] (p. 477); for exam-
ple, when a test result is completed, the ordering
provider is notified.
5) Physical proximity among team members: for
example, where are the referring provider and the
subspecialist located–in the same building, and/or
affiliated with the same institution?
These five basic mechanisms operate in various ways (e.g.,
by facilitating direct information sharing, developing agree-
ment, creating common perspectives) to allow teams to
achieve three integrating conditions, that is, the means by
which people collectively accomplish their interdependent
tasks: (1) accountability (clarity over who is responsible for
what), (2) predictability (knowing what tasks are involved
and when they happen), and (3) common understanding
(providing a shared perspective on the whole process and
how individuals’ work fits within the whole). How these
mechanisms and integrating conditions manifest themselves
in the referrals process is not well described in the litera-
ture. Using this framework as an analytic guide, our study
aims to provide insight into these relationships by identify-
ing barriers, facilitators, and perceived solutions for improv-
ing communication and coordination of EHR-based
referrals in an integrated healthcare system.
Method
Design and setting
This work is part of a larger study examining work-sys-
tem barriers, facilitators, and suggestions for improving
EHR-based communication.
Two large tertiary care Department of Veterans Affairs
(VA) Medical Centers (Sites A and B) from different
geographical areas served as study sites. The Computer-
ized Patient Record System (CPRS) is the EHR used at
all VA facilities (Figure 1); it integrates most aspects of
clinical care and has comprehensive e-referral manage-
ment functionality. Compared to nonintegrated systems,
the VA is an ideal environment to study referral coordi-
nation because the universal use of the EHR by those
who work in the same health system minimizes pro-
blems with information transmission [23].
We used subject matter expert (SME) interviews to
document and understand the e-referral process workflow
at four high-volume referral subspecialty clinics at Site A.
These insights guided focus groups (FGs) to identify bar-
riers, facilitators, and suggestions for improving the e-refer-
ral process at Sites A and B. Methods for this work have
been described elsewhere [24] and are summarized here.
Subject matter expert interviews
Participants and sampling frame
We purposefully sampled key informants, consisting of
subspecialists, physician assistants, and administrative
support staff, who were knowledgeable about referral
processes within their subspecialties (n = 7). We inter-
viewed one to two SMEs from each of four high-volume
referral subspecialties (cardiology, neurology, pulmonary,
and gastroenterology).
Hysong et al. Implementation Science 2011, 6:84
http://www.implementationscience.com/content/6/1/84
Page 2 of 12
Page 3
Procedure
We used a verbal protocol approach [25,26,26] with par-
ticipants to elicit the process of using CPRS to receive,
process, and complete or discontinue an e-referral.
Responses were audio-recorded, captured in field notes,
and used to create maps of the e-referral processes of
each subspecialty and to inform the FGs.
Data analysis
Process maps were created for each subspecialty to
capture the course of action for processing a referral
from its reception to final outcome. Two independent
coders (LAW and AE) analyzed the transcripts of each
of the SME interviews to identify the various steps of
all subspecialty referral processes. The coders used
standard flowchart symbols to denote the process flow.
The coders’ versions of each map were validated by
consensus to create final illustrations of each subspeci-
alty. Comparison of the maps highlighted the large
variability across specialty services; however, we identi-
fied activities shared across services based on their
sequence within the overall referral process and their
purpose. We used the final process maps as the foun-
dation for creating the FG protocol and subsequent
data analysis.
Focus groups
Participants and sampling frame
We conducted six FGs with a total of 30 participants.
We sampled purposefully to ensure a diversity of parti-
cipants (i.e., PCPs who referred patients to the four
selected subspecialties and subspecialists experienced in
their respective referral procedures). Two FGs with
PCPs (FGs 1 and 3) and two with subspecialists (FGs 2
and 4) were conducted at Site A. Subsequently, two FGs
(PCPs and subspecialists, respectively) were conducted
at Site B to triangulate findings and determine data
saturation. FGs were conducted in a private conference
room at each facility.
Procedure
An experienced facilitator conducted the FGs using a
semistructured protocol. A primary note taker (with a
background in qualitative methods) and a clinician (to
provide clarification and context as needed) were
included as part of the research team in each FG.
During the first two FGs, participants discussed barriers
to and facilitators of the e-referral process and offered sug-
gestions for improvement. Participants were encouraged
to consider organizational-, task-, and human resource-
related factors, in addition to technological issues. As part
Figure 1 Computerized Patient Records System (CPRS) referral order entry interface. This figure presents an example of the interface
where the primary care provider would place a request to refer a patient to a subspecialist. The provider can select the service needed, urgency,
and must provide a provisional diagnosis; the provider then enters free text details of the reason for the request and any pertinent details about
the patient’s case.
Hysong et al. Implementation Science 2011, 6:84
http://www.implementationscience.com/content/6/1/84
Page 3 of 12
Page 4
of the discussion, we presented the participants of FGs 3
and 4 with the themes frequently raised during FGs 1 and
2, checking for agreement and asking for additional detail
where appropriate. To promote free and open discussions
on sometimes opposing ideas from both groups, we did
not reveal the source of the ideas. We also encouraged
participants of subsequent FGs to volunteer their own bar-
riers, facilitators, and suggestions for improvement. Dis-
cussions were digitally audiorecorded and transcribed.
Data analysis
The FGs (370 minutes total) yielded a total of 216 tran-
script pages. Using techniques adapted from grounded
theory [27] and content analysis [28], two coders inde-
pendently coded the transcripts using ATLAS.ti 5.2.17
(ATLAS.ti Scientific Software Development GmBH, Ber-
lin, Germany) identifying perceived barriers, facilitators,
and suggestions for improving the referral process.
Based on this initial coding, the research team then
iteratively developed, refined, and applied a coding tax-
onomy to capture the complexities inherent in the refer-
ral process. Any final discrepancies were resolved by
consensus. This process yielded 120 individual codes
categorized as perceived barriers, facilitators, and sug-
gestions for improving the referral process using CPRS.
Next, the research team organized the code taxonomy
into salient themes (also by consensus), considering
each code’s groundedness (i.e., how often it was men-
tioned by participants) and whether single or multiple
providers mentioned the code. Finally, relationships
among themes were identified by their potential influ-
ences in the overall referral process.
Results
Subject matter expert interviews
Interview data were used to create detailed subspecialty-
specific referral process maps that captured workflow,
information transfer, and actions needed for processing
referrals. We discussed these maps in several debriefing
sessions and despite considerable variations across ser-
vices, we identified a series of shared steps (Figure 2,
steps a-i) in the referral processes based on the dis-
cussed sequences of events, goals, and tasks. These steps
were consistent with previous work on developing a
standardized model of the referrals process [29]. After
one or more primary care encounters (step a), a decision
to refer (step b) is made by the PCP. The PCP initiates
the referral request (step c) using the EHR’s order-entry
interface, which permits the use of predesigned tem-
plates requiring variable amounts of information.
Upon receipt, subspecialists review the requests (step
d) to determine appropriateness, urgency, and complete-
ness, a process that sometimes requires detailed infor-
mation retrieval from the EHR. Subsequently, the
referral review decision is communicated (step e) to the
PCP. Referrals can ultimately be (1) accepted and routed
within the service to have an appointment scheduled, (2)
discontinued, or (3) deferred for further discussion with
additional team members.
If the referral is accepted, a series of steps are initiated
that lead to coordinating the patient’s transition into the
subspecialty setting (step f), including communication
with patients to schedule appointments, providing
reminders, the referral encounter (step g) itself, the
communication of the care plan (step h) to the PCP
through appropriate EHR documentation, and finally, if
appropriate, the coordination of the patient’s transition
back into the primary care setting (step i).
Focus groups
The central emergent theme affecting coordination of e-
referrals was the lack of an institutional referral policy.
We also identified three additional themes that seem to
result from the observed lack of policy: (1) no standar-
dized practices for e-referrals, (2) ambiguous roles and
responsibilities, and (3) inadequate resources to adapt
and respond to incoming referral requests.
Lack of policies and detailed instruction on e-referrals
Both PCPs and subspecialists perceived that lack of clear
institutional policies for several critical steps of the out-
patient referral process, such as rescheduling after no-
shows and patient follow-up, was a barrier to successful
referrals. For instance, they cited that the only two pro-
cesses with an existing clear policy were mandatory
referral requests for review within seven days of submis-
sion and scheduling of referrals within 30 days. How-
ever, instructions or procedures on how to successfully
meet these requirements were lacking.
Subspecialists identified the large volume of referrals
and difficulties reaching patients to schedule appoint-
ments as barriers to complying with the seven-day
review/30-day scheduling policy. They acknowledged the
policy to be well intended but lacked clear procedures
to meet such high performance standards, which led to
its poor implementation.
Well, it’s reviewed within 7 and scheduled within 30.
Um we have played around with that quite a deal, but
it is impossible to get a patient scheduled within 30 days
and it’s not because of the triage process...but it’s getting
a hold of the patient... we contact every patient directly...
we could send letters and we would get, we would be
100% within seven days, but then we would have no-
show rates of 50% so... I think most of ours are reviewed
within 12 days I think on average. –Subspecialist, FG 5
Subspecialists also commented about the need for
clear policies and procedures for handling patients who
do not keep their referral appointments as an important
breakdown in the referral workflow.
Hysong et al. Implementation Science 2011, 6:84
http://www.implementationscience.com/content/6/1/84
Page 4 of 12
Page 5
Figure 2 Referral model based on subject matter expert interviews. We identified three shared stages of the referral process based on the
sequence and purpose of events and tasks: 1) submission of referral request by PCP; 2) referral review by the subspecialist; and 3) patient
transition into subspecialty care. Referral requests are initiated using the EHR’s order-entry interface (Figure 1). Upon receipt, subspecialists review
requests to determine appropriateness, urgency and completeness, a process that could require additional information retrieval from the EHR.
Subsequently, the referral is either: a) accepted and routed within the service to have an appointment scheduled; b) discontinued; or, c) deferred
for further discussion with additional team members. Acceptance triggers a series of steps to coordinate patient transition into the subspecialty
setting, including communication with patients to schedule appointments, followed by appointment reminders, an initial subspecialty encounter,
and finally, communication of care plan back to the PCP through appropriate documentation of the referral encounter in the EHR.
Hysong et al. Implementation Science 2011, 6:84
http://www.implementationscience.com/content/6/1/84
Page 5 of 12
End of preview.
