Journal of Obstetrics and Gynaecology
ISSN: 0144-3615 (Print) 1364-6893 (Online) Journal homepage: https://www.tandfonline.com/loi/ijog20
Potentially avoidable inter-hospital transfer for
gynaecology consultation at a tertiary care centre:
a retrospective study
Rebecca J. Mercier & Sandra Birnbaum
To cite this article: Rebecca J. Mercier & Sandra Birnbaum (2018): Potentially avoidable inter-
hospital transfer for gynaecology consultation at a tertiary care centre: a retrospective study,
Journal of Obstetrics and Gynaecology, DOI: 10.1080/01443615.2018.1468742
To link to this article: https://doi.org/10.1080/01443615.2018.1468742
Published online: 19 Sep 2018.
Submit your article to this journal
Article views: 9
View Crossmark data
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=ijog20
JOURNAL OF OBSTETRICS AND GYNAECOLOGY
https://doi.org/10.1080/01443615.2018.1468742
ORIGINAL ARTICLE
Potentially avoidable inter-hospital transfer for gynaecology consultation at a
tertiary care centre: a retrospective study
Rebecca J. Mercier and Sandra Birnbaum
Department of Obstetrics and Gynecology, Sidney Kimmel Medical College of Thomas Jefferson University, Philadelphia, PA, USA
ABSTRACT KEYWORDS
Inter-hospital transfers for consultation are common and costly in the USA. Our objective was to evalu- Gynaecology; transfer;
ate the inter-hospital transfers between the emergency departments (ED) for a gynaecology consult- telemedicine; pregnancy
ation and to identify markers for potentially avoidable transfers. We performed a retrospective chart of unknown location;
spontaneous abortion
review of all transfers accepted by a tertiary care hospital gynaecology service via the ED over two
years. Our primary outcome was the designation of the transfer as ‘potentially avoidable’, defined as a
patient discharged home directly from the ED, with no workup or treatment prior to their discharge.
The Chi-square tests were used to assess what patient characteristics and medical diagnoses are associ-
ated with potentially avoidable transfers. Of 156 patients meeting the inclusion criteria, a total of 38
(24.4%) were potentially avoidable transfers. Women with potentially avoidable transfers were more
likely to be pregnant than those whose transfers were necessary (63.2% vs. 40.7% p ¼ .02), and more
likely to specifically have a pregnancy of unknown location (PUL) or a complete abortion (p < .01).
IMPACT STATEMENT
What is already known on this subject? In the USA, the emergency department is a common site
for the evaluation of women with primary gynaecologic complaints. Many hospitals lack a specialist
consultation, and obtaining a consultation may require the inter-hospital transfer of patients. Inter-
hospital transfers overall cost more than a billion US dollars per year, and the gynaecologic care of
women may account for a significant portion of this cost. There is scant data describing the most
common diagnoses that lead to the transfer of women with gynaecologic complaints; a better
understanding of the patterns in this area could assist in designing more cost-effective and
convenient models of care.
What do the results of this study add? Our study confirms the pre-existing clinical impression,
which previously had not been quantified, that many women who are transferred for gynaecologic
consultation are stable, and these transfers may be avoidable. We demonstrate that women with a
pregnancy of an unknown location and with a complete abortion are frequent candidates for
avoiding a transfer.
What are the implications of these findings for clinical practice and/or future research?
Providers should feel confident in considering a remote or telemedicine consultation for women
seeking care in the ED for these common conditions. Future research may focus on the longitudinal
follow-up of such systems to demonstrate the patient safety outcomes and patient and provider
satisfactions.
Introduction
some facilities. Non-urgent visits specifically for obstetrical
Emergency room (ER) facilities provide a common point of and gynaecological care occur at a similar frequency (Afilalo
contact for medical care in the US health care system – 131 et al. 2004; Aksoy et al. 2015). Overall, women have a 20%
million ER visits occurred in 2011 (Weiss et al. 2014). A recent higher rate of ER visits than men, and are more likely to have
analysis estimates that the ER care accounts for 5 to 10% of their visit result in a discharge rather than in an admission to
the total national health care costs (Lee et al. 2013). ER the hospital (Weiss et al. 2014).
crowding is associated with various adverse outcomes, includ- The transfer of a patient from an emergency department
ing an increased patient mortality, transport and treatment to a different acute-care facility is common, with approxi-
delays, and financial effect (Hoot and Aronsky 2008). Women mately 1.5% of ER encounters resulting in transfer
are increasingly seeking care in the ER setting; a 6% increase (Kindermann et al. 2013). In a cost-benefit model created by
in ER visits by women was observed between 2006 and 2011 the Center for Information Technology Leadership, the annual
(Skinner et al. 2014). In general, non-urgent visits to ER facili- baseline cost of the ER to ER transfers was estimated at $1.39
ties are common, accounting for up to 82% of all visits at billion for approximately 2.2 million patient transfers
CONTACT Rebecca J. Mercier [email protected] Department of Obstetrics and Gynecology, Sidney Kimmel Medical College of Thomas
Jefferson University, 833 Chestnut Street, First Floor, Philadelphia, PA 19107, USA
ß 2018 Informa UK Limited, trading as Taylor & Francis Group
2 R. J. MERCIER AND S. BIRNBAUM
(Pan et al. 2008). The transfer is often necessary to obtain a attending from July 2013 to July 2015 at TJUH. The records
subspecialty or a higher-acuity care; however, many transfers were reviewed for all of the transfers accepted by a TJUH OB/
may be made unnecessarily, e.g. to obtain a non-urgent spe- Gyn attending. Records were included in the initial review if
cialist consultation, for the coordination of care, or simply for they were transferred for an OB/Gyn indication, or if the pri-
medico-legal reasons. If such transfers could be avoided by mary complaint was non-OB/Gyn, but the OB/Gyn service was
using other models of care, there could be significant cost contacted due to a patient’s pregnant status. Patients were
savings to the healthcare system. included in the study if they were non-pregnant or pregnant
A proxy measure for a potentially avoidable transfer is the at less than 16 weeks’ gestational age, and were evaluated
discharge of the patient directly from receiving ER with no by the OB/Gyn service in accepting the ER setting. The
further workup or evaluation. One retrospective study patients were excluded if they were pregnant and >16
reported a high post-transfer emergency department dis- weeks’ gestational age; were accepted for a direct admission
charge rate among the patients with pregnancy-related con- to the in-patient service; if the transfer happened to another
ditions. However, their analysis did not include any other service prior to OB/Gyn evaluating patient, and if the transfer
gynaecological conditions (Kindermann et al. 2015). Minimal was for a post-partum neo-natal indication.
data available regarding the frequency, the characteristics, or The data were abstracted from the electronic EMR includ-
the disposition of patients transferred between the emer- ing patient demographics, name of the transferring facility,
gency departments for gynaecological complaints. Given the pre-transfer diagnosis, laboratory testing, imaging, medica-
suspected high rates of non-urgent gynaecological care tion administration and the operative intervention performed
received in the ER, we hypothesised that a large proportion following the transfer, the post-transfer diagnosis and the
of ER transfers done for gynaecologic patients are likely to be eventual patient disposition to their admission or discharge.
avoidable. If these patterns could be clearly described it may Our primary outcome was the designation of the transfer
facilitate more efficient and cost-effective models of care in as either ‘necessary’ or ‘potentially avoidable’ (Figure 1).
the appropriate cases. We performed a retrospective study of Transfers were classified as necessary if the patient received
ER transfers accepted by the gynaecology service at an urban at least one of the following at the accepting facility: further
academic tertiary care hospital, in order to determine what workup (laboratory or imaging), treatment (operative inter-
proportion of these transfers were potentially avoidable and vention/procedure, blood transfusion, antibiotics, pain medi-
to delineate which of the patient characteristics were cation or other medical therapy), or an admission (in-patient
associated with potentially avoidable transfers. or observation). The transfers were classified as potentially
avoidable if after their arrival at the accepting facility, there
was no further medical workup or intervention of any kind
Materials and methods other than the consultation and physical exam by the
We performed a retrospective chart review based on the accepting team, and the patient was discharged directly
transfer records at Thomas Jefferson University Hospital from the ER.
(TJUH), an urban tertiary care centre. This study was The Chi-square test and the Fischer exact test were used
approved by the Thomas Jefferson University institutional to compare the patient characteristics between the groups of
review board. At TJUH, all of the transfers into the hospital necessary and potentially avoidable transfers. No power cal-
are coordinated by the Transfer Center office. The OB/Gyn culation was performed, as this was an exploratory analysis
service at TJUH routinely accepts transfers from an outlying using a convenience sample of women at a single institution.
affiliate campus and several area community hospitals which The data analysis was performed using the STATA version 12
do not have in-house OB/Gyn services. Via a transfer centre (StataCorp, College Station, TX).
call, a provider at the referring hospital speaks directly with
the TJUH OB/Gyn attending to discuss with the patient and
Results
decide the disposition. In general, all patients are accepted
for a transfer if the sending emergency department provider Three hundred and nine transfer patients were accepted by
thinks it is necessary; transfer requests are essentially never OB/Gyn attending physicians between July 2013 and July
rejected. Patients may be admitted directly to the in-patient 2015. Of these, a total of 153 were excluded (Figure 1). One
service, but in cases where it is unclear if the patient requires hundred and fifty-six transfer patient charts were included in
a full hospital admission, the transfers are accepted by the our analysis (Table 1). Of these, 87 patients (55.8%) required
OB/Gyn attending but brought to the emergency department further workup with laboratory studies or imaging, 108
for an evaluation. The OB/Gyn service completes an evalu- patients (69.7%) required one or more forms of treatment,
ation and consultation there, and the disposition to either and 89 patients (56.5%) required admission to the OB/Gyn
the in-patient or to discharge is determined. It is an institu- service. A total of 38 patients (24.4%) did not require any fur-
tional practice to evaluate all the pregnant patients under 16 ther workup, treatment or admission after an assessment by
weeks of gestation in the emergency room setting; patients OB/Gyn and were discharged home from the emergency
over 16 weeks of gestation are evaluated on the Labour and department. These were classified as potentially avoid-
Delivery triage unit instead, and are often admitted directly able transfers.
to the obstetrics service. We then compared the characteristics of patients who had
The transfer centre records were used to identify all of the necessary transfers with the patients who had potentially
transfers which were accepted by obstetric or gynaecology avoidable transfers (Table 2). The age, race and insurance
JOURNAL OF OBSTETRICS AND GYNAECOLOGY 3
All outside hospital transfers
accepted by TJUH OB/GYN from
7/2013 to 7/2015
(N = 309)
Excluded (N = 153)
Pregnancy > 16 WGA (123)
Direct admission to inpaent service (12)
Transfer to service other than OB/GYN (6)
Transfer for neonatal indicaons (4)
Le AMA prior to OB/GYN evaluaon (1)
Incomplete record (7)
Eligible for analysis
(N = 156)
Did Paent Receive
Further work-up Treatment Admission to hospital
Labs OR/Procedure Inpaent
Imaging Transfusion Observaon
Other medical therapy
Yes to any of the above No to all of the above
Necessary transfer Potenally avoidable transfer
N = 118 N = 38
Figure 1. Inclusion and classification of cases.
Table 1. Components of post-transfer evaluation. pregnancy type in the avoidable transfer group was a preg-
N ¼ 156 nancy of unknown location (PUL) (66.7%) and a complete
n (%) abortion (20.8%), while an intrauterine pregnancy and an
Further workup performed 87 (55.8) incomplete abortion were more common in the necessary
Laboratory testing 79 (50.1) transfer group (29.2 and 25%, respectively).
Imaging 38 (24.4)
Treatment provided 108 (69.7) Overall, the diagnoses differed significantly between the
Operative procedure± 40 (25.6) groups (p < .01). In the potentially avoidable group, PUL was
Blood transfusion 13 (8.3) the transferring diagnosis for 42% of patients and a pelvic
Pain control 19 (12.2)
Antibiotics 27 (17.3) mass concerning for torsion accounted for 21.1%. No single
Other medical therapy 56 (35.9) pre-transfer diagnosis dominated in the necessary transfer
Multiple interventions 36 (23.1)
group. The diagnosis assigned after the assessment by the
Admission required 89 (57)
Inpatient 79 (50.1) accepting service (the post-transfer diagnosis) also differed
Observation only 10 (6.4) between the groups. 36.8% of the potentially avoidable trans-
Transfer status
Necessary 118 (75.6)
fers had a diagnosis of PUL, and 28.9% of this group had a
Avoidable 38 (24.4) final diagnosis which was classed as ‘other’. Again, in the
Several patients received multiple procedures. necessary transfer group, no single diagnosis was dominant.
± either in OR or bedside procedure. All of the patients with a diagnosis of an ectopic pregnancy,
a pelvic mass concerning for malignancy, a post-operative
status were similar between the groups. A greater proportion complication, of post-partum complication, of opiate use in
of unnecessary transfers were seen from the affiliate campus pregnancy, of pelvic infection, and a ruptured cyst were des-
rather than the unaffiliated hospitals. Pregnant women were ignated as necessary transfers.
more likely than the non-pregnant women to have had an We also analysed whether a change in the diagnosis pre-
avoidable transfer (63 vs. 41%, p ¼ .02), and the pregnancy and post-transfer differed between the groups, and found
status did differ between the groups. The most common that a discordance in diagnosis was more common in the
4 R. J. MERCIER AND S. BIRNBAUM
Table 2. Characteristics of necessary and avoidable transfer patients.
Transfer necessary Transfer avoidable
n ¼ 118 n ¼ 38
Characteristics Mean (sd) or n (%) Mean (sd) or n (%) p-value
Age 33.3 (1.15) 29.1 (1.17) .05
Race .06
White 42 (35.6) 9 (23.7)
Black 51 (43.2) 20 (52.6)
Asian/Pacific Islander 13 (11.0) 6 (15.8)
Hispanic 10 (8.5) 3 (7.9)
Other 2 (1.7) 0
Insurance status .48
Commercial 35 (29.7) 12 (31.6)
Medicaid/Medicare 57 (48.3) 21 (55.3)
Uninsured 26 (22.0) 5 (13.2)
Transferring site <.01
TJUH affiliate 85 (72) 37 (97)
Non-affiliate community hospital 33 (28) 1 (3)
Pregnant 48 (40.7) 24 (63.2) .02
Pregnancy status <.01
Intrauterine, confirmed 14 (29.2) 2 (8.3)
Ectopic, confirmed 9 (18.8) 0 (0)
Pregnancy of unknown location 9 (18.8) 16 (66.7)
Abortion, complete 3 (6.3) 5 (20.8)
Abortion, incomplete 12 (25) 1 (4.2)
Molar 1 (2.1) 0 (0)
Specific pre-transfer diagnosis <.01
PUL 7 (5.9) 16 (42.1)
Vaginal bleeding, non-pregnant 10 (8.5) 3 (7.9)
Ectopic 10 (8.5) 0 (0)
Pelvic mass, c/f torsion 10 (8.5) 8 (21.1)
Pelvic mass, c/f malignancy 6 (5.1) 0 (0)
Postoperative complication 9 (7.6) 0 (0)
Postpartum complication 12 (10.2) 0 (0)
Opiate use in pregnancy 7 (5.9) 0 (0)
PID/TOA/Pelvic infection 13 (11.0) 0 (0)
Nausea/vomiting of pregnancy 5 (4.2) 1 (2.6)
Ruptured cyst 2 (1.7) 0 (0)
Vaginal bleeding with abortion 16 (13.6) 5 (13.2)
Other 8 (6.8) 4 (10.5)
Specific post-transfer diagnosis <.01
PUL 3 (2.5) 14 (36.8)
Vaginal bleeding, non-pregnant 9 (7.6) 3 (7.9)
Ectopic 12 (10.2) 0 (0)
Pelvic mass, c/f torsion 6 (5.1) 1 (2.6)
Pelvic mass, c/f malignancy 6 (5.1) 0 (0)
Postoperative complication 6 (5.1) 0 (0)
Postpartum complication 12 (10.2) 0 (0)
Opiate use in pregnancy 7 (5.9) 0 (0)
Complication of GYN cancer 4 (3.4) 1 (2.6)
PID/TOA/Pelvic infection 13 (11.1) 0 (0)
Nausea/vomiting of pregnancy 5 (4.2) 1 (2.6)
IUP, no other pathology 1 (0.8) 1 (2.6)
Ruptured cyst 2 (1.7) 0 (0)
Vaginal bleeding with abortion 17 (14.4) 6 (15.8)
Other 13 (11.1) 11 (28.9)
Discordant pre/post transfer diagnosis 11 (9.3) 9 (23.7) .02
Percentage of patients with pregnancy.
Table 3. Changes in diagnosis following transfer (n ¼ 20). 30%. The updated diagnosis for the patients transferred to
Pre-transfer diagnosis (n) Post-transfer diagnosis (n) rule out torsion included an ovarian cyst, fibroids, pelvic pain
Pregnancy of unknown location (6) Ectopic (2); PID (1) IUP (1) and a ruptured cyst; the updated diagnosis for PUL was gen-
Incomplete abortion (2) erally a specification of the pregnancy location.
Vaginal bleeding, non-pregnant (1) Complication of GYN cancer (1)
Pelvic mass, concerning for torsion (11) IUP (1); Other (10)
Post-operative complication (1) Other (1)
PID/TOA/Pelvic infection (1) Other (1) Discussion
Of the 156 transfer patients who were included in our ana-
potentially avoidable transfer group (23.7 vs. 9.3%, p ¼ .02). lysis, nearly 25% did not require any further workup, medical
Overall, diagnosis changed in 20 patients (12.8%) (Table 3). or operative treatment, or admission after their transfer for
The most common discordant diagnosis was a pelvic mass OB/GYN evaluation. Our data indicate that women were more
concerning for torsion, which accounted for 55% of the dis- likely to have an avoidable transfer if they were pregnant,
cordant diagnoses, followed by PUL which accounted for specifically with a pregnancy of an unknown location or a
JOURNAL OF OBSTETRICS AND GYNAECOLOGY 5
complete abortion, or if the transferring ER had a concern for challenges, since the time of this study, our institution has
ovarian torsion. implemented a protocol for stable patients with a pregnancy
Given that abdominal pain and vaginal bleeding in an of an unknown location meeting specific criteria to be dis-
early pregnancy are common presenting complaints in the charged directly from the initial ER with a planned follow-up
emergency department, it was anticipated that the pregnancy in the ‘beta book’ without an in-person OB/GYN consultation.
of unknown location and a complete abortion would be fre- Approximately, 30% of the PUL consults are now conducted
quent indications for transfer. Clinical experience suggests in this way, and there has been a reduction in the number of
that these patients are frequently stable, have received an transfers from external facilities with no increase in patient
adequate primary workup at the transferring institution, and morbidity (Internal QI data).
that a consultation with an OB/GYN often involves only coun- Telemedicine may represent another option for the care in
selling for the patient. Our finding that 70% of these PUL this situation. Interest in care provision through telemedicine
transfers and that 62% of the complete abortion transfers has grown in recent years, as it provides a way for the
were potentially avoidable supports this clinical impression. patients to directly interact with the providers without the
These findings suggest that in certain patients, a suspected difficulty of travel, or in this scenario, with a transfer.
diagnosis of PUL or complete abortion may be ideal for alter- Telemedicine has been used in many fields with a good clin-
nate methods of patient evaluation and the assumption of ical success (Pan et al. 2008; Weinstein et al. 2014). The use
care beyond the typical model of an inter-hospital transfer. of a telemedicine consult has been shown to increase ER effi-
The evaluation for ovarian torsion was the third most com- ciency with reductions in the length of stay and the time to
mon indication for a transfer in our study, and 44% of the consult completion (Southard et al. 2014). Telemedicine may
patients who were transferred with a concern for an ovarian decrease the utilisation and improve the disposition of
torsion were found to have had a potentially avoidable patients in need of transport for critical issues (Fugok and
transfer. These findings should be considered with caution. A Slamon 2018). For non-critical problems, a telemedicine con-
diagnosis of ovarian torsion often depends on the clinical sultation may conservatively alter the management decisions;
judgement of the provider more than on the lab and imaging a systematic review found that the transfer rates overall
findings. Many ER physicians may not feel confident in ruling increased but unnecessary transfers decreased (Du Toit et al.
out this diagnosis without the expert assessment of the 2017). Much work in telemedicine has focussed on rural set-
patient and this clinical diagnosis is difficult to make without tings, where the access and transfer may be more logistically
directly examining the patient. As a missed diagnosis of complex and expensive, but these findings are likely applic-
ovarian torsion may have significant sequelae, a transfer for able to the urban emergency room setting, as well (Sharma
gynaecological evaluation was likely warranted for many of et al. 2017). In considering the gynaecology patients, one
these patients, even if no other workup or treatment valuable component of the OB/GYN consultation may be the
was performed. counselling, reassurance, and management of patient expect-
The ability to transfer a patient when needed is essential ations and the understanding. Especially when the transfers
in providing emergency care. However, in many scenarios, involve providers at an outside institution, and patients are
several approaches to avoiding a transfer are possible, and not previously known to the consulting specialist, telemedi-
are likely safe models of patient care. For many diagnoses, a cine may improve the provider and patient comfort with a
patient with a non-urgent presentation may not need evalu- remote consultation. The incorporation of these modalities
ation by a specialist immediately in the ER; an outpatient into the care of the gynaecologic patient could be associated
assessment made in a timely manner may be adequate. The with significant cost savings for the healthcare system and
direct and expedited referral to primary care following ER vis- reduce the clinical burden on the accepting institutions.
its is associated with the improved rates of outpatient follow Our study has several strengths. We were able to compre-
up (Doran et al. 2013) and interventions to improve the refer- hensively review charts representing all transfers to a large
rals to primary and specialist care have been shown to academic institution that accepts transfers from both its own
reduce the future ER utilisation in some settings (Murnik remote campus and several area community hospitals. While
et al. 2006; Block et al. 2013). In the UK and Australia, early our experience is likely comparable to other urban academic
pregnancy units – outpatient walk-in clinics which do not institutions, our study is limited in that our patient popula-
require a referral or appointments – often provide the neces- tion and the findings may not be more precisely generalis-
sary follow-up care for the PUL patients seen after-hours in able to other rural areas. Our analysis is limited by a lack of
the emergency department. For non-urgent gynaecologic prospective information; we were able to ensure that patients
complaints, rather than a transfer for evaluation, a phone did not have complications documented in their TJUH
consultation between providers, with arrangements for expe- records, but we cannot be certain if other events subse-
dited outpatient care may meet the patient needs safely. In quently happened at other area facilities. Future research
settings which lack a universal health coverage such as the should include similar analyses in other practice settings, and
USA, the coordination of accessible outpatient care can be the prospective follow-up of patients who are cared for under
logistically challenging. Patients without insurance may not alternative modalities such as telemedicine to ascertain both
be able to pay required up-front fees to be seen at an ambu- the safety and patient and provider satisfaction. Such
latory centre. Patients may also be insured, but have an insur- research could provide a basis for the development of local
ance type that is not accepted at the clinics affiliated with or regional guidelines to enhance the cooperation and coord-
the facility where they were initially seen. Despite these ination between institutions.
6 R. J. MERCIER AND S. BIRNBAUM
A substantial number of the ER transfers for an OB/GYN Fugok K, Slamon NB. 2018. The effect of telemedicine on resource utiliza-
evaluation may be potentially avoidable. Avoiding an inter- tion and hospital disposition in critically ill pediatric transport patients.
Telemedicine Journal and e-Health 24(5):367–374.
hospital transfer may have benefits both to the patient and
Hoot NR, Aronsky D. 2008. Systematic review of emergency department
the larger healthcare system in terms of the cost, conveni- crowding: causes, effects, and solutions. Annals of Emergency
ence and satisfaction. Pregnancy increases the likelihood for a Medicine 52:126–136.
potentially avoidable transfer, with the most frequent diagno- Kindermann D, Mutter R, Pines JM. 2013. Emergency department trans-
ses including a pregnancy of unknown location and a com- fers to acute care facilities, 2009: Statistical Brief #155. Healthcare
plete abortion. In making decisions regarding patient care, Cost and Utilization Project (HCUP) Statistical Briefs. Rockville (MD):
Agency for Healthcare Research and Quality.
when clinically appropriate, providers should consider alterna-
Kindermann DR, Mutter RL, Houchens RL, Barrett ML, Pines JM. 2015.
tive methods of care such as expedited outpatient care and Emergency department transfers and transfer relationships in United
telemedicine provision beyond the traditional transfers. States hospitals. Academic Emergency Medicine: Official Journal of the
Society for Academic Emergency Medicine 22:157–165.
Lee MH, Schuur JD, Zink BJ. 2013. Owning the cost of emergency medi-
Disclosure statement cine: beyond 2%. Annals of Emergency Medicine 62:498–505.e3.
Murnik M, Randal F, Guevara M, Skipper B, Kaufman A. 2006. Web-based
No potential conflict of interest was reported by the authors. primary care referral program associated with reduced emergency
department utilization. Family Medicine 38:185–189.
Pan E, Cusack C, Hook J, Vincent A, Kaelber DC, Bates DW, et al. 2008.
References The value of provider-to-provider telehealth. Telemedicine Journal and
Afilalo J, Marinovich A, Afilalo M, Colacone A, Leger R, Unger B. 2004. e-Health: The Official Journal of the American Telemedicine
Nonurgent emergency department patient characteristics and barriers Association 14:446–453.
to primary care. Academic Emergency Medicine: Official Journal of the Sharma R, Fleischut P, Barchi D. 2017. Telemedicine and its transform-
Society for Academic Emergency Medicine 11:1302–1310. ation of emergency care: a case study of one of the largest US inte-
Aksoy H, Aksoy U, Ozturk M, Ozyurt S, Acmaz G, Karadag OI, et al. 2015. grated healthcare delivery systems. International Journal of Emergency
Utilization of emergency service of obstetrics and gynecology: a cross- Medicine 10:21.
sectional analysis of a training hospital. Journal of Clinical Medicine Skinner HG, Blanchard J, Elixhauser A. 2014. Trends in emergency depart-
Research 7:109–114. ment visits, 2006-2011: Statistical Brief #179. Healthcare Cost and
Block L, Ma S, Emerson M, Langley A, de la Torre D, Noronha G. 2013. Utilization Project (HCUP) Statistical Briefs. Rockville (MD): Agency for
Does access to comprehensive outpatient care alter patterns of emer- Healthcare Research and Quality.
gency department utilization among uninsured patients in East Southard EP, Neufeld JD, Laws S. 2014. Telemental health evaluations
Baltimore? Journal of Primary Care and Community Health 4:143–147. enhance access and efficiency in a critical access hospital emergency
Doran KM, Colucci AC, Hessler RA, Ngai CK, Williams ND, Wallach AB, department. Telemedicine and e-Health 20:664–668.
et al. 2013. An intervention connecting low-acuity emergency depart- Weinstein RS, Lopez AM, Joseph BA, Erps KA, Holcomb M, Barker GP,
ment patients with primary care: effect on future primary care linkage. et al. 2014. Telemedicine, telehealth, and mobile health applications
Annals of Emergency Medicine 61:312–321.e7. that work: opportunities and barriers. The American Journal of
Du Toit M, Malau-Aduli B, Vangaveti V, Sabesan S, Ray RA. 2017. Use of Medicine 127:183–187.
telehealth in the management of non-critical emergencies in rural or Weiss AJ, Wier LM, Stocks C, Blanchard J. 2014. Overview of emergency
remote emergency departments: a systematic review. Journal of department visits in the United States, 2011: Statistical Brief #174.
Telemedicine and Telecare 1357633X1773423. http://dx.doi:10.1177/ Healthcare Cost and Utilization Project (HCUP) Statistical Briefs.
1357633X17734239. [Epub ahead of print]. Rockville (MD): Agency for Healthcare Research and Quality.
