a b s t r a c t

Objective: We aim to evaluate the effectiveness of a broadly inclusive, comparatively low intensity intervention linking ED patients to a primary care home.

Methods: This retrospective cohort study evaluated ED patients referred for primary care linkage in a large, urban, academic ED. A care coordination specialist performed a brief interview to gauge access barriers and provide a clinic referral with optional scheduling assistance. Data were abstracted from program records and the electronic medical record. The primary outcome was the proportion of referred individuals who attended at least one primary care appointment.

Secondary outcomes included return ED encounters within one year, and factors associated with linkage outcomes. Results: There were 2142 referrals made for 2064 patients; 1688/2142 accepted assistance. Linkage was successful for 1059/1688 (63%, CI95 60% to 65%). Among patients accepting assistance, those without successful linkage were younger (41 vs 45 years, difference 3 years, CI95 2 to 3), more often male (62% vs 55%,difference 7%, CI95

2% to 12%), and less likely to have a Chronic medical condition (37% vs 45%, difference 8%; CI95 3% to 12%) or to have had an appointment scheduled within two weeks (26% vs 33%, difference 7%, CI95 2% to 12%). Insurance status and self-reported barriers to care were not associated with linkage success. Patterns of subsequent ED use were sim- ilar, regardless of referral status or linkage outcome.

Conclusion: Low intensity, broadly inclusive, ED care coordination linked nearly 50% of patients referred for intervention, and two-thirds of willing participants, with a primary care home.

(C) 2018

1. Introduction

Emergency departments (EDs) are intended to be readily accessible and always available. As such, they are frequently used for conditions that could be appropriately evaluated in a non-ED setting. Avoidable use of the ED increases with barriers to primary care, such as limited availability of prompt appointments, lack of health insurance, and a lack of reliable transportation [1-4].

Using the ED for clinical presentations amenable to primary care is suboptimal in several respects. It is commonly believed that EDs are a more costly venue than other alternatives [5-7]. Given a predisposition to search for emergent conditions and lack of familiarity with the

? Prior Presentations: Presented at the Society for Academic Emergency Medicine Annual Meeting, Dallas, May, 2014.

* Corresponding author at: 51 N. 39th Street - M01, Philadelphia, PA 19104, USA.

E-mail address: [email protected] (S.D. Foster).

patient, emergency clinicians may initiate an extensive diagnostic evaluation in situations where a primary care provider would not [8,9]. Episodes of ED care are also typically fragmented from other ongo- ing healthcare evaluations [8-10]. Problems created by care fragmenta- tion, including Hospital readmissions, Medication errors and redundant medical testing, are increasingly recognized in discussions of U.S health policy [11-13].

Care coordination interventions have been proposed as a partial solu- tion to the problem of care fragmentation [12]. Care coordination can be defined as “…the deliberate organization of Patient Care Activities between two or more participants…to facilitate the appropriate delivery of health care services. Organizing care involves the marshaling of per- sonnel and other resources needed to carry out all requirED patient care activities, and is often managed by the exchange of information among participants…” [13]. A recent study found the United States to have the highest rate of poor primary care coordination among a group of 11 high-income countries, but that having an established primary care phy- sician was associated with significantly better care coordination [14].

https://doi.org/10.1016/j.ajem.2018.04.005

0735-6757/(C) 2018

The role of EDs in care coordination has yet to be fully elucidated. A recent systematic review identified 23 articles encompassing a broad range of interventions, study designs and efficacies [15]. Two-thirds demonstrated a positive effect on the primary outcome measure, gener- ally focusing on post-discharge treatment plans and obtaining primary care follow-up. Seven studies described a change in subsequent ED uti- lization patterns, of which four demonstrated a paradoxical increase [16-19]. However, the majority of studies included in this review were randomized, controlled trials, limiting external validity. Further, only four included a general, adult, ED population [20-23], of which two were limited to low income or Uninsured patients only [20,21]. As such, the potential impact of an ED-based care coordination initiative remains unknown for the majority of ED patients.

Located in an urban environment, our regional safety net hospital and ED serve a population with substantial barriers to Primary care access. As an effort to increase access to primary care for ED patients, in 2009 our ED began employing a dedicated, full-time care coordination specialist to facilitate primary care follow-up after ED discharge. In contrast to many of the previously descriBed interventions, our program was insti- tuted on a clinical rather than research basis. Accordingly, our program, while created primarily to target uninsured patients with chronic medi- cal conditions, acceptED referrals purely on the basis of clinician discre- tion rather than requiring highly specific selection criteria.

We sought to evaluate the effectiveness of this low intensity, broadly

inclusive, ED-based care coordination intervention in achieving linkage to primary care. Secondary outcomes included return ED encounters within one year and patient and program-level factors potentially asso- ciated with linkage outcomes.

1. Methods
   1. Study design

This retrospective cohort study was approved by the local Institu- tional Review Board with a waiver of informed consent.

Setting and population

The care coordination program was implemented in 2009 in an urban, academic ED with approximately 80,000 almost exclusively adult encounters per year. It is the only adult Level I trauma center in the region and provides care for many of the region’s uninsured. The study population included all patients presenting to the ED, including the subset of patients referred for the care coordination program.

Care-coordination program

The program employed a care coordination specialist during week- day business hours. If ED staff referred a patient during business hours, the care coordination specialist interacted with the patient face-to-face concomitant to their clinical team’s diagnostic evaluation. If a referral was requested outside of regular business hours, the care coordination specialist contacted the patient by phone the next business day. In either case, patients who accepted assistance completed a structured interview pertaining to demographics and self-perceived barriers to primary care. Based on this information, the care coordinator sought to identify the pri- mary care location most likely to meet the patient’s needs. For example, a clinic might be selected based on geographic location, foreign language competency, or copay requirements. Patients were then provided, ac- cording to their preference, with either an appointment time or with in- structions for obtaining an appointment. If the appointment time was known (i.e. made by the care coordinator) and greater than 48 h after the interview, an attempt was made to contact the patient 24 h prior to the appointment to confirm the appointment details. The care coordi- nator also contacted the clinic after the scheduled time to determine whether the appointment was kept and, if not, contacted the patient to

determine their reason for missing it. Patients were offered a one-time rescheduling opportunity following the same protocol if they did not at- tend their initial appointment. If the patient failed to attend the second scheduled appointment, no further interventions were made.

The care coordination program was implemented on a clinical rather than research basis. It was designed to be highly feasible (i.e. relatively low intensity) and to target those with chronic medical conditions and barriers to primary care. However, aside from the program announce- ment at inception, providers received no ongoing training regarding its utilization and the care coordination specialist did not exclude pa- tients based on their reason for referral. As such, the program reflected practice rather than a research protocol, and referral was at the discre- tion of treating providers alone.

Data collection

Evaluation data for the care coordination program were prospec- tively and systematically collected as a part of routine operations. Subjects for this study were identified from those records. This included all patients who were referred for the intervention during their ED visit from January 2010 through December 2011. We excluded those under the age of 18, those identified as prisoners, and patients who cited “legal troubles” as a barrier to primary care access. These subgroups were considered to be distinct from the general population and were too small to justify subgroup analysis. Other patients presenting to the ED who were not referred for care coordination were identified by elec- tronic query of the ED’s electronic medical record and cross referencing with the care coordination database.

Data analysis

Analysis involved four patient groups: 1) those referred for care co- ordination and successfully linked, 2) those referred for and accepting care coordination but not successfully linked, 3) those referred for but declining care coordination, and 4) those presenting to the ED during the study period but not referred for care coordination.

The Chi-Square test was used to test for differences in proportions and independent samples t-tests were used to test for differences in means. Effect sizes and 95% confidence intervals were calculated. All sta- tistical analyses were conducted using SPSS 22.0 (IBM Corporation, Armonk, NY). All figures were created using R (base, 3.2.5). To better characterize the referral population, a set of chronic medical conditions felt to be amenable to primary care was defined (Appendix 1) and the prevalence of these conditions among the patient groups was assessed.

Outcome measures

The primary outcome measure was the proportion of patients successfully linked to primary care. Successful linkage was defined as having attended at least one appointment with the outpatient provider as scheduled or arranged during the intervention. Secondary outcome measures included basic demographics and the proportion of patients with return visits to the ED within one year of the index visit, as well as patient and program-level factors associated with success or failure of linkage.

1. Results

During the study period, there were 157,389 visits to the ED made by 85,701 unique patients. Of these unique patients, 2064 were referred to the care coordination specialist. Of referred patients, 1987 (96%) were referred only once; 76 patients were referred twice, and one patient was referred three times, for a total of 2142 patient referrals. Because patients who were referred more than once had the potential for differ- ent outcomes on subsequent referrals, referrals rather than unique pa- tients were used as the primary unit of analysis. Analysis of ED

Table 1

Characteristics of referred and non-referred patients at initial visit.

Referred Non-referred

Successful linkage Unsuccessful linkage Assistance declined

Age - mean (SD)	45	[12]	41	[12]	43	[12]	40	[17]
Race African-American	646	(61.0)	403	(64.1)	262	(57.7)	34,581	(41.3)
Caucasian	338	(31.9)	187	(29.7)	154	(33.9)	39,386	(47.1)
Other/not documented	63	(5.9)	33	(5.2)	30	(6.6)	8061	(9.6)
Hispanic	6	(0.6)	5	(0.8)	7	(1.5)	1146	(1.4)
Asian	6	(0.6)	1	(0.2)	1	(0.2)	463	(0.6)
Male	589	(55.6)	392	(62.3)	273	(60.1)	43,100	(51.5)
Discharged home	1024	(96.7)	603	(95.9)	440	(96.9)	60,008	(71.8)
Payor Self-pay	770	(72.8)	455	(72.3)	366	(80.8)	35,005	(42.1)
Medicaid	177	(16.7)	101	(16.1)	41	(9.1)	16,616	(20.0)
Medicare	56	(5.3)	31	(4.9)	8	(1.8)	11,369	(13.7)
Commercial	16	(1.5)	9	(1.4)	9	(2.0)	15,166	(18.2)
Other	39	(3.7)	33	(5.2)	29	(6.4)	5055	(6.1)
Diagnosesa
Any chronic medical condition	475	(44.9)	235	(37.4)	163	(35.9)	24,765	(29.6)
Chronic obstructive lung disease	38	(3.6)	19	(3.0)	5	(1.1)	2509	(3.0)
Asthma	92	(8.7)	61	(9.7)	37	(8.1)	6125	(7.3)
Coronary Artery Disease	18	(1.7)	5	(0.8)	3	(0.7)	3143	(3.8)
Congestive Heart Failure	15	(1.4)	6	(1.0)	2	(0.4)	2186	(2.6)
Hypertension	338	(31.9)	160	(25.4)	116	(25.6)	15,740	(18.8)
Seizures/epilepsy	0	(0.0)	0	(0.0)	1	(0.2)	115	(0.1)
Diabetes mellitus (Type 1 or 2)	134	(12.7)	62	(9.9)	42	(9.3)	7618	(9.1)
	1059		629		454		83,637
a Multiple conditions are possible.

utilization patterns over time was necessarily conducted using unique patients as the unit of analysis. For unique patients, the index ED visit was defined as the First visit leading to referral to the care coordination program or the first ED visit within the study period for patients not re- ferred. Demographic characteristics of the four groups used for analysis are depicted in Table 1.

Primary outcomes

Of the 2142 patient referrals, 1688 accepted assistance (79%, CI₉₅ 77% to 81%) from the program. Linkage was successful for 1059/1688 (63%, CI₉₅ 60% to 65%).

Secondary outcomes

1. Patient factors associated with linkage

Table 2 compares patients who were and were not successfully linked, among those who accepted care coordination assistance. The two groups were similar with regards to race, insurance status, and self-reported barriers to care. However, unlinked patients were slightly younger, with a mean age of 41 (SD 12) than linked patients, mean age 45 (SD 12) (difference in means 3 years, CI₉₅ 2 to 3) and more often male (62% compared to 56%, difference 7%, CI₉₅ 2% to 12%). Unlinked patients were also less likely to have a chronic med- ical condition compared to linked patients (37% vs 45%, difference 8%; CI₉₅ 3% to 12%).

Table 2

Patient and program characteristics of program participants successfully and unsuccessfully linked to care.^a

	Linkage		Linkage		Difference	95% CIb
	Successful		Unsuccessful			Lower	Upper
	N	%	N	%
Patient characteristics
Age - mean (SD)	45	[12]	41	[12]	-2.5	-3.2	-1.9
White	338	(31.9)	187	(29.7)	(-2.2)	(-6.7)	(2.4)
Male	589	(55.6)	392	(62.3)	(6.7)	(1.9)	(11.5)
Insured	289	(27.3)	174	(27.7)	(0.4)	(-4.0)	(4.8)
Any chronic medical conditionc	475	(44.9)	235	(37.4)	(-7.5)	(-12.3)	(-2.7)
Patient-reported barriers to care
Financial	449	(42.4)	286	(45.5)	(3.1)	(-1.8)	(8.0)
Transportation	98	(9.3)	55	(8.7)	(0.5)	(-3.3)	(2.3)
Language	25	(2.4)	11	(1.7)	(-0.6)	(-2.0)	(0.8)
Scheduling	13	(1.2)	10	(1.6)	(0.4)	(-0.8)	(1.5)
Program characteristics
Same day contact with care coordination specialist	515	(49.7)	334	(54.2)	4.5	-0.5	9.5
Two weeks or less to appointment	348	(33.2)	163	(26.2)	-7.0	-11.5	-2.5
	1059	(62.7)	629	(29.4)

^a For patients referred to and accepting of care coordination intervention.

^b Difference in proportions or means, 95% confidence intervals are presented.

^c From list of conditions in Appendix 1.

Program factors associated with linkage to care

Among program participants, same-day contact with the care coor- dinator was initiated for 54% of unlinked patients and 50% of linked pa- tients (difference 5%, CI₉₅-0.5 to 9.5). Unlinked patients were less likely to have an appointment scheduled within two weeks compared to linked patients (26% compared to 33%, difference 7%, CI₉₅ 2% to 12%)

(Table 2).

ED utilization

We determined ED use for each of the three years prior to the index visit and for the year after the index visit for each patient. Patients re- ferred to the care coordination program were more often repeat users (2-3 visits per year) or frequent users (4+ visits per year) in the years prior to the index visit and the year after, compared to non-

referred patients. These differences were particularly pronounced in the year of the index visit and the year after the index visit. For referred patients, ED use was similar in the year after referral, regardless of the referral outcome. Notably, nearly 50% of all referred patients had only one visit the year that they were referred. (Fig. 1).

1. Limitations

The results of our study are subject to several limitations. The gener- alizability of the findings may be questioned given that all subjects were recruited from a single center. This center is the regional safety net hos- pital, and so it is possible that these subjects possessed higher-than- average barriers to care, which would add to the challenge of any coor- dination effort and weaken any observed effect. Additionally, the high

Fig. 1. ED utilization by referral and linkage status prior to and after index encounter.

prevalence of uninsured individuals utilizing the study site, and the limited availability of alternative primary care referral options, may have led to a high rate of program referral as a means of obtaining follow-up for isolated acute, rather than chronic, medical conditions.

1. Discussion

Given the ED’s central role within the healthcare system and ready access to patients whose healthcare is neither coordinated nor well- managed, surprisingly little emphasis has been placed on the role of the ED in promoting care coordination. To date, a variety of ED-based initiatives targeting specific populations, under highly-protocolized conditions, have been reported. In contrast, here we have evaluated the effectiveness of a low intensity and broadly inclusive ED care coor- dination initiative intended to connect individuals with chronic medical illness and barriers to primary care access with a primary care home. Overall, patients frequently accepted assistance. Consistent with prior studies, our program demonstrated a positive effect on our primary outcome measure. Given the relatively low intensity and broadly inclusive nature of the intervention, successful linkage of nearly 50% of all referred subjects and nearly two-thirds of all willing participants represents a meaningful achievement.

The intervention evaluated by this report is notable in terms of clin- ical implementation and relatively low resource requirements. Care co- ordination interventions are often time, labor and personnel intensive, aiming to achieve improvement in specific areas of health or wellness. Many previously described interventions have occurred in the setting of clinical trials, and have incorporated significant additional resources not generally present in the ED setting [15]. Here, we have shown that a single worker, equipped with an awareness of the issues surrounding primary care access and an understanding of the local primary care community, can successfully connect a majority of willing participants with a primary care provider. Although the significance of this finding is tempered by the fact that we did not evaluate the rate of primary care linkage in the absence of this intervention, previous studies have documented a low rate of primary care follow-up after ED discharge, particularly among those with primary care access barriers [24,25]. Moreover, the failure to link a significant number of patients despite direct support in making appointments and encouraging attendance illustrates a problem for which the ED has little remedy, particularly in the absence of significant resource investment.

This intervention is also notable for its intended focus on the general

population of ED patients without a primary care provider. Many care coordination efforts have specifically focused on patients with highly in- tensive use of Healthcare resources or specific diagnoses (e.g. asthma). We are only aware of a single US-based study evaluating the impact of a care coordination intervention on what would reasonably be consid- ered a general ED cohort, which was reported by Kyriacou in 2005 [23]. In their prospective trial, 250 patients presenting to a single, aca- demic center were randomized to an intervention in which a research assistant scheduled a post-ED follow-up appointment, or to Standard care wherein the patient was instructed to arrange their own appoint- ment after discharge. Similar to our findings, they found a higher rate of follow-up among the intervention group. Our study compliments this finding in a number of ways. By virtue of our study design, we were able to include a much larger cohort of patients, and thus, to eval- uate the outcome of referral for a broader range of conditions. In addi- tion, although the inclusion criteria in the study by Kyriacou were broad as compared to the Existing literature base, our inclusion criteria were much broader and thus our results more pragmatic and reflective of the “real-world” potential of such an initiative.

Some of our findings related to factors associated with successful linkage proved intuitive, such as the negative effects of delayed appoint- ment times. This is consistent with previous studies demonstrating a substantial decline in follow-up rates as time from ED discharge lengthens [24,25], a finding which has also been demonstrated in

other healthcare settings [26,27]. Additionally, the positive association between the presence of a chronic medical condition and linkage success is not surprising, given that such patients possess a greater need and likely greater motivation to achieve primary care follow-up. We are not aware of any prior studies specifically addressing this association. Other findings proved more surprising, such as the lack of association between insurance status and self-perceived barriers to care with linkage success. While the retrospective nature of the present analysis limits our ability to draw definitive conclusions, this likely rep- resents the effects of our program’s individualized approach and the beneficial effects of employing a specialist with an intricate understand- ing of the local primary care community. Alternatively, it may suggest that motivation is more of a barrier than are financial and resource issues.

The effects observed in our study occurred without restriction or significant guidance regarding criteria for referral or follow-up plan- ning. Education about the program occurred largely through “word of mouth” and direct outreach from the care coordination specialist to individual providers, rather than formal education or training. As a result, the prevailing culture of program referral evolved to fit clinician needs, ultimately serving as a mechanism to overcome barriers to primary care access regardless of the clinical scenario. This feature is likely responsible for several of the demographic features observed in the referral population, such as the lower than expected prevalence of chronic medical conditions and higher than expected prevalence of government or private health insurance among the referral population. However, in this way the present study provides an estimate of the real- world impact of such a program, and extends generalizability beyond the majority of previously published care coordination studies.

Successful primary care linkage was not associated with a measur- able change in repeat ED utilization or hospitalization in the ensuing 12 months. Although the study design does not allow us to identify a cause for this observation, we have considered several possible explana- tions. First, it is possible that identification of a primary care provider and ED utilization are independent. Indeed, prior literature has suggested that the majority of ED patients can identify a primary care provider [28] and that heavy ED utilizers often have better-than- average access to primary care [29]. Independence between availability of primary care and ED utilization could be observed if the majority of visits were unavoidable, regardless of the provision or accessibility of primary care. This would be consistent with literature demonstrating that, contrary to popular belief, the majority of ED visits are appropriate [30]. It could also occur if having a primary care provider does not equate with sufficient access to that provider. Second, this finding may simply reflect our study population and setting. The overall frequency of ED visits and hospitalization among patients referred for care coordination was low, which may have limited our ability to detect a difference in ED use between pre and post-intervention. This may have been, in part, due to the relatively low prevalence of chronic medical conditions among the referral population. Given that such pa- tients are likely to achieve a greater benefit from primary care than those with acute or encapsulated issues, this feature likely weakened any observed effect on subsequent hospital-basED resource utilization. Additionally, our analysis was limited to a single site, thus limiting our ability to detect overall changes in Hospital-based services utilization. Previous research conducted in the study region has demonstrated that a significant number of patients use multiple emergency depart- ments [31].Thus, it is possible that exposure to the intervention im- pacted subsequent ED usage patterns in such a way that our study design was not able to capture, such as leading to preferential use of the study site. Third, although primary care provider continuity has a demonstrated association with decreased emergency department utili- zation [32,33], it may be that our intervention, focused mainly on the initial appointment, failed to address the ongoing issues that contribute to utilization of hospital-based resources. Finally, our cohort consisted entirely of those already in the emergency department, and it may be

that such a group behaves fundamentally differently than those re- cruited in the non-hospital setting.

In summary, our experience has demonstrated that a low intensity and broadly inclusive, ED-based care coordination intervention can be modestly effective in linking many patients possessing barriers to care. While we did not demonstrate a subsequent reduction in ED utilization, linkage to primary care has other benefits including the potential for other interventions to promote adherence and continuity of care, im- proved health, and greater confidence on the part of ED clinicians when discharging patients from the ED.

Funding sources

The research component was supported in part by an Institutional Clinical and Translational Science Award, NIH/NCRR Grant Number 8UL1-TR000077.

Disclosures

No conflict of interest - SF, KH, CL, CM and MS report no conflicts of interest.

Author contributions

All authors contributed to the conception, design, and interpretation of the study. KWH managed the data and performed statistical analysis. SDF drafted the manuscript, and all authors contributed substantially to its revision. SDF takes responsibility for the paper as a whole.

Acknowledgements

The clinical program was supported in part by a grant from the Health Foundation of Greater Cincinnati and administratively by the nonprofit organization healthcare access Now (HCAN). The authors acknowledge Michelle Long, the care coordination specialist, for her exceptional work, and Judith Warren, CEO of HCAN, for her dedication and support. The research component was supported in part by an Insti- tutional Clinical and Translational Science Award, NIH/NCRR Grant Number 8UL1-TR000077.

Appendix 1. Chronic medical conditions with ICD-9 codes evaluated for among the referred and non-referred ED population

1. Cheung PT, Wiler JL, Ginde AA. Changes in barriers to primary care and emergency department utilization. Arch Intern Med 2011;171(15):1397-8.
2. Shaw EK, Howard J, Clark EC, Etz RS, Arya R, Tallia AF. decision-making processes of patients who use the emergency department for primary care needs. J Health Care Poor Underserved 2013;24:1288-305.
3. Afilalo J, Marinovich A, Afilalo M, et al. Nonurgent emergency department patient characteristics and barriers to primary care. Acad Emerg Med 2004;11(12):1302-10.
4. Cheung PT, Wiler JL, Lowe RA, Ginde AA. National study of barriers to timely primary care and emergency department utilization among Medicaid beneficiaries. Ann Emerg Med 2012;60(1):4-10 [e2].
5. Baker LC, Baker LS. excess cost of emergency department visits for nonurgent care. Health Aff 1994;13(5):162-71.
6. Coventry JA, Weston MS, Collins PM. Emergency room encounters of pediatric pa- tients with asthma: cost comparisons with other treatment settings. J Ambul Care Manage 1996;19(2):9-21.
7. Mehrota A, Liu H, Adams JL, et al. Comparing costs and quality of care at retail clinics with that of other medical settings for 3 common illnesses. Ann Intern Med 2009; 151:321-8.
8. Carrier E, Yee T, Holzwart RA. Coordination between emergency and primary care physicians. NIHCR. Res Briefs 2011;3(February):1-11.
9. Stiell A, Forster AJ, Stiell IG, van Walraven C. Prevalence of information gaps in the emergency department and the effect on patient outcomes. Can Med Assoc J 2003;169(10):1023-8.
10. Bodenheimer T. Coordinating care: a perilous journey through the healthcare sys- tem. N Engl J Med 2008;358(10):1064-71.
11. The fragmentation of US health care: causes and solutions. Oxford. In: Elhauge E, ed- itor. internal-pdf://2515134372/Harvard Elhauge the fragmentation of US Health. pdf; 2010.
12. McDonald KM, Sundaram V, Bravata DM, et al. Closing the gap: a critical analysis of quality improvement strategies. 2007;7(4).
13. Penm J, MacKinnon NJ, Strakowski SM, Ying J, Doty MM. Minding the gap: factors as- sociated with primary care coordination of adults in 11 countries. Ann Fam Med 2017;15(2):113-9.
14. Katz EB, Carrier ER, Umscheid CA, Pines JM. Comparative effectiveness of care coor- dination interventions in the emergency department: a systematic review. Ann Emerg Med 2012;60(1):12-23.
15. McCusker J, Debdukuri N, Tousignant P, Verdon J, Poulin de Courval L, Belzile E. Rapid two-stage emergency department intervention for seniors: impact on conti- nuity of care. Acad Emerg Med 2003;10(3):233-43.
16. Mion LC, Palmer RM, Meldon SW, et al. Case finding and referral model for emer- gency department elders: a randomized clinical trial. Ann Emerg Med 2003;41(1): 57-68.
17. Wong FK, Chow S, Kea Chang. Effects of nurse follow-up on emergency room re- visits: a randomized controlled trial. Soc Sci Med 2004;59(11):2207-18.
18. Zorc JJ, Scarfone RJ, Yea Li. Scheduled follow-up after a pediatric emergency depart- ment visit for asthma: a randomized trial. Pediatrics 2003;111:495-502.
19. Horwitz SM, Busch SH, Balestracci KM, Ellingson KD, Rawlings J. Intensive interven- tion improves primary care follow-up for uninsured emergency department pa- tients. Acad Emerg Med Off J Soc Acad Emerg Med 2005;12(7):647-52.
20. Chan TC, Killeen JP, Castillo EM, et al. Impact of an internet-based emergency depart- ment appointment system to access primary care at safety net community clinics. Ann Emerg Med 2009;54(2):279-84.
21. Wallis M, Hooper J, Kerr D, Lind J, Bost N. Effectiveness of an advanced practice emergency nurse role in a minor injuries unit. Aust J Adv Nurs 2009;27(1):21-9.
22. Kyriacou DN, Handel D, Stein AC, Nelson RR. Factors affecting outpatient follow-up compliance of emergency department patients. J Gen Intern Med 2005;20:938-42.
23. Magnasson AR, Hedges JR, Vanko M, McCarten K, Moorhead JC. Follow-up compli- ance after emergency department evaluation. Ann Emerg Med 1993;22(3):560-7.
24. Vukmir RB, Kremen R, Dehart DA, Menegazzi J. Compliance with emergency depart-

Chronic Obstructive Lung Disease

490, 491, 491.1, 491.2, 491.21, 491.22, 491.8, 491.9, 492.8,

494, 491.1, 496

ment patient referral. Am J Emerg Med 1992;10(5):413-7.

1. McMullen MJ, Netland PA. Lead time for appointment and the no-show rate in an ophthalmology clinic. Clin Ophthalmol 2015;9:513-6.

Asthma	493, 493.01, 493.02, 493.1, 493.11, 493.12, 493.2, 493.21,	[27] Gallucci G, Swartz W. Impact of the wait for an initial appointment on the rate of
	493.22, 493.84, 493.9, 493.91, 493.92	kept appointments at a mental health center. Psychiatr Serv 2005;56(3):344-6.
Coronary Artery	411.1, 411.81, 414, 414.01, 414.07, 414.2-4	[28] Weber EJ, Showstack JA, Hunt KA, et al. Are the uninsured responsible for the in-
Disease		crease in emergency department visits in the United States? Ann Emerg Med
Congestive Heart	428, 428.1, 428.2, 428.22, 428.23, 428.3, 428.31-33, 428.4,	2008;52(2):108-15.
Failure	428.41-43, 428.9	[29] Blank FS, Li H, Henneman PL, et al. A descriptive study of heavy emergency depart-
Hypertension	401, 401.1, 401.9	ment users at an academic emergency department reveals heavy ED users have bet-

References

250.41-43, 250.5, 250.51-53, 250.6, 250.61-63, 250.7,

Seizure/epilepsy	345.4, 345.41, 345.5, 345.51, 345.8, 345.81
Diabetes mellitus	249, 249.01, 249.1, 249.11, 249.2, 249.21, 249.3, 249.31,
	249.4, 249.41, 249.5, 249.51, 249.6, 249.61, 249.7, 249.71,
	249.8, 249.81, 249.9, 249.91, 250, 250.01-03, 250.1,
	250.11-13, 250.2, 250.21-23, 250.3, 250.31-33, 250.4,

250.71-73, 250.8, 250.81-83, 250.9, 250.91-93

ter access to care than average users. J Emerg Nurs 2005;31(2):139-44.

1. Sommers AS, Boukus ER, Carrier E. Dispelling myths about emergency department use: majority of medicaid visits are for urgent or more serious symptoms. Res Briefs 2012;23:1-10.
2. Fertel BS, Hart KW, Lindsell CJ, Ryan RJ, Lyons MS. Patients who use multiple EDs: quantifying the degree of overlap between ED populations. West J Emerg Med 2015;16(2):229-33.
3. Gill JM, Mainous III AG, Nsereko MN. The effect of continuity of care on emergency department use. Arch Fam Med 2000;9:333-8.
4. Ionescu-Ittu R, McCusker J, Ciampi A, et al. Continuity of primary care and emer- gency department utilization among Elderly people. Can Med Assoc J 2007;177 (11):1362-8.
5. Cantlupe J. Volume, flow and safety issues in the ED. Health Leaders Media Council; 2012(May).