Article, Psychiatry

Care plan program reduces the number of visits for challenging psychiatric patients in the ED

Unlabelled imageAmerican Journal of Emergency Medicine (2012) 30, 1061-1067

Original Contribution

Care plan program reduces the number of visits for challenging Psychiatric patients in the ED

Arthur Abello Jr. MD a, Ben Brieger MD b, Kim Dear LCDC-CI c, Ben King MPH d,?,

Chris Ziebell MD b,d, Atheer Ahmed MD d, Truman J. Milling Jr. MD b,d

aTexas A&M Health Sciences Center, College of Medicine, Corpus Christi, TX

bUniversity Medical Center at Brackenridge, Austin, TX

cSeton Healthcare Family, Austin, TX

dHospital Physicians in Clinical Research, Austin, TX

Received 23 June 2011; accepted 5 July 2011

Abstract

Background: A small number of patients representing a significant demand on emergency department (ED) services present regularly for a variety of reasons, including psychiatric or behavioral complaints and lack of access to other services. A care plan program was created as a database of ED high users and patients of concern, as identified by ED staff and approved by program administrators to improve care and mitigate ED strain.

Methods: A list of medical record numbers was assembled by searching the care plan program database for adult patients initially enrolled between the dates of November 1, 2006, and October 21, 2007. Inclusion criteria were the occurrence of a psychiatric International Classification Diseases, Ninth Revision, code in their medical record and a care plan level implying a serious psychiatric disorder causing harmful behavior. Additional data about these patients were acquired using an indigent care tracking database and electronic medical records. Variables collected from these sources were analyzed for changes before and after program enrollment.

Results: Of 501 patients in the database in the period studied, 48 patients fulfilled the criteria for the cohort. There was a significant reduction in the number of visits to the ED from the year before program enrollment to the year after enrollment (8.9, before; 5.9, after; P b .05). There was also an increase in psychiatric Hospital visits (2%, before; 25%, after; P b .05).

Conclusion: An alert program that identifies challenging ED patients with psychiatric conditions and creates a care plan appears to reduce visits and lead to more appropriate use of other resources.

(C) 2012

Introduction

Emergency departments (ED) in the United States have experienced a 32% increase in number of visits between

* Corresponding author. Tel.: +1 512 610 0372; fax: +1 512 610 0380.

E-mail address: [email protected] (B. King).

1996 and 2006 [1]. A small group of patients use a large, disproportionate amount of ED resources [2,3] for medical, social, and psychiatric reasons [4-7]. Emergency department visits for primary mental health concerns increased by 40% between 1992 and 2000 [8], and psychiatric conditions have been shown to influence frequency of ED use [9-11], and evaluation of repeat visits to the ED is an important quality improvement tool [12-14].

0735-6757/$ – see front matter (C) 2012 doi:10.1016/j.ajem.2011.07.002

The challenges in care for US psychiatric patients may be attributed to deinstitutionalization, decreased funding for mental health care, substance abuse [8,15,16], or lack of health insurance. Texas, the state in which this study took place, has the highest percentage of uninsured people in the United States [17]. Psychiatric patients who excessively and/ or inappropriately use the ED may be better served if identified and directed to more appropriate resources such as Primary care clinics and psychiatric hospitals.

The High Alert Program (HAP) is a care plan database created in 2001. The program identifies patients with a history of excessive use of the ED, particular Medical issues of concern, dangerous behavior toward self or others while on hospital property, or other problematic behavior that requires tailored care.

Goals of this investigation

To determine whether enrollment in the HAP care plan program reduced the number of ED visits for challenging psychiatric patients, to establish whether patients used other appropriate avenues of psychiatric treatment in lieu of ED visits, and to identify characteristics of the groups that correlate with response to the program

Materials and methods

Extending to other hospitals in the same network over subsequent years, the HAP program identified patients by recommendations from ED staff, which are subsequently reviewed by social workers, case managers, and the medical director. care plans, which are categorized into 4 levels (Table 1), are generated and discussed with the patients on their next ED visit by the nurse and case manager.

These levels are used as visual cues on patient identifying stamps and stickers to inform providers of care plan availability. The care plans are easily accessed in a subprogram of the hospital electronic medical record interface. Providers

may make comments that await case manager and medical director approval before enrollment in the official care plan. Care plans are reviewed on a regular basis and are either modified according to subsequent encounters or deleted if issues resolve and/or the patient ceases to return.

This retrospective cohort study was conducted at 3 Central Texas hospitals (University Medical Center at Brackenridge, at the time a level II trauma center, Seton Medical Center-Austin, and Seton Northwest Hospital). Located in Austin, these hospitals serve more than 1 million citizens in the city and surrounding area. The total ED activity for these sites was 184 000 patients during the year studied. Patients’ medical record numbers were assembled by searching the HAP database for adult patients admitted to the HAP between the dates of November 1, 2006, and October 31, 2007, resulting in a list of 501 patients. From this, we selected all patients with psychiatric International Classification Diseases, Ninth Revision, codes (290-312) excluding those with childhood developmental or mental retardation disorders, as those patients present different challenges than adult psychiatric patients.

As a result, 42 patients fulfilled the criteria for the study. This list was added to by examining a list of cases with HAP codes 1 or 2, implying serious psychiatric pathology as evidenced by harm to self or ED staff. Subjects with these codes who were not already in the study database were evaluated for psychiatric International Classification Dis- eases, Ninth Revision, codes in documented care plans. This resulted in additional 6 patients bringing the total sample size to 48 patients (see Fig. 1 for diagram of cohort eligibility). Additional admission, medical, and demographic data were acquired using an archived data gathering tool (Decision Support Solutions v4.2; Siemens, Munich, Germany), an indigent care tracking database, and electronic medical records for the year before and the year after enrollment. The date of initial enrollment in the HAP was used as the center point that separates the year before admission from the year after. The primary outcome measure was the number of visits during those 2 years. Although patients could only be enrolled in HAP at the 3 aforementioned hospitals, previous

Table 1 HAP program levels

Level 1 Documented history of violence or abusive behavior while on hospital property. These patients may have been identified by an incident report or been involved in a Code Gray (violent or combative patient). These patients have a written treatment plan developed by the ED medical director; case manager/social worker; and, if applicable, the primary care physician with the patient’s knowledge and agreement.

Level 2 Document history of self-directed violence while on hospital property. This does not apply to patients that present to triage and state that they feel suicidal or seek treatment for a suicide attempt or gesture outside the hospital. These patients have a written treatment plan developed by the ED medical director; case manager/social worker; and. if applicable, the primary care physician with the patients knowledge and agreement.

Level 3 These patients have a written treatment plan developed by the ED medical director; case manager/social worker; and, if applicable, the primary care physician with the patient’s knowledge and agreement.

Level 4 All patients who present for the first time and return patients who do not have any other level assigned.

11/1/2006- 10/31/2007

501 patients enrolled in HAP

3 sites

184,000 ED patients

48 subjects selected cohort

6 subjects with events of violence with psychiatric conditions mentioned in Case management notes

42 subjects with psychiatric ICD-9 codes

Fig. 1 Cohort eligibility diagram.

and repeat visits at any Seton hospital or St David’s hospital (the other major hospital system in Austin) were available from a county indigent care tracking database. Two smaller area hospitals are not included in that database, and visits there would not have been captured, although these centers represent a small fraction of total area ED visits. High Alert Program-specific information recorded included HAP level, receipt of care plan letter, and site of ED visit that prompted HAP enrollment. Patient care plans were coded by content (Table 2). Other information collected included health care funding source, changes to funding while in the HAP, enrollment in Mental Health Mental Retardation benefits program, Medical Assistance Program enrollment, and visits to an indigent care program before and after enrollment.

High use of ED care was defined as 6 or greater visits in the year before enrollment based on a review of prior studies that produced a range of ED visit rates defined as high use from 3 to 12 per year and taking the mean [18-25]. Because lengths of stay were not uniformly recorded at the different hospitals, durations were manually calculated by subtracting

Table 2 HAP care plan codes

  1. Behavior modification recommended to reduce secondary gain in hospital
  2. Security alert
  3. Controlled substance abuse alert
  4. Primary care referral
  5. Psychiatric care referral

the admission time on the triage paperwork from the discharge time on the nursing note.

Results

Descriptive results are listed in Table 3.

Of the 48 patients, 9 were homeless, whereas 58% of the patients in the cohort were designated high users: 62% of males and 55% of females. The single, uninsured, and underinsured subjects were more likely to be high-users. Of the 9 homeless subjects, 6 qualified as such.

Reduction in ED visits

Overall, there was an average reduction of 3, from 8.9 to 5.9, ED visits between the year before and the year after HAP enrollment (P b .05). Male subjects saw a nonsignificant visit reduction of 2; female subjects had a significant average reduction of 4.3 (P b .05) (Table 3).

There was a significant reduction in ED visits for subjects who were not homeless (P b .005). Homeless patients did not show the same effect, and the difference in mean change of visits between the 2 groups was significant. High users had a significant average reduction of 6.3 visits (12.6-6.3; P b

.005). There was also a significant decrease in visits for the uninsured group of subjects in the cohort (P b .01), of which 8 of 11 were high users before enrollment. Relatively, there were nonsignificant decreases in visits by insured subject groups:

Table 3 (continued)

Characteristics

Total

Mean change in ED visits

P a

New psychiatric visit, after (%) Yes

No

11 (22.9)

37 (77.1)

(-) 4.6

(-) 2.6

.539

CHC indicates community health care. MHMR indicates mental health & mental retardation benefits.

* Statistical significance.

a Group means were compared by 2-sided Student t tests. For cate- gorical variables, mean change in visits before postenrollment are compared against all other categories. Binary variables are tested for difference in mean change between groups (P values in line with variable labels) and for significant differences between previsit and postvisit counts (P values in line with variable categories). Statistical significance was set at P b .05.

b Underinsured: Medicaid, Medicare, county insurance coverage; 2- sided Student t tests of mean change in visits from pre- to post-enrollment.

c Insured: private insurance and VA benefits.

Medicaid (-1.75), Veterans Affairs (-6.5), and county safety net insurance (-3.7). Medicare and privately insured subjects showed a small, nonsignificant increase in visits.

Table 3 Characteristics of cohort by change in ED visits pre- enrollment to postenrollment

Characteristics

Total

Mean change in ED visits

P a

Patients, n

48

(-) 3.0

.034

Median age at enrollment (y)

38.5

Sex (%)

.423

Male

26 (54.2)

(-) 2.0

.333

Female

22 (45.8)

(-) 4.3

.035 ?

Race (%)

White

33 (68.8)

(-) 2.4

.497

Black/African American

8 (16.7)

(+) 0.25

.296

Hispanic/Latino

6 (12.5)

(-) 9.3

.088

Other

1 (2.1)

(-) 13

marital status (%)

.366

Single (divorced,

42 (87.5)

(-) 3.5

.023 ?

widowed, etc)

Married

6 (12.5)

(+) 0.33

.937

High users (%)

.005 ?

Yes

28 (58.3)

(-) 6.3

.004 ?

No

20 (41.7)

(+) 1.5

.310

Homeless (%)

.017 ?

Yes

9 (18.8)

(+) 3.8

.262

No

39 (81.3)

(-) 4.6

.003 ?

Financial class (%)

(n = 46)

Uninsured

11 (23.9)

(-) 10.0

.005 ?

County safety net insurance

7 (15.2)

(-) 3.6

.936

Medicaid

16 (34.8)

(-) 1.8

.419

Medicare

4 (8.7)

(+) 2.75

.180

VA

2 (4.4)

(-) 6.5

.628

Private insurance

6 (13.0)

(+) 2.2

.128

Underinsured b

27 (58.7)

(-) 1.6

.357

Insured c

8 (17.4)

0.0

1.000

Primary Diagnosis at

(n = 44)

enrollment (%)

Pain

11 (25.0)

(-) 6.1

.248

Headache

3 (8.8)

(+) 0.7

.506

Injury

1 (2.3)

(+) 1.0

chronic illness

2 (4.6)

(+) 13.5

.014 ?

Chest pain

5 (11.4)

(+) 1.0

.336

Other acute condition

6 (13.6)

(-) 3.5

.910

Psychiatric

9 (20.5)

(-) 5.0

.519

Suicide attempt

3 (6.8)

(-) 9.3

.262

Primary care

4 (9.1)

(-) 2.3

.865

Care plan type (%)

Behavior modification

2 (4.2)

(+) 9.5

.060

Security alert

10 (20.8)

(-) 4.2

.675

Controlled substance

39 (81.3)

(-) 2.6

.530

abuse alert

Primary care referral

10 (20.8)

(-) 5.3

.412

Psychiatric care referral

9 (18.8)

(-) 2.0

.724

Limiting narcotics only

21 (43.8)

(-) 1.9

.460

CHC visits pre-enrollment (%)

14.8 +- 21.8

CHC visits postenrollment (%)

8.1 +- 12.9

MHMR enrollment, after (%)

(n = 43)

.810

Yes

23 (53.5)

(-) 4.0

No

20 (46.5)

(-) 3.3

Within the context of causes for enrollment in the HAP, an episode of violence in the ED (n = 6) before enrollment predicted a significantly larger reduction in visits (P b .05). Although violent episodes were not correlated with high use of the ED, this group had a reduction of 10.8 visits compared with 1.9 less visits in subjects without a history of Violent behavior. There were also differences in average visit changes for subjects with different types of care plans (Table 3).

Psychiatric care seeking

The number of subjects admitted to an inpatient psychiatric service rose from 2% (1) to 25% (11) in the year after enrollment. Detailed results of factors impacting community health center and inpatient psychiatric care services are listed in Table 4. Patients with a care plan that included a psychiatric care referral were more likely to have a psychiatric inpatient visit in the year after enrollment.

Two patients in this cohort have died since enrollment. One of these patients had an increase of 8 visits from the previous year, and the other had a decrease of 13 visits. Both subjects died more than 6 months after their date of enrollment in the study from complications of renal failure. Both patients were receiving dialysis care at sites outside the facilities included in this study.

Limitations

This is a retrospective study of a database created for other purposes and had limitations of restricted detail and potential documentation errors. The database used to track patient visits does not include 2 local hospitals, so any patient visits

Table 4 Characteristics of cohort by change in community health care and psychiatric care visits

Characteristics

CHC visits

Mean change in CHC visits (n = 43)

(-) 6.7

P a

No. of new P b

psychiatric care visits (n = 48)

Total cohort Sex

Male Female Race White

Black/African American Hispanic/Latino Other

Marital status Single (divorced, widowed, etc) Married

High-users Yes

No Homeless

.064

.438

.112

.368

11

.002

.302 c

(-) 9.3

(-) 3.8

4

7

.323 c

(-) 2.4 .399 9

(+) 0.25 .287 0

(-) 8.2

0.0

.874 2

0

(-) 7.9

.389

.053

.153 c

11

(+) 1.8

.756

.066

.561

.072

.833

0

.772

(-) 1.7

(-) 15.2

6

5

.095 c

Financial class (n = 46)

Uninsured

(n = 41)

(n = 10) .470 c

(-) 1.5

.235

1

.410 c

Yes

(-) 8.2

.431

0

No

(-) 6.4

.098

11

Primary care

(-) 6.3

.953 1

.416 c

referral

Psychiatric care

(-) 13.8

.352 5

.020 c, ?

referral

Limiting narcotics

(-) 6.5

.957 2

.083 c

only

MHMR enrollment,

(n = 43)

.053 (n = 9)

.002 c, ?

after (n = 43)

Yes

(-) 13.1

.027 ? 9

No

(+) 0.6

.872 0

New psychiatric

.598

visit, after

Yes

(-) 10.4

.038 ? 11

No

(-) 5.8

.195 0

CHC indicates community health care.

MHMR indicates mental health & mental retardation benefits.

* Statistical significance.

a Group means were compared by 2-sided Student t tests. For cate- gorical variables, mean change in visits before postenrollment are compared against all other categories. Binary variables are tested for difference in mean change between groups (P values in line with variable labels) and for significant differences between previsit and postvisit counts (P values in line with variable categories). Statistical significance was set at P b .05.

b ?2 Test.

c Fisher exact test.

d Underinsured: Medicaid, Medicare, county insurance coverage; 2-sided Student t tests of mean change in visits from pre- to post-enrollment.

e Insured: private insurance and VA benefits.

there would not have been captured and this could conceivably confound the significance in the change in number of visits after HAP enrollment. However, those hospitals comprise a small fraction of the total visits in the community. It is also possible that patients sought care in EDs outside the Austin community. Disparate databases were manually merged using name and date of birth, creating the

Table 4 (continued)

Characteristics

CHC visits

Mean change in CHC visits (n = 43)

P a

No. of new P b

psychiatric care visits (n = 48)

County safety net

(-) 20.7

.084

2

.636 c

insurance

Medicaid

(-) 8.9

.854

3

1.00 c

Medicare

(-) 7.3

.949

2

.201 c

VA (n = 1)

0.0

1

.391 c

Private insurance

(-) 2

.611

1

1.00 c

Underinsured d

(-) 11.9

.026 ?

7

.395 c

Insured e

Primary Dx at

(-) 1.5

(n = 40)

.595

1

(n = 10)

.546 c

.470 c

enrollment

(n = 44)

Pain

(-) 4.8

.707

4

.237 c

Headache

(-) 3.0

.798

0

1.00 c

Injury (n = 1)

(+) 1.0

0

Chronic illness

(+) 3.0

.540

0

1.00 c

Chest pain

(-) 3.0

.673

0

.573 c

Other acute

(-) 12.0

.612

3

.120 c

condition

Psychiatric

(-) 6.9

.954

2

1.00 c

Suicide attempt

(-) 4.7

.845

0

1.00 c

Primary care

(-) 30.0

.090

1

1.00 c

Care plan type

Behavior

(-) 32

.117

0

1.00 c

modification

Security alert

(-) 0.6

.376

3

.675 c

Controlled

(-) 5.1

.294

9

1.00 c

substance abuse

alert

potential for typos and data entry errors and the possibility of missing visits in which patients gave a different name or date of birth at ED triage or at the psychiatric facility. We also would not have known if a patient had moved out of the area. We only captured psychiatric visits at 1 major psychiatric hospital and were unable to obtain records from the other 2, although this would tend to underestimate the effect that has already attained significance. It would be impractical to gather visit data from every clinic and physician’s office in the county, so those visits would have been missed, but one

of the strengths of our approach is the county wide tracking

database, which includes visits to all county clinics and nearly all EDs. Any patient who presents to any Seton or St David’s ED or any of the county clinics is automatically entered in this database, making this a very rare resource.

It is also possible that outliers, that is, high-using patients, may have regressed to the mean from year to year naturally, a well-described phenomenon [9,23,26-29], and not as a consequence of the HAP program, al- though this would not explain the increase in psychiatric hospital visits.

Hawthorne effect can influence the study findings because there was neither blinding of the providers nor the patients to the intervention. The HAP process likely affected both provider and patient behaviors, and in this model, it is not possible to determine to what degree each influenced the outcome. However, the ultimate clinical goal of HAP is to modify patient behavior. What was demonstrated in this study is that patient behavior was modified if the patients as well as the ED were aware of high use of the ED. Although this study looked at a small cohort and at a particular metric (ie, number of ED and psychiatric visits by high-using psychiatric patients), it showed that there was significant change in number of visits. There may be unintended consequences of identifying a patient as a high user and developing a care plan. They may feel unwelcome in the ED and, thus, not present for care when they have a serious medical emergency. In our cohort, 2 patients died of renal failure. We have not been able to determine the chain of events that lead to their deaths, in particular if their being enrolled in the HAP had any role. These may have been expected deaths considering their underlying disease, but one could postulate a scenario in which the patients hesitated to present for care because of being flagged as high users. Patients in HAP are still guaranteed a medical screening examination (by federal law and hospital policy) and nonnarcotic pain relief.

The correlation between the patients who had a reduction in visits and those who had an increase in psychiatric hospital visits failed to attain significance, which could be also be explained as a nonsignificant finding in an underpowered small sample, that is, ? error. This might be further explained by the lack of access to the other psychiatric hospitals’ visit data.

Discussion

Our data suggest that a relatively simple intervention, that is, enrollment in a program that generates a care plan and discusses it with psychiatric patients and then generates a chart flag for future care providers, has the potential not only to reduce ED visits but also increase visits to more appropriate psychiatric facilities. The cohort had a reduction in ED visits, an increase in psychiatric admissions, and no significant change in community clinic visits.

The strength of our data relies on the comprehensive nature of the county patient-tracking database, which eliminates many of the confounders of previous studies. Fuda and Immekus [23] used statewide ED visit data in

Massachusetts to characterize high-using ED patients but were limited by lack of clinic data.

Previous studies on the impact of ED care plans on Health care use have yielded mixed results [10,30,31]. Although some previous studies could not demonstrate significance [10], the most recent study demonstrated an impact on ED use rates, ED costs, and psychosocial outcomes but not on other medical service use or costs [30].

Notably, these studies included all high-using patients, a heterogeneous group with a broad range of pathologies, whereas we chose to isolate the care plan effect on the narrower range of true psychiatric ED patients.

Further studies will be necessary to evaluate HAP for its safety and effectiveness for modifying the behavior of our high-using psychiatric patients. In our small cohort of psychiatric patients, we saw promising trends toward more appropriate dispositions to psychiatric facilities rather than the ED.

References

  1. Pitts SR, Niska RW, Xu J, Burt CW. National hospital ambulatory medical care survey: 2006 emergency department summary. National health statistics reports; no 7. Hyattsville (Md): National Center for Health Statistics; 2008.
  2. Magnusson G. Association between health status, social factors and level of emergency department use. Med Care 1980;4.
  3. Malone RE. Almost “like family”: emergency nurses and “frequent flyers”. J Emerg Nurs 1996;22(3):176-83.
  4. Andren KG, Rosenqvist U. Heavy users of an emergency department: psycho-social and medical characteristics, other health care contacts and the effect of a hospital social worker intervention. Soc Sci Med 1985;21:761-70.
  5. Mandelberg JH, Kuhn RE, Kohn MA. Epidemiologic analysis of an urban public emergency department’s frequent users. Acad Emerg Med 2000;7(6):637-43.
  6. Hansagi H, Allebeck P, Edbag O, Magnusson G. Frequency of emergency department attendances as a predictor of mortality: nine- year follow-up of a population-based cohort. J Public Health Med 1990;12(1):39-44.
  7. Salazar A, Bardes I, Juan A, et al. high mortality rates from medical problems of frequent emergency department users at a university hospital tertiary care centre. Eur J Emerg Med 2005;12:2-5.
  8. Larkin GL, Glassen CY, Emond JA, et al. Trends in US emergency department visits for mental health conditions. Psychiatry Serv 2005;56:671-7.
  9. Kne T, Young R, Spillane L. Frequent ED users: patterns of use over time. Am J Emerg Med 1998;16:648-52.
  10. Spillane L, Lumb E, Cobaugh D, Wilcox SR, Clark JS, Schneider SM. Frequent users of the emergency department: can we intervene? Acad Emerg Med 1997;4:574-80.
  11. Rask KJ, Williams MV, McNaguy SE, et al. Ambulatory health care use by patients in a public hospital emergency department. J Gen Int Med 1998;13:614-20.
  12. DePiero AD, Ochsenschlager DW, Chamberlain JM. Analysis of pediatric hospitalizations after emergency department release as a quality improvement tool. Ann Emerg Med 2002;39:159-63.
  13. Zimmerman DR, McCarten-Gibbs KA, DeNoble DH, Borger C, Fleming J, Hsieh M, et al. Repeat visits to a general emergency department. Ann Emerg Med 1996;28:467-73.
  14. Chamberlain JM, Pollack MM. A method for assessing emergency department performance using patient outcomes. Acad Emerg Med 1998;5:986-91.
  15. Hazlett SB, McCarthy ML, Londner MS, et al. Epidemiology of adult visits to US emergency departments. Acad Emerg Med 2004;11:193-5.
  16. Kalney R, Thomas L, King D. Changing demand for mental health services in the emergency department of a public hospital. Aust N Z J Psychiatry 2005;39:74-80.
  17. De Navas-Walt C, Proctor BD, Smith J. Income, poverty, and health insurance coverage in the United States. 2006 Current Population Reports. No P60-233. Washington, DC: US Census Bureau; 2007.
  18. Hunt K, Weber E, Showstack J, Colby D, Callaham M. Characteristics of frequent users of emergency departments, 30 March 2006. Ann Emerg Med 2006;48(1):1-8.
  19. Zuckerman S, Shen Y. Characteristics of occasional and frequent emergency department users: do insurance coverage and access to care matter? Med Care 2004;42:176-82.
  20. Byrne M, Murphy A, Plunkett P, et al. Frequent attenders to an emergency department: a study of primary health care use, medical profile and psychosocial characteristics. Ann Emerg Med 2003;41: 309-18.
  21. Blank F, Li H, Smithline H, et al. A descriptive study of heavy emergency department users at an academic emergency department reveals heavy ED users have better access to care than average users. J Emerg Nurs 2005;31:139-44.
  22. Aisiku I, Smith W, McClish D, Levenson J, Penberthy L, Roseff S, et al. Comparisons of high versus low emergency department utilizers in Sickle cell disease. Ann Emerg Med 2009;53(5):587-93.
  23. Fuda K, Immekus R. Frequent users of Massachusetts emergency departments: a statewide analysis. Ann Emerg Med 2006;48(1): 16.e1-8.
  24. Chan B, Ovens H. Frequent users of emergency departments: do they also use family physicians‘ services? Can Fam Physician 2002;48: 1654-60.
  25. Herbert T, Shuih T, Barth R, Szilagyi P. High utilization of the emergency department by children on Medicaid. Assoc Health Serv Res Meet 1997;14:278-9.
  26. Mandelberg JH, Kuhn RE, Kohn MA. Epidemiologic analysis of an urban, public emergency department’s frequent users. Acad Emerg Med 2000;7:637-46.
  27. Cook LF, Knight S, Junkins Jr EP, et al. Repeat patients to the emergency department in a statewide database. Acad Emerg Med 2004;11:256-63.
  28. Kennedy D, Ardagh M. Frequent attenders at Christchurch Hospital’s emergency department: a 4-year study of attendance patterns. N Z Med J 2004;117:U871-9.
  29. Genell AK, Rosenqvist U. Heavy users of an emergency department: a two year follow-up study. Soc Sci Med 1987;25: 825-31.
  30. Shumway M, Boccellari A, O’Brien K, Okin R. Cost-effectiveness of clinical case management for ED frequent users: results of a randomized trial. Am J Emerg Med 2008;26(2):155-64.
  31. Okin R, Boccellari A, Azocar F, Shumway M, O’Brien K, Gelb A, et al. The effects of clinical case management on hospital service use among ED frequent users. Am J Emerg Med 2000; 18(5):603-8.

Leave a Reply

Your email address will not be published. Required fields are marked *