The Journal of Emergency Medicine, Vol. -, No. -, pp.

1–7, 2019
Ó 2019 The Authors. Published by Elsevier Inc. This is an open access article under the
CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
0736-4679/$ - see front matter

https://doi.org/10.1016/j.jemermed.2019.10.006

Administration of
Emergency Medicine

THE GAP-ED PROJECT: IMPROVING CARE FOR ELDERLY PATIENTS PRESENTING

TO THE EMERGENCY DEPARTMENT

Tara Liberman, DO,* Regina Roofeh, MPH,* Natalie Sohn, BS,*† Martina Brave, BA,* Alison Smith, LMSW,*
Helena Willis, RN,* Robert Silverman, MD,* and Nancy Kwon, MD*
*Long Island Jewish Medical Center, Northwell Health, New Hyde Park, New York and †Donald and Barbara Zucker School of Medicine at
Hofstra Northwell, Hempstead, New York
Reprint Address: Regina Roofeh, MPH, Long Island Jewish Medical Center, 270-05 76th Ave, Research Building, New Hyde Park, NY, 11040

, Abstract—Background: Older adults presenting to the
emergency department (ED) represent a highly vulnerable
patient population with complex conditions and multiple
comorbidities. The introduction of a Geriatric and Pallia-
tive (GAP)-ED partnership may be an effective strategy to
avoid unneeded admissions and improve outcomes for this
population. Objectives: The primary objective was to
decrease 30-day revisit and hospitalization rates in this
population through identifying patients that could be
safely sent home with connection to community resources.
Secondary outcomes included achieving high patient and
family satisfaction scores assessed through follow-up inter-
views. Methods: The GAP-ED intervention included the
placement of a Specialist in the ED to coordinate care
for older adults presenting to the ED who were likely to
be discharged home. Independent t-tests and chi-squared
tests were used to assess for changes in outcomes between
the intervention group and a blocked matched historical
usual-care group. Results: There was no significant differ-
ence in 30-day ED revisits between the two groups, but
there was a statistically significant reduction in hospital
admissions from these 30-day revisits. Patient and family
satisfaction with the presence of the GAP-ED Specialist
was high. Conclusion: The implementation of a GAP-ED
partnership and use of a GAP-ED Specialist is an effective
means of reducing hospitalization in older adults revisit-
ing the ED. Ó 2019 The Authors. Published by Elsevier
Inc. This is an open access article under the CC BY-NC-
ND license (http://creativecommons.org/licenses/by-nc-nd/
4.0/).
, Keywords—older adults; geriatric emergency medicine;
discharge coordination
INTRODUCTION
Older adults in the emergency department (ED) setting
are a highly vulnerable patient population who frequently
present with complex conditions and multiple comorbid-
ities. It is estimated that 58% of individuals aged 75 years
and older visit the ED at least once during a 1-year period
and use disproportionately more ED services than other
age groups, leading to longer overall time spent in the
ED (1,2). In addition to widespread ED admission of in-
dividuals aged 75 and older, there was a 34% increase in
ED visits for patients aged 65–74 years between 1993 and
2003 (3). Overall, estimates indicate that those aged
65 years and older represent 15–20% of all visits to the
ED and 36% of hospitalizations (4,5).
Older adults who present to the ED are often at
increased risk for complications and readmissions. It is
estimated that 25% of older adults admitted to the ED
present with impaired mental status, up to 30% present
as a result of falls, and 11% present due to adverse drug
interactions (6). Admissions to the ED have been associ-
ated with return visits, hospitalization, and death in this
population (7). Between one-third and one-half of older
adults who are treated in the ED are admitted to the

RECEIVED: 3 June 2019; FINAL SUBMISSION RECEIVED: 4 October 2019;

ACCEPTED: 13 October 2019

1
2 T. Liberman et al.
hospital, and up to 26% of those who are discharged from
the ED return within 90 days (1,4,5).
These patients often present with challenges that some-
times require a hospital admission for nonmedical reasons,
including caregiving needs or potential elder abuse and
neglect (6). Of particular concern are older adults living
in the community who have minimal social infrastructure
for caregiving and extensive social needs. These ‘‘elder or-
phans’’represent a vulnerable portion of this population that
can be at risk for additional needs in the community (8).
As the population’s proportion of older adults in the
United States is expected to reach 20% by 2030, health
care systems should be equipped to provide care and man-
agement tailored to their needs (9). Because EDs are the
safety net for vulnerable patient populations including
older adults, focus should be placed on the ED setting to
address the needs of these patients. Currently, widespread
geriatric training is lacking in Emergency Medicine resi-
dency programs, and those who work in the ED often
experience higher levels of burden and stress when treat-
ing older adult patients (10,11).
A Geriatric and Palliative Emergency Department
(GAP-ED) team was implemented as a 2-year pilot at a
tertiary medical center to provide comprehensive geri-
atric and palliative care assessment in the ED setting.
The goal of the GAP-ED team was to optimize the treat-
ment of older adults presenting to the ED and identify pa-
tients that could be safely sent home with community
resources to avoid unnecessary admissions and returns
to the ED that so often result in poor outcomes. The
objective of this study is to examine the implementation
of this program through analysis of 30-day revisit rates,
30-day hospitalizations, and patient satisfaction of at-
risk older adults in the ED.
METHODS
To improve care of older adults in the ED, we imple-
mented a GAP-ED team comprised of a Geriatric and
Palliative Medicine physician, an Emergency physician,
a Social Worker specialized in Geriatrics and Palliative
Medicine (GAP-ED Specialist), nursing leadership, and
administrative and research support. The team developed
relationships within the ED and with community organi-
zations that would serve as resources for discharged pa-
tients. The project was approved by the facility
institutional review board as both quality improvement
(for the pilot project of the GAP-ED team) and chart re-
view (for the retrospective chart review).
Intervention
This study was conducted at an academic, urban, tertiary
care ED with an approximate annual volume of over
100,000 adult visits. A significant portion of the patient
population that utilizes this hospital are residents of
Queens County, the most ethnically diverse large county
in the United States (12,13). The intervention portion of
the study was conducted from November 2015 through
July 2017. The target population for this initiative was
developed based on literature and previous trends from
the local ED population. Community-dwelling Queens
County residents presenting to the ED who were 65 years
or older, owned a telephone, had at least one prior ED
visit within the last year, and were likely to be discharged
home were included for analysis.
The GAP-ED team identified appropriate patients
through the ED tracking board and by referral from other
members of the ED team. Once the patients were identified,
the GAP-ED clinical team coordinated with the GAP-ED
Specialist to determine concerns and a plan of care. This
coordination of a geriatric-minded interdisciplinary team
allowed for the de-escalation of emergency conditions
and a comprehensive geriatric assessment that identified
and addressed issues common in older and medically com-
plex patients. Further consultation by Geriatric and Pallia-
tive Medicine physicians was available as needed. Patients
or their family members were provided with education,
emotional support, coordinated links to community re-
sources, and conversations on advanced care planning.
The GAP-ED Specialist formed relationships with or-
ganizations that provide resources and assistance to older
adults living in the community, including home care,
outpatient practices, senior centers, and food assistance
programs. Ultimately this intervention was done in hopes
of decreasing future unnecessary hospitalizations and re-
turns to the ED by improving post-ED care coordination
and linkage to outpatient and community-based re-
sources.
Follow-Up
Intervention patients were called by the Specialist at 3, 7,
and 30 days post discharge. During these follow-up calls,
the GAP-ED Specialist offered additional assistance in
connecting patients to community resources, encouraged
follow-up with primary care providers, and provided
emotional support. The GAP-ED research assistant called
patients or family members at 30 days post discharge
from the ED. The 30-day interview assessed whether pa-
tients or their caregivers remembered the interaction with
the GAP-ED team, whether they had successfully con-
nected to the resources recommended by the Specialist,
and whether they had returned to the ED since the initial
visit. The interview also surveyed patients and caregivers
about their satisfaction with the intervention and asked
whether they felt a GAP-ED team should be available
to all older adults in the ED.
The GAP-ED Project 3

Table 1. Comparison of Demographic Characteristics of Table 2. Comparison of ED Presentation Chief Complaint

Patients Included in the Intervention and Usual- and Past Medical History of Patients Included in
Care Groups* the Intervention and Usual-Care Groups*

Intervention Usual-Care Intervention Usual-Care

(n = 283) n (%) (n = 283) n (%) (n = 283) n (%) (n = 283) n (%)

Age, years Chief Complaint

65–75 30 (11) 30 (11) Fall 96 (34) 96 (34)
76–85 113 (40) 114 (40) Pain 54 (19) 54 (19)
86–94 124 (44) 125 (44) Other† 42 (15) 42 (15)
95+ 16 (5) 14 (5) Weakness 20 (7) 20 (7)
Sex Altered mental status 17 (6) 17 (6)
Female 220 (78) 220 (78) Dizziness 14 (5) 14 (5)
Male 63 (22) 63 (22) Bleeding 11 (4) 11 (4)
Ethnicity Shortness of breath 11 (4) 11 (4)
African American/Black 115 (41) 102 (35) Swelling 8 (3) 8 (3)
Asian 29 (10) 13 (4) Abdominal pain 8 (3) 8 (3)
White 134 (47) 128 (44) Past medical history
Hispanic or Latino 5 (2) 28 (10) Falls 106 (37) 112 (40)
Other or unknown 0 (0) 19 (7) Dementia 53 (19) 47 (17)
Primary language Stroke 39 (14) 38 (13)
English 265 (94) 256 (9) ED visit in 30 days prior
Spanish 7 (2) 8 (3) to index visit
Other (Chinese, Portuguese, 11 (4) 19 (7) At least 1 visit 47 (17) 50 (18)
Bengali, Creole, Italian,
Punjabi, Russian, ED = emergency department.
Tagalog, French) * These groups were matched based on these factors to usual-
care for confounding variables.
* These groups were matched based on these factors to usual- † Other chief complaints included constipation, bed sores, hypo-
care for confounding variables. tension.

Outcomes matched by demographic factors, including age, sex,

race/ethnicity, and primary language (Table 1). To assess
The primary end point of the study was 30-day ED revisit for medical complexity and risk for readmission, the
rate or hospitalization, measured by data pulled from the matched groups were analyzed for chief complaint
electronic medical record. Any visit to a health system upon ED presentation, past medical history, including
ED within 30 days from index visit was counted toward falls, dementia, and stroke, and ED visits in the 30 days
this metric, regardless of whether the patient was prior to index visit (Table 2). Variation in mean 30-day
admitted or discharged on revisit. The secondary end revisit rates between the usual-care and intervention
point of the study was patient satisfaction with the
GAP-ED intervention, assessed by the surveys adminis-
tered by phone at 30 days post discharge.
Assessed PaƟents
N=535
Data Analysis
AdmiƩed
The patients assessed by the GAP-ED team from N=165
November 2015–July 2017 formed the intervention
group. To assess changes in outcomes, we created a Did not reside in
matched historical usual-care group. Prior to the GAP- Queens county
N=19
ED program, usual care in the ED included care by the
ED team and connection to community resources by the
Transfer or discharge to
social worker or case manager on duty. This care team facility
did not have the additional geriatric and palliative N=63
training of the GAP-ED team. Patients in the historical
usual-care group visited the ED from June 2014–October Expired before follow up
2015 during the days and times the full GAP-ED team N=5
was available (Monday through Friday, 10 AM–6 PM)
and met all inclusion criteria used in the intervention Eligible PaƟents
group. To account for confounding variables that impact N=283
readmission rates, the usual-care group was block- Figure 1. Flow diagram of patients included for analysis.
4 T. Liberman et al.
Table 3. Number of Hospital Revisits in 30 Days After Index
Visit Within the Health System
Intervention Usual-Care
Number of ED Visits (n = 283) (n = 283) p-Value
30 days after index visit 0.20 0.22 0.34
ED = emergency department.
groups was assessed through independent t-tests. A chi-
squared analysis was conducted to determine if there
was a statistically significant difference between rates
of admission and discharge on revisit between the inter-
vention and usual-care groups.
RESULTS
The initial intervention group was composed of 535 pa-
tients. After application of inclusion criteria, a final inter-
vention group of 283 patients remained, representing 53%
of all patients seen by the GAP-ED team (Figure 1). A
usual-care group (n = 283) was matched based on the
same inclusion criteria. The distribution of ages was nearly
equivalent between both groups, as was the predominance
of female, compared with male, patients. The breakdown
of ethnicity varied somewhat between the groups. The
usual-care group was composed of relatively fewer pa-
tients who identified as African American, Asian, or
White, and more patients who identified as Hispanic,
Latino, or a different ethnicity. Similarly, a greater propor-
tion of patients in the intervention group spoke English in
comparison with the usual-care group. The distribution of
chief complaints between the two groups was identical,
with falls (34%) and pain (19%) being the most common.
The differences in demographics, medical history, and
average number of ED visits in the 30 days prior to index
visit were determined to be nonsignificant.
Primary Outcome
An ED revisit was considered to be any visit to a
Health System ED (15 total), regardless of whether
the patient was subsequently admitted to the hospital.
In both groups, over 80% of patients had no revisits
to any system ED within 30 days of the index visit.
In the intervention group, 15% of patients had at least
Table 4. Comparison of Revisits that Lead to Admission vs. Discharge for Patients Who Returned Within 30 Days of Index Visit
Intervention (n = 57)
Admission Status of 30-Day ED Revisits n %
Admitted 23
Discharged 34
ED = emergency department.
one ED revisit within 30 days, whereas 17% of patients
in the usual-care group had at least one ED revisit
within 30 days of index visit. The average number of
ED revisits in the 30 days after the index visit was
0.20 in the intervention group and 0.22 in the usual-
care group (p = 0.34) (Table 3).
For both intervention and usual-care groups, revisits
were stratified by whether the patient was admitted or
discharged from the ED. Of the intervention patients
who revisited the ED within 30 days of index visit
(n = 57), 23 were admitted to the hospital on revisit.
Of the usual-care patients who revisited the ED within
30 days of index visit (n = 61), 35 were admitted to
the hospital on revisit, representing a statistically sig-
nificant change in hospital admission upon ED revisit
(p = 0.01) (Table 4).
Secondary Outcome
The GAP-ED team attempted to contact 509 intervention
patients at 30 days post discharge. Although this cohort
contained the subset of 283 patients used for the t-test
analysis above, it also included patients who did not fulfill
criteria for the statistical analysis. Of the 509 patients
called, 242 patients or their caregivers were successfully
reached and completed the survey, yielding a 47.5%
response rate. In 57% of these cases, a caregiver
completed the survey. Of all patients or caregivers sur-
veyed (n = 242), 95% of the respondents recalled the
ED visit in question and 75% recalled being helped by
the GAP-ED team that day. When asked whether they fol-
lowed up with potential resources discussed with the
Specialist, 56% responded ‘‘yes.’’ When asked whether
the patient had returned to the ED since the index date,
79% responded ‘‘yes’’ and 19% responded ‘‘no.’’ Patients
were also asked whether they think EDs should have a
GAP-ED team to consult patients and caregivers, to
which 90% responded ‘‘yes.’’ Finally, 91% of respondents
‘‘strongly agreed’’ or ‘‘agreed’’ that they found the GAP-
ED Specialist helpful in providing support and resources.
DISCUSSION
As the population of older adults increases, there will be a
corresponding increase in their use of EDs and services.
Usual-Care (n = 61)
n % p-Value
40% 35 57% 0.01
60% 26 33%
The GAP-ED Project
The presence of a GAP-ED team can be an effective
method of providing specialized care for this population
to avoid unnecessary readmissions. Although this pilot
program was unable to decrease repeat ED visits, we
were able to decrease the number of hospital admissions
resulting from these revisits. The data suggest that this
reduction is due to the presence of the GAP-ED team
and the coordination of the GAP-ED Specialist to appro-
priately assess each patient’s individual needs. This
assessment and advocacy for older adults in the busy
ED environment delivers a deeper level of care that can
be needed to untangle the complex comorbidities that
often present in this population. The closer connection
to resources in the community may have also had an ef-
fect on the reduction of hospital admission during ED re-
visits. The relationships established during the first visit
may have prevented or slowed further health decline be-
tween ED visits and allowed ED providers to determine
safe discharge on the second visit, rather than an auto-
matic readmission.
Increasingly, patients are expected to leave the hospi-
tal at an earlier stage in recovery with more caregiving
needs. This expectation can be taxing on patients and
their families due to the limited resources available
within the community. These partnerships can streamline
the discharge process, be included as part of the ED
discharge, and potentially address medical and social is-
sues in the community before they grow to require emer-
gency care.
Providing information on community resources is vi-
tal to this initiative. However, many times it is the
emotional support provided to patients and their family
members that leaves a lasting impact. The value of the
GAP-ED team remains in the ability to address the com-
plex needs of older adults while in the ED (e.g., reorien-
tation, food, warm blankets) and while in the community
by connecting the patient to community resources. As
evidenced by the high patient and family satisfaction
scores, this additional layer of support provides a
welcome resource and connection during a particularly
vulnerable time.
Limitations
The use of a historical usual-care group is an important
limitation of the pilot study. Even though the usual-care
group was thoroughly matched, it does not account for
the expertise of the GAP-ED team, who thoughtfully
chose which patients to approach based on inclusion
criteria. Additionally, the usual-care group relied on the
information available in the system electronic medical re-
cord, which can often be incomplete and less nuanced
than personal interaction. Ideally, the study will be
repeated as a randomized, controlled study.
5
The time criteria limited the scope of patients who
were approached by the GAP-ED team, as we used
only a sample of the hours that the full GAP-ED team
was available. Although this sample did not affect the
outcomes of this study, it may limit generalizability, as
patients presenting to the ED during this time period
may be different than those who present at other times
of the day.
Furthermore, the ED readmission data captured by this
study allowed us to see only patients who returned to an
ED within our health system, not to other EDs in the area.
We expect that patients in both the usual-care and inter-
vention groups visited EDs in other health systems within
the 30 days post index visit.
Finally, the single site of the GAP-ED presence is a
limitation. Future attempts at replication of this program
will require a close relationship between Emergency
Medicine and Geriatrics and Palliative Medicine repre-
sentatives, in addition to the presence of a GAP-ED
Specialist to establish community links for this patient
population.
CONCLUSION
Health care institutions have begun to establish care
programs specifically tailored to older adults to appro-
priately care for this growing population. The inclusion
of a GAP-ED partnership is an effective first step in di-
recting older adults and their families to appropriate
levels of care. Although this pilot study was not able
to decrease ED revisits, the reduction of hospitaliza-
tions has provided an impactful level of care for our
community. In addition to providing high quality of
care, this program received high satisfaction scores
and improved the overall ED experience for patients
and families.
Acknowledgments—This work was supported by the Fan Fox
and Leslie R. Samuels Foundation. The sponsor did not have
any role in the study design, collection, analysis, or interpreta-
tion of data, article preparation or submission.
REFERENCES
1. Gruneir A, Silver MJ, Rochon PA. Review: emergency department
use by older adults: a literature review on trends, appropriateness,
and consequences of unmet health care needs. Med Care Res Rev
2010;68:131–55.
2. Hwang U, Morrison RS. The geriatric emergency department. J Am
Geriatr Soc 2007;55:1873–6.
3. Roberts DC, McKay MP, Shaffer A. Increasing rates of emergency
department visits for elderly patients in the United States, 1993 to
2003. Ann Emerg Med 2008;51:769–74.
4. Albert M, Rui P, McCaig L. Emergency department visits for injury
and illness among adults aged 65 and over: United States, 2012–
2013. NCHS Data Brief 2017;1–8.
6 T. Liberman et al.
5. Lo AX, Flood KL, Biese K, Platts-Mills TF, Donnelly JP,
Carpenter CR. Factors associated with hospital admission for older
adults receiving care in U.S. emergency departments. J Gerontol A
Biol Sci Med Sci 2016;72:1105–9.
6. Samaras N, Chevalley T, Samaras D, Gold G. Older patients in
the emergency department: a review. Ann Emerg Med 2010;56:
261–9.
7. Aminzadeh F, Dalziel WB. Older adults in the emergency depart-
ment: a systematic review of patterns of use, adverse outcomes,
and effectiveness of interventions. Ann Emerg Med 2002;39:238–
47.
8. Carney MT, Fujiwara J, Emmert BE, Liberman TA, Paris B. Elder
orphans hiding in plain sight: a growing vulnerable population.
Curr Gerontol Geriatr Res 2016;2016:4723250.
9. Ortman JM, Velkoff VA, Hogan H. An aging nation: the
older population in the United States. Current population re-
ports: population estimates and projections. Washington DC:
U.S. Census Bureau; 2014. Available at: https://www.census.
gov/prod/2014pubs/p25-1140.pdf. Accessed October 22,
2019.
10. Jones JS, Rousseau EW, Schropp MA, Sanders AB. Geriatric
training in emergency medicine residency programs. Ann Emerg
Med 1992;21:825–9.
11. Schumacher JG, Deimling GT, Meldon S, Woolard B. Older adults
in the emergency department: predicting physicians’ burden levels.
J Emerg Med 2006;30:455–60.
12. New York State NY. New York City: Queens. Available at: https://
www.ny.gov/counties/queens. Accessed September 16, 2019.
13. Gamio L. Where America’s diversity is increasing the fastest. 2019.
Available at: https://www.axios.com/where-americas-diversity-is-
increasing-the-fastest-ae06eea7-e031-46a2-bb64-c74de85eca77.
html. Accessed September 16, 2019.
The GAP-ED Project 7

ARTICLE SUMMARY
1. Why is this topic important?
Older adults in the emergency department (ED) setting
are a highly vulnerable patient population who frequently
present with complex conditions and multiple comorbid-
ities. As the population of older adults grows in the United
States, providing specialized care for this population in
the ED is becoming vitally important to avoid the
increased risk of morbidity and mortality associated
with hospital admission.
2. What does this study attempt to show?
This study attempts to show the benefit of a Geriatric
and Palliative Medicine–Emergency Department (GAP-
ED) partnership to care for older adults presenting to
the ED. The goal of the GAP-ED team was to optimize
the treatment of older adults presenting to the ED and
identify patients that could be safely sent home with com-
munity resources to avoid the unnecessary admissions and
returns to the ED that often result in poor outcomes.
3. What are the key findings?
Although the pilot did not reduce 30-day ED revisits,
there was a statistically significant reduction in hospital
admissions from these revisits. The implementation of a
GAP-ED Specialist resulted in high patient and family
satisfaction.
4. How is patient care impacted?
This pilot reduced the rate of hospital admission during
30-day ED revisits. Patient satisfaction with ED visit was
improved through additional care and connection with the
GAP-ED team.