Gastroenterology

Transition-of-care program from emergency department to gastroenterology clinics improves follow-up

a b s t r a c t

Objectives: Patients discharged from the emergency department (ED) with gastrointestinal (GI) symptoms need to appropriately transition their care to a GI outpatient clinic in a timely manner to have their health needs met and avoid significant morbidity. When this transition isn’t optimal, patients are lost to follow-up, potentially plac- ing them at risk for adverse events. We sought to study the effectiveness of implementing an electronic medical record (EMR) based transition-of-care (TOC) program from the ED to outpatient GI clinics.

Methods: We performed a retrospective single center cohort study of patients discharged from the ED of a tertiary care academic medical center referred to outpatient GI clinic before (Pre-TOC patients) and after implementation of an EMR based TOC program (TOC patients). We further stratified patients based on the Distressed Communi- ties Index (DCI), which is a composite measure of economic well-being. We compared rates of appointment scheduling and appointment attendance between the two groups, as well as 30-day readmission rates to the ED. We also performed a subgroup analysis to determine if socioeconomic status would affect patient follow- up rates.

Results: We included 380 Pre-TOC and 399 TOC patients in our analysis. TOC patients were found to both schedule appointments (50% vs 27% p-value <0.01) as well as show up to appointments (34% vs 24% p-value <0.01) at sig- nificantly higher rates compared to Pre-TOC patients. There was no significant difference between 30-day read- mission rates between the two groups. In addition, TOC patients from At-Risk and Distressed Communities were over 22 times more likely to schedule an appointment compared to Pre-TOC patients from similar neighborhoods (OR 22.18, 95% CI 4.23-116.32).

Conclusion: Our study shows that patients who are discharged from the ED with outpatient GI follow-up are more likely to both schedule and show up to appointments with implementation of an EMR-based direct referral pro- gram compared to no patient navigation, particularly among patients of Lower socioeconomic status.

(C) 2023

  1. Introduction

Patients discharged from emergency departments (ED) require co- ordinated transitions of care to outpatient medical clinics to ensure pa- tient safety and clinical improvement. Studies have shown increased

Abbreviations: Pre-TOC, Before Transition of Care Program was Implemented; DCI, Distressed Community Index; EMR, Electronic Medical Record; ED, Emergency Department; GI, Gastrointestinal; IBD, inflammatory bowel disease; TOC, Transition of Care.

* Corresponding author at: Department of Medicine, Division of Gastroenterology and Hepatology, 101 Nicolls Road, Health Science Tower, Level 17, Room 060, Stony Brook, NY 11794-8173, United States of America.

E-mail addresses: [email protected] (J. Mizrahi), [email protected] (J. Marhaba), [email protected] (E. Sun).

rates of adverse events including higher rates of Hospital readmissions and repeat ED visits when such transitions are not optimized [1]. Much of the literature regarding potential solutions has focused on the transition from the ED to Primary care settings [2,3]. There is limited data regarding transitions of care from the ED to specialty care.

Gastrointestinal (GI) symptoms represent a large proportion of pre- senting complaints to the ED, with abdominal pain alone comprising anywhere from 5%-10% of all ED visits [4,5]. Given the high number of patients presenting with GI symptoms to the ED that may not need ad- mission to the hospital, a systematic process transitioning care to outpa- tient GI clinics from the ED is critical to ensuring these patients receive appropriate gastroenterology follow-up, as well as avoid quick readmis- sion to the ED if they were not able to follow-up in a timely manner. This orderly transition is even more important in patients of lower

https://doi.org/10.1016/j.ajem.2023.04.030

0735-6757/(C) 2023

socioeconomic status, as numerous studies have shown this population to be particularly susceptible to poor follow-up and inappropriate ED utilization [6,7]. This study examined the effectiveness of implementing an electronic medical record (EMR) based transition-of-care (TOC) pro- gram from the ED to outpatient GI clinics.

  1. Methods

We performed a retrospective single center cohort study at a tertiary care academic medical center comparing patients discharged from the ED who were referred to outpatient GI clinic before implementation of an EMR based TOC program (Pre-TOC patients) to those patients discharged from the ED after implementation of the TOC program (TOC patients).

    1. Transition of care program

The goal of the TOC program was to create a systematic layer of pa- tient navigation to ensure that patients presenting to the ED with GI symptoms could see an appropriate GI subspecialist (Hepatologist, In- flammatory Bowel Disease (IBD), Pancreatico-Biliary, etc.) within a timely manner befitting their clinical situation. Prior to implementation of the TOC program, patients who presented to the ED with GI symp- toms and were deemed stable for outpatient follow-up were given the phone number to the GI clinic in their discharge paperwork and sent home. The patient was left with the responsibility of calling the GI clinic and scheduling their own appointment. Often patients fail to call, and even if they do so, they sometimes scheudle an appointment with the wrong GI subspecialist or are not given an appointment date suitable for their GI symptoms.

The TOC program was initiated in July of 2019 and was the result of a multidisciplinary effort involving the GI Division, the ED, and the Information and Technology Division at our institution. A “GI TOC” order in our EMR was created and linked to ED discharge orders. When this order was placed, a message was generated that would populate a “GI TOC Patient List” in a message pool of our EMR. This message auto-populated information such as the patient’s name, date of birth, medical record number, insurance, and reason for GI consultation. The ED physicians were educated to place this order in the EMR upon discharging a patient needing outpatient GI follow-up. Two GI qualified health care professionals (a GI attending

hospitalist and a nurse practitioner who specializes in GI hospitalist medicine) monitored the GI TOC Patient List on a regular basis, reviewing each message and all relevant documentation including ED visit notes, labs, and imaging. They then sent a message to the GI clinic staff with recommendations regarding appropriate GI sub- specialty clinic where applicable, and the appropriate timeframe for GI clinic follow-up. GI clinic staff would then check the patient’s insurance and would subsequently call the patient to schedule an ap- pointment with the appropriate GI subspecialist within a medically appropriate timeframe (Fig. 1).

    1. Inclusion criteria for pre-TOC and TOC patients

To assess the efficacy of the TOC program, we compared patients re- ferred to outpatient GI clinic via the TOC program (TOC patients) with patients prior to implementation of the TOC program (Pre-TOC pa- tients). Inclusion criteria for Pre-TOC patients were discharges from the ED between 7/30/2018 and 6/30/2019, age 18 years or older, and having the GI clinic phone number included in a patient’s discharge pa- perwork. The COVID-19 pandemic began in March 2020, which signifi- cantly affected both outpatient GI availability and patients’ willingness to seek medical care, both in the ED and in outpatient clinic. Thus, the inclusion criteria for TOC patients consisted of patients discharged from the ED from 7/30/2020 to 6/30/2021 (to avoid the peak of COVID-19 disruptions), age 18 years or older, and having a “GI TOC” order placed upon discharge from the ED. Given the uneven numbers of patients likely in these two groups, we used Simple Sample Random- ization to randomly select 400 patients in each cohort to compare, thus providing similar numbers of patients in both cohorts for our final analysis.

    1. Primary and secondary endpoints

The primary endpoint of the research was the number of success- fully scheduled appointments that TOC patients made compared to Pre-TOC patients. Secondary endpoints consisted of the number of pa- tients who showed up to those scheduled appointments as well as the number of patients who were readmitted to the ED within 30 days of discharge. Other data collected included patient demographics, which GI subspecialty clinic patients were seen in, and any labs, imaging, or procedures ordered as a result of the GI clinic visit.

      1. Pre-TOC Implementation Referral to Outpatient GI Clinic from Emergency Department

ED discharges Patient requiring Outpatient GI Follow-up

ED provides phone number to GI Clinic in discharge paperwork

Patient calls to schedule appointment

      1. EMR based TOC Program Referral to Outpatient GI Clinic from Emergency Department

ED discharges Patient requiring Outpatient GI Follow-up

ED places “GI TOC” order in EMR

Order triggers message sent to GI TOC Pool, adding Patient to list in EMR

GI Clinic checks Patient’s insurance and calls Patient to schedule appointment

GI Healthcare Provider sends message to GI Clinic for specific subspecialist and time frame for appointment

GI Healthcare Provider reviews message and Patient documentation, labs, and imaging

Fig. 1. Transition of Care (TOC) vs Pre-Transition of Care (Pre-TOC) Referrals to Outpatient GI Clinic from Emergency Department.

Given our particular interest in how socioeconomic factors affect pa-

Table 1

Patient characteristics of transition of care (TOC) vs pre-transition of care (Pre-TOC) patients.

tient follow-up, we also recorded a metric called the Distressed Com-

Variable

Level

Total

Pre-TOC

TOC

P-value

munities Index (DCI) for each patient. The DCI is a composite score

(N = 779)

(N = 380)

(N = 399)

based on seven socioeconomic metrics that can assess a zip code’s over-

all socioeconomic well-being [8]. The scores are divided into five overall tiers from least to most prosperous zip codes: Distressed Communities, At Risk Communities, Mid-Tier Communities, Comfortable Communi- ties, and Prosperous Communities. We recorded the DCI score for each patient based on the zip code they reside in, and ultimately divided our patients into three groups based on their scores – At Risk and Distressed, Mid-Tier, and Comfortable and Prosperous.

Age (year) Gender

Race

0.06

<0.01

Unit = 1

41.00

+- 25.00

40.00

+- 28.50

43.00

+- 24.00

Female

443 (57%)

235 (62%)

208 (52%)

Male

336 (43%)

145 (38%)

191 (48%)

Asian

33 (4%)

15 (4%)

18 (5%)

Black or

African

55 (7%)

25 (7%)

30 (8%)

American

Other/Not

listed

166 (21%)

74 (19%)

92 (23%)

White

525 (67%)

266 (70%)

259 (65%)

Commercial

369 (47%)

186 (49%)

183 (46%)

Medicaid

255 (33%)

128 (34%)

127 (32%)

Medicare

97 (12%)

45 (12%)

52 (13%)

Self-Pay

58 (7%)

21 (6%)

37 (9%)

0.51

    1. Statistical analysis

Chi-square tests with exact p-values based on Monte Carlo simula- tion were utilized to examine the marginal association between cate- gorical variables and group (Pre-TOC vs. TOC), as well as between categorical variables and scheduled outpatient appointment, and be- tween categorical variables and showed outpatient appointment. Wilcoxon rank sum test was used to compare unadjusted marginal dif- ferences in continuous variable (age) across groups, or between pa- tients who did and who didn’t schedule appointment, between patients who did and who didn’t show to the appointment. For sched- uled appointment and showed appointment, multivariable logistic re- gression model was used to compare the outcome between Pre-TOC

Insurance

Distressed Communities Index (DCI)

Preexisting GI Diagnosis?

Previous

Follow-up with Outpatient GI?

At Risk and Distressed Mid-Tier Comfortable and Prosperous

Yes 338 (43%) 162 (43%) 176 (44%)

No 441 (57%) 218 (57%) 223 (56%)

Yes 122 (16%) 57 (15%) 65 (16%)

No 657 (84%) 323 (85%) 334 (84%)

0.21

0.75

52 (7%)

23 (6%)

29 (8%)

140 (18%)

67 (18%)

73 (18%)

587 (75%)

290 (76%)

297 (74%)

0.68

0.62

and TOC group with further adjustment of demographic factors. An in- teraction term between TOC and DCI was used to examine if the differ- ence of outcome rates between TOC and Pre-TOC was different between DCI levels. Statistical analysis was performed using SAS 9.4 (SAS Insti- tute, Inc., Cary, NC) and significance level was set at 0.05.

  1. Results

We identified 3280 patients who were referred to outpatient GI clinic via a phone number from the ED from 7/30/2018 to 6/30/2019, and 646 patients who were referred to outpatient GI via the TOC order in the EMR from 7/30/2020 to 6/30/2021, with simple sample random- ization used to select 400 patients in each cohort. After removing pa- tients who were <18 years old or who had incomplete data, there were 380 Pre-TOC patients and 399 TOC patients included in the final analysis.

    1. Patient characteristics

Table 1 shows patient characteristics for Pre-TOC and TOC patients. In terms of patient demographics, TOC and Pre-TOC patients had no sig-

nificant differences between age, race, insurance status, or DCI. Only

TOC (Transition of Care); GI (Gastrointestinal); Significance = p value <0.05

regarding inpatient GI consultation or the outpatient ordering of labs, imaging, or procedures between Pre-TOC and TOC patients.

    1. Results of outpatient appointments

Overall, 74% of our patient population who scheduled an outpatient GI clinic appointment (305 patients) completed the appointment (227 patients). Pre-TOC patients were statistically more likely to complete an appointment when compared to TOC patients (90% vs 66% p-value

<0.01). We also recorded the breakdown of the specific GI subspecialty clinic attended by those who completed their appointments (Table 3). Pre-TOC patients compared with TOC patients were statistically more likely to complete General GI Clinic (65% vs 53%) and Motility GI Clinic (16% vs 4%) appointments, while TOC patients were statistically more likely to complete Hepatology Clinic (18% vs 9%), IBD Clinic (14% vs

Table 2 Comparison of results of emergency department visits between transition of care (TOC) vs pre-transition of care (Pre-TOC) patients.

gender was found to be significantly different between the two groups, with 62% of Pre-TOC patients being female compared to 52% of TOC pa-

tients (p-value <0.01). In addition, there were no significant differences between the two cohorts regarding preexisting GI diagnosis or previous

follow-up with outpatient GI.

Scheduled for Outpatient Appointment?

Show Up at

Appointment?

Yes No Yes

No

305 (39%)

474 (61%)

227 (29%)

552 (71%)

103 (27%)

277 (73%)

93 (24%)

287 (76%)

202 (51%)

197 (49%)

134 (34%)

265 (66%)

Seen by Inpatient GI

Yes

51 (7%)

20 (5%)

31 (8%)

Variable Level Total

(N = 779)

Pre-TOC (N = 380)

TOC

(N = 399)

P-value*

<0.01

<0.01

0.16

    1. Results of ED visits

Table 2 highlights the results of the ED visits between the two co-

Team? Readmission Within

30 days to ED

Were Any Labs, Imaging,

No 728 (93%) 360 (95%) 368 (92%)

Yes 42 (5%) 16 (4%) 26 (7%) 0.15

No 737 (95%) 364 (96%) 373 (93%)

Yes 178 (23%) 83 (22%) 95 (24%)

horts and contains the primary and secondary endpoints of the project. Regarding the primary endpoint of the research, TOC patients were found to schedule outpatient appointments at a significantly higher rate compared to Pre-TOC patients (50% vs 27% p-value <0.01). TOC pa- tients were also found to show up to appointments significantly more often compared to Pre-TOC patients (34% vs 24% p-value <0.01). 30- day readmission rates to the ED were not found to be significantly dif-

or Procedures Ordered in Clinic?

Clinic

Labs Ordered in Clinic

Yes

No

74 (10%)

705 (91%)

36 (9%)

344 (91%)

38 (10%)

361 (90%)

Imaging Ordered in

Yes

48 (6%)

27 (7%)

21 (5%)

Clinic No

731 (94%)

353 (93%)

378 (95%)

Procedure Ordered in Yes

123 (16%)

57 (15%)

66 (17%)

No 601 (77%) 297 (78%) 304 (76%)

No 656 (84%) 323 (85%) 333 (83%)

0.51

0.98

0.29

0.56

ferent between the two groups. In addition, there was no difference

TOC (Transition of Care); GI (Gastrointestinal); ED (Emergency Department).

Subspecialty GI clinic visits of transition of care (TOC) vs pre-transition of care (Pre-TOC) patients

Complete Appointment (Yes)

Total (N = 227)

Pre-TOC (N = 93)

TOC (N = 134)

P-value*

General

131 (58%)

60 (65%)

71 (53%)

Hepatology

32 (14%)

8 (9%)

24 (18%)

Type of Subspecialty GI Clinic

Inflammatory Bowel Disease

25 (11%)

6 (6%)

19 (14%)

<0.01

Motility

20 (9%)

15 (16%)

5 (4%)

Pancreatico/Biliary

18 (8%)

4 (4%)

14 (11%)

TOC (Transition of Care).

6%), and Pancreatico/Biliary Clinic (11% vs 4%) appointments compared to Pre-TOC patients. Differences in completing appointments by GI sub- specialty between Pre-TOC and TOC patients were found to be statisti- cally significant (p-value <0.01).

    1. Results of outpatient appointments broken down by DCI group

To explore the effect of socioeconomic status between TOC and Pre- TOC patients, we broke down these two cohorts of patients by their DCI and assessed their rates of scheduling and showing up to appointments. Most patients in both the Pre-TOC and TOC cohorts were from Prosper- ous or Comfortable communities (76% and 74%, respectively), with fewer patients in Mid-Tier Communities (18% and 18%, respectively) and At-Risk and Distressed Communities (6% and 8%, respectively). However, when looking at the proportion of patients who scheduled and showed up to appointments within each DCI group among TOC and Pre-TOC patients, we see significant differences most notably in lower socioeconomic tiers (Table 4). Of the 29 TOC patients from At-Risk and Distressed Communities, 66% of patients scheduled ap- pointments and 41% of patients showed up to appointments. This is compared to the 23 Pre-TOC patients from At-Risk and Distressed Com- munities, where only 9% of patients scheduled appointments and 0% of patients showed up to appointments. Similar trends were evident for both patients in Mid-Tier Communities and from Comfortable and Prosperous Communities.

Given the various demographic factors that influence a patient’s ability to schedule an appointment in addition to their DCI tier, we per- formed a multivariable regression model to compare the outcomes be- tween Pre-TOC and TOC patients while adjusting for various demographic factors and used the DCI tiers as an interaction term be- tween TOC and Pre-TOC patients. We found that TOC patients from every DCI tier were statistically more likely to schedule an appointment compared to Pre-TOC patients, with the difference between tiers being significant (p-value 0.03) (Table 5). Most notably, TOC patients from At Risk and Distressed Communities were >22 times more likely to schedule an appointment compared to Pre-TOC patients (OR 22.18, 95% CI 4.23-116.32). This trend, albeit to a lesser degree, was consistent in TOC patients from Mid-Tier Communities (OR 2.88, 95% CI 1.41-5.88) and from Comfortable and Prosperous Communities (OR 2.29, 95% CI 1.62-3.25). While age was also found to be significantly different be- tween TOC and Pre-TOC patients, none of the other variables – gender, race, insurance, or preexisting GI diagnosis – were found to be different between the two cohorts.

  1. Discussion
    1. Significance of problem regarding transition of care from ED to outpatient clinic

While there are numerous studies evaluating the transition from ED to Primary care clinic, there is a paucity of literature regarding the tran- sition from the ED to outpatient GI clinics specifically. There is some data relating to IBD patients and their ED utilization [9], as well as how IBD patients of lower socioeconomic status having increased ED utilization and decreased outpatient clinic visits [10]. Separately, there is data regarding predictors for poor attendance at outpatient GI clinics [11]. Otherwise, the literature on this important topic is lacking.

Our results demonstrate that patients discharged from the ED who are expected to call the GI office to make an appointment themselves often do not follow up – only 27% of Pre-TOC patients were able to suc- cessfully schedule an appointment. Due to a variety of factors – the com- plexity of the healthcare infrastructure in this country, lack of resources, low patient health literacy – patients face significant challenges both scheduling a timely appointment or showing up to their scheduled ap- pointment, leading to patients being lost to follow-up, or possibly need- ing to return to the ED prior to their appointment given the severity of their symptoms.

    1. TOC program’s efficacy

The GI TOC program at our institution addressed these roadblocks by adding patient navigation as well as provider insight into securing ap- propriately timed and triaged appointments for patients. Placing the im- petus on the healthcare system, which is more equipped both resource wise and knowledge wise, to get patients to the right outpatient GI clinic in an appropriate amount of time, rather than on the patient them- selves, we hypothesized would improve patient safety and get patients into our clinic more efficiently. Indeed, our analysis did show that TOC patients were statistically more likely to both schedule and show up to an outpatient GI appointment compared with Pre-TOC patients. This is the first study to our knowledge assessing the utility of a protocol-driven, EMR-based, transition-of-care program optimizing pa- tient follow-up from the ED to outpatient GI clinic.

While within the overall patient population, more TOC patients showed up to appointments compared to Pre-TOC patients (134 pa- tients vs 93 patients), a higher percentage of Pre-TOC patients com- pleted their scheduled appointments compared to TOC patients (90%

Table 4

Proportion of patients who scheduled and showed up to appointments within each DCI cohort among transition of care (TOC) vs pre-transition of care (Pre-TOC) patients.

Method of Referral to Outpatient GI Clinic

DCI Cohort

Total Number of Patients

Successfully Scheduled Appointment

Successfully Showed to Appointment

At Risk and Distressed

29

19 (66%)

12 (41%)

TOC

Mid-Tier

73

39 (53%)

21 (29%)

Comfortable and Prosperous

297

144 (49%)

101 (34%)

At Risk and Distressed

23

2 (9%)

0 (0%)

Pre-TOC

Mid-Tier

67

19 (28%)

18 (27%)

Comfortable and Prosperous

290

82 (28%)

75 (26%)

TOC (Transition of Care); DCI (Distressed Community Index).

Table 5 Estimated OR with 95% CI Of explanatory variables for scheduled appointment based on a multivariable logistic regression model with an interaction term between transition of care (TOC) vs pre-transition of care (Pre-TOC) patients.

Variable

Levels

OR (95% CI)

P-value*

TOC * DCI

TOC vs Pre-TOC at DCI _ At Risk and Distressed TOC vs Pre-TOC at DCI = Mid-Tier

22.18 (4.23-116.32)

2.88 (1.41-5.88)

0.03

TOC vs Pre-TOC at DCI = Comfortable and Prosperous

2.29 (1.62-3.25)

Age (year)

Unit = 1

1.01 (1-1.02)

0.03

Gender

Female vs Male

0.85 (0.63-1.15)

0.30

Asian vs Black or African American

0.81 (0.32-2.07)

Asian vs Other/Not listed

1.10 (0.48-2.48)

Race

Insurance

Asian vs White 1.14 (0.53-2.43)

Black or African American vs Other/Not listed 1.35 (0.7-2.61)

Black or African American vs White 1.40 (0.76-2.58)

Other/Not listed vs White 1.04 (0.69-1.56)

Commercial vs Medicaid 0.87 (0.61-1.24)

Commercial vs Medicare 1.18 (0.68-2.03)

Commercial vs Self Pay 0.97 (0.53-1.77)

Medicaid vs Medicare 1.35 (0.76-2.4)

Medicaid vs Self Pay 1.11 (0.6-2.05)

Medicare vs Self Pay 0.82 (0.39-1.75)

0.74

0.76

Preexisting GI Diagnosis? Yes vs No 1.29 (0.95-1.75) 0.11

OR (Odds Ratio); TOC (Transition of Care); DCI (Distressed Community Index); GI (Gastrointestinal).

vs 66%). We suspect that Pre-TOC patients, who made the appointments on their own volition and called the phone number to the clinic them- selves, were more determined to complete an appointment since they themselves initiated the phone call. On the other hand, TOC patients likely had a higher no-show rate due to the clinic offering specific ap- pointment times to these patients upon calling them, that patients in the end perhaps could not keep. Regardless, acknowledging this limita- tion is important yet does not supplant the fact that the TOC program overall got more patients into clinic, which indeed was the primary goal of the program.

We also found that patients referred to outpatient GI clinic through the TOC program had more visits to various subspecialty clinics – Hepatology, IBD, and Pancreatico-Biliary clinics – compared to Pre- TOC patients, who had more General GI and Motility clinics. Patients coming to the ED with abdominal pain that may be more related to ei- ther a Biliary or IBD cause may not be able to articulate that to the clinic scheduler, as opposed to a GI qualified health care professional who can make this determination based on the patient’s symptoms, history of present illness, and ED workup. Most patients referred before imple- mentation of the TOC program would thus be referred to General GI, as patients likely would simply call the clinic office and say they were told they needed outpatient GI follow-up. Patients following up with appropriate subspecialists ensures their time is well spent, doesn’t waste the provider’s time seeing a patient they cannot fully help and must refer to one of their colleagues, and most importantly optimizes the care that each patient receives.

Interestingly, we did not see a significant difference in 30-day read- mission rates between Pre-TOC patients and TOC patients (4% vs 7%, p-value 0.15). Our hypothesis was that TOC patients would have a lower 30-day readmission rate than Pre-TOC patients since those pa- tients would ideally be set up with outpatient GI appointments more ef- ficiently, thereby potentially heading off a return to the ED. We suspect that most of the GI appointments made were >30 days after the ED visit, thus likely not affecting the 30-day readmission rate between the two groups.

    1. TOC program’s efficacy per DCI group

There is a plethora of data regarding the impact of social determi- nants of health on healthcare, particularly in the overutilization of the ED [12,13] and the underutilization of ambulatory clinics [14,15]. We hypothesized that patients of lower socioeconomic status would benefit most from our extra layer of patient navigation, given these disparities. We used the DCI score, which was created by the Economic Innovation

Group, a bipartisan public policy organization that studies the diversity of the American economy. This scoring system has been used exten- sively throughout the literature to measure how socioeconomic status affects health related outcomes [16,17]. However, to our knowledge this is the first instance of its use to assess outpatient scheduling and follow-up.

Both TOC and Pre-TOC patients had similar numbers of patients in each DCI group, denoting both cohorts of patients having similar socio- economic makeups. Given our hospital’s location in a relatively affluent

suburb of New York City, ~75% of patients lived in Prosperous and Com- fortable communities with the remaining ~25% living in less affluent neighborhoods. Consistent with most studies showing that increased

patient navigation improves outpatient followup [18,19], in every DCI group TOC patients scheduled more appointments and showed up to more appointments compared to Pre-TOC patients who lacked any post ED discharge navigation.

Our results are even more striking when considering only two pa- tients from At Risk and Distressed Communities scheduled appoint- ments and zero patients showed up to their appointment from the Pre-TOC group. This highly vulnerable population benefitted the most from the TOC program’s patient navigation, as TOC patients from this group were over 22 times more likely to schedule an appointment com- pared to Pre-TOC patients after controlling for demographic factors. The wide confidence interval (4.23-116.32) was likely due to the relatively low number of patients from At Risk and Distressed communities in our cohorts (29 TOC and 23 Pre-TOC), however overall rates of scheduling appointments between all DCI groups was still found to be significant. This suggests that while patients from every socioeconomic level benefitted from the TOC program, those from lower socioeconomic neighborhoods seemed to benefit most from the TOC program com- pared to more affluent patients.

This finding is important. Patient navigation requires a significant

amount of work from the healthcare system’s GI team, as it requires a medically proficient healthcare practitioner to manually go through many patient charts and make individualized Clinical decisions re- garding which clinic to refer to and in what timeframe. Applying this patient navigation across the board to all patients is not feasible in a large healthcare system with limited resources, thus targeting specific patient populations who would stand to benefit the most from this patient navigation would enable the program to target more vulnerable patients. Given our findings, we suggest focusing ef- forts on patients from lower socioeconomic neighborhoods rather than patients from more affluent communities, who benefitted less from the TOC program.

    1. Study limitations

There are several limitations to our study. This was a retrospective, single center cohort study involving an affluent suburb of New York City where most patients were from Prosperous and Comfortable com- munities, thus the generalizability of the results is limited – especially in more urban areas with a higher proportion of patients in lower socio- economic situations. There are other hospitals and GI practices around our institution, and thus patients discharged from the ED who elected to follow up at another hospital could not be known to us. Sample sizes between the pre-TOC and the TOC population varied greatly. We suspect there may have been challenges to adoption of this new process despite providing education to the ED providers inherent with imple- menting any new process. Finally, the Pre-TOC method of referring pa- tients to outpatient GI clinic is specific to our institution, and thus the success of the TOC program can only be judged based on our specific institution’s process and not to other hospitals that refer patients from the ED to outpatient clinic in a different way.

  1. Conclusion

Our study shows that patients who are discharged from the ED with outpatient GI follow-up are more likely to both schedule and show up to appointments with implementation of an EMR-based direct Referral program with proactive calls from the outpatient GI office to facilitate appointment scheduling, especially for patients from At Risk and Distressed Communities. Future plans involve codifying this system throughout the ED, expanding it to other specialties, and instituting the same program to hospitalized inpatients who are discharged ensur- ing that all patients who leave the hospital – whether from the ED or the inpatient ward – receive the appropriate follow-up care they need to ensure their continued health and wellness.

Sources of support or funding

This research did not receive any specific grant from funding agen- cies in the public, commercial, or not-for-profit sectors.

Presentations

Digestive Diseases Week Scientific Meeting 2022; May 22, 2022, San Diego, CA. Selected for Oral Plenary during Session: Understanding and Addressing Disparities in GI Disease.

CRediT authorship contribution statement Joseph Mizrahi: Writing – review & editing, Writing – original draft,

Methodology, Investigation, Formal analysis, Data curation, Conceptual-

ization. Jade Marhaba: Writing – original draft, Formal analysis, Data curation. William Buniak: Writing – review & editing, Formal analysis, Data curation. Edward Sun: Writing – review & editing, Supervision, Methodology, Conceptualization.

Declaration of Competing Interest

JM, JM, WB, or ES have no conflicts of interest to report.

References

  1. Snow V, Beck D, Budnitz T, et al. Transitions of care consensus policy statement: American College of Physicians, Society of General Internal Medicine, Society of Hos- pital Medicine, American Geriatrics Society, American college of emergency physi- cians, and Society for Academic Emergency Medicine. J Hosp Med. 2009;4(6): 364-70.
  2. Carmel AS, Steel P, Tanouye R, et al. Rapid primary care follow-up from the ED to re- duce avoidable hospital admissions. West J Emerg Med. 2017;18(5):870-7.
  3. Atzema CL, Maclagan LC. The transition of care between emergency department and primary care: a scoping study. Acad Emerg Med. 2017;24(2):201-15.
  4. Kamin RA, Nowicki TA, Courtney DS, Powers RD. Pearls and pitfalls in the emergency department evaluation of abdominal pain. Emerg Med Clin North Am. 2003;21(1): 61-72. [vi].
  5. Natesan S, Lee J, Volkamer H, Thoureen T. Evidence-based medicine approach to ab- dominal pain. Emerg Med Clin North Am. 2016;34(2):165-90.
  6. Carlson LC, Zachrison KS, Yun BJ, et al. The Association of Demographic, socioeco- nomic, and geographic factors with potentially preventable emergency department utilization. West J Emerg Med. 2021;22(6):1283-90.
  7. 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.e12.
  8. Group EI. Introduction to the distressed communities index (DCI). https://eig.org/ distressed-communities/; 2022.
  9. Fudman D.I., Perez-Reyes A.E., Niccum B.A., Melmed G.Y., Khalili H. Interventions to decrease unplanned healthcare utilization and improve quality of Care in Adults with Inflammatory Bowel Disease: a systematic review. Clin Gastroenterol Hepatol 2022;(9):1947-1970.e7
  10. Horton N, Kaur R, Sadiq O, Abu Ghanimeh M, Singh M, Kaur N. P036 socioeco- nomic disparities in IBD: outcomes in emergency department utilization and outpa- tient follow-up. Official journal of the American College of Gastroenterology. ACG; 2019. p. S9-S10.
  11. Shrestha MP, Hu C, Taleban S. appointment wait time, primary care provider status, and patient demographics are associated with nonattendance at outpatient gastro- enterology clinic. J Clin Gastroenterol. 2017;51(5):433-8.
  12. Davis CI, Montgomery AE, Dichter ME, Taylor LD, Blosnich JR. Social determinants and emergency department utilization: findings from the veterans health adminis- tration. Am J Emerg Med. 2020;38(9):1904-9.
  13. McCarthy ML, Zheng Z, Wilder ME, Elmi A, Li Y, Zeger SL. The influence of social de- terminants of health on emergency departments visits in a Medicaid sample. Ann Emerg Med. 2021;77(5):511-22.
  14. Aday LA, Andersen RM. The national profile of access to medical care: where do we stand? Am J Public Health. 1984;74(12):1331-9.
  15. Kangovi S, Barg FK, Carter T, Long JA, Shannon R, Grande D. Understanding why pa- tients of low socioeconomic status prefer hospitals over ambulatory care. Health Aff (Millwood). 2013;32(7):1196-203.
  16. Mehaffey JH, Hawkins RB, Charles EJ, et al. Socioeconomic “distressed communities index” improves surgical risk-adjustment. Ann Surg. 2020;271(3):470-4.
  17. Bilfinger T, Nemesure A, Pyo R, et al. Distressed communities index in patients un- dergoing Transcatheter aortic valve implantation in an Affluent County in New York. J Interv Cardiol. 2021;2021:8837644.
  18. Dobbs RW, Stinson J, Vasavada SR, et al. Helping men find their way: improving prostate Cancer clinic attendance via patient navigation. J Community Health. 2020;45(3):561-8.
  19. Percac-Lima S, Ashburner JM, Rigotti NA, et al. Patient navigation for lung cancer screening among current smokers in community health centers a randomized con- trolled trial. Cancer Med. 2018;7(3):894-902.