A pilot project of a Post Discharge Care Team for firearm injury survivors decreases emergency department utilization, hospital readmission days, and cost

Abstract

BACKGROUND

Persons of low socioeconomic status are overrepresented in the firearm injury patient population and may experience challenges in accessing complex outpatient health systems. Consequently, outpatient care for these patients is plagued by poor follow-up and increased emergency department (ED) utilization. We developed a Post Discharge Care Team (PDCT) consisting of a dedicated trauma nurse navigator and medical social worker to bridge the gap between hospital discharge and outpatient care to improve recovery.

METHODS

Adult firearm injury survivors admitted to the trauma service were randomized 1:1 to receive either PDCT services or standard of care (SOC) workflows. The PDCT nurse provided education and set expectations regarding injuries, wound care, and outpatient follow-up. The PDCT social worker performed a comprehensive assessment to identify concerns including housing and financial instability, food insecurity, or transportation issues. The primary outcome was ED utilization, with secondary outcomes including readmissions and overall health care costs compared between groups.

RESULTS

In the first 6 months of the study, a total of 44 patients were randomized to PDCT and 47 to SOC. There were 10 patients who visited the ED in the PDCT group compared with 16 in the SOC group (p = 0.23) for a total of 14 and 23 ED visits, respectively. There were 14 patients in the PDCT and 11 patients in the SOC groups who were readmitted (p = 0.31), but the PDCT group was readmitted for 27.9 fewer hospital days. After accounting for programmatic costs, the PDCT had a hospital savings of $34,542.71.

CONCLUSION

A collaborative, specialized PDCT for firearm injury survivors consisting of a dedicated trauma nurse navigator and medical social worker decreased outpatient ED utilization, readmission days, and was cost effective. Trauma centers with high volumes of penetrating trauma should consider a similar model to improve outpatient care for firearm injury survivors.

LEVEL OF EVIDENCE

Therapeutic/Care Management; Level III.

A six-month interim analysis of a Post Discharge Care Team for firearm survivors has shown a decrease in outpatient emergency department utilization, readmission days, and cost through use of a dedicated team of providers to bridge the gap from inpatient to outpatient care.

Outpatient care after traumatic injury is plagued by poor outpatient follow-up and emergency department (ED) utilization. Socioeconomic factors such as insurance status, lower levels of education, and lower income have been associated with poor outpatient follow-up attendance.^1,2 One particularly vulnerable patient population is firearm injury survivors, who are disproportionally young, Black men of lower socioeconomic status.^3–6 These patients represent a population that has been historically marginalized by the health care system as well as society,^7–9 and are impacted by the intersection of low resources, individual and structural racism, and residing in neighborhoods with high levels of violence. All of these factors perpetuate health disparities for firearm injury survivors.

Along with the complex multidisciplinary nature of trauma care, health care systems and insurance concerns can be challenging for patients to navigate, and socioeconomic barriers such as lack of transportation or lost wages can deprioritize outpatient clinic attendance.¹⁰ In addition, discharge processes and education are non-individualized, and follow-up care for trauma patients is variable.¹¹ In qualitative interviews assessing the posttrauma discharge process, patients reported feeling rushed out of the hospital as well as emotionally or physically unprepared for discharge due to poor communication regarding their injuries and outpatient treatment plans.¹² All of these factors may contribute to recovery challenges within the community and poor outpatient follow-up. Moreover, penetrating injury is also a predictor of ED utilization after hospital discharge,¹³ and firearm-injured patients have been shown to experience long-term pain, as well as poor functional and mental health outcomes at rates higher than those who were injured in a motor vehicle crash.¹⁴

In other multidisciplinary specialties caring for similarly complex patients, patient nurse navigators have been utilized successfully to improve outpatient engagement through increased education, appointment arrangement, and generalized support for disease process understanding.^15–17 However, the potential benefits of a patient nurse navigator are less understood in the trauma setting, particularly for those with firearm injury. Our objective was to evaluate patient outcomes after implementation of a Post Discharge Care Team (PDCT) consisting of a dedicated, trauma-specialized nurse navigator and medical social worker to assist in addressing the complex social and resource needs of this patient population. We hypothesized that the PDCT would reduce utilization of the ED, reduce readmissions, and reduce overall health care costs for these measures.

METHODS

Population and Procedure

This project as conducted as an approved quality improvement project and was exempt from review by the Institutional Review Board at the primary authors’ institution. The findings are reported in accordance with the SQUIRE guidelines¹⁸ (Supplemental Digital Content, http://links.lww.com/TA/D628). It was conducted with a prospective, randomized controlled design from November 2022 through April 2023. All adult (≥18 years old) patients admitted to the trauma surgery service at an urban, Midwest adult Level 1 trauma center with a firearm-related injury were screened for inclusion in the study. As the trauma nurse navigator and social worker were integrated into the trauma surgery service teams, patients who were admitted to subspecialty services were excluded. Patients that were admitted solely over the weekend or for an observation period of ≤24 hours were also excluded as the PDCT could not consistently see them prior to discharge due to time constraints. In addition, after commencement of the study, patients who suffered from intentional self-inflicted injury, those that had a new cervical spine injury resulting in quadriplegia, as well as those who had severely impaired neurologic outcomes resulting in the need for long term acute care were excluded as these represent patient populations with unique, complex health care needs that were deemed beyond the scope of this intervention.

Eligible patients were randomized 1:1 to either the PDCT or standard of care (SOC) groups each weekday using block randomization. Patients were randomized in order of admission time, and the allocation sequence was determined by a different author from the author who randomized, therefore blinding the randomizing author to allocation sequence. The PDCT was comprised of a dedicated trauma nurse navigator and master's level-trained medical social worker. Once the patient had clinically stabilized, the patient was evaluated by the PDCT as soon as possible, ideally within 24 hours of admission. The patients were seen by the PDCT throughout their hospital stay with the number and length of visits dictated by individual needs.

Intervention

The nurse navigator educated the patient on their injuries, multidisciplinary medical plans, wound care, pain control strategies, and milestones needed to reach discharge. The objective was to increase the patient’s understanding of their injuries, expectations for the disease process, and follow-up plans of care. If desired by the patients, the nurse navigator included family members or caregivers in discussions to provide increased support upon discharge. Just prior to discharge, the nurse navigator again met with the patient to review their injuries, ensure understanding of wound care needs and medication regimens, elucidate upcoming care needs that had not been addressed, and to review follow-up plans including future clinic appointments. At a minimum, patients received a phone call from the nurse navigator within 48 hours after discharge and at 1 month. The nurse navigator was available at a unique phone number with the ability for patients to leave messages with any questions. The nurse navigator triaged concerns, and when appropriate, arranged earlier follow-up in clinic for concerns that could not be addressed by phone.

The medical social worker performed a comprehensive psychosocial assessment to evaluate for concerns with substance use, community safety, finances, employment, family support, legal issues, a history of chronic pain, and mental health. They provided ongoing support with transportation for appointments and financial resources, including application to the Crime Victim Compensation Program, which is an individual state-led program where victims of crime may apply to have crime-related expenses reimbursed by the state. The social worker also worked closely with the hospital violence interruption program to coordinate resource referrals and to utilize emergency housing resources if there were safety concerns upon the patient’s return to the community.

Both the nurse navigator and social worker had dedicated time to see the patient in-person at the Trauma Quality of Life clinic,¹⁹ the hospital's multidisciplinary clinic dedicated to firearm injury survivors. At this first scheduled outpatient appointment, the PDCT followed up on the patient's progress and provided additional resources (Fig. 1). In addition, the PDCT's longitudinal interactions with the patients allowed them to provide critical information to the outpatient care team which included a trauma nurse practitioner, a trauma psychologist, a physical therapist, and a member of the hospital violence interruption team regarding the patient's clinical course and ongoing needs.

Figure 1.

A description of the PDCT vs. SOC care continuum models to highlight additional interventions and resources provided by the PDCT.

Data Collection and Statistical Analysis

The primary outcome was 60-day ED utilization. Secondary outcomes were 60-day readmissions, health care costs related to ED utilization and readmissions, and outpatient follow-up attendance. Emergency department visits resulting in a readmission were counted solely as a readmission, they were not counted as both an ED visit and a readmission. Study outcomes were abstracted from the electronic health record. Financial data was obtained through the institution, and total emergency department and hospital readmission costs was summed together and compared across groups for each of the outcomes.

Study data were collected and randomization allocation was performed using REDCap (Research Electronic Data Capture) electronic data capture tools.^20,21 Statistical analysis was performed using Stata (StataCorp. 2021. Stata Statistical Software: Release 17. College Station, TX: StataCorp LLC). Descriptive statistics were used to evaluate the two groups. For categorical variables, χ² was used to compare groups. For continuous variables, the Kruskall-Wallis test was used to compare medians as outcomes of interest were not normally distributed.

To complement the statistical analysis, geospatial mapping was conducted to visualize areas of highest overall patient volume as well as volume of postdischarge ED utilization. Home addresses abstracted from the electronic health record were geocoded to Zip Code Tabulation Areas (ZCTAs) obtained from the US Census Bureau 2020 Census data for Milwaukee County²² using QGIS 3.32.3-Lima. Zip Code Tabulation Areas were utilized as they permit the visualization of the area of a zip code. Choropleth maps displayed the variation of these two volumes by ZCTA for overall admission rates as well as the number of patients who returned to the ED after discharge. Of particular focus were the northside neighborhoods of Milwaukee County, which have a high Area Deprivation Index,²³ and the impact of neighborhood disadvantage and lower socioeconomic status is associated with risk for traumatic injury and recovery factors.^3,5,24,25 Patients who lived outside of the hospital’s county or who were discharged to jail were excluded as the purpose of mapping was to identify high-risk locations for ED utilization. Health care access for patients who live in different counties and who are discharged to jail are inherently different.

RESULTS

Over the program's first 6 months, 114 patients were admitted to the hospital for a firearm injury. After applying the exclusion criteria, a total of 91 patients were randomized into the study, of which 44 (48%) were assigned to the PDCT and 47 (52%) were assigned to the SOC (Fig. 2). In each group, the majority of patients were male (84% in PDCT and 81% in SOC) and Black (73% in PDCT and 87% in SOC). Additional baseline characteristics were similar between the two groups (Table 1).

Figure 2.

CONSORT diagram showing algorithm of study flow.

TABLE 1.

Demographics of the Post Discharge Care Team and Standard of Care Cohorts

	PDCT	SOC	p
n	44	47
Age*	29.0 (24.0; 34.0)	30.0 (22.0; 35.0)	0.66
Male gender	37 (84%)	38 (81%)	0.68
Race/ethnicity			0.29
Black	32 (73%)	41 (87%)
Hispanic	6 (14%)	3 (6%)
White	3 (7%)	3 (6%)
Asian	2 (5%)	0 (0%)
Other	1 (2%)	0 (0%)
Insurance			0.15
Medicaid	40 (91%)	37 (79%)
Private	3 (7%)	8 (17%)
Uninsured	0 (0%)	2 (4%)
Workers compensation	1 (2%)	0 (0%)
Operative intervention	12 (27%)	15 (32%)	0.63
Discharge disposition			0.39
Home	31 (71%)	33 (70%)
Rehabilitation	9 (21%)	5 (11%)
Correctional facility	3 (7%)	7 (15%)
Other**	1 (2%)	2 (4%)
Law enforcement involved	18 (41%)	24 (51%)	0.33
Hospital LOS*	6.5 (3.0–10.8)	5.0 (3.0–9.0)	0.39
ICU stay	24 (54%)	19 (41%)	0.21
ICU LOS*	1.5 (0.0–3.0)	0.0 (0.0–3.0)	0.51

*Median, IQR.

**Other includes homeless shelter, rooming house, or group home.

Outpatient Health Care Engagement

There were 10 patients who visited the ED in the PDCT group compared to 16 in the SOC group (χ² = 1.43, p = 0.23), and there was a total of 14 ED visits in the PDCT group and 23 in the SOC group. Although there were no statistically significant differences in demographics of patients who visited the ED (Table 3), half the number of Black patients visited the ED in the PDCT (7 vs. 14 patients, p = 0.18) and the median days to first ED visit was 14 days in the PDCT vs. 10 days in the SOC group (p = 0.38). For readmissions, there were 11 patients who were readmitted in the PDCT group compared to 9 in the SOC group (χ² = 0.45, p = 0.5) for a total of 14 readmissions in the PDCT group and 11 in the SOC group (Table 2). Despite the higher number of readmissions, PDCT patients were readmitted for a total of 34.4 days, and SOC patients were readmitted for a total of 62.2 days, saving nearly 28 fewer total hospital readmission days in the PDCT group. A greater number of PDCT patients attended their trauma-related follow-up compared with the SOC group (84% vs. 68%, χ² = 3.18, p = 0.07), although this was not statistically significant. In addition, PDCT patients were more likely to engage with a primary care physician (PCP) (39% vs. 19%, χ² = 4.23, p = 0.04) (Table 2).

TABLE 3.

Demographics of Patients Who Returned to the Emergency Department Compared by Intervention Groups

	PDCT	SOC	p
n	10	16
Age*	27.5 (23.3; 30.3)	29.0 (22.3; 36.5)	0.51
Male gender	9 (90%)	13 (81%)	0.55
Race			0.18
White	2 (20%)	0 (0%)
Black	7 (70%)	14 (88%)
Hispanic	1 (10%)	2 (13%)
Insurance			0.31
Medicaid	8 (80%)	12 (75%)
Private	1 (10%)	4 (25%)
Workers compensation	1 (10%)	0 (0%)
ICU Stay	7 (70%)	11 (69%)	0.95
ICU LOS*	2.0 (0.0; 2.5)	2.5 (0.0; 6.0)	0.44
Hospital LOS*	10.0 (5.8; 15.5)	8.0 (5.0; 14.0)	0.65
Required index operation	8 (80%)	14 (88%)	0.61
Disposition
Home	5 (50%)	11 (67%)	0.15
Rehab	4 (40%)	1 (6.3%)
Correctional facility	0 (0%)	2 (13%)
Other	1 (10%)	2 (13%)
Days to first ED visit*	14.0 (7.5; 20.3)	10.0 (3.0; 21.8)	0.38

*Median, IQR.

TABLE 2.

Comparison of Outpatient Health Care Engagement Between the Post Discharge Care Team and Standard of Care Groups. ED Visits and Readmissions Were Within 60 Days of Discharge

	PDCT	SOC	χ2	p
n	44	47
Outpatient follow-up	37 (84%)	32 (68%)	3.18	0.07
Saw primary care	17 (39%)	9 (19%)	4.23	0.04
Total ED visits	14	23
Patients who visited ED	10 (23%)	16 (34%)	1.43	0.23
ED visits per patient*	0 (0; 0)	0 (0; 1)		0.35
Total readmissions	14	11
Patients readmitted	11 (25%)	9 (19%)	0.45	0.5
Readmissions per patient*	0 (0–1)	0 (0–0)		0.62
Days per readmission*	2.0 (1.0; 3.0)	3.0 (2.0; 9.0)		0.07
Total readmission hours	825.2	1493.5
Hours per readmission*	46.5 (30.4–72.5)	72.3 (57.0–227.2)		0.07
Total hours difference	668.4
Total days difference	27.9

*Median, IQR.

Hospital Costs

The total ED visit cost was $11,722.28 in the PDCT group and was $16,893.74 for the SOC group. The total readmission cost was $151,019.28 in the PDCT group and was $252,654.53 in the SOC group. The overall total return-to-hospital cost was $162,741.56 in the PDCT group and was $269,548.27 in the SOC group for a total PDCT cost savings of $106,806.71. After factoring in total PDCT program costs of $72,264.00 for the time period, the PDCT group had a net savings of $34,542.71 (Table 4).

TABLE 4.

Comparison of Hospital Cost Data Between the PDCT and SOC Groups

	PDCT	SOC	p
N	44	47
Total ED visits	14	23
Total ED cost	$11,722.28	$16,893.74
Cost per ED visit*	$697.91 (419.53; 1368.00)	$594.25 (405.66; 1175.87)	0.47
Total readmissions	14	11
Total readmission cost	$151,019.28	$252,654.53
Cost per readmission*	$8,416.58 (3941.81; 154.42)	$11,290 (6980.36; 36784.58)	0.14
Total cost	$162,741.56	$269,548.27
Total cost difference	$106,806.71
Program costs	$72,264.00
Net return	$34,542.71

*Median, IQR.

Geospatial Mapping

After exclusion of patients who lived outside the hospital county (n = 4) and who were discharged to jail (n = 10), there were 77 patients included in the geospatial mapping. A choropleth map of admissions (Fig. 3A) shows high rates of enrollment in several ZCTAs on the north side of the county. Figure 3B illustrates ED utilization postdischarge in which only three ZCTAs on the north side represented more than half of the patients who returned to the ED. ZCTA 53206 had 5 of 9 patients present to the emergency department postdischarge, while 53216 had 4 of 7 patients return to the ED and 53208 had 3 of 5 patients return (Fig. 3).

Figure 3.

Map of Milwaukee County. (A) the number of randomized patients per zip code tabulation area; (B) the number of patients who visited the ED from each zip code tabulation area. In zip code tabulation areas 53216, 53206, and 53208, over half of the admitted patients returned to the ED.

DISCUSSION

Our PDCT was implemented as a pilot program to provide a bridge from inpatient to outpatient care by leveraging a dedicated nurse navigator and medical social worker. We demonstrated improvements in outpatient health care engagement with trauma-related follow-up, ED utilization, health care costs, and readmission days. Most notably, the PDCT group had a total cost savings of $106,806.71; even after accounting for program costs, the PDCT saved a net $34,542.71 in 6 months. The majority of savings were seen in readmissions, as the PDCT patients had nearly 28 fewer readmission days.

Successful nurse navigator roles are well described in other patient populations that involve multidisciplinary care, including advanced heart failure and oncology.^15–17 In addition, in an orthopedic trauma population, improved follow-up at the first outpatient appointment was seen in patients who received a phone call from a trauma recovery coach 3 days to 5 days after hospital discharge, suggesting that these practices are translatable.²⁶ These early postdischarge phone calls from health care team members have been shown to identify complications in patients^27,28 and to improve follow-up,²⁹ further supporting the idea that consistent communication can lead to early intervention and prevention of further, possibly more complicated health issues. The PDCT patients had a median of 14 days to the first ED visit compared with 10 in the SOC group, demonstrating that our early outpatient phone calls might be preventing some early ED visits. On the other hand, this may indicate that another follow-up call slightly later in the postdischarge phase near 14 days could also be helpful to identify patient issues. Regarding readmissions, it is of note that not all of the PDCT readmissions were via the ED. Some were a result of the PDCT nurse identifying the problem on a phone call with the patient, allowing for expedited care through the outpatient clinic and direct admission to the hospital, therefore, bypassing ED wait times, which are a significant source of patient dissatisfaction.³⁰ Overall, these concepts of improved outpatient care were clearly reflected in our increased outpatient healthcare engagement regarding trauma-related follow-up and PCP visits, decreased readmission days, and significant improvements in health care readmission costs.

The PDCT included not only a nurse navigator, but also a medical social worker tasked with identifying socioeconomic needs during hospitalization and providing ongoing connection to resources throughout the recovery process. Firearm-injured patients often come from low socioeconomic status and have complex discharge needs that are not limited to health care navigation..^3,5,6,10 The geospatial analysis identified that a majority of patients who visited the ED came from a few ZCTAs which are known to be persistently marginalized secondary to historic redlining practices in Milwaukee County, and have a high proportion of Black residents.^31,32 This highlights the importance of a social worker to provide resources for socioeconomic needs after discharge. In a prior study of discharge needs for firearm survivors, transportation was identified as a major barrier to accessing health care after injury, as well as financial concerns, need for help with insurance and financial paperwork, and safety concerns with returning to the community where they were injured.¹⁰ In Black male violence survivors, additional barriers to care include a disconnect between their injuries and medical understanding, a high value on self-reliance in problem solving, as well as issues with transportation and finances in the setting of new functional limitations.³³ We saw half the number of ED visits for Black patients in the PDCT group compared with SOC, indicating that we may be successful in targeting this high-risk patient population.

Medical mistrust, particularly when law enforcement is involved, is also known barrier to care for Black patients.^33,34 In academic centers, it is not uncommon for patients to have difficulty identifying and connecting with their providers as care teams are large and may change frequently. The PDCT program created a consistent care team focused on the patients’ individualized health care and social needs. Anecdotally, our patients reported increased feelings of trust and engagement with the PDCT. Patients often requested that the bedside nurses call the PDCT to come meet with them. One patient, when discussing ostomy reversal with a senior surgeon, informed the surgeon that they would have to consult with the PDCT team before making such a big medical decision. Aside from the successes of reducing ED utilization and health care costs, these moments displayed improved communication and trust, and were a highlight of the program.

Similar multidisciplinary care models for trauma patients have been successful, but have not focused on the unique needs of firearm-injured patients. Livingston et al. described a longitudinal comprehensive care model for severely injured trauma patients using an ICU LOS > 2 days and injury severity score ≥16 as inclusion criteria. The care team included a nurse practitioner, social worker, and health care navigator. Patients were identified for their Center for Trauma Survivorship program in-hospital and then seen comprehensively as an outpatient. They saw improvements in ED utilization, but not readmissions, similar to our results with the PDCT pilot.³⁵

At our institution, we currently have a multidisciplinary outpatient clinic, the Trauma Quality of Life (TQOL) Clinic, which focuses specifically on the care of gun violence survivors.¹⁹ The PDCT provided the essential link in the continuum of care between the inpatient and outpatient environments to address the vulnerable time period between discharge and clinic follow-up. The nurse navigator served as a contact point for questions and concerns, provided support for self-care, was able to identify potential complications early, and facilitated follow-up in the outpatient setting. If urgent evaluation was needed as determined by the nurse navigator, the social worker was available to arrange transportation.

The PDCT medical social worker was previously integrated into the outpatient TQOL Clinic, which significantly benefited our patients by providing a consistent presence and promoting a trusting relationship for this population that traditionally has been marginalized and mistreated by the health care system.^7,8 By moving up the time of initial evaluation to the inpatient setting, she was able to identify patient needs and provide resources earlier, thereby alleviating potential financial crisis in the initial recovery weeks. As a result of the overall interventions from the PDCT, there were more PDCT patients who followed up in the outpatient clinic and who saw a PCP. We hypothesize that these critical interventions in the PDCT patients were the reason for increased health care engagement, less reliance on the emergency department, and fewer hospital readmission days.

Our pilot study should be interpreted within the context of its limitations. We were only able to include patients that were admitted to the trauma surgery service. Therefore, we missed a large population of patients who were either discharged from the ED or admitted to other specialty services, as well as those who met our additional exclusion criteria of self-inflicted intentional injury or new quadriplegia. As this was a pilot study, our patient cohort is small, which limited the ability to detect statistical differences in our outcomes. However, we still saw successes in health care engagement, emergency care utilization, and readmission days which demonstrate an opportunity for increasing the scope of postdischarge trauma services to serve all firearm-injured patients. We were only able to include patient data from visits at our institution, therefore we likely missed ED visits, readmissions, or follow-up that was done at outside institutions, which may be influenced by insurance status or discharge disposition. Finally, this intervention is best generalizable to centers with high rates of firearm injury, however this model has the potential to be applied to other high risk patient populations such as those with prolonged ICU stays. Further work is ongoing to evaluate the PDCT’s true impact on medical mistrust and patient satisfaction through validated survey measures, as well as its impact on bedside nurses and other health care professionals who cared for patients during the pilot study.

The preliminary findings have resulted in the PDCT nurse navigator and social work positions becoming permanently funded by the hospital system. Therefore, future work will focus on identifying the high-risk patient groups which would most benefit from PDCT services to provide care for increased numbers of patients without diluting the successes of the PDCT. This will include analysis of injury patterns and in-depth reasons for ED visits and readmissions. In addition, the geospatial analysis provided insight into residential areas in the city that are associated with emergency health care utilization. Patients from these areas will be targeted for engagement in the program. Finally, ongoing work will aim to identify any unmet needs of this patient cohort that can be addressed through additional offerings.

The pilot PDCT team was successful in increasing outpatient health care engagement and in reducing 60-day emergency department visits and total readmission days in our patients while being cost-effective. This was accomplished through the team’s consistent, trusting, and therapeutic relationship with patients to empower them to engage with the health care system after injury. Health care systems should consider the use of similar inpatient to outpatient multidisciplinary care models to improve recovery after injury, especially for high-risk patient populations such as firearm injury survivors.