The Health and Social Needs of Patients Discharged From the Emergency Department With Suspected COVID-19

Abstract

Health-related social needs (HRSNs), such as food or housing insecurity, are important drivers of disparities in outcomes during public health emergencies. We describe the development of a telehealth follow-up program in Boston, Massachusetts, for patients discharged from the emergency department after coronavirus disease 2019 (COVID-19) testing to identify patients with worsening clinical symptoms, to screen for unmet HRSNs, and to deliver self-isolation counseling and risk-reduction strategies for socially vulnerable people. We prioritized telephone calls to patients with public health insurance and patients without primary care physicians. In the first 43 days of operation, March 30–May 12, 2020, our intervention reached 509 patients, with 209 (41.1%) patients reporting an HRSN, most commonly related to food, housing, or utilities. Thirty-one (6.1%) patients required assessment by a clinician for clinical worsening. This public health intervention may be useful for other institutions developing programs to address the social and health needs of patients discharged with suspected COVID-19.

Public health emergencies such as the coronavirus disease 2019 (COVID-19) pandemic disproportionately affect socially vulnerable populations. Socially vulnerable populations are more likely to die during public health emergencies than well-resourced populations. ¹ During the 2009 H1N1 influenza pandemic, hospitalization and mortality rates were higher among racial/ethnic minority groups and those who had low incomes and lived in poor neighborhoods compared with people who were White, in higher income groups, and living in wealthier neighborhoods. ^2,3 Similarly, traditionally marginalized groups such as Black and Latinx populations have borne the brunt of COVID-19 and its sequelae in the United States. ^{4 -7}

Health-related social needs (HRSNs), such as food or housing insecurity, are important drivers of disparities in morbidity and mortality outcomes during public health emergencies. ^7,8 The comorbidities that predispose people to severe infection, such as asthma, diabetes, chronic kidney disease, and heart disease, are also prevalent among non-White, elderly, and low socioeconomic status populations. ^{6,9 -12} Limited access to healthy food options, common in poor neighborhoods that may also have low levels of walkability and limited opportunities for exercise, results in a higher incidence of heart disease and diabetes—both of which have been linked to increased risk of mortality from COVID-19—than among the general population. ^12,13 Compared with groups with higher socioeconomic status, groups with lower socioeconomic status and racial/ethnic minority groups are more likely to live in neighborhoods with poor air quality that results in lung disease, which is also linked with higher rates of COVID-19 mortality. ¹⁴

Socially vulnerable patient populations also face unique challenges to self-isolation and social distancing. These challenges, which are associated with the patient’s HRSNs, include cohabitation with multiple family members or friends, unstable housing, food insecurity, poor access to private transportation, limited social networks, lack of childcare, and reliance on income from low-wage and low-benefit essential jobs that pose additional health risks in the setting of an infectious pandemic. ^3,8 Compared with patients with adequate resources, vulnerable patients living in crowded housing or multigenerational homes may have more barriers to accessing care when they or someone in their household becomes ill enough to require evaluation. ³ Patients with suspected and confirmed COVID-19 who are well enough to be instructed to self-isolate at home can infect more vulnerable cohabiting family members who later become critically ill. Early epidemiologic evidence has shown that household contacts of people with COVID-19 are at high risk of secondary infection. ^{15 -17}

National and international guidelines recommend that health care providers ensure that patients’ living conditions support self-isolation and that patients have access to critical resources (eg, food) when making the decision to discharge a patient home. ^18,19 Moreover, the World Health Organization (WHO) recommends that “a communication link with health care provider or public health personnel, or both, should be established for the duration of the home care period.” ¹⁸ Despite these recommendations, no currently established national standard clinical protocols exist on the safe discharge of COVID-19 patients to their homes. Moreover, during surge periods, traditional outpatient follow-up systems may be less robust than during non-surge periods given ambulatory practice closures and possible staff redeployment. ²⁰ Given resource constraints, telemedicine encounters can serve as a convenient and practical strategy for conducting follow-up assessments of COVID-19 patients. Follow-up via telephone or video-enabled virtual visits can facilitate referral to care during a pandemic.

We describe the development of a telehealth follow-up program for patients discharged from the emergency department (ED) after COVID-19 testing. The goal of the program is to screen and provide referral for unmet HRSNs, to identify patients with worsening clinical symptoms who require in-person reevaluation, and to reinforce self-isolation counseling and risk-reduction strategies for vulnerable people.

Methods

Conceptual Framework

Our intervention is rooted in the theoretical framework of the WHO’s Conceptual Framework on Social Determinants of Health. This framework establishes that individual material circumstances (ie, access to food and housing), psychosocial circumstances, and the health system itself are among the key intermediary determinants of people’s health outcomes. ²¹ The primary objective of our intervention is to address these determinants of health by conducting HRSN screening and referral to community resources. Our secondary objective is to leverage the health system’s ability to mediate the risk of severe or critical illness by actively surveilling for clinical symptom progression and reinforcing counseling on symptom surveillance and risk-reduction strategies to minimize exposure of others in the home.

Operational Framework

We operationalized our intervention at Massachusetts General Hospital in Boston, Massachusetts, in March 2020 by building on the existing infrastructure of a pre-pandemic program that employed ED navigators to address the HRSNs of Medicaid accountable care organization (ACO) patients, with the goals of improving health outcomes and reducing rehospitalization rates. ED navigators are nonclinical staff members who assist ED patients, traditionally Medicaid ACO patients at our institution, in accessing health care and community resources, such as connecting patients with primary care providers or mitigating social barriers to care. Our initiative uses ED navigators and ED clinicians to call patients who are discharged after testing for COVID-19.

Target Patient Population

Our intervention targets patients discharged from our large, urban academic ED after undergoing exposure and symptom-based testing for COVID-19. We include patients of all payer groups, expanding the reach of ED navigators beyond Medicaid ACO patients. Although we prioritize patients with public or no health insurance and patients without primary care physicians, we also include patients with private health insurance, because previous experiences have shown that public health emergencies exacerbate the population’s need for social and health services, even among populations that typically do not require social support. ^1,8,22 We included patients with both positive and negative COVID-19 test results. We based this decision on early pandemic uncertainty about the sensitivity of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) reverse-transcription polymerase chain reaction (RT-PCR) assays and, thus, the potential for false-negative results.

Structure of Intervention

Since March 30, 2020, patients tested for COVID-19 are called by an ED navigator 2 days after discharge (Figure 1). ED navigators conduct screening for unmet HRSNs. Using open-ended and closed-ended questions, ED navigators assess patients for needs relevant to the context of home isolation, including food security, housing stability, access to medications and primary care, and paying for utilities. Patients who have unmet HRSNs are referred to local community programs and social services that can meet the patients’ need. Patients are referred to a wide variety of community resources, such as food pantries, housing rental assistance programs, and legal organizations that can assist with unemployment assistance applications.

Figure 1.

Emergency department coronavirus disease 2019 discharge follow-up program workflow, Massachusetts General Hospital, 2020. Abbreviations: ED, emergency department; EDC, emergency department clinician; EDN, emergency department navigator; EMR, electronic medical record; HRSN, health-related social need; PCP, primary care physician; SDH, social determinant of health.

ED navigators are also trained to survey patients’ general clinical well-being. If, during conversation with the ED navigator, patients have a clinical concern or question about themselves or others in the household, the ED navigator triages them to the program’s on-call clinician, who then calls the patients directly. The on-call clinician uses a standardized institutional screening tool to assess the patient for worsening symptoms that might require a return visit to the ED (Figure 2). The screening tool uses criteria for evaluation based on institutional recommendations and Centers for Disease Control and Prevention guidance. ²³ If highly concerning “red flag” (ie, shortness of breath, chest pain, hemoptysis, bluish lips, presyncope, confusion) symptoms are present (eg, persistent chest pain, new confusion), the patient is directed to return to the ED. If worsening non–red flag symptoms are identified, especially with substantial medical comorbidities, the patient may be referred to our mobile response program, a recently developed initiative that provides typically same-day, in-home assessment of patients by a paramedic team. If patients do not require in-person reassessment, they are instructed to continue to isolate and follow up with their primary care physician. Repeat telephone calls to patients with confirmed COVID-19 are done on day 8 post-discharge to evaluate for clinical worsening in the second week of illness. Upon discharge, patients are also provided with a number for the hospital’s COVID-19 hotline, through which patients can have questions about COVID-19 answered and be directed to additional care if needed. Because one objective of our intervention is to help reduce the transmission of COVID-19 to household contacts, we screen patients for the presence of symptomatic household contacts with the intention of referring people who are at the highest risk of acquiring COVID-19 to care.

Figure 2.

Emergency department (ED) clinical navigator script used to assess for worsening symptoms in patients discharged home with suspected coronavirus disease 2019 (COVID-19), Massachusetts General Hospital, 2020.

Results

In its first 43 days of operation, from March 30 through May 12, 2020, the program contacted more than 723 patients who had been discharged from the ED after testing for COVID-19. Twelve percent of the approximately 2000 COVID-19 tests performed on patients ultimately discharged from the ED during this period were positive. We reached 70.4% (n = 509) of patients by telephone. Of the 509 patients, 41.1% (n = 209) reported an HRSN; the most common HRSNs were food, reported by 16.5% (n = 84); maintaining housing or paying rent or mortgage (13.9%, n = 71); and concerns about paying utilities (7.7%, n = 39). Thirty-one (6.1%) patients were referred to ED clinicians for clinical evaluation.

Discussion

Several factors facilitated our institution’s ability to implement this program. Before the COVID-19 pandemic, our ED had an infrastructure of Medicaid ACO navigators familiar with performing HRSN assessments and connecting patients with trusted community resources. During the pandemic, the ED navigators were retasked with performing follow-up of patients discharged from the ED with suspected COVID-19, regardless of whether those patients were part of the Medicaid ACO. This retasking was possible given ED leadership buy-in, the flexibility of the ED navigators, and a decrease in the overall number of Medicaid ACO patients during the COVID-19 surge at our hospital. This program highlights the value of expanding the scope of ED navigators beyond an ACO, particularly during public health emergencies.

In addition, the follow-up program benefited from the excess capacity generated by staffing changes instituted by the hospital in response to the COVID-19 pandemic. For example, resident physicians in our department were redeployed from nonessential rotations to generate back-up capacity for the ED to manage the surge of cases as needed. Given their decreased clinical demands, residents were able to volunteer their time for ED-based COVID-19 initiatives. Furthermore, medical students affiliated with our institution, who were not engaged in in-person training because of COVID-19, provided additional support to the initiative from home.

Another factor that facilitated the rollout of the program was the ability to leverage parallel initiatives. As this project was starting, our institution launched its mobile response program, providing in-home assessment of patients with COVID-19 by a paramedic. We integrated this program into our workflow to provide another avenue for clinical reassessment of patients discharged from our ED with suspected COVID-19 infection. In addition, the Department of Medicine at Massachusetts General Hospital launched a similar follow-up for patients discharged from the inpatient service with COVID-19. To maximize resources, staff members involved in the HRSN and clinical reassessment of discharged patients were shared between the Department of Medicine and ED programs.

In implementing this initiative, we identified several challenges. First, this undertaking was labor intensive and demanding. It required a substantial time commitment, some of which was volunteered, on the part of ED navigators and clinicians. Many telephone calls required the use of an interpreter, which added considerable time to each patient encounter. Often, navigators needed to call multiple times to reach the patient. The excess staffing capacity the program benefited from during the initial surge in cases may not be available in the later stages of a surge or in additional waves of the pandemic if, for example, ED and Medicaid ACO patient volume unrelated to COVID-19 returns to baseline levels or seasonal influenza activity spikes. We are currently examining ways to institutionalize this outreach program to decrease its reliance on volunteer and surplus labor. In addition, because recent data suggest a relatively low rate of false-positive results with SARS-CoV-2 RT-PCR testing, prioritizing patients with a positive test result as opposed to contacting all patients tested could reduce the number of calls. ^24,25 Furthermore, because patients may continue to have positive RT-PCR test results despite recovering from their acute infection, recovered patients can potentially be excluded from follow-up to prioritize patients with current infections, thereby reducing demand. ¹⁹

An additional challenge identified was occasional reluctance on the part of patients to return to the ED if an assessment was made during a call that the patient required prompt in-person evaluation. Concerns included fear of additional exposure to COVID-19 by the patient and his or her family and the additional cost of transport to the ED and the ED visit itself. Expanding home visits may help to mitigate some of these concerns. Finally, although the ED navigators leveraged their experience working with community partners to connect patients with local resources, we did not follow the patients beyond the 8-day follow-up period to assess long-term use of these resources. It may be beneficial in future iterations of this program to integrate leaders of the local community organizations to which patients are most often referred into planning and review meetings.

This intervention had several limitations. First, the results of our intervention are not necessarily generalizable to other hospitals, particularly less resourced facilities that, although serving a population that may most benefit from this type of intervention, do not have the staff members to execute this labor-intensive undertaking and are experiencing substantial revenue loss secondary to the pandemic. ^26,27 Second, we did not survey our target patient population on the acceptability or feasibility of our intervention, and we did not follow up on HRSN outcomes after social service referral; thus, the perceived utility of the program to our patient population is unknown.

We anticipate the need for this program to grow as the pandemic evolves. Facilitating isolation at home through periodic symptom screening and HRSN referral will be important even as the number of cases decreases and businesses reopen to attenuate additional waves of infections and maximally and safely use the outpatient environment to ensure hospital resources are available for the most acutely ill. This public health intervention may be useful for other institutions developing programs to address the heightened social and health needs of patients discharged with suspected COVID-19. Further research will be needed to assess the direct effect of this program on the transmission and clinical outcomes of COVID-19 among vulnerable populations.