Caring for older adults' social needs in emergency departments: Where to draw the line?

Abstract

The popular conception of emergency departments (EDs) is that they primarily care for critically ill patients with sudden illness. It is all gunshot wounds, heart attacks, and sepsis. EDs are exceptionally adept at treating these illnesses, and many emergency clinicians chose their specialty due to an interest in addressing these acute life threats. In reality, ED clinicians and staff also work to address social determinants of health and help patients navigate increasingly complex medical and social care systems. On any given ED shift, one is much more likely to meet a person unable to access primary care due to homelessness, addiction, or social challenges, as to diagnose a heart attack or treat a gunshot wound.

In EDs, social needs are essentially bottomless—to address them all would devastate the ED's ability to complete its core mission. For older adults who are both medically and socially complex, EDs may be expensive and inefficient solutions for unmet care needs. Others have suggested several solutions: creating geriatric certified EDs, embedding pharmacists and physical therapists in the ED,^{1, 2} screening for social determinants of health, and using ED navigators³ to help patients establish care with a clinic or primary care clinician. One currently popular approach, identifying social concerns and referring to external resources, has the advantage of minimal impact on the ED's core mission; however, it is often ineffective for older, vulnerable adults who face barriers to following up on referrals due to cognitive, hearing, visual, and other functional impairments.⁴

In this issue, Southerland et al. demonstrate the effectiveness of one approach to address the social needs of vulnerable elders.⁵ As a result of a unique partnership with the local Office on Aging (OA), they embedded OA case managers within their ED to connect vulnerable older adults to nutrition services, emergency response systems, transportation, and other services, as needed. In this model, OA case managers work with ED social workers to identify community-dwelling older patients and perform in-person intake assessments during daytime hours whereas older patients are in the ED. Case workers arrange needed community services, often starting them immediately.

The advantages of the program are twofold. First, by matching the right professional to the task, the program bypasses several known barriers to hospital-to-community transitions, such as: the ED clinicians forgetting to screen or refer, patients forgetting to reach out, and failure to engage family members.^{6, 7} Second, the program eliminates inefficient communication between older adult patients and OA programs, as when successful service provision relies on the older adult to reach out for care, and to be available and willing to answer the phone when the OA worker calls.

Southerland's results are encouraging. In the seven-month implementation period, program staff performed 252 full assessments, and they frequently identified eligible patients who had not yet been linked to services. In these cases, patients were accepting services and could often be linked to relevant services during the ED visit. Importantly, ED length of stay was not increased and there was no increase in hospital admission rates. By all standards, Southerland's team was successful in achieving their stated goal; they identified and helped vulnerable older adults without negatively impacting other ED care.

The next question is whether this model can be replicated and whether it should be a model of care nationally. Hospital executives will have additional questions that remain to be answered: whether the program is cost-effective, whether it is significantly better than ED referral to traditional services, and whether ED social workers redirected their efforts from other vulnerable patients to provide these services. Further, this program did not reduce the need for prolonged hospitalizations among vulnerable elders (a core hospital executive goal). Hospital executives could also reasonably wonder if this program would encourage others to preferentially transfer the most difficult and vulnerable older patients, often referred to as “dumping,” to their facility.

OA program leaders will wonder whether this program required redirection of resources from other settings, and whether graduate-level case managers were needed, or if the aims could be accomplished by patient navigators with less training and expertise. With expansion in federal community health worker models,⁸ this may be an alternative worth exploring. Finally, professionals will continue to struggle with ethical questions regarding the proper role of clinicians and social support organizations. In this program, the hospital executives decided that OA case managers should not have access to the electronic health record in order to maintain health data privacy. If programs like these continue to identify cases with concern for elder mistreatment, they will need to confront issues about how closely clinicians can partner with organizations that can exercise the power of the state to protect vulnerable citizens from their families and caregivers, or even to punish mistreatment. A future study, comparing this program to alternatives or to usual care, would need to measure outcomes relevant to patients as well as to EDs, hospitals, and social service organizations.

Southerland et al. should be applauded for overcoming many barriers to address the thorniest issues for some of the most vulnerable ED patients. Their model balances well the often-competing priorities of various stakeholders in the provision of health care and social services (see Figure 1). It challenges ED clinicians to identify other areas where social support can be provided even in the midst of busy, sometimes overwhelmed EDs. Answering that question will require creative solutions to clinical, logistical, ethical, and financial questions. Two things are clear: (1) the ED cannot function if it tries to be all things to all people and (2) treating acute life threats without addressing social determinants of health seems to be a wasted opportunity if not a dereliction of duty. Between these extremes, much work remains.

EMG conceived of the idea and wrote the first draft of the manuscript. EB and DML made substantial contributions to the conception and design, and aided in the writing of subsequent versions of the manuscript. All authors have approved the final manuscript to be published.

The authors declare no conflicts of interest.

Research support was provided by the National Institute on Aging (K76 AG059983, PI Goldberg). This grant was provided after undergoing peer review by the National Institutes of Health. The content is solely the responsibility of the authors and does not reflect the official views of the National Institutes of Health.

This work was supported by National Institute on Aging, K76 AG059983.