The Emergency Room Is Often a Failure of Belonging

What if the most important predictor of an emergency room visit is not a medical emergency at all, but the absence of a reliable place to go before the emergency?

About 70 percent of hospitalizations begin in the ER, yet only 13 percent of ER visits result in hospitalization. That contrast reveals something important. The ER is not merely a gateway to inpatient care. It is also a default response to uncertainty, poor access, untreated mental health needs, low income, weak social support, and the feeling that no other part of the system will reliably respond.

This changes the central question in health care. Instead of asking only, “How do we coordinate care after a patient enters the system?” we should ask, “What would make this person choose a different door in the first place?”

That is the deeper promise of accountable care. It is not simply a payment arrangement, a network, or a set of quality measures. It is a commitment to become responsible for the conditions that produce repeated crises, including the conditions that do not look like medicine until they become impossible to ignore.

The ER Is a Signal, Not Just a Site of Service

Health systems often treat emergency utilization as a volume problem. More visits mean more cost, so the proposed solution is to reduce unnecessary visits through triage, education, or financial incentives. Those tools can help, but they miss the underlying pattern: high ER use is frequently a signal that the rest of the care system is difficult to access, difficult to trust, or poorly matched to the patient’s actual life.

Consider two patients with the same symptom: recurring abdominal pain. One has a primary care clinician who can see her tomorrow, a nurse who returns calls, transportation, and enough health literacy to know what can wait. The other works irregular hours, has no regular clinician, cannot afford a surprise bill, feels anxious about the symptom, and has learned that the ER is open when every other door is closed. On paper, both patients have abdominal pain. Operationally, they inhabit different health systems.

The ER wins in the second case because it offers three things at once: immediate access, diagnostic authority, and psychological certainty. It may not be the most appropriate setting, but it is the most dependable setting available. Telling a person to use primary care instead does not create an alternative. It merely transfers the burden of coordination to someone who may already lack the time, money, confidence, or support to coordinate anything.

The demographic predictors of high ER utilization make this point difficult to dismiss. High utilizers are disproportionately likely to be younger or middle aged adults, have lower incomes, have no college education, and report poor perceived health. Mental health conditions also appear prominently among the predictors, with mental health concerns present in a substantial majority of high utilization patterns reviewed across studies.

These are not signs of irresponsible consumers making irrational choices. They are clues about friction. Every gap between a symptom and a timely response increases the probability that the patient will seek the one service designed to absorb uncertainty at any hour.

When the ER becomes the most reliable part of the health system, the problem is not only overuse. It is misplaced trust.

This does not mean every ER visit is preventable, nor that emergency departments should be judged for treating patients who have nowhere else to turn. It means utilization data should be interpreted as a map of unmet needs. A surge in ER visits may indicate inadequate behavioral health access, unstable housing, inaccessible primary care, untreated chronic disease, or an organization that has never made itself available at the moment patients actually need help.

Accountability Begins Before the Episode

The language of accountable care is often associated with managing a population or coordinating an episode. Both are necessary, but they leave an important question unanswered: Where does accountability begin?

A narrow answer says it begins when a patient is attributed to a clinician, enrolled in a program, admitted to a hospital, or assigned to a care pathway. A more useful answer says it begins when a pattern of risk becomes visible, even if the patient has not yet entered a formal episode of care.

This distinction matters. An episode based model might focus on what happens after a hip fracture, a heart failure admission, or a surgical procedure. A population based model might examine total cost and quality for a defined group over time. But high ER utilization often sits upstream of both. It is a recurring expression of problems that cross formal boundaries: medical and behavioral health, clinic and community, patient and institution.

If accountability begins only after hospitalization, the system is managing the consequences of failure. If it begins when repeated ER use, missed appointments, mental health symptoms, and social instability start to cluster, the system has a chance to change the trajectory.

A useful way to think about this is through the accountability horizon. Every organization has a point beyond which it says, implicitly or explicitly, “That is not ours.” The primary care office may say the patient’s housing problem is outside its role. The behavioral health clinic may say the chest pain belongs in the ER. The hospital may say the patient has been discharged and is now someone else’s responsibility. The insurer may say the patient has been offered a benefit. The patient experiences none of these distinctions. She experiences one continuous life.

Accountable care expands the horizon. It does not require one organization to solve every social problem. It requires organizations to recognize where their decisions create predictable downstream consequences and to build reliable handoffs across those boundaries.

That requires more than good intentions. It requires capabilities.

Five Capabilities That Turn Accountability Into Prevention

A health system can claim responsibility for a population and still behave reactively. The difference lies in whether it can perform five practical functions.

1. Detect patterns before they become labels

The first capability is not simply identifying “frequent flyers,” a term that subtly turns a system pattern into a patient identity. It is recognizing combinations of signals: repeated ER visits, missed primary care appointments, medication gaps, behavioral health diagnoses, unstable contact information, and complaints that seem medically minor but socially urgent.

The goal is not surveillance for its own sake. It is early recognition. A single visit may reveal little. A pattern can reveal a broken pathway.

2. Interpret utilization with context

Data can tell a system that a patient visited the ER six times. It cannot, by itself, explain why. Interpretation requires asking whether the patient lacked a clinician, feared a symptom, experienced panic, needed a prescription refill, had nowhere safe to sleep, or encountered an appointment system that required more flexibility than the patient possessed.

This is where behavioral health and social conditions become central rather than supplemental. A care team that responds to repeated visits with only medical instructions may be treating the visible output while ignoring the engine producing it.

3. Absorb uncertainty somewhere other than the ER

People often go to emergency departments because they cannot tell whether a condition is serious. A system that wants to redirect them must offer a credible alternative for uncertainty: same day appointments, phone or video access, nurse advice, behavioral health response, direct primary care, or a clinician who can say, “We do not yet know, but we will help you figure it out.”

This is a design problem. The alternative must be easier to reach than the ER for the situations it is meant to handle. A pamphlet about appropriate utilization is not an alternative. Neither is a primary care office that schedules three weeks out.

4. Convert handoffs into relationships

A referral is not the same as a connection. Sending a patient to a mental health provider may satisfy a process measure while leaving the patient alone with a phone number, a long wait, and uncertainty about what to say.

A stronger handoff has ownership on both sides. The referring team knows whether the appointment occurred. The receiving team knows why the referral matters. The patient knows who will call if the plan fails. Accountability becomes real when responsibility survives the boundary between organizations.

5. Learn from recurrence instead of merely documenting it

Repeated ER use should trigger a learning loop. What happened after the last visit? Did the patient understand the discharge plan? Could she obtain the medication? Did anyone contact her? Was the diagnosis correct? Did anxiety, loneliness, pain, or transportation make the recommended plan impossible?

The system should treat recurrence as feedback about its own design. A patient who returns is not only someone who needs another intervention. She may be showing the organization exactly where its previous intervention broke.

The Counterintuitive Economics of Being Available

There is a temptation to view access as a cost center. More clinicians answering messages, offering evening appointments, coordinating behavioral health, and following up after discharge appear to require additional expense. But the relevant economic comparison is not between “more service” and “less service.” It is between planned capacity and crisis capacity.

The ER is expensive partly because it must be ready for everything. When routine uncertainty, medication questions, panic symptoms, and fragmented chronic care are routed there, the system pays emergency prices for problems that might have been resolved earlier and more humanely. The result is not only financial waste. It is clinical distortion. Emergency clinicians must work in a setting optimized for acute danger, while patients with complex but nonacute needs receive episodic attention.

A small investment upstream can therefore have a nonlinear effect. A clinician who spends twenty minutes clarifying a treatment plan may prevent a night in the ER. A behavioral health appointment may interrupt a cycle of panic driven by repeated unexplained symptoms. A direct line to primary care may keep a low income worker from choosing between lost wages and delayed treatment.

The key is not to promise that every ER visit can be eliminated. The key is to lower the threshold for appropriate help before the crisis becomes the only visible option.

This also explains why accountable care cannot be reduced to financial risk. Risk may motivate organizations to act, but motivation is not capability. If a system is financially responsible for avoidable utilization but lacks timely access, behavioral health integration, usable data, and trusted relationships, it may respond with denial management rather than better care.

The sequence matters: see the pattern, understand the cause, create an alternative, assign ownership, and learn from what happens next. Payment reform can support that sequence, but it cannot substitute for it.

From “Frequent Flyer” to “Frequent Signal”

The most useful conceptual shift is to stop treating high ER utilization as a fixed trait of certain patients. People are not inherently high utilizers in the way they might be tall or left handed. Utilization is a relationship between a person and a system under particular conditions.

Someone may use the ER repeatedly because the system has not offered a stable place to belong. That word matters. Belonging in health care does not mean friendliness alone. It means the patient knows where to start, expects a response, and trusts that the response will not disappear after one encounter.

A patient with low income and poor perceived health may need more than clinical expertise. She may need an institution willing to make contact in several ways, tolerate missed appointments, coordinate with social services, and distinguish inability from unwillingness. A patient with mental health needs may need a care team that does not force her to translate distress into the language of physical symptoms before anyone responds.

This approach also protects against a common error: assuming that reducing ER utilization is automatically a sign of success. A lower rate could mean better access, but it could also mean patients are delaying care, giving up, or receiving care elsewhere. The meaningful outcome is not fewer visits alone. It is fewer avoidable crises alongside better access, patient trust, health outcomes, and continuity.

The goal is not to make the emergency room harder to enter. The goal is to make the rest of health care dependable enough that the emergency room is no longer the default home for uncertainty.

Key Takeaways

1. Read ER utilization as system feedback. Repeated visits often reveal gaps in access, behavioral health, transportation, medication support, or trust. Ask what the pattern says about the care environment before blaming the patient.

2. Move accountability upstream. Do not wait for hospitalization or a formal episode to begin coordinating care. Use recurring utilization and clustered risk signals to intervene before crisis becomes inevitable.

3. Build a real alternative to the ER. Same day access, responsive phone support, integrated behavioral health, and clear escalation pathways are more effective than education alone.

4. Make handoffs measurable and human. A referral is incomplete until someone knows whether the patient connected, understood the plan, and has a person to contact when the plan fails.

5. Judge success by avoided crisis, not utilization alone. Pair ER metrics with access, continuity, patient experience, mental health outcomes, and evidence that patients are receiving appropriate care elsewhere.

The future of accountable care will not be decided only by which organization accepts financial risk. It will be decided by which organizations accept relational responsibility. The decisive question is whether a patient facing uncertainty encounters a maze of disconnected services or a system that has already noticed, prepared, and made room for her.

An emergency department will always be essential. But it should not have to be the place where the health system first proves that it is willing to help.