USA Today ran two pieces the other day about the crisis in ERs. An editorial told an alarming story about a toddler having a seizure. The ambulance was diverted from a nearby but too-crowded ER where the family's doctor was waiting. The piece called for changes in how ERs are run. The other was an opinion piece written by the head of a leading hospital organization saying the way to solve the ER crisis is to cover the 47 million uninsured.

Both are right. But even if we followed all their recommendations, we'd still have a missing piece—adequate primary care in the United States.

The story of 23-month old Bella Nannini and her seizures, (she has insurance, by the way) is not a rare occurrence. Thankfully she was not harmed. Not everyone is so lucky. In Houston and Los Angeles, for example, the average hospital waves away or "diverts" ambulances away about one-quarter of the time. Nationally one ambulance is diverted from an emergency room a minute. Diversion is not always benign.

In Houston, Texas, the chance of dying from severe trauma is increased on days when the trauma centers are both on diversion for more than 8 hours. In New York City, the odds of death from a heart attack jump when hospitals are diverting ambulances. Patients with life-threatening emergencies of many kinds (stroke, heart attacks, brain injury, for instance) are at risk in any community where ambulances divert frequently. A delay in receiving definitive care in the critical minutes after a medical emergency can mean the difference between life and death.

The cause of ambulance diversion is overcrowding of emergency rooms which is ubiquitous in the United States. ERs become overcrowded for two reasons. First, patients seek care for conditions that could be treated in regular clinics. Second, full hospital beds, which means that really sick patients are "boarded" in emergency rooms until a bed can be found in an intensive care unit or a regular hospital floor.

The editorial included recent recommendations by the American College of Emergency Physicians, who say that hospitals can serve the best interests of the patients by changing bed usage patterns, for instance by such steps as scheduling non-emergency surgeries throughout the week (not just Monday, Tuesday and Wednesday—preferred in part because more patients can go home by the weekend) and moving patients to wait somewhere in the hospital other than the emergency room itself.

Rich Umbdenstock, president of the American Hospital Association, said the main solution will be found in covering the 47 million uninsured Americans. (And we share that goal, for many reasons.)

The ER doctors propose an immediate solution, and the hospitals a long-term solution. Boston hospitals have relieved ER crowding by changing surgery schedules. Poor management of hospital beds is very common. For example, doctors often discharge patients at the end of the day when they make their rounds, even if a patient was ready hours earlier. Hospitals are remarkably inefficient and the ER doctors point that out.

But, hospital efficiencies will not solve the whole ER problem. One reason that there are not enough hospital beds for an emergency is because there are so many uninsured; the hospitals are focused on solving this problem. Hospitals are required by federal law to admit anyone with an emergency regardless of insurance status if the hospital has the capacity (an empty bed). In regions where there is a high percentage of uninsured, hospitals restrict the number of beds available for emergency care for everyone in order to reduce their losses from admitting the uninsured. That is, in regions with many uninsured, hospitals lose so much money on the care of uninsured emergency patients that they cannot recoup it from the admission of insured emergency patients. So they cut their losses by eliminating beds, even below the capacity a community would need in an emergency. The ER problem is no respecter of personal wealth, education, or insurance status.

There is, however, a third problem that neither the hospitals nor the ER doctors are discussing and it is a big one. We don't have enough primary care in the United States. While it is true, as the editorial points out that only about 14 percent of ER patients did not need care within 24 hours, the story is not that simple. Dr. John Billings has studied patients that visit New York hospitals and found that a whopping 40 percent did not need care within 12 hours, another third could have been readily treated in a clinic, and seven percent would likely never have gone to an emergency room if they had received primary care in a doctor's office. The conclusion is that only about 20 percent of ER patients need the kind of care that can only be delivered in an ER, if our healthcare system were able to provide good primary care to everyone.

The ERs have become the meeting place for three of the great failures of the US medical system—too many uninsured, inefficient hospitals, and too little primary care. ERs will remain overburdened, often dangerous, and frequently inaccessible until all three of these ills are addressed.