America's ER Crisis

Although typically Americans have greater and more rapid access to surgical procedures than people in other countries, we do not possess a uniform superiority in the speed of health care access. One excellent example of this is visiting the Emergency Room. ER wait times have been increasing steadily over the last decade as indicated by Wilper et al. publishing in the journal Health Affairs.

Wilper et al. performed the best and most comprehensive analysis to date of wait-time in ERs around the country. The looked at wait-times from 1997 to 2004 using the National Hospital Ambulatory Medical Care Survey (NHAMCS) database.

What they found was that wait times have been slowly increasing over the last decade.

This data is shown below (Figure 2 from the paper, Click to enlarge):

The chart shows the median wait-time for all adults admitted to the ER went from 22 minutes in 1997 to 30 minutes in 2004. Similarly, patients with Acute Myocardial Infarction (AMI) -- otherwise known as a heart attack -- saw their wait-times increase from 8 to 20 minutes over that trial period. The "emergent" group -- or patients classified in triage as must-treat ASAP -- also showed increases in wait-time from 10 to 14 minutes.

Frankly, I find this data very disturbing. When you are dealing with AMI, you need to get that patient up to interventional radiology for stenting as soon as possible. This is not a condition where you have time to fuck around. Heart muscle is dying while you wait.

The authors performed further analyses on the data to determine what predicted longer wait-times. Unsurprisingly, urban hospitals had much longer wait-times then nonurban hospitals. (This isn't surprising because they generally have much busier ERs. They accounted for about 80% of the ER admits during this period.) Hispanics, blacks and women also had longer wait-times than whites and men. This is explained in part by the fact that they are much more likely to go to urban ERs, but the story is probably more complicated. (The racial difference, for example, was only present for all adults admitted to the ER. The difference vanishes in emergent cases and for patients with AMI, suggesting that in severe cases race may not be a factor.)

Fortunately method of payment does not seem to affect wait-time. Patients with private insurance end up waiting about the same amount as those with Medicare and Medicaid. The only financial predictor of longer wait-times were those patients listed as "no charge" -- presumably pro bono cases although I don't understand how that is determined. (Again the difference in "no charge" patients vanishes for the AMI and emergent groups. It is only present in all adults.)

How do we interpret this data?

The most important question is why are wait-times increasing, and I think we can attribute this change largely to the increase in ER volume over the last decade. The authors note that while ER visits have increased over the last decade the number of ER has actually decreased due to closure:

Between 1997 and 2004, the median ED wait time increased 36 percent. Patients needing emergent attention waited 40 percent longer, while median waits for AMI patients increased 150 percent. A variety of factors probably contributed to longer waits. Most important was likely crowding as a result of ED closures and an increase in total ED visits; between 1994 and 2004 the number of ED visits increased from 93.4 million to 110.2 million annually, while the number of EDs fell by as much as 12.4 percent. Other likely contributors include inpatient bed shortages leading to bottlenecks in the ED; increasing uninsurance; population aging; shortages of staffing, space, and interpreters; and difficulties assuring non-ED follow-up care. (Citations removed. Emphasis mine.)

You have to ask why more and more people are going to the ER.

Aside from population growth, part of the answer is that we have large numbers of uninsured individuals in this country. Because of that large numbers of people -- particularly in urban settings -- are seeking day-to-day care in the ER as opposed to with a primary-care physician. This statement is supported by the fact that the authors observed an overall decline in the proportion of emergent visits to the ER from 26.9 percent to 15.2 percent during the survey period. This means a lot of people are coming to the ER for non-emergency reasons. (The survey also indicates that the uninsured (self-paying) wait about as long as the insured, suggesting that an increase in the volume of uninsured would cause an across the board increase in wait-time.)

Now clearly there is a quality of care issue here. We need to lower this wait-times across the board, or people are going to die. However, we also need to be very careful about how we address the question of increased ER volume, so that we don't shoot ourselves in the foot.

For example, the recognition that uninsured individuals going to the ER for non-emergent reasons is one cause of the increase might cause some to advocate for wider government funding of health care. However, while this might be effective at limiting the numbers of uninsured, it would not in my opinion solve the underlying problem of ER closing: namely that running an ER is not a very good business.

Part of the financial reason that ERs are closing is that the reimbursement from Medicare and Medicaid patients -- patients the ERs are legally required to treat regardless of chief complaint -- does not cover what it actually costs to treat them. This is one reason for ERs closing up: it has become really hard to make ends meet. If you want to expand government coverage of individuals that would be great, but if you don't increase the ER reimbursement it still means that ERs are going to go out of business and continue to limit the overall supply of care. You could respond that giving people medical insurance will encourage them to go to the primary care as opposed to the ER, and you would probably be right in degrees. However, health care utilization is a complex issue, and giving people free care doesn't always imply that they will use it.

This is why, I think, alternative solutions might be much more effective for this problem. I would recommend increasing the Medicare and Medicaid reimbursement and possibly subsidizing opening new emergency rooms. This will solve the supply-side of the problem, and lead to lower wait-times. It does not solve the underlying uninsured problem in this country, but it will I think address the quality of care problem. You could also do what some analysts have suggested in opening up ER visits to more competition in the market. This would add financial incentives to increase both the quality and the quantity of the care provided. Finally, you could allow ER doctors to refuse treatment to clearly non-emergent cases. This would help remove the "coming to get an aspirin at 3 am" problem, but you also have an issue of liability in identifying whether a patient is truly ill or not ill.

The authors suggest that the problem should be countered in multiple ways, and I think they make reasonable suggestions:

Reversing the trend of longer ED waits would likely require multiple reforms. Possible interventions include expanding insurance coverage and access to primary care resources to increase alternatives to ED visits; directing hospital resources to medical need (for example, the expansion of ED resources) instead of profitable but unnecessary services; increasing available ED space, staff, and specialty consultation services; and modified management of elective surgery scheduling.

This paper was covered on NPR.

Wilper, A.P., Woolhandler, S., Lasser, K.E., McCormick, D., Cutrona, S.L., Bor, D.H., Himmelstein, D.U. (2008). Waits To See An Emergency Department Physician: U.S. Trends And Predictors, 1997-2004. Health Affairs DOI: 10.1377/hlthaff.27.2.w84