ER Bills Have No Relationship To The Actual Cost Of Services

Sarah Kliff at the Washington Post reviewed an NIH-funded study that looked at medical bills for common emergency room visits: 

They found, essentially, two things. First, huge variation in prices: Bills sent out for sprained ankles ranged from $4 to $24,110. Second, overall, really high prices: The average emergency room visit now costs 40 percent more than a month's rent.
Researchers used data from the Medical Expenditures Panel Survey, which looks at how much American families spend on, well, medical expenditures. They looked just at what they found to be the 10 most common reasons listed for trips to the emergency room — designated by billing codes — and how much each one cost. The chart below shows a huge amount of variation.

A headache could cost $15 — or $17,797. There was a difference of more than  $70,000 between the most and least costly treatments for a urinary tract infection.
Pulling all that data together, researchers found that the average charge for an emergency room trip for all these conditions came out to $1,233, which is 40 percent higher than the average American rent right now, $871 per month.
These costs could go down for an individual with insurance; as lead author Renee Hsia explained to me over e-mail. "We looked at charges, meaning what patients were billed.  We did not look at the amounts insurers paid, or what individuals paid," Hsia said.
...Stil, Hsia wanted to show the sticker price for the emergency services.
"We feel that while charges are usually much higher than what insurers pay, for uninsured patients, or for patients who receive care in an out-of-network hospital (like one of my friends, which really started my whole interest in this area a few years ago!)," Hsia wrote, "they actually are faced with the sticker prices."

Is moral hazard bigger under Medicaid or private insurance?

UPDATE: I posted an updated version of this article at my Medianism blog.
Some health professionals perceive greater moral hazard among Medicaid recipients than among the rest of us.  All the research I have been able to find about the matter indicates that Medicaid does increase health expenditures, but by less than private insurance does.  One study in the Journal of Health Economics found no moral hazard in nursing home care when Medicaid is more generous than when it is stingy. Another study in Inquiry, found that health care spending is significantly lower under Medicaid than under private insurance due to lower provider payment rates and found little difference in services used. A paper in the American Journal of Public Health found that children on Medicaid have more physician visits than uninsured children, but less than privately insured children. 
Evidence in a recent HSC Research Brief suggests that there is greater moral hazard among Medicaid recipients in their use of Emergency Room (ER) vists.  The only category of care in the figure below that is likely to indicate moral hazard is non-urgent care because this does not require care within 2 hours and so is not worthy of being called an emergency.  But even non-urgent care is not always wasteful.  I have been forced to go to an ER when my child needed antibiotics on a Friday night (my kids always seem to get sick and have accidents on weekends and evenings) and there was no alternative until Monday morning, 60 long hours later.  Medicaid recipients are more concentrated in places with less access to convenient doctor office hours and that could explain some of these results. The US healthcare system does rely excessively on ER care, but that is not due to the moral hazard of insurance.  It is because our system has a shortage of primary-care physicians (insufficient office hours) and too many uninsured people who lack access to care except in an ER. 
The big moral hazard headline in the data below is that Medicaid recipients use 45% more non-urgent care than people with private insurance! Forty-five percent! 

Forty-five percent sounds like a huge percentage increase in moral hazard, but compared to the total amount of ER care, it is insignificant.  Medicaid recipients only get 3 percentage points more non-urgent care than people with private insurance so if all the extra non-urgent care were eliminated, there would only be 3% fewer ER visits for Medicare patients.  And non-urgent care is the cheapest care so the dollars lost are probably much less than one percent of total ER expenditure.  A few expensive urgent-care patients cost almost all of the money.  In healthcare, 20% of patients typically account for 80% of the spending and non-urgent care is the very cheapest kind of care. 

But the number of non-urgent patients sometimes looks overwhelming if you look around in any busy ER.  That is because the non-urgent care is triaged, so those patients have to wait.  A busy ER will typically have many non-urgent patients impatiently waiting for care until the ER doctors have the free time to deal with them.  That will make them much more visible than the urgent care that gets dealt with immediately.  Even though non-urgent patients are less than 10% of the people coming into an ER, they might make up more than 50% of the patients who are present in the ER because they have to wait longer, sometimes many hours (in my own experience).  

Many economists think that moral hazard on the part of patients produces most of the expensive waste in the US healthcare system, but I have never seen statistical evidence that moral hazard are a significant problem compared with high administrative costs or over-treatment by providers.  This data seems to confirm that the moral hazard of Medicaid is not particularly important in the big picture.  The use of ER care for Medicaid recipients and people with private insurance is broadly similar. Neither group uses much frivolous, non-urgent care compared with the total volume of care according to this data and in dollars spent, it is trivial.  And remember, much of the non-urgent care is not frivolous.  Much is useful care that happens to be outside of office hours.  It isn't completely clear if any of this is really moral hazard at all, but even if all non-urgent care were moral hazard, it would not be anywhere near the biggest source of waste in our ER system.

Economists like Mark Pauly who focus on 'moral' hazard are usually using a warped ethical system which defines morality according to ability to pay.  According to this version of the theory, insurance should never cause anyone to spend more on healthcare than they would spend without insurance.  Any expenditures that a patient would not pay out of pocket is moral hazard.  Under this definition of moral hazard, poor people are always more guilty of moral hazard than wealthier people because poor people have lower ability to pay than wealthy people.  But the whole point of health insurance is to increase your ability to pay if you get a mortal illness and need more money to survive.  People who are wealthy enough to pay for any possible health problem out-of-pocket have little need for insurance.  Insurance is the most useful when the ability to pay for healthcare could make a life-or-death difference.  This warped morality underlying 'moral' hazard explains why economists like Mark Pauly see greater moral hazard in insurance for poor people like Medicaid than in insurance for the rest of us.  Under the 'moral' hazard view, elites like Bill Gates should be able to get whatever they can afford, (whether insurance pays or not) but healthcare should be rationed for the middle classes and healthcare for the poorest people is pure moral hazard that should be eliminated.  According to the 'moral' hazard view, a billionaire alcoholic whose habit burns through liver transplants at a black-market boutique hospital is exercising his consumer sovereignty and efficiently maximizing consumer surplus whereas a homeless guy who gets a lifesaving $10 antibiotic at an ER is squandering society's resources because his life is worth less than $10 if he isn't willing to pay that much to save it.  

If there is more actual wasteful treatment among Medicaid recipients, (as opposed to 'moral' hazard), I have yet to see it measured in any statistical work.  There is wasteful treatment for people with every kind of insurance and I don't know why providers would want to do more for Medicaid recipients than for the rest of us. 

Set The Nurses Free

Licensure raises costs which can only be justified if it also raises quality.  However, the evidence makes it clear that licensure restrictions on nurse practitioners is too strong.  If we set the nurses free, we can get lower costs.  Yglesias:

In 18 states (disproportionately but not exclusively rural ones) and the District of Columbia, a nurse practitioner can examine, diagnose, and treat patients in a primary care context. Nurse practitioners also get paid less than doctors. Medicare reimburses them at 85 percent of the doctors' rate, for example, but they also charge less to insurance companies and out-of-pocket patients. And since the "blue states" in this sense are a pretty diverse lot, we can get pretty good quasi-experimental data as to whether cheaper nurse practitioners are actually any worse than primary care doctors in this regard. The answer is a resounding no:

There is a growing body of research demonstrating that patients perceive that receiving primary care and having a usual source of care is more important than who it was that provided these services. Studies comparing the quality of care provided by physicians and nurse practitioners have found that clinical outcomes are similar. For example, a systematic review of 26 studies published since 2000 found that health status, treatment practices, and prescribing behavior were consistent between nurse practitioners and physicians.

What's more, patients seeing nurse practitioners were also found to have higher levels of satisfaction with their care. Studies found that nurse practitioners do better than physicians on measures related to patient follow up; time spent in consultations; and provision of screening, assessment, and counseling services. The patient-centered nature of nurse practitioner training, which often includes care coordination and sensitivity to the impact on health of social and cultural factors, such as environment and family situation, makes nurse practitioners particularly well prepared for and interested in providing primary care.

It isn't surprising that nurse practitioners get higher quality ratings from patients than physicians do.  Patients typically judge quality according to bedside manner and medical schools don't select students based on their warm-and-fuzziness!  They want the academic elites of society.  In contrast, nursing programs have a long tradition of seeking to provide comfort.  The very word 'nurse' means to take care of. 

Why Is Disability Expenditure Increasing?

NPR's Planet Money did a story about how disability rolls have been increasing and linking it to declining welfare rolls and rising unemployment.  There is some truth to this, but the bigger story is that demographics have had a bigger impact on the following graph.  NPR did not even adjust the graph for population growth nor our aging population which accounts for much of the increase. 

 
If you look at the red line on the graph in Figure 1 below, it doesn't look like the recession of 2008 caused disability rolls to spike upwards. Brad Plumer:

Three big demographic factors are at work [increasing disability rolls]. First, as the U.S. population ages, people are more likely to become physically disabled, especially if they’re working in manual labor. Second, Congress has hiked the retirement age for Social Security over the years — in fact, about 5 percent of current disability recipients would simply be on ordinary Social Security had Congress not changed the rules.
Third, many more women have been qualifying for disability insurance since the 1980s because there are many more women in the workforce, period.
If one controls for those three demographic factors, the rise looks somewhat more gradual, with the percentage of workers on disability going from 3.5 percent in 1995 to 4.6 percent in 2012:

And these three factors won’t continue forever: Ruffing writes that the disability program is currently at “its peak demographic stress,” with the rolls expected to shrink in the coming years as many current beneficiaries age into Social Security and Medicare.
That said, demographics aren’t the whole story. The program is still growing even after controlling for age and gender dynamics. In her recent testimony to Congress, Ruffing runs through some of the other possible factors. For instance, the changing economy has made it harder for older workers with less education to find jobs. (States with lower education levels tend to have far more workers on disability.) And the surging cost of health insurance over the years has pushed more people with ailments to seek out federal disability insurance in order to get care.
Interestingly, however, Ruffing is skeptical that the current recession created a surge of beneficiaries. “[E]conomists generally find that while a sour economy significantly boosts applications to the program, it has a much smaller effect on awards,” she notes. ”The implication is that economic downturns tend to attract more marginal, partially disabled applicants, but their applications are more likely to be denied.”
Now, that still leaves a pressing ...issue here. The disability insurance program is underfunded and is expected to exhaust its trust fund by 2016. At that point, benefits will get cut sharply unless Congress makes alterations. ...Ruffing, ...offers some recommendations for fixing disability insurance here.

 Brad Plumer has another graph:

Pricing Everything and Value of ???

An undergraduate student, Jaime Rosenthal, spent her summer working for some University of Iowa professors and published her summer research in JAMA.  She called 102 hospitals, usually multiple times, to request pricing information about hip surgery for her uninsured grandmother (ficticious).  Only about half could come up with any price at all and the prices varied hugely. 

“Among both top-ranked and non-top-ranked hospitals, total price estimates ranged from $10,000 to well over $100,000; for reference, available data suggest that Medicare and other large payers frequently pay between $10,000 and $25,000 for primary joint replacement surgery.”

Sara Kliff notes that medical prices are usually nontransparent and vary widely:

Rosenthal’s findings are not exactly an anomaly in medical research: A study published in the Archives of Internal Medicine in the spring found even greater variation in prices for an appendectomy. Depending on where it was performed, the very simple procedure could cost anywhere from as little as $1,529 to as much as $182,955.

Moneybox notes that the health care industry is similar to other industries with price discrimination like hotels (which a hospital resembles) and airlines:

[The health care industry lacks price transparency and has wide variance in prices.] I often hear this discussed by health care wonks as driven by unique attributes of the health care industry... But nontransparent pricing is a fairly common feature of economic life outside the realm of standardized commodities traded in highly competitive markets. Look at the range of different prices residents of any given city are paying for identical cable television packages at any given time. Or look at airline pricing, where all kinds of crazy stuff happens. Connecting flights are sometimes cheaper than direct ones despite higher costs, how many days you stay is often a factor, one-way and round-trip price differentials are wacky, nobody understands the alchemy behind when you can redeem frequent flier miles, etc.

Or perhaps the most directly relevant comparison would be hotels, since hospitals and hotels feature the same basic dynamic that an empty room is pure waste. Hotel revenue management is a very complicated issue that you can study at Cornell or hire consultants for and is undertaken with the assistance of complicated computer models. A small, charming B&B may have a pricing scheme that's as simple as a two-by-two grid assessing whether it's a weekend or a weekday and whether it's peak or off-peak season. But a sophisticated hospitality firm is going to have a pricing formula that's extremely hard to summarize, and there's no reason to expect hospitals to be any different from hotels in this regard.