Wednesday, April 30, 2014

An unleashed demand for health care services

It slipped by so fast that I almost missed it.  It wasn't, after all, the main point of this news report by NPRs John Ydstie on All Things Considered.  It was on the economy as a wholeHere's the audio recording.

Listen at minute at 2:57 and hear the story of the one growth area of the economy:

It looks as thought the advent of Obamacare and the provision of insurance to people who previously didn't have it has unleashed demand for health care services that simply wasn't there before.

In fact, spending on health care grew in the first three months of the year at an astonishing annual rate of nearly 10%.

This is no surprise to many of us.  As we saw in Massachusetts when universal access was provided, if you give people health insurance, they will use it.  In fact, at heart, that's why we want to give them insurance.  In the past, people without health insurance would avoid important visits to the doctor, or important visits to the hospital, or important consumption of drugs.

There are some who might argue that the fee-for-service rate design by which many providers get paid accounts for this.  But, no.  There is actually less fee-for-service care than there was in the past, as a number of insurers and providers have moved to more risk-based payment plans.

There are some who might argue that consolidation in the health care industry accounts for upward price pressure.  While that should be a concern over the coming years, I don't believe that it has yet had the full effect that it will.

No, what's happening now is much more simple.  People are going to the doctor sooner when they have symptoms.  Some portion of those people are sick and need more extensive care.  In the past, this group would not have gone to the doctor, and they would not have gotten the extensive care.  Some percentage would have gotten sicker and sicker at home and eventually would have shown up in emergency departments.  Without the full course of treatment they are now getting as insured patients, we would have "saved" a lot of money in treatment. Some, though, would have died prematurely or suffered from extended morbidity.

Others are having regular preventative diagnostic tests, like mammograms.  Early detection of breast cancer results in earlier treatment--radiation, surgery, chemotherapy--adding to the nation's health care bill.  (We saw this in Massachusetts.)  In the past, their disease would not have been noticed until it was too late, again "saving" a lot of money in treatment.

The human cost under the old regime was unconscionable, and we have done the correct thing to give people access to insurance and care.  The dollar cost, though, had to increase.

The problem here is not the result.  The problem was that the Administration, as part of its political strategy to get the legislation passed, glossed over this fact.  In fact, the promise was that access to insurance would be paired with lower costs.  The Administration's plan, all along, was that these costs would be covered by the imposition of a value-added tax.  But the political environment has changed, and the chance of such a tax is very low.  So the tension between access and costs will grow.  Over time, choices will be made.

In the short run, the solution imposed will be to charge consumers directly for a greater portion of their health care costs, with high deductible health plans (HDHPs).  This direction creates important issues and can interfere with our achievement of several of the Institute of Medicine's six aims of care--safe, effective, patient-centered, timely, efficient and equitable.  As the American Academy of Pediatrics notes in this policy statement:

Because HDHPs require out-of-pocket payment in the initial stages of care, primary care and other outpatient services as well as elective procedures are the services most affected, whereas higher-cost services in the health care system, incurred after the deductible is met, are unaffected. HDHPs promote adverse selection because healthier and wealthier patients tend to opt out of conventional plans in favor of HDHPs. Because the ill pay more than the healthy under HDHPs, families with children with special health care needs bear an increased cost burden in this model. HDHPs discourage use of nonpreventive primary care and thus are at odds with most recommendations for improving the organization of health care, which focus on strengthening primary care.


Aditya Pathak said...

Paul, has the Mass cost curve stabilized at a higher level, or was it a spike ?

Paul Levy said...

Not sure. CHIA would know. Will try to get from them.

Paul Levy said...

Actually, you can look here and draw your own conclusions:

Paul Levy said...

Looks to me pretty stabilized at the higher level, i.e, not a spike, a permanent change after the post-legislation uptick (as a percentage of GSP.)

Peter Yang said...

I'm interested in seeing whether that initial spike helps to curtail additional spending in the future. That is, can we count these expenses as investments in preventative care, or are we merely seeing a moral hazard increase in healthcare utilization without it reducing future care needs?

Barry Carol said...

While the data is not as robust as it could be, there appears to be some pent-up demand for healthcare services among people who age into Medicare, especially if they were uninsured or underinsured previously. At the same time, according to the CBO’s Monthly Budget Review, Medicare spending for the first six months of fiscal 2014 increased only 0.6% despite at least a 2% increase in the number of beneficiaries.

Also, so far in the first quarter, the large publicly traded health insurance companies did not experience a meaningful spike in healthcare utilization though there is an issue regarding the high cost of the new hepatitis C drug, Sovaldi, recently introduced by Gilead Sciences. They generally reaffirmed their estimate of medical cost trend growth for 2014.

To the extent that newly insured people are getting care they need but previously avoided due to lack of insurance or high out-of-pocket costs, that’s a good thing and may quickly run its course once the pent-up demand is satisfied. On the other hand, one positive aspect of HDHP’s is that help to create an increasing demand for robust price and quality transparency tools so people can make more informed choices as both patients and referring doctors can more easily identify the most cost-effective high quality providers in real time. We still have a long way to go in this area though.

As services become an increasing percentage of U.S. GDP at the expense of manufactured goods and the number of employees working for large companies shrinks as a percentage of the total workforce, I’ve become increasingly skeptical of data coming out of the Department of Commerce. I remain an optimist about the prospects for slowing growth of healthcare costs even as more people get health insurance.

Anonymous said...

Your comments are good ones (minus a minor quibble about mammograms, which studies have shown to be less effective as a screening tool than we thought). As Barry Carol once observed, people who die earlier are a smaller expense for any health care system. However, let us not forget that there are big savings out there in the delivery of care more efficiently and the cessation of overused or ineffective care, such as proton beam radiation and robotic surgery for prostate cancer. As your previous post pointed out, the problem is how to persuade providers to stop concentrating on these profit-makers and learn how to deliver care more cheaply.

nonlocal MD

Paul Levy said...

Thanks, Peter,

I've never seen any evidence that there is the kind of effect you describe, that these are investments that reduce long-term expenses.

Paul Levy said...

Thanks, Barry,

The theory suggests that what you hope for could happen, i.e., by affecting factor prices as people go shopping for the deductible portion of their care. But, as we know, a lot depends on the particular configuration of the market. For example, the point I make about the influence of concentration in the marketplace--and the higher prices that can result from that--might offset the potential gains from "shopping around."

Also, there might be an odd bounceback: Once I pay my deductible I REALLY try to get my money's worth out of my insurance company by becoming totally price insensitive for additional services beyond that point.

In short, there are lots of moving parts here, and it is hard to know which will prevail more than the others.

Paul Levy said...

Thanks, nonlocal. I didn't mean to suggest that mammograms make a generalized difference in outcomes over time. Indeed, as I read the studies, they suggest that they often lead to overtreatment. Nonetheless, I don't yet see that their use is likely to diminish among people with insurance. (In fact, doesn't the ACA demand that they be included as a nondeductible/non-copay component of preventative care?)

Paul Levy said...

Ah yes, "Under the Affordable Care Act, women’s preventive health care – such as mammograms, screenings for cervical cancer, prenatal care, and other services – generally must be covered by health plans with no cost sharing."

Barry Carol said...

I’ve read recently that doctors, especially in the fields of oncology and cardiology, are starting to view it as part of their job to be wise stewards of society’s limited resources by providing cost-conscious and cost-effective care. While their primary focus will and should remain doing the best they can for their patients, there is a lot of room for reducing the use of futile or low value care, doing a better job of providing an honest diagnosis and fully explaining side effects and the quality of life implications of various cancer drugs, and stop holding out false hope as often as they have in the past. We could probably also do a far better job of getting more people, especially the elderly, to execute living wills and advance directives and store that information on registries so it’s available to providers and family members when needed.

On preventive care, there is confusion around the potential for cost savings. High value preventive care could save money for insurers and taxpayers in the short term. However, if patients live longer lives and many of them have dementia or Alzheimer’s at the end, both kill very slowly and leave patients needing lots of expensive custodial care. To the extent that good preventive care helps people to live longer, healthier and more productive lives, that’s great. However, the downside is likely to be higher costs over the long term for the healthcare system.

Separately, my PCP tells me that people don’t need to get colonoscopies or mammograms after age 75 unless they have a history personally or within their family of colon or breast cancer. My wife reports that her doctor says that he has quite a few patients coming in for mammograms well into their 80’s, including people with Alzheimer’s or dementia. As a society, why are we paying for this when resources are finite and there are lots of other worthwhile priorities both in the public and private sectors?

Anonymous said...

Well, the ACA was passed before a lot of these studies were publicized. The same could be said for any preventive procedure which is later found to be obsolete. I do think the use of mammograms will decline with this knowledge, despite the desperate attempts of its financial stakeholders (like the American College of Radiology) to sustain them. I myself stopped getting them some time ago.

You make an interesting point in your comment about your behavior once your deductible is satisfied. I think a lot of people feel that way about various types of insurance (auto, home, etc.) But unless you have 100% coverage after the deductible there is still an incentive to keep your costs down.


Josh said...

Has anybody considered allowing balance billing for high-cost services like robotics or proton beam? It might make the provider think twice before offering this, knowing they could be subjecting their patient to some of the "real costs" of these treatments with little to no additional benefit.

Paul Levy said...

Balance billing is a political third rail. No legislative body would permit it to happen.

Paul Levy said...

BTW, another good story here:

Paul Levy said...

Here's an excerpt from that post:

This all said, a few more things must be said about this 9.9% annual increase in health utilization figure:

--if it holds up, then it shows that in fact there are many people who obtained health insurance right away in 2014, which contradicts all the nay-sayers who have been saying that the ACA has not led to an increase in coverage;
--this of course implies that advocates for the ACA, who have been predicting (and hoping) that many of the uninsured gained coverage must also accept or even embrace results like this;
--either one side is right or the other! (there was no net increase in coverage, but if so, then these numbers will be revised downward; or there was a big increase in coverage (with the resulting increase in demand/use of medical care).
--my guess? I expect the number will eventually be revised downward, to perhaps 5-7% annually once Commerce has some private plan data. And that is just a hunch, probably informed as much based on prior estimates from CMS, and recent work by Altarum and others, and the fact that ACA enrollment has been fairly robust.

Finally, if these results indicate that many of the newly-insured ran out right away and obtained medical care (a pent-up demand effect), then it could also be explained not just by a moral hazard effect (price of medical care falls, use rises), but of evidence of adverse selection among the previously uninsured and those first to obtain coverage (which is of course possible, because evidence exists that they were the first to sign up for the Marketplaces and Medicaid).

Anonymous said...

Paul - One key driver that United Health mentioned in their first quarter report was the cost of Gilead's Sovaldi Hepatitis C drug that runs about $80,000 per patient. United singled out the cost of this drug treatment as a reason for a higher than expected medical costs in the first quarter of 2014. I imagine this is also the case with other insurers as well as Medicare and Medicaid.

dennis byron said...

The way I read the actual BEA report (as opposed to the links which gives opinions about the report, maybe after reading it or not), I believe this healthcare spending estimate is totally modeled. It may be based on some Jan/Fed data that is not specified but the BEA report specifically says it does not reflect the March hockey stick in PPACA-related insurance enrollments.

It seems totally against human nature that what we know to be a very small percentage of the population (under 1%) even under the rosiest Obama-administration scenario for "previously insured vs. not previously insured" and "paid premium vs. did not pay premium," received insurance on Jan 1 or Feb 1 for the first time (or for the first time in a while) and rushed out to the doctors to the extent that it would drive spending like that. You can't even get an appointment that fast if you have not previously had a doctor.

So either something else drove the spending or the model is wrong (maybe so as to not show that the economy deflated?).

akhan13 said...

We keep going back to the wrong issue- FFS vs capitation! Either way we work out what the 'right' amount to pay docs should be, it is just realized slightly differently. There is much less difference between these systems than people think! What would be a change is changing the training time for docs which leads to the massive salary expectations- not by reducing residency (which would compromise skill) but by changing it to work instead of training. Pay every resident 150-200k (depending on specialty), and now they see their training time as 4 years of med school, not as 10 years total. For the cost of a few hundred thousand, we will increase the talent pool able to join the ranks of physicians AND be able to pay a bit less over 30-40 years of practice, which will far exceed the up front investment.

dennis byron said...

On May 3, in response to a lot of wondering and glad-handing above by the blog post's author and among commenters about the supposed large upswing in health spending in the first quarter of 2014, I wrote commented

"The way I read the actual BEA report... (instead of all the liberal comments about the BEA report), I believe this healthcare spending estimate is totally modeled. (The supposed growth did not make sense for a lot of reasons, in my opinion, for reasons I stated) So either something else drove the spending or the model is wrong (maybe so as to not show that the economy deflated?)."

And the answer is...

... the model was wrong, I guess so as not to show that the economy deflated (hoping that subsequent revisions would save the day). And it turns out the biggest problem with the model is that healthcare spending did not grow. There was no pent-up demand. There was no rush for life-saving (or meaningless) mammograms and other preventive tests. There was no surge in co-pays or balance billing. There was no 6% growth instead of 10% growth. Health care spending deflated (See