Thursday, September 18, 2014

CMS Releases Selected Data from ACO Year 2 Results.

Today’s Managing Health Care Costs

Number is $372 million

CMS reported Tuesday on the financial and quality results of the second year of the Pioneer ACOs and the financial results of the first year of the shared savings ACOs.    The press release doesn’t make it easy to figure out exactly what happened. It’s headlined “Medicare ACOs continue to succeed in improving care, lowering cost growth.”  The press release claims a total of $372 million in savings after accounting for $445 in shared savings payments and overages from some ACOs.

The quality results appear good – with Pioneer ACOs improving in 28 of 32 quality measures and 14.2% across all quality measures.  Still,  many of the process oriented measures might have little to do with the actual quality of patient care, and the data don’t compare the Pioneer ACOs with nonparticipating groups.  The actual scores are not released.  Six of seven patient experience measures were improved in the Pioneer ACOs, although CMS doesn’t share results or benchmarks.

The Kaiser Health Network headline “One-Quarter Of ACOs Save Enough Money To Earn Bonuses,” tells the financial story.   CMS emphasizes the success of the provider organizations which did well, and mentions briefly that a single hospital system had $10 million in expenses beyond budget, and owes CMS $4 million back.  A pdf is available that shows the ACOs that have costs that are significantly higher or lower than budget – but doesn’t show the results of the ACOs that were in the corridor with financial results that were not significantly different than their budgets.

CMS offers aggregate numbers – but then they don’t add up.  See this for example:

During the second performance year, Pioneer ACOs generated estimated total model savings of over $96 million and at the same time qualified for shared savings payments of $68 million. They saved the Medicare Trust Fund approximately $41 million

68+41 = 109, not 96. That $13 million is a mystery.   Of course, it’s a small mystery – representing under 0.003% of the total Medicare budget. 

Release of the full data set – as opposed to cherry-picked data in a press release – will better serve public policy.  It’s no wonder that no major news outlet picked up this story!

Source for both graphics: CMS Press Release and my calculations 

Monday, September 15, 2014

US has high hospital administrative costs

Today’s Managing Health Care Costs

Number is $150 billion

David Himmelstein and colleagues report in this month’s Health Affairs that US hospitals spend more on administrative expenses than  hospitals in seven other developed countries.   This is especially surprising because we have fewer hospital days than any of the countries studied.   Sarah Kliff of Vox notes that this means that $1.43 of every $100 of our GDP is spent on hospital administration. The authors say that the US could save $150 billion if our hospital administrative costs were more like the comparator countries. 

This was a hard study to complete – other countries don’t have Medicare cost reports –and their reporting doesn’t mesh perfectly with the US reporting.   Hospitals have different scopes – some include ambulance services and post acute services, and it’s difficult to tease out what is spent administratively to support physicians and what is spent to support hospitals The authors acknowledge that they don’t have data on military or veterans hospitals – which could lower the costs.  The US also has a lower percentage of elderly, which should lower the cost of hospital administration when it’s expressed as a per capita number.

Himmelstein is right – we spend too much on hospital administration.   Years ago in a Physician for National Health Program presentation, he used to show a photo of the accounts receivable department at a Boston hospital – and compare it to a single person at a desk at a Canadian hospital.  This article is the research to back up that image – but the actual differences are smaller than that PowerPoint slide made it appear.  

The US has a complicated finance system – and there is a striking amount of shifting costs from one party to another; hospitals which don’t have to do billing and receive a standard budget as in Scotland obviously have far lower administrative costs.

I believe the total spending on hospital administration is tightly related to the total dollars flowing through the system, as opposed to the number of people receiving care.   We couldn’t have the expensive health care we have and dramatically lower hospital administrative costs. On the other hand, if we lowered overall costs – hospital administrative costs would likely look more like those in the comparator countries. 

Health Care Cost Inflation Increases Income Inequality

Today’s Managing Health Care Costs

Number is 60%

I’ve periodically revisited the issue of healthcare costs and social justice.   High health care costs affect everyone and all businesses – but they are exceptionally bad for the working poor. The poorest Americans are likely to be eligible for Medicaid if they are not undocumented immigrants.  However, those above Medicaid eligibility (which is as low as 25% of the Federal Poverty level in some southern states) are especially harmed by high health care costs. That’s because these costs represent a substantial portion of their total income. 

David Blumenthal and David Squires of the Commonwealth Fund posted the above figure last week – noting that health care costs represent a whopping 60% of average annual income for low income families – up from 13% in 1998.   He writes:

Economists generally agree that employers for the most part treat workers’ compensation in all forms—wages and benefits—as a single expense. When health insurance premiums go up, employers may reduce take-home pay to keep overall compensation in check.

He also notes that an additional 20 million Americans are in poverty when their out of pocket health care costs are considered. 

There are nuances to this – the best source of jobs with good compensation and good benefits for many low wage workers have historically been in health care, and efforts to lower health care costs are likely to lead to hospital closures and loss of good working class jobs.    I trumpet the importance of disruptive innovation – but this usually involves some replacement of labor by capital – which is bad for those with the lowest skills whose jobs are most likely to be displaced.   An app to get women prompt UTI treatment is great for patients and can improve quality, but it also means fewer clerks answering the phone.

My Towers Watson colleague Steve Nyce published projections of how health care cost increases would affect Americans of different income deciles.  This research was completed when health care cost trends were higher- but the idea remains. Subsequent research showed that health care inflation had wiped out all wage increases for the median American household between 1999-2009.  Health care cost increases widen the income gap, and contribute to growing income inequality. 

This is one more reason that we need to push hard to lower health care costs.

Thursday, September 11, 2014

Narrow Networks Save Big Bucks

Today’s Managing Health Care Costs

Number is 4.2%

Jonathan Gruber and Robin McKnight of MIT and Wellesley report that the Group Insurance Commission (GIC), which purchases insurance for Massachusetts public employees, was able to save over 4% in total medical costs by offering large incentives for its beneficiaries to choose a narrow network plan that didn’t include access to the highest cost providers.

The GIC purchases insurance on behalf of state employees and some municipal workers; state employees were offered a 3 month premium holiday for going on the narrow network plans, while municipal employees saw lower premiums but no premium holidays.    The researchers show that adoption of the narrow network plan was higher among those with the incentive (no surprise).

Some news reports have focused on the wildly different utilization of those in the narrow network plans.  For instance,  Fierce Healthcare trumpets “ Narrow networks lower spending by a third.” The article goes on to report emergency department visits down by 36% and specialty care services down by 45%.”  That’s true, but the populations are entirely different.  The narrow networks attracted the healthiest patients, and their overall costs were almost 40% lower than those who chose the wider network.  But if the only impact of narrow networks was that the healthiest people switched to the narrow network plan, that wouldn’t have saved the taxpayers any money.  

The researchers showed that when you add the costs of the PPO population (about 90% -and sicker) and the narrow network population (10% and much healthier) – there were still 4.2% less costs  - meaning that the narrow networks decreased both cost per unit and utilization over the entire population – not merely the population that chose that option.

This is a powerful validation of narrow or high performance networks.  These networks are likely to provide the most savings for employers with populations in metropolitan areas that have substantial provider competition, and where high cost providers can be excluded from a network.   I suspect that these networks will also exert downward price pressure on all providers, including those who don’t participate in the narrow network products.

Wednesday, September 10, 2014

Physicians Behaving Badly

Today’s Managing Health Care Costs

Number is $4.9 million

It’s been a bad few days in terms of reports of physician conflicts of interest that are likely to increase our overall health care costs.

The Wall Street Journal reported on a laboratory which bundles cardiac risk tests and bills Medicare for these. This company likely violated federal anti-kickback laws by paying physicians a $20 fee for each blood sample they sent to the lab.   It’s not a large fee – but remember how little drug companies have paid to gain loyalty of physicians.  Also- for some physician offices total payment was tens of thousands of dollars.   This lab also stored samples of blood, and asked physicians for orders retrospectively to add new (expensive) tests to the old samples as these tests became available.  It doesn’t appear physicians were paid extra for this –but it’s a good example of bad behavior.  In some instances, physicians authorized these retrospective tests on all the previously-drawn blood samples –even though it’s hard to believe that this was necessary for the care of these patients.

McClatchy reported on a more vile scheme

During a meeting that was secretly recorded, a salesman for Reliance Medical Systems promised that within a month or two of joining its illicit kickback scheme, spinal surgeons could collect enough money to pay for their kids’ college educations, Justice Department lawyers charge.Taxpayers were the multimillion-dollar sugar daddies in this plot

Physicians invested a token amount in a medical device distributorship, $5000, and were paid over $20,000 at the end of the first month after they became investors.  One physician was paid almost half a million dollars.  Of course, he billed Medicare and private insurers for his surgical services while getting what was essentially a commission on the implantable devices (hips, knees, back fusion devices) that the hospital was purchasing for his patients. Physician investors were paid a total of $4.9 million in fees from 2007 to 2012. 

Aaron Carroll of the Incidental Economist has an excellent column in the NY Times Upshot reporting on a study showing that academic physicians on FDA panels were more likely to approve drugs when they disclosed that they were paid advisors on panels for the relevant drug company.   Carroll, a pediatrician and professor, writes

Some physicians, especially those opposed to the Sunshine Act, believe that they should be responsible for regulating themselves. But our thinking about conflicts of interest isn’t always rational. A study of radiation oncologists found that only 5 percent thought that they might be affected by gifts. But a third of them thought that other radiation oncologists would be affected.Another study asked medical residents similar questions. More than 60 percent of them said that gifts could not influence their behavior; only 16 percent believed that other residents could remain uninfluenced.
This “magical thinking” that somehow we, ourselves, are immune to what we are sure will influence others is why conflict of interest regulations exist in the first place. We simply cannot be accurate judges of what’s affecting us. 
We need more public disclosure of potential conflicts of interest.  We need aggressive enforcement of downright fraud, and the medical community should enforce stricter conflict of interest rules.   Health care would be expensive even without fraud and unconscionable self-dealing. But then we could have the debate about how to cover the great new technologies that can make our lives longer and better –not how to police our worst instinct.


Monday, September 8, 2014

Hospitals shouldn’t partner with mobile screening firm that peddles unnecessary tests

Today’s Managing Health Care Costs

Number is 1 Million

I like the idea of finding disease early – and treating people when their conditions are mild and reversible.  Who wouldn’t rather have a simple outpatient cure today rather than endure a complicated inpatient stay that would at best keep a disease at bay?

But screening people at low pre-test risk is fraught with danger.   For an unselected population, the likelihood of a false positive far outweights the likelihood of a true positive.

Last week’s JAMA highlights a letter sent by Public Citizen to 20 hospital systems which called on them to stop direct to consumer screening efforts marketed by HealthFair, a Florida-based company which promotes screening tests including carotid ultrasound, EKG, echocardiograms, ultrasound to look for abdominal aortic aneurisms (AAA) and blood pressures to screen for peripheral arterial disease.  Their advanced packages include testosterone and PSA for men, and TSH and CRP for women.  There is a role for all of these tests when patients have specific symptoms – and the AAA test is indicated  once in men 65-75 with a history of smoking – but that’s it. 

The Annals of Internal Medicine had a good explanation about the ethics of health fair screening in 2012, concluding

Misuse of preventive services, under the guise of saving lives and saving costs, may actually lead to increased cost and harm due to unnecessary follow-up testing and treatment with associated avoidable complications. We suggest that medical entities and physicians withdraw from the unethical business of promoting unproven and potentially harmful screening tests.

HealthFair claims to have screened a million people, The cost of their services ranges from $187 to $325 per person.  However, the indirect costs – including followup invasive tests and surgical procedures that might well have no benefit for the individuals, are substantially higher.

Friday, September 5, 2014

The Health Plan Competition Paradox

Today’s Managing Health Care Costs

Number is 11%

Two of my favorite health care wonks, Sarah Kliff of, and Austin Frakt of the Incidental Economist, come to very different conclusions this week about whether more competing health plans lowers health care premiums.

Frakt reviewed academic papers by Jonathan Gruber and others which show an association between health plan competition and lower health care premiums.

UnitedHealthcare, the nation’s largest insurer with 84 million policies in force in 2010, did not participate in any exchanges. Had it done so, Ms. Dafny and colleagues estimated that premiums would have been 5.4 percent lower. Had all insurers in each state’s 2011 individual market participated in that state’s exchange in 2014, premiums would have been 11 percent lower, saving $1.7 billion in federal premium subsidies.

One reason the cost of American health care is so high is that insurers are so weak. Having hundreds of different carriers, for example, means no one insurer has lots of negotiating power — hence those high prices drug and device makers can charge.

So – who’s right?    Are a dozen health plans fighting for your business likely to successfully get the lowest prices?   Or is mega-plan with market dominance likely to obtain the best prices from providers frightened of losing all of their business.   Could they both be right?

Here is how health plans can lower the cost of health care premiums:

(1)     Obtain lower unit prices
(2)     Lower overall utilization
(3)     Lower administrative costs
(4)     Sell health insurance at a loss

That’s pretty much it.  If the prices don’t come down or the utilization doesn’t come down, the premium cannot get and stay lower.   

Health plans can skinny their administrative costs (3)  – but remember that these costs are only a small portion of the total health care spend – so a 10% decrease in insurer administrative cost is only worth 1-2% of total costs.  Further, smaller health plans should generally be more expensive on a per member per month basis administratively because they have fewer subscribers to bear the cost of expensive IT systems and executives.

A health plan fighting for market share could accept losses (4)  – but only for a little while before we would start worrying about its financial stability.  State regulators don’t look fondly on health plans losing money – for if they go out of business – it causes a regulator’s (and a subscriber’s) nightmare.  Some of the studies that Frakt cites could have included some new plans "buying business" and losing money on entry to a new market - but that would only lower premiums for a year or two - by which time the health services researchers would have already moved on. 

So this leaves us with (1) and (2).    Health plans with little competition in markets with provider competition can lower unit costs.(1)   It's not clear why the degree of health plan competition should impact plans' ability to lower utilization.  (2)

So – will the entry of a new health plan lower premium costs?   I think it depends.   A market with no meaningful provider competition will probably have high premiums regardless – and it makes no difference how many health plans compete.  Therefore, markets with little provider competition end up with little health plan competition, too.  What health plan would want to enter a market where there was no chance to succeed?   That's why we see little health plan competition in many rural markets with little or no hospital competition.

A market with meaningful provider competition allows health plan differentiation –so could lower prices.  This is especially true in metropolitan areas that have 3 or more viable provider health care systems.  But if the health plans all look the same and each has higher administrative costs and less leverage to lower prices, don't expect sustainable cost savings!