Thursday, September 3, 2015

Cadillac Tax will raise out of pocket costs for the sick

Today’s Managing Health Care Costs Number is 46%

Debate continues on whether to junk the Affordable Care Act’s excise tax (“Cadillac Tax). The tax would raise $87 billion over 10 years – not a whole lot of money compared to the overall cost of the ACA subsidies ($1.35 trillion over that time period).  About a quarter of employers would have at least one plan that would be subjected to this tax in its first year (2018), and this could go up to 30% in 2023 according to a recent Kaiser Family Foundation simulation study.

But this doesn’t capture the full scale of the problem.   Among employers with over 200 employees, 46% will hit the excise tax threshold in 2018, and this will go up to 56% by 2023.   The KFF simulation shows that employers who would not be liable to pay this tax for their overall benefit are highly likely to have to pay it for at least a portion of their population – those who have an open network plan with lower member cost sharing.   The excise tax is levied on dollars paid by employer and those paid by the employee through payroll deductions; dollars paid by members as out of pocket costs are not included in the calculation.

Employers will continue to make difficult choices to avoid the excise tax.   The likely changes between now and 2018 include:

1)     Eliminate flexible savings account (FSA) contributions, which are fully funded by employees but would count toward the tax liabities.
2)     Decrease health plan choice – eliminating high actuarial value health plans which would be more likely to trigger the tax
3)     Move to high deductible health plans – and decrease what is covered by the health plan
4)     Limit employer “seeding” of account based health plans (HRAs, HSAs)
5)     Decrease breadth of network to exclude higher cost providers.

This plan will work much as the ACA’s drafters expected.   It encourages a decrease in the generosity of health plans, which helps address the problem of patients overusing health care since they have generous health insurance. This is good, to the extent we’re convinced that patients drive the increasing cost of health care.  But this means that patients are “on their own” more and more when it comes to paying for care.  Employees face a gauntlet of bureaucratic structures on the provider side that seek to extract revenue from them.  

We also tend to look at average costs for patients – and this misses the fact that there is a long tail of patients with exceptionally high costs. For these patients, the average out of pocket cost is irrelevant –annual out of pocket maximum really matters. The out of pocket maximums are $6600 (Individual) and $13,200 (family) for health exchange plans and the maximums don’t apply to out of network care.  Out of pocket maximums can be substantially higher in non-exchange health plans; I’ve seen plan designs with out of pocket maximums that exceed $25,000 annually for a family.

I’m increasingly worried that health care costs will be increasingly out of reach for those who have extensive medical problems. This KFF simulation helps support efforts to eliminate the excise tax. Maintaining the excise tax as it is currently being interpreted will diminish access to health plans that are most meaningful for those with serious illnesses.

Recent posts on excise tax 

Tuesday, September 1, 2015

A New Taxonomy for Health Care Payments

Today’s Managing Health Care Costs Number is 8

My favorite quote on provider payment is from James Robinson (2001)

There are many mechanisms for paying physicians; some are good and some are bad. The three worst are fee-for-service, capitation, and salary. Fee-for-service rewards the provision of inappropriate services, the fraudulent upcoding of visits and procedures, and the churning of "ping-pong" referrals among specialists. Capitation rewards the denial of appropriate services, the dumping of the chronically ill, and a narrow scope of practice that refers out every time-consuming patient. Salary undermines productivity, condones on-the-job leisure, and fosters a bureaucratic mentality in which every procedure is someone else's problem.

The late August issue of Annals of Internal Medicine has a thoughtful article proposing a new way to look at provider payment.    Author Kevin Quinn proposes 8 ways to pay providers (see table above).   As we move down the list, provider risk decreases and payer risk increases.   Quinn distinguishes between epidemiologic risk (prevalence of medical conditions) and performance risk (treatment of medical conditions).  I’ve often heard the epidemiologic risk called “insurance risk,” and it makes sense that insurance risk should not be borne by providers.   Excellent risk adjustment, of course, can decrease insurance risk – or on the other hand can encourage aggressive diagnosis upcoding to harvest additional risk adjustment revenue.

Quinn notes that the bottom four payment methods are all “fee for service,” but the service is defined as a day (5), a unit of service (6), proportional to cost (7) and proportional to charges (8).  He notes that most recent health payment reform has moved risk toward providers (moving up in this hierarchy). 

I’m not sure that I agree with all elements on this taxonomy. For instance, I think “per time period” (1) means that providers have the highest risk only to the extent that they are responsible for an entire population. A dermatologist who is on salary is paid a set amount per year, but the “risk” to the provider organization for this is only high when the organization is responsible for the care of an entire population.  If the physician organization feels no obligation to meet all a population’s dermatologic need, the epidemiologic risk is not borne by the provider organization.

Quinn concludes with four questions he believes will determine the success in payment reforms:

First, as more payments are based on capitation and episodes, will providers avoid costly patients? The answer depends on the accuracy of case-mix adjustment and the effectiveness of nonpayment mechanisms to minimize risk selection.
Second, will an overabundance of uncoordinated reforms create a muddle of incentives? Some accountable care organizations have raised this concern already.
Third, will providers rebel against the administrative burden and clinical oversight inherent in many initiatives? For example, the American Medical Association describes the current scene as a “regulatory nightmare”.
Fourth, will provider mergers subvert efforts to control health care spending? Accountable care organizations and other reforms encourage providers to band together, but consolidation helps providers negotiate higher prices from commercial payers.

I've given up on the idea of finding the "ideal" method to pay providers.   Too much health care is not amenable to capitation, whether because the care is rare or the provider volume is small.  This taxonomy is helpful in that it makes it clear that per episode payment approaches can also drive more accountability to providers.  Fee for service clearly induces too much utilization of high margin services, which tend to be low value when deployed to additional patients.   

Many high performing provider organizations benefit from having a portfolio of different types of payment, including those which involve more provider risk and those which do not.  Understanding stakeholder implications of each provider payment methodology can help us design better provider payment approaches.  Provider organizations should realize that there is no provider payment nirvana that guarantees success- and provider payment is just one lever to improve organizational performance. 

Wednesday, August 26, 2015

Colonoscopies Save Lives

Today’s Managing Health Care Costs Number is 29.7%

Deaths per 100,000

My post on Monday about the lack of mortality improvement from detection of thousands of cases of ductal carcinoma in situ (DCIS) made me look for some data on the changing rate of colorectal cancer mortality in the US.  The CDC has published data on this since 1999, during which time our population has aged – so we’d expect an increase in colon cancer mortality.   We’ve also had a steep fall-off in death from heart disease, leaving more people alive to get colon cancer.

Nonetheless, the unadjusted mortality rate for colorectal cancer has decreased every year – for a total decrease of 29.7%

Now, don’t think for a minute that this means that doing more colonoscopies saves money.   The cost per quality adjusted life year saved for colonoscopy is between $50-100,000.   That means that saving all these lives has an associated cost of billions.   But the purpose of medical care is to prevent death and improve the quality of life, not to lower costs.

Are you 50 or over?  Get your colonoscopy, if you haven’t already!  Repeat every 10 years.  We need to see more bar graphs like this.

Monday, August 24, 2015

Data show little or no benefit to treating "early" breast cancer (DCIS)

Today’s Managing Health Care Costs Number is 5

It’s been 5 days since JAMA Oncology published its landmark article on ductal carcinoma in situ (DCIS), thought to be a precursor to breast cancer.   And the top five articles in the New York Times Health section this morning were all about this article.   It’s a big deal – important to the treatment of women with breast calcifications on mammograms, and also important to health policy.

Researchers used the SEER (Surveillance, Epidemiology, and End Results) database to show that the 20 year likelihood of dying of breast cancer among the 110,000 women who had DCIS was a bit over 3% - very similar to the likelihood among women who were not diagnosed with DCIS.   The likelihood of local recurrence went down with mastectomy or lumpectomy, but the likelihood of dying of breast cancer was unmoved. Radiation therapy with lumpectomy was not significantly better at decreasing mortality compared to lumpectomy alone; removal of the entire breast was associated with statistically significantly higher rate of breast cancer mortality than lumpectomy with or without radiation therapy.

DCIS is increasingly common; it represented 3% of all breast cancers found before mammography became widespread, and was usually diagnosed on autopsy only.   It now represents 20-25% of all breast cancers detected – 50 to 60,000 women a year.

The accompanying editorial observes:
1.     The risk of death from breast cancer in those with DCIS is low
2.     The risk of dying from breast cancer in black women and women under 40 diagnosed with DCIS is substantially higher
3.     Aggressive treatment doesn’t lower mortality
4.     The risk of invasive breast cancer is almost the same in the other breast, making it likely that DCIS is an indicator of higher risk as opposed to a precursor lesion itself.

Screening for cancer makes intuitive sense. Find it early, remove it, and lengthen life.   This works for at least two cancers – colon cancer, where death rates are falling in the era of colonoscopy, and cervical cancer, where pap smears decreased death rates dramatically.

But we’re increasingly finding that screening isn’t nearly as effective for many other cancers.  Prostate cancer screening raised diagnosis dramatically, but few lives were saved, and many suffered from complications of therapy.  I’ve written earlier about the epidemic of diagnosis of thyroid cancer, unaccompanied by any decrease in diagnosis of invasive cancer, or any decline in death from thyroid cancer.

We should look at population data before deploying population-wide screening. The goal of medical care is to give people more life and better quality life – not merely to make diagnoses that subject patients to potentially harmful therapy without benefit. We also need to recognize that there are subpopulations where screening and therapy recommendations can differ, as their risks are substantially higher.

JAMA Oncology, a journal that just began publishing this year, is offering access with no paywall to the original research and to the editorial.

Friday, August 21, 2015

Direct to Consumer Pharmaceutical Advertising

Today’s Managing Health Care Costs Number is 2

There are two developed countries that allow direct to consumer (DTC) marketing of prescription pharmaceuticals – the United States and New Zealand. The FDA’s power to regulate drug company commercial speech is on the wane; a district court found earlier this month that the FDA couldn’t stop a company from communicating that its fish oil pills could be used for non-approved indications unless the company knew the claims were false.    A federal appeals court ruled earlier that the FDA couldn’t mandate that tobacco companies  admit their previous misstatements on cigarette packages, and plans to require use of plain packaging with large graphic warnings have been thwarted. 

The US began to allow DTC pharmaceutical marketing in newspapers and magazines in 1985, and television in 1997.  The rationale was that they could help educate Americans about various conditions which could be helped by drugs, and DTC advertising was consistent with a move to make health care less paternalistic.   Why shouldn’t patients know about the drugs available to them?  

But DTC marketing has made television a wasteland of ads for treatments for erectile dysfunction, insomnia, and hepatitis C.  And many of the well-meaning regulations restricting how drug companies communicate actually obfuscate the real dangers and tradeoffs of medications. For instance, mandating the presentation of a raft of uncommon side effects makes it harder to focus on common side effects. Everything eventually looks like an iTunes user agreement – and who pays any attention to those.

What’s always really bothered me is advertising for expensive brand name drugs (Yes, I’m thinking of Nexium, the purple pill) that are absolutely equivalent to omeprazole at 16 times the price. Until Nexium lost its patent protection last year, it was among the top cost drugs for most employers whose data I reviewed – which certainly doesn’t speak to increasing value.

There are two articles in yesterday’s New England Journal focused on DTC drug advertising.  “Rethinking DTC Advertising.. “ points to the failure over decades to develop meaningful “patient package inserts” (PPIs) to help patients understand the risks and benefits of a drug.  The FDA initially intended to mandate PPIs – but gave up during the Reagan administration.  The authors conclude
For all its capacity to encourage overdiagnosis and overmedication, DTCA's virtue is that it treats consumers as people who deserve to know something about the compounds they take into their bodies. After 30 years of DTCA, it's not clear that advertising is the best medium for communicating risk information,5 but marketers should at least be required to try to communicate risk information as effectively as they do their promotional messages.

New DTCA Guidance — Enough to Empower Consumers?” points to the fundamental inequivalence between regulatorily mandated disclosure of risk and vigorous company marketing of potential benefit.

If disclosure is to work, as others have argued, it must be done right, in a format that's designed to be usable. As an example of success, Fung et al. cite the simple, salient, and familiar “A, B, C” system used to rate restaurants on the basis of public health inspections, with the results posted prominently by the door. The new FDA guidance is a move in this direction, at least if it gives companies more liberty to construct readable disclosures. However, even revised disclosures written by the companies themselves are unlikely to be simple and candid enough to steer patients away from drugs that are inappropriate for them. One can imagine a system that would grant drugs an “A” rating if they proved a substantial advance over the previous standard of care in treating a serious medical condition, with minimal risks or side effects. Regrettably, many of the most widely advertised drugs would not secure that golden ring.

I’m pretty sure we won’t see restaurant style A, B and C grades for drugs that include safety cost and effectiveness any time soon – and in fact a drug might be a “c” for some patients and an “a’ for others (think about high dose statins for those who had a heart attack, compared to those with a 3% risk of a heart attack in the next decade).

But we do desperately need better consumer information on pharmaceuticals – and given the strong incentives for pharmas to emphasize benefits, we need a standardized way to portray risk and overall value.

Tuesday, August 18, 2015

Air pollution in China and “Poop Water” in my Hometown

Today’s Managing Health Care Costs Number is 4400

A report in this weekend’s New York Times notes that air pollution in China is more widespread than previously believed -  3/8 of the population lives in areas with air pollution rated  on the average “unhealthy” in the US; 92% of the population.   1.6 million people die prematurely each year in China due to air pollution. That’s a staggering number; I think it might have more impact to say that air pollution kills 4400 people a day. 

Here’s a link to the full report.  

I’ve noted before that environmental protection saves substantial dollars in the economy, and we’ve made huge strides in the United States.   China is focused right now on maintaining its high growth rate.   The economy (and health care) require better environmental stewardship.  

Speaking of environmental stewardship, I’m pretty embarrassed to be a citizen of Belmont, MA, where our inadequate efforts to control water runoff mean that our streams, which feed the Mystic River, just received an “F” from the Environmental Protection Agency.   According to a Boston Globe columnist,

On many days, just getting splashed with the water coming out of the culvert flowing into Little Pond, which in turn flows into Alewife and the Mystic, poses a health hazard.

Clean water, like clean air, is important to the health of the community.    Both are a community good  -  the air I breathe and the water I swim in, or don’t, is profoundly influenced by activities far from my hometown.  Funding environmental protection locally makes it more likely that some towns won’t make a big enough investment. Cap and Trade for sulphur dioxide has also shown us that we can use the market to be sure that incremental dollar spend on cleaning air (or water) are spend where they can do the most good.

So, I’m pretty embarrassed to live in a town whose streams are called “poop water” in our newspaper.  There is a health and economic price we pay for sending raw sewage into our streams, or nitrogen and sulphur dioxide and particulates into the air.

Wednesday, August 12, 2015

Should we repeal the "Cadillac Tax?"

There's a  bipartisan move to eliminate the Affordable Care Act's excise tax on employer sponsored health insurance plans that cost more than $27,500 (family) or $12,700 (individual) in 2018.   Current bills to repeal this tax have been cosponsored by 132 Representatives in the House and 7 Senators.  The New York Times editorialized against repeal in today's paper.

The broad based coalition that opposes the excise tax includes:

·         Employers.  Who wants to be taxed?
·         Labor:  The excise tax will make it tougher to negotiate higher benefits, and will take dollars away from potential raises.
·        Health plans, which don't want to administer this and be blamed for it
·         Anti-tax advocates:  who don't want the feds to collect extra taxes
·         Equality advocates: There's no adjustment - so employers with older employees, sicker employees, or with workers in expensive areas will now be doubly penalized
·         Urban advocates: Health care costs in many metro areas are high, with large portions of the population cared for by academic medical centers
·         Rural advocates: Health care costs on single hospital rural areas also tend to be high since there is no competition

There are serious problems with this excise tax as it will be administered.   

The first problem is fairness. A “fair” excise tax would either account for demographics, illness, and geography.  All these adjustments would increase fairness, but they would also increase complexity and the propensity to game the system, though.  Another way to be sure that the excise tax didn’t penalize employers for elements beyond their control would be to levy the tax if the total actuarial value of the health plan was high.  That would penalize plans that were “too rich”, and many economists believe that these plans are responsible for excess spending.  It would also discourage narrow network plans or some staff-model plans like Kaiser that accomplish high actuarial value without an especially high price.

The second problem is that the excise tax has helped encourage substantial cost shifting to employees.  One way to pretty much guarantee an employer’s health plan will cost less is to “buy down” to a lower value plan.  These low value plans, though, are bad news for those of modest means, for whom a $3000 or higher family deductible is an unbearable barrier to obtaining care.

My sense is that the threat of the excise tax has helped convince employers to substantially decrease the value of their health plans already, and I don’t think we need this tax to further discourage overly generous plans.  Making it fair through adjustments would be complicated, and pegging it to actuarial value would discourage plans with meaningful coverage for those of modest means.

It’s not likely that Congress can pass a nuanced bill to address these problems with the excise tax, nor is it likely that the current Congress can replace the estimated $84 billion in revenue from the excise tax over its first decade. Therefore, a repeal would likely add to the projected deficit. This would cost angst to those who dislike allowing the government to run a deficit, but will not be a significant economic drag at this point.  

Congress eliminated the physician “sustainable growth rate” pay cuts without coming up with respecting budget neutrality;  I suspect they do the same for the excise tax.  Opponents of the ACA can declare that they have repealed a piece of the law, and supporters can breathe another sigh of relief because this tax will make the ACA less popular and it is not critical to the ACA's success.

This shows that the portion of Americans facing high deductibles is already quite high.  The exercise tax threshold would  only increase with the consumer price index, pushing employers to increase cost sharing each successive year to avoid the tax.  Source: KFF HRET Employer Sponsored Health Benefit survey, 2014.