The Incidental Economist

A battle is heating up between public health advocates and corporations that use the First Amendment as a legal defense. I’ve posted on this before concerning pharmaceutical data mining and tobacco (additional NEJM commentary here).

The Interagency Working Group on Food Marketed to Children includes the USDA, FDA, FTC and the CDC. They are drafting VOLUNTARY guidelines for industry (prior TIE coverage on the IWG and obesity here). Some junk food lobbyists are waiving waving the First Amendment, claiming the guidelines won’t be voluntary and will therefore infringe on their freedom to sell Twinkies to toddlers.  (That must be why we fought the American Revolution).

A group of law professors just published an open letter to the IWG dismissing this constitutional argument. Just 2 pages of text, uncharacteristically brief for law professors. I signed the letter as well.

Yesterday, I posted on 3 studies in The Lancet on 9/11 health issues from survivors at the WTC. Quick summary: 9/11 rescue and recovery workers and other WTC-exposed people are living longer, but possibly sicker, than average. They are not dying in greater numbers from heart & lung diseases or cancer, but the 8927 FDNY responders at the WTC may have from 4 – 38 additional cancers since 9/11. 

My media score card:

How did the media report on these studies? Only the Associated Press (David Caruso), Medical News Today, Medscape and Sarah Boseley at The Guardian seem to have reported the all-cause mortality study prominently, but the WSJ coverage comes very close and gets bonus points for its graphic.

How did the NYT report these 3 studies?

Study Suggests Higher Cancer Risk for 9/11 Firefighters

By SYDNEY EMBER

A new study says firefighters who toiled in the wreckage of the World Trade Center in 2001 were 19 percent more likely to develop cancer than those who were not there, the strongest evidence to date of a possible link between work at ground zero and cancer.

The two other studies are buried near the end, with the mortality study inappropriately minimized:

The Lancet also published other reports related to the Sept. 11 attacks on Thursday, including one by researchers at Mount Sinai Medical Center showing the persistence of respiratory, gastrointestinal and mental illness among rescue and recovery workers. Another report, by the New York City Department of Health and Mental Hygiene, however, indicated a lower mortality rate among recovery and rescue workers compared to the general population of New York City. Its authors attributed the lower rate to the “healthy-worker effect” — those present at the site were more likely to be employed, a group typically healthier than the general population.

The data from the mortality study did not find the differences to be due to the “healthy worker effect.” That was one possible hypothesis offered in the discussion section. What the study found was remarkably lower SMR. I’m also troubled by the final paragraph in the NYT story, where the chronic disease study PI said he “had no doubt that time would reveal a connection between certain cancers and exposure to the trade center site.”

Other media reporting on the Lancet articles, mainly the FDNY cancer study: Bloomberg; LA Times (cancer and chronic, but no mortality coverage); MSNBC (mistakenly claiming that “ those exposed as a 9/11 rescuer had a 19 percent rate of any type of cancer.”); HuffPo (which noted that the cancer study was “conducted” by the FDNY; several authors hold positions at FDNY but funding was NIOSH); HuffPo also discussed the push to get NIOSH to reverse the July decision to exclude cancer from the Zadroga Act’s list of covered conditions); WSJ (with an excellent graphic on the prostate & lung cancer data); and the KHN list.

As we approach the 10th anniversary of 9/11, The Lancet published 3 studies on the health of people at the World Trade Center site on 9/11. We should pause and remember that while civilians were trying to escape, members of the FDNY and other first responders ran towards the danger. That’s what they do – run into burning buildings. Heroes.

Our mission at TIE is research-based health policy. The key issues in these Lancet studies are the control groups and surveillance bias. We rely on research, even for heroes.

The first study reports all-cause mortality for the 41,930 people enrolled in the World Trade Center Health Registry cohort (2003-2009 data; Jordan HT et al.) The surprising findings include a much lower standardized mortality ratio (SMR):

All-cause SMRs were significantly lower than that expected for rescue and recovery participants (SMR 0·45, 95% CI 0·38—0·53) and non-rescue and non-recovery participants (0·61, 0·56—0·66). No significantly increased SMRs for diseases of the respiratory system or heart, or for haematological malignancies were found.

You read that right:  9/11 workers are dying at a 55% lower rate and are not dying at a higher rate from heart/lung diseases or cancer. The differences are huge – any drug with this effect on mortality over 10 years would be worth >$100 billion. Note that 10 years might not be long enough for some long-latency diseases to impact SMR.

The chronic disease study (by Wisnevsky JP et al.), tracked 27,449 participants in the WTC Screening, Monitoring, and Treatment Program, and found a substantial burden of physical and mental chronic diseases, including asthma, sinusitis, GI disease, depression, PTSD and panic disorder:

We reported that about 10–30% of rescue and recovery workers still had persistent medical disorders 9 years after the WTC attacks. More than a fifth had multiple physical and mental health problems. Most health disorders were more common in rescue and recovery workers with highest levels of exposure to dust and smoke than in those with lower levels of exposure.

This study did not use an external control group, noting problems with finding a control group that was as healthy as the study population. The study found significant dose-response, but does not tell us whether the rescue and recovery workers are sicker (or how much sicker) than any other group of Americans.

The final study, the FDNY cancer study got the most press, by far.  This study (by Zeig-Owens R et al.) found a 10% increase in cancer incidence (not mortality) in the 8927 WTC-exposed FDNY fire fighters, compared to the US population.

But these fire fighters don’t have normal American health care, they are perhaps the best-screened population in history, due to the various funds established to care for 9/11 first responders. Correction for this surveillance bias is important, because 30 of the increased cancers in the exposed group were prostate cancer (observed 90, expected 60), which may be due to better screening rather than 9/11-related etiology. You might expect lung cancer to be the WTC leader, but lung cancers were actually much lower than expected:  (observed 9, expected 21). The causal story is unclear at this point. After correcting for surveillance bias, the net increase in FDNY cancers compared to the general population? 4 cancers, SIR of 1.02 (2% increase) with a 95% CI of 0.90 – 1.15:

Higher rates of cancer were found when the control group was shifted to the FDNY members who didn’t go to the WTC (1.32 without correction for surveillance bias, 1.21 above after correction, 1.19 after full lagging), mainly because this control group had a much lower cancer incidence than expected (perhaps fire fighters are in better shape and smoke less than average). 19% higher incidence rate = perhaps 39 38 additional cancers out of 8927 exposed FDNY members. The choice of control group makes a huge difference. Remember also that this cohort does not have higher cancer mortality at this time (first study), perhaps due to increased surveillance, medical attention and long latencies.

Tomorrow I’ll post my media score card on how these studies were reported. On Thursday, the NIOSH decision to not include cancer in the Zadroga fund.

Not all Supreme Court rulings are created equal. Many can be overturned by Congress or even a regulatory agency; others require a constitutional amendment or perhaps new Justices. Some health sector examples:

The Constitution. The only people with a constitutional right to a minimum standard of health care in the US are prisoners.  Estelle v. Gamble, 429 US 97 (1976). This right was found in the Eighth Amendment, so it will take a constitutional amendment or a change of heart at SCOTUS to overrule it.

Other constitutional cases can by addressed by revising the law. In Sorrell v. IMS Health, the Court found pharmaceutical data mining to be protected by the First Amendment and struck down Vermont’s poorly worded law. A revised version might be blessed by the Court.

Or consider the various challenges to the individual mandate under the Commerce Clause; if the decision goes against health reform, Congress could re-enact it as a tax. The politics might be toxic, but it would be constitutional.

Federal statutes.  One primary SCOTUS job is interpreting federal statutes. In FDA v. Brown & Williamson Tobacco Corp, 529 US 120 (2000) the Court decided (5-4) that the Food, Drug and Cosmetics Act did not give the FDA authority to regulate tobacco. Congress overruled the 5 conservative Justices 9 years later in the Family Smoking Prevention and Tobacco Control Act of 2009.

Federal regulations. Many cases claim that a federal regulation is beyond the agency’s authority and therefore invalid. In Alexander v. Sandoval, 532 US 275 (2001), the Court invalidated (5-4) a regulation under Title VI of the Civil Rights Act of 1964, a key statute prohibiting discrimination by federal contractors such as hospitals. The Court said the regulation went beyond the language of the statute. Striking down the regulation made it more difficult to bring private discrimination suits against federal contractors. Congress could reverse this decision with a statute, but hasn’t, even in PPACA. In other cases, the agency could perhaps overturn a SCOTUS decision with a more carefully worded regulation.

Pre-emption of state law.  If a federal law and a state law are in conflict, the feds win under the Supremacy Clause. But the most interesting cases are when the federal law is unclear. Some examples:

• PhRMA v. Walsh (2003) – Maine’s prescription drug program did not conflict with Medicaid; HHS agreed with Maine on this one.
• Aetna Health v. Davila (2004) – the Texas Health Care Liability Act is pre-empted by ERISA §502.

Responses at the state level include drafting the law carefully to avoid stepping on federal toes. Revision to the federal statute would almost always work, but can be politically difficult. ERISA was passed in 1974 and the Court suffered through more than two dozen full decisions before Congress amended some aspects of ERISA in PPACA. Finally, it might be possible to revise federal agency guidance in order to change a pre-emption ruling. I’ll discuss this option in a future post on drug and device pre-emption.

Conflicts of interest are a perennial problem across all of medicine, but the solutions tend to be silos – one set of rules in Stark II, another standard in biomedical research, a third in medical journals, a fourth in teaching.  Drug companies are treated differently than device companies; pediatricians more strictly than surgeons; Medicare makes some P4P plans a potential felony. Sometimes these distinctions make sense, but generally not. It is time to think about “COI best practices” across medicine. 

Is there a conference for that?  Funny you should ask, because on October 27-28, the American Society of Law, Medicine & Ethics (ASLME) will co-sponsor the upcoming “Conflicts of Interest in the Practice of Medicine: A National Symposium.”  Click here to view the conference agenda.  Online registration here. Special preview of the event brochure, click here.

Disclosure:  I’m on the ASLME board; conference papers may be published in the Journal of Law, Medicine & Ethics, where I serve as EIC; and the conference is generously funded by a grant from the Highmark Foundation of Pittsburgh, Pennsylvania, with additional funding contributed by the Jewish Healthcare Foundation, the University of Pittsburgh Schools of the Health Sciences, and by the University of Pittsburgh School of Law.  Funders had no role in the selection of topics or speakers. 

At KHN, Susan Jaffe reports:

A provision of the 2010 federal health law seeking to increase Medicare beneficiaries’ share of health care costs is meeting resistance from an unlikely group of 33 state insurance regulators, health insurers and consumer advocates charged with revising Medigap insurance policies that cover most out-of-pocket expenses.

What caught my eye was “resistance” from “consumer advocates.”

The overall top seller of Medigap and Part D policies is AARP, in its insurance program with UnitedHealth. The relationship with AARP is a very big deal, even for an insurance company with 2010 revenues exceeding $94 billion (2010 Form 10-k here).  The latest annual 10-k report filed with the SEC mentions AARP 57 times and states that the “loss of our AARP relationship could have an adverse effect on our business and results of operations.”   Note 12 of the 10-K gives the numbers: “Premium revenues from the Company’s portion of the Program for 2010, 2009 and 2008 were $6.3 billion, $6.0 billion and $5.7 billion, respectively.” That’s 6 to 7% of UnitedHealth revenues – an important chunk.

But some relationships are asymmetric, in business as well as life. How important is the relationship to AARP?

As a tax-exempt entity, AARP files an IRS Form 990, which is available here. AARP revenues from the most recently available year (2009) are summarized below. United pays AARP royalties for the joint insurance program.  It’s a huge deal for AARP.

AARP does many, many excellent things, but 60% of their revenues come from selling insurance, including Medigap and Part D policies.  AARP wasn’t mentioned in Susan Jaffe’s article, but this financial conflict of interest is a relevant point, given their prominence.

Our federal government is an intricate policy machine. Amateurs make almost no impact in Washington bureaucracy. To understand and influence federal policy, you need a permanent presence in DC. The Administrative Procedure Act generally requires that federal agencies finalize regulations only after a “notice and comment” period. Washington lobbyists track such items assiduously. When these moments come, lobbyists need ready-made policy studies to influence the rulemaking process.

The topic is childhood obesity.  TIE is a research-based blog, but in this post, focus on process, not policy.

On April 28, 2011, the federal government published draft guidance on a voluntary program restricting advertising of unhealthy foods to children (the FTC blogged it here). Who could be opposed to such a thing, with all the concerns about childhood obesity?

For starters, anyone who sells junk food to kids, and believe me when I say that junk food is well represented in DC. 150 trade associations such as the Alabama Grocers Association and the Frozen Potato Product Institute wrote a joint letter attacking the proposal. The short letter mainly asks for delay, claiming that more data is needed before government acts. But the letter also illustrates how lobbyists plan ahead. The group funded a consultants’ report which finds that fighting childhood obesity puts “74,000 jobs”  “in jeopardy.” It’s not peer reviewed, but it sows doubt and gives opponents something to talk about.  With a few months lead time and a six-figure budget, such studies are easy to procure.

But lobbyists also play long ball, games that last years, even decades. Like a cricket match in hell.

For more than a decade, the Washington Legal Foundation has filed legal cases and briefs advancing the “commercial speech” theory of the First Amendment, an idea that was unanimously rejected by the Supreme Court in 1942. But today, commercial speech is a profound de-regulatory tool, striking down regulation of tobacco ads, pharmaceutical data mining, liquor stores, and other products. Millions of dollars, invested strategically over many years, are now bearing fruit as judges use “commercial speech” to strike down public health rules.

And now they come for childhood obesity. On the Sensible Food Policy Coalition website, two distinguished law professors argue that the First Amendment forbids voluntary guidelines to combat childhood obesity. One of the professors acknowledges financial support from Viacom, who must make good money from Twinkies ads.

My point isn’t to dispute either of these papers (at least not today) nor to disparage the distinguished authors, but to just note what success the commercial speech movement has made since the 1980s, and how this long-planned strategy is now attacking public health efforts against childhood obesity.

But who is working the long-term game on behalf of public health? Academics are amateurs in this game. Many foundations and think tanks have strategic vision, but many shy away from litigation. The friends of tobacco, booze and fatty foods haven’t been shy. And they are winning.

Research says: Obesity is a problem, but solving it doesn’t fix our health care spending problem, just a very small step in the right direction.

After Aaron’s fat-laden post on the “food” at the Indiana State Fair, The Lancet released a thoughtful issue on obesity, including estimates that obesity-related health costs in the US will reach $28 billion in 2020.  Wang YC, McPherson K, Marsh T, Gortmaker SL, Brown M.  Health and economic burden of the projected obesity trends in the USA and the UK.  The Lancet 2011;378:815-25 (gated). This sounds like a lot until you compare it to US NHE in 2020:

Sources:  Wang YC, et al. Lancet 2011;378:815-25 ($28 billion); US NHE, 2020, table 1.

More obesity news from The Lancet:  experts have underestimated how significant calorie reductions need to be in order to lose sustained weight.  The new models call for daily caloric reductions in range of 30 – 40% for very significant weight loss (ie, a 220 pound sedentary man reducing his 3000 calorie/day diet down to 1800 per day, in order to lose 44 pounds in 6 months).  Hall KD, Sachs G, Chandramohan D, Chow CC, Wang YC, Gortmaker SL, Swinburn BA.  Quantification of the effect of energy imbalance on bodyweight.  The Lancet 2011;378:826-837 (gated).

Bottom line:  very significant reductions in US caloric intake will save 0.6% of health care costs by 2020.

Aaron’s 2010 posts (the blame du jour) and (I’m not saying obesity isn’t a problem) reached similar conclusions.

UPDATE:  Sarah Kliff, writing over at Ezra’s place, covered the same studies. Everything she writes correctly reflects the articles. And yet, our posts are very different.

UPDATED – the case remands back to the district court – see at the bottom

Certificate of Need (CON) laws are a holdover from President Nixon’s foray into health planning:

The concept of certificate of need regimes, which many states enforce, is to avoid private parties making socially inefficient investments in health-care resources they might make if left unregulated. A certificate of need program corrects the market by requiring preapproval for certain investments and, in theory, thereby ensures that providers will make only necessary investments in health care.  Yakima Valley Memorial Hospital v. Washington, 2011.

President Reagan signed the repeal of federal support for CONs in 1986, but many states kept CON statutes on the books (see the detailed 50-state survey by NCSL), including the State of Washington.

The Yakima Valley Memorial Hospital occasionally performed percutaneous coronary interventions (PCI) in emergencies, but wanted to offer the procedure on an elective basis.  Examples of PCIs include moneymakers such as stent implantation and laser angioplasty.

Responding to evidence that higher procedural volumes improve quality, the State of Washington gave PCI licenses only if it projected 300 procedures a year at the facility. For Memorial, that meant that the only local PCI license would remain in the hands of its competition, the for-profit Yakima Regional Medical and Cardiac Center.  Unless Memorial could prove an unmet need for at least 300 additional PCIs a year in Yakima, it will never get a license to compete.

Health care lawyers see situations like this every day, with CON laws essentially defending market power and preventing competition, but also arguably improving quality by concentrating complex procedures into focused factories. Memorial loses this case under state law.

But Memorial had exceedingly clever attorneys, who won last week at the 9^th Circuit with a novel argument:  state CON laws violate the dormant commerce clause:

Here, the barrier to interstate commerce is the requirement of a certificate of need to offer elective PCI to all patients, instate or out-of-state. By virtue of the certificate of need requirement, the Department prevents Memorial from soliciting out-of-state patients and competing in an interstate market to offer elective PCI services, activities that clearly involve interstate commerce. See Summit Health, Ltd. v. Pinhas, 500 U.S. 322, 329-30 (1991). Under Pike such incidental effects on interstate commerce are an unconstitutional barrier to trade if they are “clearly excessive in relation to the putative local benefits.” Pike, 397 U.S. at 142.

Unless the Supreme Court takes the case, or the 9^th Circuit reverses en banc (ie, the entire 9th Circuit agrees to hear the case instead of just the 3 judge panel), many CON laws are now unconstitutional in the 9^th Circuit, which is a radical earthquake to many local health care markets.

UPDATE #2:  As pointed out in the comments, this case remands to the district court now, where the case will be reheard to see whether the PCI license restriction violates the dormant commerce clause under the standard articulated by the 9th Circuit. We’ll have to wait and see what the district court says, but the case opens a constitutional front for all CON litigation in the 9th Circuit (especially in Washington), which is still a big deal.  Apologies for my error and thanks for the comment.

UPDATED: To explain en banc

The University of Cambridge is so quirky – I love it. In October, Cambridge will choose a new Chancellor of the University, to replace HRH the Duke of Edinburgh (Prince Philip, Queen Elizabeth’s husband), who has held the post since 1976.

The insider candidate is Lord Sainsbury of Turville, who was Chairman of a large British grocery story chain – wait for it – Sainsbury’s. This is where it gets silly. Any 50 Senate members can nominate candidates, and the Senate includes not only faculty, but almost anyone with a graduate degree from Cambridge. Believe it or not, I get to vote.

Lord Sainsbury faces competition, which now includes a well-known local grocer from Mill Road, Cambridge, Mr. Abdul Arain. I lived around the corner from his shop and can attest to his fine selection and friendly service. His prime reason for running for Chancellor of an institution with more than eight centuries of history? To oppose a new Sainsbury’s grocery store near Mill Road.

The two other nominees are Michael Mansfield, described as a republican, socialist, vegetarian, radical lawyer and British actor Brian Blessed, well known to all Black Adder fans [video].

This isn’t an election to sit out, but you have to vote in person, at the Senate House, in a gown (any black gown will do, “full academical dress” is not required, according to the FAQs). So medieval.