The Incidental Economist

After documenting the rapid rise in number of clinical studies, authors of the Institute of Medicine’s Knowing What Works in Health Care (page 36) conclude,

For physicians—and patients—who are motivated enough to read through and assess all of the relevant individual clinical studies on their own, keeping current is an arduous, if not impossible, task. Given the variable quality of the research and its limited generalizability, these providers and patients are faced not only with reconciling vastly different research findings but also with scrutinizing each study’s methodology in detail to ensure that the study has been well designed, that the analyses have been well performed, and that the results apply to their particular clinical circumstance (Abramson, 2004). This expectation is unrealistic, especially given that today’s medical residents frequently lack the knowledge in biostatistics necessary to interpret the findings of published clinical research (Windish et al., 2007). These findings illustrate the need for a system that can make sense of all of the data that currently exist, as well as the new knowledge that is now being generated.

Naturally, systematic reviews and clinical guidelines help, but the report documents an explosive number of those too, or none at all in some areas. This strikes me as a real challenge to the notion of consumer-directed health care, at least for consumers who want to make evidence-based choices. Those that just want to make it up as they go along need not be concerned about the volume of evidence.

But it’s worse than that. This strikes me as a real challenge for the very experts on which many of us rely, the physicians we consult for advice and care. Is it possible to assess the degree to which one’s physician is keeping up with the literature? There can’t be very many of them that do so thoroughly, given the challenge of the task and the fact that they must spend some of their day actually delivering care. They probably sleep some too, and eat, and have lives, as they should!

I’m not sure what to do about all this, since the volume of  clinical studies is likely to keep rising as we fund more of them. We need those studies. But we also need the information to be accessible and usable. I know PCORI is all over this concept, but the jury is out as to whether it can do anything substantial about it.

No doubt there are information technology solutions. That’s also probably another domain into which we could sink a lot of money for little return. Just as throwing an iPad at a high school student doesn’t cause him to learn calculus, putting everything into a searchable database won’t make doctors and patients sudden masters of the evidence.

There’s a real need for answers. The wealth of our nation and our well being depend on them. I don’t have any right now.

AF

Declines in Employer-Sponsored Insurance between 2000 and 2008: Examining the Components of Coverage by Firm Size, by Jessica Vistnes, Alice Zawacki, Kosali Simon and Amy Taylor (Health Services Research)

Objective:  To examine trends in employer-sponsored health insurance coverage rates and its associated components between 2000 and 2008, to provide a baseline for later evaluations of the Affordable Care Act, and to provide information to policy makers as they design the implementation details of the law.

Data Sources:  Private sector employer data from the 2000, 2001, and 2008 Medical Expenditure Panel Survey-Insurance Component (MEPS-IC).

Study Design:  We examine time trends in employer offer, eligibility, and take-up rates. We add a new dimension to the literature by examining dependent coverage and decomposing its trends. We investigate heterogeneity in trends by firm size.

Data Collection:  The MEPS-IC is an annual survey, sponsored by the Agency for Healthcare Research and Quality and conducted by the U.S. Census Bureau. The MEPS-IC obtains information on establishment characteristics, whether an establishment offers health insurance, and details on up to four plans.

Principal Findings:  We find that coverage rates for workers declined in both small and large firms. In small firms, coverage declined due to a drop in both offer and take-up rates. In the largest firms, offer rates were stable and the decline was due to falling take-up rates. In addition, enrollment shifted toward single coverage and away from dependent coverage in both small and large firms. For small firms, this shift was due to declining offer and take-up rates for dependent coverage. In large firms, offers of dependent coverage were stable but take-up rates dropped. Within the category of dependent coverage, the availability of employee-plus-one plans increased in all firm size categories, but take-up rates for these plans declined in small firms.

LESS IS MORE. Overuse of Health Care Services in the United States: An Understudied Problem, by Deborah Korenstein, Raphael Falk, Elizabeth A. Howell, Tara Bishop and Salomeh Keyhani (Archives of Internal Medicine)

Background:  Overuse, the provision of health care services for which harms outweigh benefits, represents poor quality and contributes to high costs. A better understanding of overuse in US health care could inform efforts to reduce inappropriate care. We performed an extensive search for studies of overuse of therapeutic procedures, diagnostic tests, and medications in the United States and describe the state of the literature.

Methods:  We searched MEDLINE (1978-2009) for studies measuring US rates of overuse of procedures, tests, and medications, augmented by author tracking, reference tracking, and expert consultation. Four reviewers screened titles; 2 reviewers screened abstracts and full articles and extracted data including overuse rate, type of service, clinical area, and publication year.

Results:  We identified 172 articles measuring overuse: 53 concerned therapeutic procedures; 38, diagnostic tests; and 81, medications. Eighteen unique therapeutic procedures and 24 diagnostic services were evaluated, including 10 preventive diagnostic services. The most commonly studied services were antibiotics for upper respiratory tract infections (59 studies), coronary angiography (17 studies), carotid endarterectomy (13 studies), and coronary artery bypass grafting (10 studies). Overuse of carotid endarterectomy and antibiotics for upper respiratory tract infections declined over time.

Conclusions:  The robust evidence about overuse in the United States is limited to a few services. Reducing inappropriate care in the US health care system likely requires a more substantial investment in overuse research.

How Can We Know So Little About Physician Referrals? by Mitchell H. Katz (Archives of Internal Medicine)

A First Look at the Volume and Cost of Comparative Effectiveness Research in the United States, by AcademyHealth

Massachusetts Health Reforms: Uninsurance Remains Low, Self-Reported Health Status Improves As State Prepares To Tackle Costs, by Sharon K. Long, Karen Stockley and Heather Dahlen (Health Affairs)

The Massachusetts health reform initiative enacted into law in 2006 continued to fare well in 2010, with uninsurance rates remaining quite low and employer-sponsored insurance still strong. Access to health care also remained strong, and first-time reductions in emergency department visits and hospital inpatient stays suggested improvements in the effectiveness of health care delivery in the state. There were also improvements in self-reported health status. The affordability of health care, however, remains an issue for many people, as the state, like the nation, continues to struggle with the problem of rising health care costs. And although nearly two-thirds of adults continue to support reform, among nonsupporters there has been a marked shift from a neutral position toward opposition (17.0 percent opposed to reform in 2006 compared with 26.9 percent in 2010). Taken together, Massachusetts’s experience under the 2006 reform initiative, which became the template for the structure of the Affordable Care Act, highlights the potential gains and the challenges the nation now faces under federal health reform.

A number of you have written me today, asking why I haven’t commented about the NYT piece yesterday on how insurance costs have risen so much from 2003-2007.

Not to be snarky, but the reason I didn’t was because I already wrote about it when the Commonwealth Fund Issue Brief the NYT piece is based on was released. Last year.

Please, go read my post from November 18, 2011 (which followed the publication of the report on Novermber 17). More evidence that blogging has value.

Did you ever pull some yarn on an old sweater and end up making it worse? Before long, you could unravel the entire sweater if you kept at it.

That is the legal strategy of Florida and the other states challenging the ACA, according to their “severability” brief filed with SCOTUS. If the Court finds one section of the ACA unconstitutional, the Justices must then decide whether the rest of the law survives (or is “severable”).

Florida’s starting point is the minimum coverage provision (aka the individual mandate) – and from there they start pulling the yarn. First to fall – according to Florida’s brief – will be the guaranteed issue and community rating rules, mainly because the insurance industry wouldn’t have accepted these bitter pills without the sweetener of 32 million more customers. The insurance industry makes the same argument in its briefs. The 11th Circuit agreed, but stopped here.

Once these three provisions are gone, Florida’s brief sees no way to keep any of the private insurance reforms in the ACA:

Simply put, without guaranteed issue and community rating, the impetus for the ACA would disappear, and the Act’s whole private insurance expansion would unravel, for insurance companies would remain free to turn away millions of the very same individuals to whom the Act promised insurance. (at 48)

As these “core components” fall, so does every last bit of the Act:

The ACA is a delicate balance of inextricably intertwined provisions, none of which can survive without the Act’s core components. (at 35)

Even the clearly constitutional revenue-raising provisions should fall if the Medicaid expansions are struck down. Why? Because Congress only wanted the money to pay for Medicaid expansion:

The massive expansion of Medicaid was a costly endeavor that Congress attempted to counterbalance with projected cost savings.  If the Medicaid expansion is invalidated directly or falls as a consequence of invalidation of the individual mandate, then these offsetting provisions cannot survive while respecting Congress’ intent. (at 26)

The upshot (according to Florida) is if any word in the ACA is unconstitutional, all of it must burn:

…calling into serious question whether Congress would have passed health insurance legislation at all if even a single word of the ACA was altered… (at 39)

Meanwhile, in unrelated news, Republicans in Congress are drafting legislation just in case the Supreme Court strikes down the ACA.

KO

I want to add a note to Austin’s post on peer-reviewed literature versus blogging. I think that there is also a long-term benefit to blogging as well as a short-term one. More people read things that are timely and on point. The more people that read what we write here, the more likely that we are able to influence people with our arguments.

We’re all trying to make the world a better place through research. But unless that research, and the knowledge it generates, actually makes it into discussion, it might as well not have been done.

As Mark Twain said, “80 percent of life is showing up.” In many ways, this work allows us much more an opportunity to “show up” than being published in journals does. That matters.

From the debate last night:

ROMNEY: Rick, I make enough mistakes in what I say, not for you to add more mistakes to what I say. I didn’t say I’m in favor of top- down government-run health care, 92 percent of the people in my state had insurance before our plan went in place. And nothing changes for them. They own the same private insurance they had before.

And for the 8 percent of people who didn’t have insurance, we said to them, if you can afford insurance, buy it yourself, any one of the plans out there, you can choose any plan. There’s no government plan.

And if you don’t want to buy insurance, then you have to help pay for the cost of the state picking up your bill, because under federal law if someone doesn’t have insurance, then we have to care for them in the hospitals, give them free care. So we said, no more, no more free riders. We are insisting on personal responsibility.

Either get the insurance or help pay for your care. And that was the conclusion that we reached.

SANTORUM: Does everybody in Massachusetts have a requirement to buy health care?

ROMNEY: Everyone has a requirement to either buy it or pay the state for the cost of providing them free care. Because the idea of people getting something for free when they could afford to care for themselves is something that we decided in our state was not a good idea.

SANTORUM: So, in Massachusetts…

(APPLAUSE)

That last line there was the audience, made up of Republican primary voters, applauding a full-throated defense of the individual mandate. I’m somewhat amazed at how well Governor Romney can defend the inherent structure of health care reform.

This should be an interesting election year.

Something rare happened to me on Wednesday. In the span of a few hours a post I co-authored with Aaron Carroll published on JAMA’s blog, and a commentary I co-authored with Henry Aaron published in NEJM. This turned out to be a great, admittedly small sample, natural experiment of blogs vs. journal publication.

Both pieces were short, ungated, and accessible to a general health policy audience. Both were promoted in essentially the same way on the authors’ end (emails to journalists after the pieces appeared, announcements on this blog [here and here] and on Twitter). But one received a lot more attention, as measured by links from prominent journalists and institutions, than the other. Can you guess which?

The answer is that the blog post at JAMA received more attention. It’s not surprising, actually. The JAMA piece was in sync with the media;  it was about the State of the Union Address, which had occurred the night before. Aaron Carroll (I have to use his full name in posts that also mention Henry Aaron) had drafted it just after the Address on Tuesday night, and he and I refined it Wednesday morning. JAMA approved and posted it by early Wednesday afternoon. The wider media was still interested in the subject that afternoon, evening, and the next day.

In contrast, the NEJM piece with Henry Aaron was not on a topic of immediate interest, premium support. When Henry and I wrote it, in December, premium support was more topical. The holidays and the process of review and preparation for publication delayed its appearance until a time when premium support wasn’t as hot. It published at 5PM on Wednesday. That’s not NEJM’s fault. It’s just the way peer-review, journal publishing works. It takes time.

Had President Obama mentioned in his speech on Tuesday anything about Medicare that hinted at a political contrast over premium support, the issue might have been current enough for the NEJM piece to garner some attention. But, to my surprise, he did not. Again, not NEJM’s fault. The editors could not have known the issue climate at the time of publication. That’s my point. Journal articles basically come out randomly with respect to the timing of policy debate. Blogs can be better timed.

Premium support will be hot again, and I trust that, in time, the NEJM piece will get its due. Nevertheless, this was a beautiful example of how blogs can disseminate relevant information much more efficiently than journals. That’s not to say traditional journal publication, or something like it, is not important. Slower, deliberate, peer review is important. I would not want to lose it, or something like it. But there is a cost.

There’s a gain too. I get to list the NEJM piece in the peer-reviewed portion of my CV. It counts for promotion. Blog posts don’t, or if they do, they count much less. In the long run it matters more, at least to me, that I have a peer-reviewed piece in NEJM (or anywhere). But in the short-term the JAMA blog post got the attention of the day. When that’s important, if that’s important, blogging wins.

AF

• NASI: Social Insurance in a Market Economy, 9-10am 1/27 National Press Club Wash DC (5 concurrent panels)

Today in TIE: Aaron on the persistent link between unemployment and uninsurance, and Austin on clinical guidelines (too many, too few, or both?).

DT

Just as there is, simultaneously, over- and under-use of health care in the US (about which, more another time), we also “enjoy” both too many and too few clinical guidelines. This passage, from page 8 of the Institute of Medicine’s Knowing What Works in Health Care explains:

The National Guideline Clearinghouse (NGC) maintained by the Agency for Healthcare Research and Quality includes clinical guidelines from about 360 different organizations. [...]

One of the challenges inherent in having a highly decentralized, pluralistic process for developing clinical guidelines is that multiple groups will produce guidelines in the same clinical topic area. Currently, for example, the NGC contains 471 guidelines relating to the topic of hypertension and 276 guidelines related to stroke. Despite the abundance of clinical guidance for some topics, there is little clinical guidance on other important topics.

This is, sadly, typical of the US health system. We have lots of duplication and inefficiency along side a dearth of information accompanied by tremendous need. This, despite many improvement efforts by dozens of agencies and organizations over many decades and at great cost. It’s no wonder that some throw up their hands and suggest that we just let the free market, price system sort it all out. After all, the information conveyed by prices orchestrates much of the rest of the economy beautifully, all by “invisible hand.” What could possibly go wrong?

On the other hand, it’s rational to think nothing of the sort could work in health care, or not health care as we know it anyway. The market failures are too numerous and severe. The allure of central planning, regulation, and guidelines is completely reasonable, to a point.

But we don’t have to be stupid about it. We don’t have to have hundreds of overlapping guidelines for disease X and zero for Y. If you’re not a clinician – and I am not — can you imagine keeping up with “the latest” when “the latest” is such a mess? If you are a clinician, I bet you can tell me some tales of woe on this front. Some already have.

I’m not pointing fingers here. I’m not sure who to point to. It’s a collective failure or a lack of vision. Maybe it’s built into a path-dependent, policy trap we somehow set for ourselves. Maybe we made some huge mistake decades ago, and now we’re locked into a system that fails us while strongly resisting improvement. Maybe it’s just the American way, baked in the cake.

I don’t know. Some days I just despair. Too much knowledge can be a dangerous thing, especially when it’s the same thing over and over. We’re doing it wrong. We’re doing it wrong. We’re doing it wrong.

AF

The CDC has released a report entitled, “Health and Access to Care Among Employed and Unemployed Adults: United States, 2009–2010“. Key finding:

Overall, in 2009-2010, more than 80% of employed adults age 18-64 had health insurance. But only 48% of unemployed adults in the same age group had insurance. But it gets worse. Less than 30% of unemployed adults had private insurance (less than a third!), versus more than three-quarters of employed people

Many people like to think that being uninsured is a “choice”. And they’re correct, in the sense that you can “choose” not to buy insurance. I get that. But many people “choose” not to buy insurance for the sole reason that it’s crazy expensive. The average – not gold plated, but average – employer sponsored insurance plan for an individual plan in the United States last year was $5429. And that was just the premium. It didn’t include deductibles, co-pays, or co-insurance. The average family plan was $15,073. The median salary in the US, on the other hand, was less than $50,000 for households. For individuals, the median paycheck is $26,364.

When you’re making that amount, and you lose your job, paying for that insurance plan is no longer possible. Paying for COBRA is even harder, as it’s usually more expensive. So sometimes you get poor enough to go on Medicaid. Or, if you’re in the majority of states that offer no Medicaid benefits whatsoever to adults who aren’t parents of children, you go uninsured.

In case you haven’t noticed, the economy is hurting. Unemployment is very high, especially in groups that tend to also need health care the most. At times like this, it’s hard to imagine attacking the safety net programs that protect the poorest among us. Without them, the uninsurance rate among uninsured adults would be much, much worse.

UPDATE: Fixed to make one point clearer.