Could Price Be Right?

If confirmed as Secretary of HHS, Tom Price will oversee a $1 trillion budget – roughly one-third of all health expenditures.  His proposed legislation “Empowering Patients First” seeks to control costs by giving patients more choices and providing the information required to make them. He calls for publicly available standardized information on the price and quality of physicians, hospitals and other health care institutions.

It sounds like Dr. Price is prescribing a single data system. 

Medicare has had a single data system on the over-65 population for decades.  Since 2005, these data have informed Hospital Compare, a consumer oriented website comparing the quality of over 4000 hospitals.  And while prices in Medicare are relatively fixed, these same data have shown substantial variation in costs because the quantity of service – the number of hospital admissions, procedures and physician visits – varies substantially from place to place.

But Medicare is only one piece of the data puzzle.  A National Bureau of Economic Research report[nber.org] added another piece last year with data from large insurance companies like Aetna and United.  For the under-65 commercially insured population, it’s not just the quantity of services that are all over the map – it’s also the prices. 

We all have an interest in a sustainable health care system: one that provides high quality, needed care at a cost that doesn’t bankrupt public budgets – or worse, individual patients.  We need to learn from those communities that have found the right balance.  Rochester NY, for example, is among the least expensive health care markets in the country, for both Medicare and commercial patients.

But the only reason we know this is because a few insurers have been willing to share some of their data with researchers.

Unfortunately, much health information is still hidden from view – behind a proprietary curtain of privately held data. We rarely get to peek under that curtain, as researchers did last year.   Mind you, it took a heroic effort and still they were only able to look at less than a third of the private insurance market.  More than two-thirds remains a black box. 

That’s no way to run a railroad.  You want patient choice to encourage a better health care system in the future? Then the first step is for them to know what’s going on out there right now.  We need a single data system.

A single data system would make both price and quantity more transparent.  That alone might have the desired effect of dampening some of the price/quantity extremes.  It could also save billions in administrative costs by moving to a single uniform insurance claim. Furthermore, it would provide the opportunity to explore who is delivering great care at low cost – both to help consumers make choices and all of us learn how to improve the system. 

And a single data system is important for more than just saving money.  It’s important for the quality of clinical care.  Right now it’s hard to know what happens when a new drug, new procedure, or new device is introduced into the system.  No one can see the complete picture since different patients have different insurers. Then patients change employers and get a new insurer or they don’t change employers, but their employer changes insurers.  Or their insurance company merges with another one.

It’s a mess.  It’s a system that not only can’t monitor costs – it can’t monitor safety, something every orthopedic surgeon, including Dr. Price, cares deeply about.  The United States has probably implanted more metal on metal hip replacements than any other country in the world.  But we can’t say for sure, since we don’t have a single data system.  But we can say one thing for sure: we didn’t learn that metal-on-metal hips are much more likely to fail from US data, we had to import that knowledge from the United Kingdom – a country with a single data system. 

Dr. Price’s legislation envisions state-based portals for consumers to get information.  But why make 50 states build 50 different data systems? Patients move across state lines; they might appreciate one stop shopping. More importantly, so do insurers.  Four insurance companies – Blue Cross, Anthem, United and Aetna – now comprise of three-quarters of the national private insurance market.  That they would prefer a single uniform reporting requirement over 50 disparate requirements would seem to be a safe bet. 

We hope Dr. Price does the right thing: construct a single data system – so we can all know how our health care system does, and doesn’t, work.  

Dr. Welch is professor of medicine at the Geisel School of Medicine and author of “Less Medicine, More Health”.  Dr. Fisher is the Director of the Dartmouth Institute for Health Policy and Clinical Practice.

Categories: Uncategorized

10 replies »

  1. By the way, “a single uniform insurance claim” already exists and has existed for many years. The level of administrative inefficiency is small. As a Primary Physician for 40+ years, my receivable days ‘including patient payments’ for the last 5 years varied from 27 to 30 days because of the electronic claims process, honestly. And, the financial payment was posted directly to my corporate checking account by all institutional payers. An EOB was sent to our business office by Email from each payer.

    Defining the diagnostic label is quite another factor. Last year, the payer process (mandated by Congress) changed from a diagnosis coding system with @ 7,000 alternatives to one with 10 times as many. Other than the fact that all of the other developed nations used the new set of codes, I personally could really find no basis for any possible contribution to the management of our nation’s healthcare other than its negative impact on the cost of healthcare. We now are employing many people, just “to code” a healthcare encounter. Yup, believe it or not!

    Fraud is much more likely to be a problem than administrative inefficiency, especially for Medicare. Of the likely $800 Billion excess cost (as a portion of its $3 Trillion cost in 2015) within our nation’s healthcare, its administrative inefficiency was probably less than $0.5 Billion, significant but much less an issue than the fact that our nation’s healthcare is neither justly efficient nor reliably effective. The poorly definable lapses in quality are not resolvable by the isolated bursts of high quality healthcare, unimaginable 10 years ago, as in the replacement of the outlet valve of the heart without chest surgery.

  2. Historically, our national economy has a recession every 10 years. We are now 8 years into the current economic cycle. My perception of our healthcare industry suspects that a few major health systems are marginally stable and occasionally borrow funds for pay-roll to sustain their cash flow. My essential concern is that their lender also may be marginally unstable as in, the level of their capital reserves. With a recession, they may be suddenly forced to “call their short-term loans.” Just as suddenly, one or more health systems would close because our nation’s health care demands always diminish during a recession, a change in demand that is NOT fully understood. The hospital closings might be announced with a 30 day notice, initially by one then several more health systems. It is likely that a Federal “too big to fail” loan guarantee program would appear. Then, what would occur next. The chaos would be profound, especially on top of everything else.

    Clearly, nothing currently occurring within our public arena of healthcare reform has any chance to substantially improve any of its over-all needs: EQUITABLY available, CULTURALLY accessible, JUSTLY efficient and RELIABLY effective. The level of genetic diversity underlying healthcare is profound. Just remember there are 7+ Billion people living on this earth, 10 billion by 2050 (so says the UN). Other than twins et al, each one of these folks are genetically different. The impact on our healthcare industry is that the level of this genetic diversity is largely undefinable, with a relatively few well-known exceptions. Any attempt to define this for each person, for every change in HEALTH, is likely to be financially impossible. The technology for this may advance to solve the financial issues, but what about our ability to build the “caring relationships” necessary to accomplish this, let alone any unified national information system?

  3. “We hope Dr. Price does the right thing:’

    Is anybody in this regime going to do the “right” thing? Isn’t “imposing” a single data system what Repugnacans hate from big government?

  4. Anthem is Blue Cross, oops do you actually look at your EOBs, understand billing?
    United you mean United Healthcare? This is yet another unrealistic idea. There are MANY forces that would never let this happen. Ever. First being a lack of understanding of how healthcare is delivered by the authors (besides the Anthem/Blue Cross mistake). The system is going to fail and when it does, that is the ONLY time it will be ready for a fix. MACRA may push us over, but most will take penalties. 96% serious concern about MD burnout in NEJM is a big red flag, yet ignored. There is a new administration, and we have yet to see how they act. If its more of the same buzzwords and puffery language vs real regulation reduction we are in trouble. And in big trouble soon.

  5. It’s probably my ignorance, of which there is much, but I don’t understand what a single data system might be. As usual, God is in the details. Are we looking to get a clear look (transparent) into price, quality, and outcomes? Is the only way to do this is to have a uniform set of definitionals etc.? It sounds like an enormous federal project with potentially far reaching consequences.

    The first step may be simply to require price transparency. Should that be via state or federal statute? I’d probably opt for state so that there can be the federalist experimentation to see what works and what doesn’t. There’s something about a single federal repository that makes me antsy. When the authors and Tom Price say “standardized,” I suppose that reflects that even prices are difficult to compare when you get to complex bundles of services.

    Next I could see merit in having a single set of quality of care and outcome measures for CMS and the commercial insurers. Much easier on the providers and easier to compare.

    I wondering if the authors assumed that readers like me knew a lot more about the single data system than I do.

  6. UCSF had an hour long grand rounds on this very problem: how can we get everyone in health care to use the same language?…specifically, as a for-instance, “what is hypertension”. They were also thinking that uniform international definitions would be nice too.

    As I recall, the discussants rather gave up. Too many variables. Ages. Races. Genetics. Ambulatory vs non-ambulatory. Cultures who ate salt, for example. Occupation and background level of cortisol and stress. Too many instruments measuring blood pressure. Too much changing reasearch as to what is beneficial blood pressure compared to what is statistically normal for a population. “If we have to do this for every datum in health care we will be old men by the time we have anything useful.”

    I think, if we got rid of third party payers*, the providers themselves could work on this slowly and calmly–from a medical usefulness point of view. Then, big data and the mining thereof might be very useful.

    *E.g. everyone pays for ambulatory health care with Medi-bucks and the “insurers” for hospital care are [hopefully small] governments or hospital districts paying for a public good. No billing occurs from hospitals.

  7. Well, yes, this is a very nice research project, although if I may add, the Health Affairs reference is very old and most likely no longer accurate. We do have one common claim form and one common electronic standard for all the things mentioned there. We also have more than four insurance companies, and Blue Cross is not one insurer. Claim processing rules are complex and vary because that’s how insurers make more money. I would also be interested in calculating how much this standardization is going to cost taxpayers, insurers, clearinghouses, hospitals, doctors, and all other facilities, because those costs will be offloaded to patients in one form or another.

    The mention of Rochester, NY reminds me of how we wasted eight years of policy on similar “observations” regarding McAllen and ElPaso, which turned out to be inaccurate, to put it mildly. Another wild goose chase as a new administration takes the reins seems a bit ill advised to me.

    Furthermore, transparency is obviously a good thing, but it belongs to the tinkering at the margins category when it comes to making health care more affordable for people.

    I also suspect that Optum probably has all the information academics may need in this area. So why not ask them to turn it over to the public? The answer to this question is the same as the reason why we will not have “transparency” in billing as long as we have profiteering as motive in health “care” operations.

  8. “Single data system” wold be fine. I’ve been arguing for that for years. I’ve called long and loud for a “standard data dictionary” as part of ONC EHR certification. If “data are the lifeblood of health care,” they should be “Type-O blood.”

    But, the GOP Ayn Ryan wing’s larger and priority goal is to turn all “entitlement” programs into means-tested spend-down-to-poverty programs (with a lovely side dish of outlawing women’s reproductive rights while they’re at it — Price himself is one of those “Personhood-at-Conception” zealots). http://regionalextensioncenter.blogspot.com/2016/12/the-price-is-right-wing-that-is.html

  9. Drs. Welch and Fisher, thanks for discussing a single data system, I believe that perhaps now the time has come and hope your article leads down this path. In 2006 I wrote about it as well, essentially recommending that we establish a “Manhattan Project”.

    Here is an excerpt:

    “Imagine the possibilities A major component of this effort should be the funding and development of a National Health Care Informatics Center similar to the Lawrence Labs. This information center should house a supercomputer with a database of all patient records in the U.S., including claims, laboratory results, medical records and other services. All providers of services including physicians, hospitals and ancillary providers such as pharmacy benefits managers, laboratories, imaging centers and regional health information organizations should be required to submit data from claims, electronic medical records and other sources. The center should be organized along multiple divisions targeted towards unique areas of research and analysis, including evidence-based medicine, fraud detection, chronic illnesses management, prevention, pharmacy utilization, infection monitoring, surgical outcomes, and many others. To staff this national center we should recruit the best and the brightest in medicine, informatics, epidemiology, fraud and abuse detection, disease management, and statistics. A facility of this magnitude would become a gold mine of information. The outcomes from this center could then be given to the providers, patients’ government entities and others to implement changes and make adjustments to the system. Imagine if we had data on every diabetic in the U.S. and knew what medications they were taking, how often, what provider they were seeing and what their laboratory results were. The ability to determine if certain therapies were working or failing would be amazing. Besides being able to identify problem areas, we could target them for intervention and measure the results. Such a database would undoubtedly present many problems that would have to be overcome–including privacy concerns–but I believe these could be addressed.”