Tuesday, May 29, 2007

Uncomfortable growth

An editorial in today's Globe raises questions of several types with regard to a plan by Massachusetts General Hospital to expand its emergency department, surgical facilities, and beds. Charlie Baker, CEO of Harvard Pilgrim Health Care, also raises a number of issues on his blog.

I had trouble knowing what lessons to draw from this commentary. I understand the authors' concerns about the cost of new hospital facilities, but rest assured, no one in the hospital business builds expensive new space unless there is a reasonable assurance of the demand being there to use it.

Here are the trends we see. Even if population growth in Massachusetts is minimal or flat for the coming decade, demand for the services of tertiary hospitals is likely to grow. Why? As the baby boomers age beyond 50, they have an increasing demand for hospital services. In addition, their parents are living longer than ever, and they, too, are heavy users of hospitals. Not only are both groups displaying greater utilization of hospitals, but their ailments are of greater acuity, resulting in an increased demand for tertiary care. (And by the way, if the people in the younger generation do not change their ways with regard to weight control, they will be in the hospitals also for care of diabetes-related health problems like vascular and heart disease.)

While I have not asked the folks at Partners Healthcare about this, I am guessing it is this demographic wave that leads MGH and Brigham and Women's Hospital to plan expansion of tertiary care facilities. The same is true for BIDMC.

I know people get uncomfortable with an additional aspect of the issue: There is also a business imperative for this kind of construction. The insurance reimbursement system rewards increases in volume, especially high-end procedures and surgery. It does not reward preventative care, nor cognitive specialties in which doctors examine you and then make judgements about paths of care. It does not, in particular, provide sufficient income to primary care doctors to spend the extra time with patients that might avoid hospitalization.

So if you are running an academic medical center and you would like to be financially healthy to sustain and enhance your tri-partite mission of clinical care, research, and teaching, your business plan is simple: Grow, and expand your tertiary lines of care in particular. If you stand still, the inflationary costs affecting your hospital will soon outstrip your revenues.

Gee, in this regard, hospitals are not all that different from other businesses. After all, who would complain if any corporation decided to expand capacity in its more profitable lines of business in anticipation of increased customer demand?

But hospitals are different in that they do not sell directly to the public and are not judged every day on the quality of their service. The intermediaries in this field -- insurance companies, the federal and state government, and employers -- mask the costs of health care that you or I actually use. And, there is virtually no way for a consumer to compare the actual quality of care delivered by hospitals: Almost all the publicly available data is out of date, based on administrative rather than clinical information, and embedded in hopelessly confusing websites. So there is really no market-based set of checks and balances on hospitals of the sort faced by other types of corporations.

I fear all this is leading to an unsustainable situation and that the academic medical centers are the ones that will feel the public's wrath once that is apparent. Why? Because we are the high-cost part of the health care system. There are legitimate reasons for that, which we could discuss at another time. But those reasons will not hold sway when the bill to employers, subscribers, and taxpayers gets just too high to pay.

To me, the remedy is clear. AMCs have to become the places that set the standard for quality improvement and cost efficiencies. They need to demonstrate that their value to society goes beyond the clinical care, research, and teaching they offer. On the quality side, they need to establish the science of care as an academic discipline that informs health care providers everywhere. On the cost side, they need to engage and adapt principles of organizational efficiency from other industries to make a structural change in their production model. Meanwhile, insurers and government payers have to support both components of the solution by rewarding hospitals that improve quality and reduce costs.


Anonymous said...

Paul -

Enjoyed your comments and wholeheartedly agree with them. I would love to see academic healthcare centers lead the way in terms of cost efficicency at the same time as they lead the way in other paradigms. Measuring outcomes among hospitals is a very difficult thing because the populations served are so varied, but running a better business shows up pretty clearly in the bottom line. I enjoyed your post very much.


Anonymous said...


General comment, not related to this post - you might find this interesting.

Anonymous said...

Paul – That was a very interesting and informative post.

To help us better understand the economics of running a large AMC, I wonder if you could address the following questions:

1. What percentage of your costs are attributable to medical research, and how much of those costs are covered by research grants and other funding sources not related to payments for healthcare services?

2. What percentage of your costs (including those related to the satellite facilities) are attributable to graduate medical education, and how much of those are covered by tuition payments from medical school students? How much of the costs must be covered from health insurers and patients? Also, what exactly did Charlie Baker mean when he suggested that funding graduate medical education with general tax revenue instead of through the payment system probably wouldn't work for AMC's?

3. What percentage of your inpatient admissions come through the ER, where patients, presumably, are in no position to make choices based on cost and quality information even if it were available? How much of the inpatient revenue is attributable to these patients?

4. What percentage of your costs are attributable to uncompensated care that is not offset by disproportionate share payments or other government reimbursements and how does that compare to other AMC's and non-AMC affiliated community hospitals in the Boston area?

Separately, as a society, I think we need to address utilization, especially within hospitals, in the following ways: (1) make living wills universal, (2) rethink what constitutes good, sound, medical practice at or near the end of life. For example, should we provide anything other than comfort care to patients whose dementia or Alzheimer's is so advanced that they can no longer even recognize family members? (3) Pass sensible tort reform like health courts to reduce defensive medicine, (4) Build a network of interoperable electronic medical records to reduce duplicate testing and adverse drug interactions, especially in hospitals, and (5) Offer robust price and quality transparency to help both patients and doctors access the most cost-effective providers.

With respect to health insurance financing, I think every employer should send every employee a statement each year that lists the employee's salary as well as the employer's cost for all benefits including health insurance as well as the employer's share of FICA taxes, 401-K matching contribution, disability insurance, etc. The cost of the health insurance policy should also be unbundled into a catastrophic coverage piece (costs above $5,000 per person per year) and an insulation piece (the first $5,000 per person of healthcare costs). That way, people would at least understand how much their health insurance is costing and that it is part of their total compensation that could otherwise be paid in wages if health insurance costs could be reduced. Finally, while I recognize it is politically unlikely, if not impossible in the current environment, I think the tax preference for employer provided health insurance should be eliminated with the savings used to reduce income tax rates, increase the standard deduction or some combination of the two.

Anonymous said...


I am not sure I can provide complete answers in this format. We'd have to go through full sets of accounts to show you the various funds flows.

Let me mull this over to see how it might best be portrayed.

Anonymous said...


As usual bc asks some excellent and very pertinent questions. However, it's not your hospital board asking, and I think guesstimates would do. I would think AMC's must have a rough idea of how training and research expenses affect them?
Another question is, how are AMC's treated differently than community hospitals in terms of reimbursement? I have heard that AMC's get a higher reimbursement for any given procedure or medical condition than community hospitals - presumably to cover their training costs.

As far as new construction goes, this is certainly not limited to Boston; it is a nationwide phenomenon that hospitals (both academic and community) are on an unprecedented building boom. Although the familiar justification of aging population makes sense, I wonder if it's not more about market share. Almost all this new construction is of private rooms, with many new amenities, making hospitals seem more like hotels. Although I am no friend of health insurance companies, I can see why that makes them uneasy. It seems similar to an employee on a business trip who picks the most expensive and nicest hotel and then bills it on his expense account - how long do you think the employer would allow that to go on? It just seems illustrative of how cockeyed our health care system has become - the patients, who do not pay for their hospitalizations, "demand" private rooms.

Regarding your last paragraph about AMC's leading the way on quality and efficiency - hear, hear! For too long they have rested on their arrogant laurels when, in reality, they tend to be less efficient, therefore more costly, in their processes than community hospitals (partly due to their training obligations). The outcome results being published by the Cleveland Clinic are a good start and I encourage you to use your influence to further your great idea.

Anonymous said...


Pursuing your comments in previous posts and in Charlie Baker's blog concerning the availability of current quality data, I just checked out the HHS website "Hospital Compare" (google it and you'll get the URL).
Just for fun, I picked BIDMC's name and that of its most well known neighbor in Boston, and found rating information on a number of quality measures for conditions such as heart attack, pneumonia, etc. I am happy to report that BID met or exceeded (largely exceeded) the Mass. average on all the measures except one. The other hospital was close but not quite as good.
First, congratulations. Second, do you not consider this to be adequate, albeit limited, quality information available to a consumer? I do confess, I lost track of the currency of the data presented. But it looked pretty useful, although one might quibble with what quality measures were presented.

Ken Farbstein said...

Thank you for your observation that "AMCs have to become the places that set the standard for quality improvement." Kudos to BIDMC as the flagship of the multi-hospital system that did the hard work and earned the first annual Premier Award for Hospital Medication Safety. Too few hospitals have invested their money in improving quality. As you say, the financial incentives are to get bigger, not necessarily better.

Anonymous said...

Mr. Levy,

I am struck by the incongruity of two points you make in your post: That patients (the people who come in the door and consume services) are essentially price indifferent and that AMCs ought to be models of cost-effeciency going forward.

I'm not knocking the latter as a laudable goal (it is), but how realistic is it for an AMC to pursue a goal that patients don't have any incentive to care about?

If I'm a boomer or the parent of a boomer, I look in the NYTimes and see a (highly misleading) write up on why tPA is best for all stroke patients or articles about why I should ignore every cost effective cancer screening guideline and get tested up the wazoo every chance I get. Every signal I receive in my daily existence reinforces the impression that maximal healthcare consumption is best and there is virtually no competing const constraint placed upon me: whether I demand 100 tests or cost-effective care I pay the same amount and I clearly prefer to consume more healthcare.

Given all this, isn't an AMC billing itself as 'cost-effective' committing suicide in the marketplace for patients?

Anonymous said...

Anon 9:26: "Hospital Compare reports on 20 measures of hospital quality of care in the areas of heart attack, heart failure, pneumonia, and the prevention of surgical infections. More than 4000 hospitals have voluntarily reported data on quality of care provided from October 2004 through September 2005." My point exactly -- data from two years ago.

Anonymous said...

bpc --

Interesting point. Take a look at the experience of Virgina mason in Seattle, which achieved marvelous operating efficiencies and started to lose money as a result. They were able, tho, to negotiate a new type of reimbursement deal with their local insurer.

You are right -- absent a proper reimbursement system, it can be a losing proposition. I am hoping we can work with our insurers on this front.

Anonymous said...


Re Hospital Compare, thanks for doing my homework for me. (: I see your point about the data being old, but honestly one year old is about the best one could expect - monthly data doesn't carry much statistical significance, and many hospitals are very slow to compile such data internally so as to have it right out there at the end of the year. If they have to then submit it to the government or other centralized site, then there goes the other year used up before it's published.

One exception is the Cleveland Clinic, at which a good friend of mine had a mitral valve repair last November. Although from the D.C. area, he picked it partially based on its published cumulative 30 day op mortality rate for that procedure of around 0.3%.
By this past March, when I had reason to check their site again, they had already posted their 2006 statistics of 0.0% mortality (including him, thank heaven.) That's about as current as you're going to get.

Anonymous said...

You will note that the numbers I have posted on this website for central line infections are about a month old.

Hospitals collect all of these kinds of numbers on a current basis. Most websites do not use that data. They base their figures on administrative data sets that are older.

Anonymous said...

I know I'm getting argumentative, but as a potential patient I wouldn't care what your infection rate is this month compared to last month; it's your longer term trends I want to see compared to others' before picking your hospital. That said, I realize hospitals' quality can go way downhill, or uphill, in a short time period and I agree with your premise that hospitals should publish this data themselves rather than older administrative data sets.
Assuming, of course, that the data are believeable; who is checking on their accuracy? I hate to say that, but in modern life faking is possible.....

Anonymous said...

Thanks very much. It is not argumentative at all. Don't worry. (And even if it were, it's the blogosphere!)

Please check my post below on central ine infections: http://runningahospital.blogspot.com/2007/05/central-line-infection-report.html. I provide month-to-month, but also the trend over time.

Is that at all helpful? I value your reaction because we are all trying to figure out what is useful to people.

Of course, it would be better if other places would "play", too, but so far there is little interest in that.

As for lying, I guess it is possible, but someday it would come out, and the hospital would be much the worse for doing that.

Anonymous said...

Well, what info would be useful to someone trying to pick a hospital is a big, big topic. (And that's not even addressing the issue of the doctor.) Let me answer your narrow question first and say monthly data on your central line infections is more than I need to know as a patient, but no doubt useful to your staff. For my purposes, every 6 months would do or even yearly, since I want to see whether you're going up, down, or straight across.

As far as the larger question of useful data for the "consumer" (I prefer to call us patients); I would say the biggest concern of the patient has to do with comparing different hospitals, optimally against some kind of regional or national average. This is where I found Hospital Compare useful, in that they provided both the national and the Massachusetts averages, and one could directly compare BIDMC to "big hospital M".
There are two major flaws in their data; one is its age, as you aptly pointed out, and the other is, as I think you have also pointed out previously, is that these are process metrics, not outcome metrics. I would imagine there is some correlation, but ultimately I'd like to know what your overall outcome is for heart attack, heart failure, etc.(the conditions for which they measure process metrics.) This is where the Cleveland Clinic has made some valuable inroads, I think. But there has to be some central moderating body that both defines the measures used in the outcome data, and establishes the national/state averages.

The bottom line is that I want to know if my local hospital is good enough to treat me for my condition, or I need to go elsewhere. Numbers in a vacuum won't help me with that.

Having said that, don't stop publishing your data, because you are making an important political statement to others about transparency. I would even expand it a la Cleveland Clinic style if you can; their website is excellent from the patient's point of view.

Anonymous said...

Thanks very much. Check out Dartmouth-Hitchcock, also.

Anonymous said...


To put my money where my mouth is, here are the CC links:



Note one of them gives the consumer advice about how to evaluate quality.

I'll check out Dartmouth-Hitchcock; I live southward. Thanks.

Anonymous said...


Having moved to Boston 20 years ago from the rural midwest, I have obtained all of my medical care including primary care from a tertiary care system despite having a 45 year history of using community based services. Reason: I wanted the best care despite my health being generally good. Nonetheless, I have experienced a missed diagnosis, and, for three years, some of my doctors have been anchored on one diagnosis and thus missed a probable co-morbidity. I suspect that if I had community based care, I'd still be consulting specialists in a tertiary care system.

Anonymous said...

For sure, that's the way it should work, i.e., getting consults with specialists in the tertiary centers when needed. But there should be no need for your PCP to be affiliated with a specific tertiary care network for you to get access to the specialist you might want. That raises yet another issue, of interoperability of electronic patient records, which is not yet in place.

Anonymous said...

Ironic that insurance companies give you more incentives to expand emergency rooms, while patients get disincentives (higher copays) to go there rather than the PCP office. And we wonder why we can't get an appointment at the PCP office!

Anonymous said...

As one who was in the middle of lots of DON debates in the 1970s and 1980s, I am stunned by the deference given to the huge expansion at MGH. Have the community hospitals given up the fight?

Anonymous said...


I don't know what goes on in Boston, but in Baltimore Johns Hopkins is undergoing an $800 plus million renovation/expansion. It helped that it bought up half the existing community hospitals in the city over the last 20 years, so there's hardly anyone left to complain; just other tertiary care centers. Does MGH own any former community hospitals?

I think it's natural for patients to assume a tertiary care center is "better." Who is educating them otherwise, other than Paul?

Anonymous said...


Speaking of quality assessment, this link just dropped on me via a "commercial" Modern Healthcare emailed me. What do you think of this AHRQ compendium of report cards?
Where are they trying to go with this?

Anonymous said...

Interesting stuff...Three comments...
1) Medicare drives all provider reimbursement policy. If you don't like the way hospitals and/or docs get paid, start there. Medicare DRG's for inpatient services, Medicare RBRVS schedules for physician services, and Medicare codes and structures for outpatient services drive all reimbursement policy. As a general rule, Medicare favors specialty services and procedure-based services, and doesn't think across specialties. So - if you want a system that pays for "team-based care," supports primary care, and isn't so focused on rewarding new technology, you need to start with Medicare. No private plan, with the exception - maybe - of the Blue plans (which are huge and have huge market share), can do anything other than ride under the umbrella of Medicare payment policies. And as the baby boomers age into Medicare eligibility, this problem is going to get worse, not better.
2) We don't support the study health care delivery in this country - we support basic scientific research. If we really want to "fix" health care and come up with a more effective and efficient way of delivering care, we should start by creating a National Institute of Health Care Delivery, and fund it the way we fund the other national institutes. Right now, the federal government spends $100 on basic scientific research for every $1 it spends on studying health care delivery. As a result, our technology has out-stripped our capacity to understand it and leverag it. You want a more cost-effective and higher performing health care system? Tell the feds to fund the study of health care delivery. It'll change medical school behavior and health care delivery faster than anything else.
3) More and better publicly available information on health care services and delivery will help, but as I say this, I recognize that this is a multi-year effort on a sustained basis to get to the point where people can feel comfortable about both the information and what it means. That said, it's never too soon to start, and I, for one, am glad that the transparency movement appears to have support across the political spectrum.