Friday, February 27, 2009

Missing the lesson(s): NEJM gets it wrong

A recent NEJM article really missed the point.

NEJM -- Slowing the Growth of Health Care Costs -- Lessons from Regional Variation

They correctly point out that there is regional variation in health care costs. This is no secret. But why?

This article simplifies the reasons. Their response is that it is some output of "how physicians and others respond to the availability of technology, capital, and other resources in the context of the fee-for-service payment system". This ultimately led, in their thinking, to "Differences in the propensity to intervene in such gray areas of decision making were highly correlated with regional differences in per capita spending."

Health care costs can vary due to a number of reasons - not just because of payment. I'm not going to go through an exhaustive list. But let me highlight a few reasons I think exist for market variation:

-Market consolidation (i.e. there's only one dominant provider, so they can demand a higher price for services rendered for a payer)
-Certificate of need regulation could drive down capacity for technology could help reduce costs
-Percent uninsured in a state can drive variation
-Greater concentration of for-profit or physician-owned hospitals could lead to greater health care consuption
-States with higher acuity service capacity (i.e. ICU) will use these services more, and drive up cost (the "if you build it they will come" phenomenon)
- To the last point - states with higher capacity of any kind (Long term care beds, inpt beds, dialysis, etc) may have higher costs
- Medicaid drug purchasing policies vary from state to state
- States with lower workforce capacity (thus higher visiting nurse needs, etc.) may have higher costs
- States with more complex insurance regulation or higher requirements for insurance reporting may drive up payor costs
- Higher state public health expenditures could account for differences in health care costs

I could go on. But you get the point. It doesn't take a rocket scientist to see that this article has reduced the problem of variation in health care costs to a laughable conclusion.

And don't even get me started on their banal conclusion that health care variation calls for more "physician involvement". Really?!?

I must say I'm very surprised that the NEJM editorial board put out such a half-baked article.

Thursday, February 19, 2009

Racial disparities in health care

A recent study on racial disparities in health care found that urvival rates after surgery at teaching hospitals are worse for black patients than for white patients

Arch Surg -- Hospital Teaching Intensity, Patient Race, and Surgical Outcomes, February 2009, Silber et al. 144 (2): 113

Another interesting finding was that patients undergoing surgery at hospitals with high teaching intensity (defined as 0.6 resident per bed) had a 15% lower risk for death and a 15% lower risk for death after complications than patients at nonteaching hospitals (P< 0.001 for both comparisons). This is a topic for another time, but I think it reinforces the critical role that academic medical centers play in providing high quality health care in America, and though we pay more for care at AMCs, it appears we are receiving higher quality for that additional cost.

But back t0 equity in health care - as an internist, I'm most aware of the literature on racial disparities in cardiology. I haven't seen literature on surgical outcomes, but this is consistent with what is published in the internal medicine literature.

The Kaiser Family Foundation and the ACC did a retrospective literature review and found that

• Of the 81 studies investigating racial/ethnic differences in
cardiac care from 1984 to 2001, 68 find racial/ethnic
differences in cardiac care for at least one of the minority
groups under study.
• Of the 68, 46 find differences in cardiac care for all of the
procedures and treatments investigated, and 22 find
differences in cardiac care for some procedures and
treatments and not others.
• The 13 remaining studies include 11 that find no
racial/ethnic differences in cardiac care,1 and two that find
the racial/ethnic minority group more likely than whites to
receive appropriate care.

In all health system reform, we must think about how we balance access, cost and quality. Our national conversation has been frenetically touching upon all three (i.e. - health insurance exchanges, cost effectiveness panels, medical homes), but this research makes me think the ingredient of equity has largely been missing, or is sitting on the back burner. We must not forget this piece, as we push forth health reform in America.

Tuesday, February 17, 2009

Comparative effectiveness comes to the US

This article has reached fairly high on the NY Times most emailed...

U.S., as Part of Stimulus Bill, Will Compare Effectiveness of Medical Treatments - NYTimes.com

I'm actually surprised that this got put into the stimulus bill. A welcome addition, for those of us thinking and working on the issue. But of course, devil is in the details.

I am here in London, and talking to folks about this development and parallels to NICE, it's interesting to hear the UK perspective. NICE (National Institute for Health and Clinical Excellence) publishes clinical appraisals of whether particular treatments should be considered worthwhile by the NHS. These appraisals are based primarily on cost-effectiveness as measured by QALYs. Anecdotal, but I am told that ~75% of NICE's recommendations are around clinical pathways, not pharmaceutical approval. And NICE is primarily an advisory body, and must rely on NHS Primary Care Trusts (essentially the UK payor) to enforce the clinical pathways. But PCTs are hesitant to employ the pathways, because they worry about disturbing their relationship with the providers.

Can easily see the same scenario playing out in the US - advisory body, little to no regulatory power - but the difference is that Obama's comparative effectiveness panel will advise CMS, not small payers like the UK's NHS PCTs (there are some 100+ PCTs for a ~75 million population! (that's very decentralized, if you're wondering)). For better or for worse, CMS has a bully pullpit from which to make major changes to the way health care is practiced in the US. I am eagerly awaiting the details...

Thursday, February 12, 2009

The Independent Physician -- Going, Going . . .

Follow up to yesterday's post - today's NEJM is about the same issue - consolidation of practices -

NEJM -- The Independent Physician -- Going, Going . . .

This article focuses on impact on quality of care, something I haven't done much thinking on.

" The only aspect of this question that has received attention from researchers has been the effect on the quality of care, and the results have so far been mixed. One recent report found that hospital-owned or health plan–owned physician organizations were more likely than independent physicians to engage in quality-improvement or health-promotion activities.2 Others have found no significant difference in quality of care between independent and employee physicians3 or that quality of care was somewhat higher among independent physicians.4"

Will be interesting to see how this literature evolves. Seems like there is a parallels from inpatient research investigating restriction to physician work hours, requiring more patient hand-offs (with mixed findings on impact on quality of care). But on the other hand, being part of a bigger practice affords you the opportunity to leverage scale to engage in quality improvement measures.

Wednesday, February 11, 2009

EMRs, operational assistance and physician practice size

Electronic Health Records: How to Spend the Money Wisely - Bits Blog - NYTimes.com

Fifty clinical leaders submitted a letter to the White House arguing that 1) EMR is necessary to improve quality 2) Most physicians are in small practices therefore 3) the "challenge is going to be all about implementation"

So are most physicians in small practices?

Yes - but there's consolidation of physician practices.

The proportion of physicians in solo and two-physician practice decreased significantly from 40.7 percent to 32.5 percent between 1996-97 and 2004-05, according to a national study in August 2007 from the Center for Studying Health System Change (HSC). Physicians increasingly are practicing in mid-sized, single-specialty groups of six to 50 physicians. Despite the shift away from the smallest practices, physicians are not moving to large, multispecialty practices, the organizational model that may be best able to support care coordination, quality improvement and reporting activities, and investments in health information technology.

What's the implication for EMRs?

The letter is correct to state that these small physician groups are going to need assistance in implementing EMRs - a process that typically a one to two year process and decreases physician productivity (thus revenue). However, again, even if you look at the Stark Law relaxation which essentially allows physicians to receive EMRs from their employers (prevented previously b/c of antikickback concerns), free has not been good enough. Another example which I can't pull the hard data from at this moment is eprescribing - a large midwest payor gave away eprescribing, and uptake was abysmal, even though free, because the product supposedly did not improve physician productivity, revenue or quality of care.

The value just isn't there for physicians to adopt EMR as is - the product itself needs to evolve to help in performance, not just in connectivity. Operational assistance is only part of the problem here. Follow the EMR tag on this blog, and you can find some more explanation on this concept.

Tuesday, February 10, 2009

What is performance?

Nupe - you bring up a good point about reforming health care reimbursement. Some random thoughts -

Leapfrog has put together a list of pay for performance programs in their Incentives and Rewards Compendium.

If you look at the list, you find that there is a dizzying array of programs. As of 2007, there were 148 different P4P programs, growing historically at a CAGR of 21% and projected to grow to 160 programs by 2009.

These programs come from all sources: federal government, employers, medicaid-only plans, commercial health plans.

And the programs utilize a number of different metrics to measure "performance"

-Experience based metrics such as volume targets for PCI and CABGs
-Outcome metrics suck as mortality, complications and readmission rates
-Binary process or structural metrics such as the presence of an electronic medical record, or submission or Rx through e-prescribing
-Classic process metrics such as percent of acute myocardial infarction patients receiving aspirin on arrival
-Process metrics with time sensitivity such as PCI within 90 minutes of acute myocardial infarction

Physicians invariably interacting with numerous payers, and payers are all defining "performance"through a wide variety of metrics .

It's only becoming more difficult as a provider when you are faced with numerous, often conflicting metrics from a number of payers.

Side note: interestingly enough, most of the metrics are around payer cost centers - i.e. pneumonia, cardiovascular disease, diabetes - but other large cost centers such as depression and GERD don't have many metrics. Harder to measure quality using most of the five broad categories from above, and unclear if you can drive down cost and increase quality by implementing metrics on these 'lifestyle' conditions.

More on this to come...

Physician Compensation

So funny you wrote about what doctors do and maybe should get paid. I was just thinking about his issue yesterday, but from a slightly different angle. I was wondering more about why cardiologists, dermatologists, and radiologists get paid so much more than PCP's. I mean, its one or two more years of clinical training but lots more money. No one has addressed why these lifestyle fields are compensated so much more. It's not necessarily because they're better or more valuable to the broader society, and you could argue that primary care is more important than another radiologist. There is much more salary parity in places like the UK and Canada, which is why there are so many more PCPs in socialized nations. Obviously this disparity has to do with how medical care is reimbursed (procedures vs. office visits, curative vs. prevenative care, etc). There are lots of health policy implications to physician reimbursement, especially with regards to the push for medical homes. Check out this interesting pilot in Arizona. They're actually (eek) paying people for preventative care...

http://www.nytimes.com/2009/02/07/business/07medhome.html?_r=1&scp=1&sq=medical%20homes&st=cse

Monday, February 9, 2009

Is physician compensation different than executive compensation?

I had dinner this weekend with a number of bankers from Goldman Sachs, this weekend in London. Inevitably conversation turned to the recession. The bankers quickly steered the conversation and their anger towards Obama's recent proposal to cap TARP recipient executive compensation

U.S. Plans $500,000 Cap on Executive Pay in Bailouts - NYTimes.com

The bankers argued that compensation is critical and paramount in recruiting and retaining top talent in banking and finance.

This got me thinking - is the same thing true with physician compensation? It can't be - or else we would be a nation of interventional cardiologists. Granted there's a supply restriction, but you see top quality students in supposedly easier professions such as family medicine, primary care and pediatrics. And we always hear about there's a rising trends of medical students gravitating towards professions high on the lifestyle-compensation matrix: "dermatology, radiology, anesthesiology". This JAMA paper brought a lot of attention to this trend.

You also get studies such as this one that show there is no association between PC residency choice and debt. The authors conclude that medical students make residency decisions on the basis of a complex set of factors.

So are physicians drawn to their specialty due to compensation, lifestyle, or some complex mix of factors? Is it specialty specific - i.e. dermatologists are all very similar in their desire for high compensation, high lifestyle?

For better or worse, in classic consulting style, some colleagues of mine did a conjoint analysis to determine what are the archetypes of physicians. We found through our interviews of over hundreds of physicians that there are broadly six types of physicians. Take with a grain of salt, but this is one classification set:

Financially oriented physicians
Wealth maximizers:
Seek ways to increase pay even if it involves risk, least interested in non-financial rewards
Financial security guards: Value security, less reward from pure act of patient care, highly risk averse
Industrious clinicians:Value rewards of hard work, committed to medicine, appreciate respect and admiration

Non-financially oriented
Caring healers:
dedicated to patient care, public health, and teaching, least value on money
Academic clinicians: Teaching and contributing to science most important, less concerned with hours or income
Lifestyle conscious: Most likely to cut back hours, less reward from patient care or academic pursuits

Each physician has a different set of values, and will be drawn to work to an institution that can best fit his or her values. By understanding your hospital's goals, you can determine what is the mix of physicians that you want to hire and what you can offer to attract the right talent.

How does this get back to executive compensation? The bankers may be right - perhaps all banking executives are "wealth maximizers". However, my gut tells me that there must be other archetypes - ones who enjoy applying intellectual rigor to solving business challenges, those who enjoy the entreprenurial nature of building a business, those who enjoy being a caretaker for a large brand and its employees... and each of these archetypes is driven by something more than just money.

The compensation argument just seems to shallow. Perhaps it is true - that is the best way to recruit and retain talent in the finance and banking sector - but somehow that would just be unfortunate on many levels. I think the industry just needs to do a bit more self reflection to find out what are the drivers of happiness amongst executives, and tailor compensation accordingly.

Sunday, February 8, 2009

Public Private Partnerships

Here's a great story about a Dutch company Vestergaard-Frandsen that is making profits selling bednets and water-filters to poor people (via aid agencies). Apparently they're profitable, which just goes to show you - market forces can be leveraged to work on problems of the poor. Now, we just have to find people who are willing to pay for this kind of thing.

My favorite line:

“Very few companies take the attitude that doing good is good money,” he said. “They make a net, or they make a ceramic filter, and sell it. But make no mistake — as soon as we’ve proven this is a good idea, they’ll come in. They’re sitting there right now, watching us.”

Friday, February 6, 2009

On Rationing

Whew, this is a big topic. What got me thinking about rationing was Ruth Bader Ginsburg’s recent Whipple procedure for pancreatic cancer. For those of you who don’t know (seeing as only like three doctors read this blog, I’m sure you do know), a Whipple is a complex, 12 hour long surgical procedure where the pancreas is removed with all of it’s surrounding structures, and then everything is reattached back together. There are only a handful of places that can do this sort of thing well, and RBG went to Memorial Sloan-Kettering. Of course, everyone is also aware of Ted Kennedy’s recent brain tumor removal at Duke. As we saw on inauguration day, things aren’t going well for Senator Kennedy, but his physicians already knew that his cancer was going to relapse because the tumor margins were positive (meaning they left some tumor in his brain to preserve as much brain tissue as possible).

Ruth Ginsburg is in a slightly different position in that her cancer is felt to be localized to the head of the pancreas and can therefore be removed with a good chance of cure. Nonetheless, pancreatic cancer has a five percent five year survival rate, and that’s with appropriate therapy including surgery and chemotherapy. What people don’t talk about is the fact that she’s 75, she’s emaciated, and she’s already battled colon cancer. Most surgeons wouldn’t touch her because her perioperative mortality is sky high. But, since she is a Supreme Court Justice and has the ability to pay for the most advanced therapies, she can do whatever she wants. The same, of course, is true for Kennedy, and he had no chance for cure.

Now, I know what you’re thinking – it’s wrong to waste precious resources on old folks who are going to die anyway. I’m not so sure. The US healthcare system, for all its faults, is the best in the world in efficiently matching patient’s demand for healthcare with doctor’s/hospitals’/pharmaceutical’s supply. There is tremendous inequity no doubt, but for those who can pay for it, care is the best in the world. I’m sure old Ruthie would argue that she should get the surgery AND the poor guy down the street should, too. There is no reason, in our affluent society, that we can’t do both.

I can feel the socialists out there welling up with anger at the notion that our healthcare system is efficient and accessible. In some ways I agree, but let’s imagine that Ruth was in Canada. She would have waited months for the CT scan that was rapidly done executive-style at the NIH. OK, so maybe the scan was excessive (she had no symptoms to suggest she had an abdominal cancer), but it would never have happened anywhere else, and with an aggressive disease like pancreatic cancer, that spells a death sentence. I’m not sure what the right answer is. Some would say, not catching early cancers is just tough luck - we should spend money on procedures that have a lasting impact on quality and quantity of life. Others would say, if I’m paying for it, why shouldn’t I have access to the very best medical technologies at my disposal? Ultimately, I think it will come to a head when our entitlement programs like Medicare and Medicaid are asked to continue providing the same level of service as private insurers. Then, the taxpayer is footing the bill for that new, great technology, and America just can’t afford it for much longer.

Tuesday, February 3, 2009

Monday, February 2, 2009

Visualizing the uninsured

For health policy to build traction, we need ways to visualize those who are affected. Classic way to influence change - expert testimony goes only so far, then you need to influence through emotional appeal. See the article :

Questions for J.D. Trout - Dr. Feel It - Interview - NYTimes.com

The relevant piece:

"Interviewer: Why do you think people tend to feel more empathy for a puppy with a hurt paw than for a person without health insurance?

J.D. Trout: Part of the reason is banal. Ease of visualization. The person without health care is likely to appear as a statistic, one among 50 million others."

Sunday, February 1, 2009

On Creating “Diagonal” Health Systems

The climate for treating the diseases of the global poor has fundamentally changed. In the past five years, multi-lateral players like the WHO, the Global Fund for AIDS, TB, and Malaria (GFATM), the Gates Foundation, the Clinton Foundation, and the President’s Emergency Plan for AIDS Relief (PEPFAR) have poured billions of dollars into impoverished nations to stem the rising tide of preventable and treatable diseases. The time of standing idly by while millions die of HIV, TB, and malaria is purportedly over, yet, mortality rates are still sky-high and all this money doesn’t really seem to be working. I think it’s ultimately a failure of well-executed healthcare delivery.

One major obstacle to providing efficient, effective care is the conflict between so-called horizontal versus vertical programs. Horizontal programs have been traditionally provided by national governments. A good example is the UK’s National Health Service. While I’m not super-familiar with the inner-workings of the NHS, it seems to me that the primary care physician serves as a gate-keeper and coordinates all aspects of care. These include preventative services (vaccines, screening tests, etc), basic curative services, social work, counseling and mental health, and referral to specialists. The NHS model is touted as the blueprint for district-health services in Anglophone developing countries, but due to scarce resources (human and capital), health care falls woefully short. Some pundits argue that international aid for heath should go to strengthening these comprehensive horizontal programs contending that providing food, social support, housing, economic opportunities like microfinance are integral parts of improving health. The major problem with these programs is they take a long time to develop, require infrastructure and human resources, and, assuming fixed health sector spend, may provide good care to few while neglecting many. Good horizontal programs are very hard to develop and spread, and because of their bulkiness, do not do well adapting to local environments.

Vertical programs, as the name suggests, focus on one disease area, like HIV/AIDS. In contrast to horizontal systems, vertical programs are agile, able to scale up quickly, and often do one thing very well. Because they don’t provide a comprehensive package of services, they are able to adapt and change to local environments, can task-shift routine processes down to less-educated workers, and are able to develop simple tracking systems to follow patients over time. A good example of a vertical program is the Center for Infectious Disease Research in Zambia (CIDRZ). Funded by PEPFAR and GFATM, CIDRZ was able to rapidly put 100,000 people on AIDS therapy in three years at a relatively low cost (~230 dollars per person per year). The coverage in some places was 80% of the affected population. What they didn’t do was provide vaccines, accompaniment (a form of directly observed therapy), economic advancement, maternal care, etc. They were solely focused on responding to the AIDS crisis, and approached the problem like a disaster-relief organization. Get the treatment out. Fast.

Obviously, both approaches have their significant drawbacks, and most places are focusing on developing and sustaining vertical programs. In Uganda alone, there are over 600 foreign-based non-govermental organizations all doing different things, often overlapping in the same places. I suggest we start thinking about diagonalizing health care in poor countries. Julio Frenk, the new Dean of the Harvard School of Public Health and former Gates Foundation big-wig, has advocated for diagonality saying that the attention focused on HIV/AIDS provides a unique opportunity to capture “spill-over.” HIV is a big problem, no doubt, but once the frenetic initial response has finished, focusing on the broader health concerns of the population has to take precedence. If successful HIV programs like CIDRZ are in an area, consolidate services for TB, malaria, vaccine delivery, bednets, food, etc and combine the NGOs personnel under a unified umbrella. Of course, most organizations won’t like this, but it will take strong national governments with the guts to rebuke NGOs that don’t fall in line to make this happen. Country coordinating mechanisms must be set up, strengthened, and supported by funders otherwise we run the risk of having individual programs each with their own little fiefdom. If things stay the way they are now, everyone will think they’re doing good for individuals, but the population won’t get any better.