Monday, December 5, 2011

An ounce of prevention is worth a pound of cure....or is it?

Is it possible that Benjamin Franklin could be wrong? I tend to trust in the wisdom of this renowned polymath. He was a statesman, scientists, printer, inventor, author, philosopher, and diplomat. The only person I have ever known who is more knowledgeable in a wider variety of topics than him is the great Jason Kroening-Roche. However, could Franklin be mistaken such that prevention is not cheaper than the costs of cure? Intuitively speaking, preventing something today is less costly and should save a lot of time, energy, and money compared to dealing with the aftermath of full on disease manifestations. Joshua Cohen, however, makes a convincing counter argument against the cost saving effects of prevention, at least on the population level. Maybe an ounce of prevention is actually worth just an ounce of cure.

As an aspiring primary care physician, I often find myself making arguments in support of primary care on the grounds that prevention is cheaper, both financially and generally. Cohen does not reject prevention altogether, but he argues that our preventative efforts are neither as cost-saving nor cost-effective as we are inclined to believe. Smoking cessation, colorectal cancer screening, and flu immunizations are good preventative measures that protect a lot of people at a relatively low cost, but many diseases, which are not as common in the population, do not yield significant benefits from a cost standpoint.

Cohen and his team did a systematic literature review to look at the overall cost-effectiveness ratio (in terms of $ per QALY) of prevention versus treatment for existing conditions. The lower the ratio, the better as that indicates less cost for one quality adjusted life year (QALY) gained. His results are surprising. He found that the cost-effectiveness ratios for prevention and treatment were relatively the same (see graph above). He makes the argument that cost of treatment of disease is nearly as prudent as investing financially in prevention. He does state that preventative interventions, particularly those aimed at high risk populations, do in fact save money, but broadly generalizing that prevention is cost-effective or cost saving is not always the case.

I am intrigued by this article and found it surprising, but I have a couple criticisms. First, what is cost-effective and what is not is a value judgment. He may not think that spending $29,000 for one QALY on combination anti-retroviral therapy is a sensible investment, but I do. Next, he includes secondary prevention efforts as treatment. It seems that a lot of secondary prevention efforts would in fact prevent many diseases from coming to full fruition. He uses the example of an intracardiac defibrillator which cost $52,000 per QALY. However, there are many secondary prevention efforts, like statins for patients after they have a heart attack to prevent a recurring event, which yield good results at a low cost. I would be interested in seeing his analysis if he included secondary prevention and primary prevention together and compared it to tertiary measures. Lastly, Cohen does not tell us how he discounts the future costs of treatment or the future benefits of prevention. This could have a large effect on the cost effectiveness ratio depending on what discount rate he used.

I am not quite ready to discount Franklin’s wisdom on prevention as of yet, but Cohen has at least made me stop and reconsider my argument.


Cohen, J., Does Preventative Care Save Money? Health Economics and the Presidential Candidate, New England Journal of Medicine, 2008; 661-663


Tuesday, November 22, 2011

An Evening with Dr. Blumenthal

I just got home from a fascinating 3 hour talk with Dr. David Blumenthal. You can read more about him here, but he is basically one of the leaders in the quality field of health care and recently spent 2 years working in the Obama administration as the head of Health Information Technology (HIT). He came to speak to our US health policy class at the Harvard Kennedy School of Government and a smaller group of us had dinner with him afterwards. I wanted to share a couple things he spoke about that I found compelling.

Dr. Blumenthal answered one question about cost containment, which has been a hot topic in the national debate of late and something I've been reading quite a lot about. His answer was not novel nor complicated but brought my attention back to the reality of our health care system. 5% of the patients account for 50% of the costs. Atul Gawande wrote about this in his The Hotspotters article in January and countless others have shed light on this as well. Blumenthal's answer was simple: "Go where the money is." He spoke about better coordinating care, keeping these patients with multiple chronic diseases out of the ER and hospital, and actively pursuing them to better manage their care. Call them at home everyday, visit them in person, lay out their 15 or 20 medications for them, and fill in every gap where their care is dropped.

My reflection is this: If we are serious about reducing health care costs, and there is no doubt that we now are, we must work tirelessly to first find these patients in our health care system, even if they are not our own. We must proactively engage them in the medical system so that we are not caught reactively responding in extremely inefficient ways. We must think system-wide about the ways in which we currently fail to fill the gaps and bring all of the sometimes dozen providers for these patients into the discussion. And we must find innovative ways to engage the patients and their communities to embrace healthier attitudes and behaviors the are community driven.

Health Information Technology
We couldn't spend three hours with Dr. Blumenthal without talking about health IT, something he spent the last two years of his life on in Washington. Here are some of his reflections:

-Health IT is rapidly expanding across the country and has doubled in the last 2 years
-Competition and decentralization of HIT is good because it is driving innovation in the field. There are now over 1300 different private companies providing HIT and the innovation they are creating will have enormous positive impacts decades into the future. Quelling that force now in favor of a more unified system is the wrong thing to do.
-That being said, cross-talk is an important part of the future of our electronic medical records (EMR). Patients need to be able to take their records with them when changing locations and health systems. The government has put in place standards that are being adopted by the private HIT enterprise. Providers will also have to achieve some standard of meaningful use in the coming years to earn the substantial savings available through the stimulus bill that passed in early 2009. It is up to providers to put pressure on the health IT companies they work with to comply with these standards
-Privacy isn't nearly as big an issue as people say it is. Technology has been developed. Barriers to the sharing of information is primarily systems and politically generated.
-He doesn't have much sympathy for small practice physicians who are complaining about the cost of switching to EMR. The previously mentioned stimulus bill has provisions that will more than pay for their implementation, provisions that amount to $100 of tax revenue per American citizen to make this happen. He says this indicates that the public prioritizes it and it is high time providers do too.
-He is very optimistic that HIT will continue to grow, and it is necessary, but not sufficient, to attending to many of the problems in our health care system.

Dr. Blumenthal is a national leader in the quality movement in health care. He spoke at length about this during our structured lecture with him. Quality essentially was put on the map in the late 1990's after the Institute of Medicine's report in 1999 called "To Err is Human." This was followed in 2001 by another report from the IOM that Blumenthal described as a seminal document called "Crossing the Quality Chasm." Prior to these reports, quality was not on the policy map as much of an issue.

Quality is something that I admit I don't think about much. I tend to be of the mindset that our health care system is pretty darn good, but I often compare it to other places I've been in the world that have far worse health systems but also shoddy electricity. This fails to ask the important question, "How can we do better?" If I truly reflect on my medical training, I'm appalled at much of what goes on in the hospital and even in the outpatient setting. Care is so fragmented that providers are often very under-informed about the stage of care delivery a given patient is in or a recent change in treatment plan proposed by a different physician or service. The bottom line is that lots of bad things happen all the time with even more near-misses to make even the most inexperienced clinician worried.

This is what Dr. Blumenthal has been trying to remedy. So while I often push quality issues to the back of my mind, they are an area we can drastically improve on. We will save money if we keep people's blood pressures and hemoglobin a1c's under better control and don't order unnecessary costly imaging studies. We will save lives if we prevent infections in the hospital by washing our hands. And we will make much better clinical decisions if we have electronic records that remind us when we are doing something outside best practices. I think we can do a lot better to improve quality and I am excited that through better care coordination and the smart use of HIT we can work to accomplish so much more than simply containing costs. After all, our primary goal is helping patients get and stay healthy.


Monday, November 14, 2011

Price Discrimination and Hidden Negotiations

Uwe Reinhardt of the New York Times blog Economix recently wrote about an issue that has been perplexing me of late. How do hospitals and providers in the same state and even the same community charge such different prices for the exact same services they provide? Similarly but in the reverse, how do health insurance companies, seemingly competing with each other in the same market, pay such different prices for the exact same services? Isn't our free market health care system supposed to lower costs through good old-fashioned market competition?

I thought so, but it doesn't seem to.

The contribution of high and rising unit costs to our overall health cost crisis was the subject of the recent September Health Affairs issue. The same cost conundrum was detailed in 2003 and has been described in many other times and places as well. In addition to describing the problem, the recent Health Affairs issue also includes a section devoted to "strategies to cut costs" which includes the following ideas: a weight loss program, telehealth innovation, successful collaborative care models, and bundled payment reform. But haven't we seen this all before?

None of these ideas address the fact that price discrimination and hidden negotiations are contributing to the rising costs in our system. There seems to me to be three (overly simplistic) options to reduce high unit costs.

  1. Improve price competition among payers and providers. Massachusetts recently introduced recommendations on provider price reform. One recommendation was to open up the secret negotiations that now take place between hospitals and insurance companies in an effort to introduce transparency to the process. The goal is to reduce some of the variability that now exists by introducing real competition back into the "market."
  2. The second idea was also approached in the same recommendation set and is considerably farther left. Regulate prices. If the market can't be relied upon to drive competition, the other option is to strip competition from the system entirely. In this setting I can imagine some immediate concerns: heavy government control, decreased innovation, lower salaries, not mention a serious lack of political feasibility. There is a third option, however.
  3. Many countries (Germany, Switzerland, Belgium) operate systems whereby providers and payers negotiate prices on a regional level, but in a coordinated fashion. The negotiations do not occur between individual hospitals and insurance companies, but rather take place between formed coalitions. This encourages active participation, allows for regional variability, but eliminates the price discrimination and hidden negotiation practices currently employed in the US.
Even with a very introductory understanding of economics I am able to see that the current system of price setting is extremely inefficient. It neither allows the free market to function (which requires widely available information among all parties involved) nor for government oversight. In the current system monopoly (in markets where large providers dominate) and monopsony (in markets where large insurance dominates) run free. Both are market failures.

We must move from a hidden process of back-room deals to a more transparent and competitive system. If we are not able to, government price regulation may be our only choice, because our high and growing unit costs are entirely unsustainable.


Wednesday, November 9, 2011

Can the United States Learn from Brazil's Primary Care Model?

Brazil has seen huge changes in the past several decades in their political and social structure. Prior to democratization in the 1980’s, the country had been under several decades of military rule. After the authoritarian rule, socio-political reformation began to take shape which included an overhaul of their health care system. The belief that health is a universal human right became more widespread throughout the Brazilian culture and with this in mind, the National Health System (SUS) was created in 1988 and implemented in 1990. Since implementation of SUS, Brazil has seen huge improvements in health outcomes including increased life expectancy, decreased infant mortality, and decreased deaths from infectious diseases.

A cornerstone of the SUS is the Family Health Program (PSF) which functions as their primary care delivery system. The PSF was established in 1994 as a way to meet universal access goals, better coordinate care, and focus on prevention. The PSF’s focus is on family and community health which encompasses prevention and public health goals. Under PSF, family health teams, which consists of one doctor, a nurse, an auxiliary nurse, and 4-6 community health workers cover a geographic area which includes no more than 5,000 patients. Community health workers are vital to the team as each community health worker is responsible for about 120 families and they make home visits to each of these families once per month. The community health workers’ primary focus is on child and maternal health, but they have been instrumental in health promotion, education, medical adherence, public health actions (i.e. sanitation), chronic disease management, and triage support.

The PSF is highly decentralized with most authority placed on the municipalities. The PSF program has been successful in getting to the majority of the population as family health teams are found in 85 percent of municipalities serving over 98 million people. Despite the large coverage, the PSF program only takes up 8 percent of the federal health care budget.

The PSF program has been shown to be very effective in improving health outcomes for Brazil. Access to care has increased by over 25 percent since its origination in 1994. Life expectancy has increased from 67 to 72 years, immunization rates for Tetanus, Diphtheria, and Pertussis is over 95 percent for all children 1 year of age, and infant mortality has decreased from 48 per 1000 to 17 per 1000. A reputable study by Guanais and Macinko focused on postneonatal mortality as a proxy to assess success of primary care delivery. It is argued that the postneonatal period, ranging from 30 days to one year, reflects success in primary care goals of nutrition, immunization, sanitation, breast feeding, and prevention of respiratory and diarrheal infections. They found that between 1998 and 2006, the postneonatal death rate decreased from 14.24 to 6.92 per 1000 children while at the same time PSF coverage increased from 8.74 percent of municipalities to 60.90 percent. Their published paper accounted for increases in clean water, decreased illiteracy rates, and other confounders and through regression analyses determined that the family health teams decreased postneonatal death rates by 0.86 per 1000 compared to communities without teams.

The PSF does not come without some problems, however, and need to be mentioned. The decentralized nature of the program makes it difficult to ensure quality between geographic regions. The federal government collects money, but the money is allocated to municipal governments who are ultimately in charge of managing their budget and health care delivery. Problems seem to arise most in the northeast region which is more socioeconomically depressed than the central or southern regions. As a result, the northeast has a higher infant mortality rate, more deaths from infectious diseases, and lower life expectancy. Not surprisingly, many municipalities in the northeast do not have family health programs, partly because it is difficult to recruit and retain health providers for this region. Additionally, corruption is a problem with some municipalities which results in allocated money not being used for health care expenditures. These problems create conspicuous disparities in access and health care outcomes between different municipalities.

Another problem faced is the difficulty of integrating the family health teams with local hospitals and specialist care. Some municipalities have more success than others in working with the private sector. Naturally, the more coordinated and integrated the care is, this results in better available health care for Brazilians. Lastly, the emphasis on improving primary care has come at the sacrifice of providing good hospital and tertiary care. Brazil’s public hospital system only provides 35 percent of hospital beds in the country with the remainder made up of by private hospitals. The public hospital system is overcrowded and has large variability in quality.

As the United States works to come up with solutions for improved primary care delivery and access, it may be prudent to look outside of Europe and Canada for ideas. Arguably, Brazil’s primary care model has seen some of the greatest health improvement of any country in the world through their innovative decentralized method. I feel that U.S. could benefit from a team based approach. Teams are seen to an extent in our patient centered medical homes, but the addition of community health workers could facilitate improved chronic disease management, health promotion, behavior change and support, and medical adherence. Most importantly, Brazil has shown us that by investing in primary care, the health of the population can be greatly improved. As we continue to work towards improved health systems and primary care delivery, let’s not forget to keep Brazil in mind.



Frederico C. Guanais and James Macinko, The Health Effects of Decentralizing Primary Care in Brazil, Health Affairs, 28, no 4 (2009): 1127:1135

Michael Kemp, Cracks Appear in Brazil’s Primary Health Programme, The Lancet, vol 372 (2008): 877

Mathew Harris and Andrew Haines, Brazil’s Family Health Program, BMJ 2010; 341: c4945

Paim, J., Travassos C., Almeida, et al., The Brazilian Health System: history, advances, and challenges, The Lancet, 2011; 377: 1788-97

Monday, November 7, 2011

Innovation in Physician Training

Patient-Centered Medical Homes (PCMH’s) are presented as a care delivery innovation that has had success in the realm of increasing quality of care while decreasing costs. At its core, the PCMH seeks to strengthen primary care, leading to coordinated care, improved access, and ultimately reduced costs. However, the dark irony of the success of the PCMH model is that as primary care demonstrates its ability to serve as a hub for care coordination while reducing cost, primary care physicians (PCP’s) are dwindling in proportion of the physician work force. Perversely designed fee schedules have created a dramatic income inequality among physician specialties, strongly steering medical students away from generalist careers. While many factors ultimately influence student specialty choice, medical students are hard pressed to choose generalist careers while income disparity remains so troublingly wide.

Further, essential to the success of PCMH’s are functional teams. Looking at the results of the 2006 Group Health Cooperative model, strong PCP leadership was identified as a critical factor for the success of organization transformation into a PCMH[1]. However, current undergraduate and graduate medical training does not search out these skills in its applicants, nor does it provide comprehensive education for how to develop these practice tools. Strong leadership capable of excelling at the challenges of practice management are necessary for PCMH’s to become a widespread national success story, yet medical education has not responded to the growing need for these additional physician skill sets. Biomedical science training and clinical knowledge, the current mainstay of medical education, is not enough to produce physicians competent for current health challenges. Physicians, especially PCP’s, must be taught how to work and lead within teams, manage practices, motivate members of the care team, integrate health IT, and adapt quickly to practice innovations. These competencies are critical for the success of PCMH’s and ultimately care delivery transformation.

The reported success of PCMH’s is exciting; though many details need to be ironed out to rigorously define the characteristics of successful models, the early victories among differing patient populations provide evidence that the key organizing principles of the PCMH are widely applicable across our nation. However, to achieve this dream of coordinated, high value care, it is necessary to rethink medical student training and physician payment. Large income disparities are not sustainable if we are to attract the number and quality of primary care doctors necessary to lead transformation in care delivery. Further, medical education itself must innovate in order to equip its graduates to practice within these new models.

[1] Robert J. Reid, Katie Coleman, Eric A. Johnson, Paul A. Fishman, Clarissa Hsu, Michael P. Soman, Claire E. Trescott, Michael Erikson and Eric B. Larson The Group Health Medical Home At Year Two: Cost Savings, Higher Patient Satisfaction, And Less Burnout For Providers Health Affairs, 29, no.5 (2010):835-843

Sunday, November 6, 2011

Thoughts on Employer-Based Insurance in U.S.

I fail to understand why corporations oppose moving away from the employer-based insurance schemes which have predominated in the United States the past 70 years. As health care premiums continue to rise and corporations are forced to spend more than ever on their employees' health, I would think that they would jump at the opportunity to shift some of the coverage responsibility to the government. This would free up time, energy, and resources to focus on their business related objectives. Additionally, the government would be better equipped to gain expertise in health coverage compared to corporations which could result in more efficient delivery of care and improved health for the population. The government would be less distracted by profit seeking goals and would instead focus on improving the health of its citizens and delivery of care through a more cost-effective approach.

Despite drastic increases in health care premiums and overall costs for health care, corporations, in general, are opposed to getting rid of the employer-based model. One opposing stance is that by providing health insurance to their employees this increases employees’ loyalty. This is probably true, but if everyone was covered through a government funded insurance plan, companies would be liberated from the expectation to provide health benefits. The fact that employers are expected to provide coverage is a historical accident dating back to the 1940’s. By shifting to a government program, this would free up businesses to increase loyalty through other means, such as increased pay or higher bonuses at year end.

A second view expressed is that corporations feel that they are better at managing health care costs than the government. If this is the case, why are companies spending so much time and energy coming up with new ways to provide coverage? Furthermore, why are corporations continuing to increase cost sharing with employees, who ultimately carry the burden as a result? To me, this suggests that companies are unable to keep up with rising costs and therefore are failing to control costs effectively. Perhaps, corporations would argue that costs would be even more outrageous if the government controlled insurance plans, but I have not seen any actual data to support this. I am not trying to suggest that corporations are at fault for the rising health care costs, but I am arguing that they are not in any better position to curb rising costs than the government.

Corporations may also believe that they would end up paying more in taxes to cover a government insurance program than the costs they incur now to provide coverage for their employees. I find it difficult to believe that the tax burden placed on corporations to fund a public insurance plan would be greater than the amount they pay for their employees right now. The car industry provides a good example of how large the health care burden is on business; Chrysler, for example, spends more on health care than they do on steel. That is probably why the CEO of Chrysler is one of the lone business leaders to support a government insurance plan over employer-based plans.

Lastly, corporations may argue against government plans on principle or ideology. In essence, those who believe this are taking responsibility to make sure that their employees are covered and that they have adequate opportunities to pursue better health. However, as health insurance premiums and overall costs continue to rise, employers are finding it more difficult to provide for their employees. As corporations are less able to provide insurance, the government will have to take on more of the coverage responsibility.

As a result of the numerous ways to reduce health care costs – from full coverage, managed care, preferred provider, cost-sharing, health savings accounts, and more – a confusing web of options have been developed by employers. However, none have proven to be the silver bullet for cost containment within the employer-based model. As fewer companies are able to keep ahead of the costs, a better option may be a single payer public plan. With one agency providing coverage, premiums could be lowered through broader risk pools and administrative costs could be reduced. Additionally, companies would see significantly reduced cost for health care and, most importantly, patients would see more money invested in their health.