Remarks Prepared for the Honorable Mike Leavitt, Secretary of Health and Human Services, on Personalized Medicine Coalition

We need to do a better job helping people understand what personalized medicine means. I suppose part of the problem is we’re all still trying to define it.

However, I’m sensing when we use the phrase “personalized medicine,” for some, it creates a mental picture of a patient having one-of-a-kind pharmacology developed specifically for them, based on their phenotype, environment, and genetic make-up.

That model, while appealing, raises doubts. Intuitively, people have questions about the scalability and sustainability of trying to treat a population of people in that fashion.

The vision we are moving toward, in my mind, is best described as mass personalization. Using a thorough understanding of a person’s genetic and clinical history, a doctor will select a combination from a group of biological and chemical treatment tools.

I sense our vision will be better understood and accepted if we begin to paint a picture more familiar and comfortable to patients, providers, and payers. As consumers we have become quite familiar with mass customization in many of the things we purchase.

When I bought my first set of golf clubs, I bought a set the golf professional had on the shelf. After many years, I decided to buy new ones. The technology had improved, and there were several aspects of my game that would fall into the category of needing treatment.

This time, I was confronted with a different experience. The golf professional and I measured and inventoried my game until we knew what the perfect length and flexibility of the shaft should be, the angle of the housel, the weight distribution of the club head, and grips to fit my touch.

The golf professional said to me, “Now that we know how you align your clubs, with your alignments, we carry five different models of this club, and the X20 Long has most of the attributes you need.” I bought a set of clubs, off the shelf that was personalized to me. This company is now engaged in mass customization.

Now, I want to say, tongue in cheek, I have a vision of the golf business in the future. It’s personalized golf. There will emerge a system of electronic golf records. These records will be interoperable between golf courses so no matter where I play, each shot will be tracked. The genetic tendency I seem to have for slicing the ball will be well documented. So, as I need golf clubs in the future, they will be able to personalize them.

In fact, because there will be so many golfers like me with electronic golf records, researchers will be able to gather data to invent new tools to cure the common slice and the three-putt green.

All kidding aside, we do need to begin defining personalized medicine in ways people can understand. We have the technology now to make health care much more personal and much more efficient. We just need to apply it to the tasks at hand —

• To develop evidence-based standards for measuring and improving quality of care.

• To build systems of health information technology to make the patient records available when and where they are needed.

• To use those same systems to gather medical evidence from the day-to-day practice of medicine so that clinicians can make the best decisions for each patient.

• And ultimately, to use our personal genetic information to tailor treatments more effectively to each patient.

We are here today to explore the emerging possibilities for evidence-based Personalized Health Care.

The potential is huge — for predicting health, for preventing and preempting disease, and for personalizing treatment according to each person’s unique biology.

The Genome Project itself has just left the starting line. Its pace of discovery is still accelerating.

Just this year, we have seen great progress in using existing data to match genetic information with known health outcomes. These genome-wide association studies have begun delivering results on almost a weekly basis.

In this month’s issue of Biological Psychiatry, there’s an article about a link between a certain genetic variant and the drug Zyban, which helps people quit smoking. It seems that people who have the variant were less likely to have resumed smoking six months after taking Zyban.

Next month, data from the NIH’s Framingham heart study will be made available to researchers. The Framingham study spans three generations. It assembles data from thousands of volunteers in a single Massachusetts town. This is some of the richest clinical data available to us. Its release for use by qualified researchers is an important step.

I’m happy to report that, throughout HHS, work is moving ahead rapidly to support genomic medicine in particular and Personalized Health Care in general:

• The FDA, the Centers for Medicare & Medicaid Services, and the CDC are laying the groundwork for Personalized Health Care and looking at the kinds of structures we need within HHS to support the development and review of genetic testing.
• CDC and AHRQ are looking at the effectiveness and appropriate uses of genetic tests.
• And the Secretary’s Advisory Committee on Genetics, Health and Society is preparing a report and options  on incorporating genetic testing into clinical practice.

I’m also pleased to announce that today HHS is releasing its first report on Personalized Health Care. This report is about developing a common vision of our opportunities in genomic medicine, health IT and evidence-based care. 

The report takes stock of the tasks ahead for bringing Personalized Health Care to fruition. It identifies the “pathways” we need to pursue. And it provides an inventory of HHS-supported activities that are important for achieving Personalized Health Care. 

The report is just glimpse of what lies ahead. I expect it will help us coordinate future work more effectively.

The report is part of the initiative I’ve launched within HHS to help us achieve Personalized Health Care for Americans. The initiative is about helping coordinate the Department’s activities. But it’s also about addressing cross-cutting issues that need to be raised and resolved.

Personalized Health Care can give us the power to live longer, healthier lives. But, as with any other power, we need to safeguard its use.

• For personal information, we need to be sure that security and confidentiality are protected.
• For our clinical trial data, we need to be sure researchers have productive access, and the opportunity to reap the rewards from true innovation.
• For genetic tests, we need the right regulatory structure and the right supports for health care professionals in using them.

As our report makes clear, a key role for government is helping achieve common standards – especially in the field of health information technology. That work is underway, through our American Health Information Community and its working group on Personalized Health Care.

One area of special importance is the field of clinical decision support. How do we translate what we learn quickly and effectively, so that doctors and patients can work together to put it into practice?

Finally, there’s another role for government — one that only government can fulfill. I’m talking about the legal framework for avoiding misuse of genetic information. This is important for building and keeping the public’s trust in this new area.

President Bush first called for legislation in 2001 to protect against genetic discrimination in employment and health insurance. This year, the House passed legislation to do that.  The Senate has passed similar legislation in previous sessions, and we have reason to hope that a final bill can be worked out and passed by both chambers this year. Events and science are moving quickly in this area, and the time has come for legislation protecting personal genetic information against inappropriate uses in employment and health insurance. 

We also need to build trust into any system of evidence-based care.

To make Value-Driven Health Care work, we need to be able to measure and compare the quality of care. That means setting standards of quality that everyone can agree to.

Setting standards must be done collaboratively — with payers, providers and consumers all at the table. It requires a strong foundation of evidence-based care so that everyone can trust that they are being treated fairly.

• We need a strong evidence base to identify best practices. 

• We need a strong evidence base to measure quality. 

• We need a strong evidence base to take health care to a new level for each patient.

In the future, health information technology can help us develop our evidence base faster and better, using the data from millions of encounters in day-to-day medical practice. This has the potential to vastly improve the safety of medical care, by detecting problems much faster. 

It also has the potential to measure different outcomes for different patient groups at a much finer level.

In his 2008 budget, President Bush proposed $15 million to construct a new electronic network to bring together data from the actual practice of medicine. This network would be a prototype for what some have called a “learning health care system.”

The system wouldn’t replace controlled clinical research. But it could add to our base of medical evidence on effectiveness and safety of various treatments. Health care delivery itself could serve as a platform for rapid progress and more effective health care.

Those are some of the potentials. There are some caveats as well. We have a lot to learn as we take our evidence development to a new level. Carolyn Clancy will be talking about all this in more depth later today. 

There are many questions still to be answered —

• When does that data become “evidence” that we should act on? 
• How do we ensure the quality of the data outside the confines of a controlled clinical trial? 
• How do we even achieve standard nomenclatures so that data can be aggregated? 
• How do we balance evidence from clinical trials with evidence from clinical practice? 
• And even when we’re confident we have good evidence — how do we get it into practice quickly?

New emphasis on evidence-based care is an important tool with great promise. It’s not a quick fix or a magic bullet. We have a lot to figure out about how to best use it. But it has the potential to make health care even more personal and effective.

Personalized Health Care has many components:

• It is genetics put into clinical practice.
• It is health information technology.
• It is the informatics to make complex information useful.
• It is evidence-based care at a new level.
• Finally, it’s the synergy we get when all of these work together.

Getting there is going to be involved and complicated. But the end result is easy enough to understand: 

Personalized Health Care means knowing what works, knowing why it works, knowing who it works for, and putting that knowledge into practice for patients.

Personalized Health Care brings together our best prospects at the cutting edge of health care. It can help us achieve not only better quality, but better value in health care.

Last revised: March 13, 2008