Data not drugs

Taking control of your health in the age of genetics

We have access to more health information now than any time in history, yet this deluge of medical data may sometimes make health decisions more difficult. The Internet has opened a Pandora’s Box of data that can easily overwhelm us. We need a way to process all this information to assist us in making better healthcare decisions. Sifting through the barrage of health information writhing across the Internet can be a challenge and new sources are continually cropping up.

There are some great online resources that can help. Search sites are now guiding consumers to safe, trusted health websites, says Susannah Fox of Pew Internet. Both Google and Bing are entering the health search arena by providing a highlight at the top of health related searches that allow you to access a wealth of information. Compare the results of a search on “type 2 diabetes” from Google and Bing. While they both present relevant articles from their libraries of health resources and present a summary related to the query, so far I have seen better results from Bing but Google is gaining fast. I expect both will continue to refine these algorithms to improve results. Google has also updated their popular Flu Trends providing flu info for 121 U.S. cities. Previously, flu trends were available on a state and country level.

WolframAlpha can calculate clinical markers on cholesterol levels, BMI and a wide range of indicators. WolframAlpha computes a breakdown of total calories, fat, cholesterol, sodium, carbohydrates, protein, and other particular nutrients of most foods. And then there is the incredible Genetics Home Reference by the National Library of Medicine which provides consumer-friendly information about the effects of genetic variations on human health. The “type 2 diabetes” search here gives some very interesting results with detailed genetic information and many links to additional valuable resources.

One strategy to cope with all this data and help to create a framework for our medical decisions is to use a decision tree. In his new book, The Decision Tree: Taking Control of Your Health in the New Era of Personalized Medicine, Thomas Goetz offers a structure to reduce uncertainty and allow us to make better choices. I was fortunate to read an advance copy of the book a few months ago and have had a series of interesting conversations with Thomas since then to discuss some aspects of the book. The decision tree is basically a flow chart to move us towards better healthcare choices. I am most impressed that he could take the rather complicated subject matter of personalized medicine and distill it into layman’s terms that make for an interesting and compelling read. The book will be released on February 16, 2010 and I highly recommend it. You can read Chapter 1 today on The Decision Tree blog.

It is basically divided into three sections: prevention, diagnosis and treatment; it is Thomas’s contention that we are all moving along this spectrum and our baseline is our DNA. One of the themes of the book is that by knowing and better understanding our genetic makeup, we can improve the medical decision making process. Spring boarding from a future of genetic medicine envisioned by Dr. George Church’s Personal Genome Project he leads us to the current state of personalized medicine with services like 23andMe and Navigenics which offer genetic testing. But with the cost of these tests, is this really for the average patient? “The price of genetic sequencing is falling rapidly,” Thomas said, “but I’m not actually calling people to start with genomics.” There are some basic starting points for using the decision tree strategy that don’t wholly rely on having your genome sequenced. The widget below gives you an idea of approach he takes:

The book describes his participation in a Quantified Self meeting, a sort of show and tell for people taking advantage of various kinds of personal tracking methods like geotracking, life-logging, DNA sequencing, etc. They track the various metrics in an effort to find quantifiable meaning to the data. These folks are “geeking out… just like the guys who stand in line for iPhones and then rush home and take them apart to see how they’re made. They’re just the same. Except in this case, the iPhones are their own bodies.” While it is important to have as complete information as possible to make better choices that will improve our health, he said, “Tracking your health with gadgets and gizmos is not for everyone. But not long ago no one even knew what their blood pressure of cholesterol level was and now tracking these metrics is quite common.” Tracking our health metrics and combining this with genetic data to use as a starting point for a decision tree can help us to make choices that will improve our health. Whether it is simply taking our blood pressure, or using the Twitter-based GetUpAndMove service started by Jen McCabe, keeping track of what our bodies are doing and what we do with them is good data to have.

One of the problems we face in dealing with health issues is finding drugs that actually help, and a chapter in Goetz’s book deals with the drug problem in healthcare today. While modern pharmacology has developed many drugs that have saved and improved lives, finding the right drug for the right condition is a challenge. And the blockbuster model used for research and development in the pharmaceutical industry is fading out. Some of the the promising drugs now in the pipeline are designed for smaller groups of patients. But as chapter 8 points out:

The pipeline of new drugs has slowed to a crawl, as one promising candidate after another has petered out in the last phases of development. “The low-hanging fruit has been picked,” says Derek Lowe, PhD, a drug discovery chemist and industry pundit. William Haseltine, PhD, a former researcher at Harvard Medical School and the founder of Human Genome Sciences and eight other biotechnology companies, notes that fewer than 1 in 100 new ideas reaches clinical trials and fewer than 10 percent of these are approved for sale.

But developing drugs for less common medical problems will require major changes for the industry. When promising molecules are found, and the chemistry works, the drug companies are anxious to get these to market. And all of those “failed” clinical trials (which may have produced valuable data) are locked away, leaving possible medicine for “lesser” conditions undiscovered. Goetz lays out a hopeful possible future where we enter into a new era of research that will rescue drugs and free the data that will help people to live healthier lives.

Drug ads are also sometimes misleading. By ambiguously defining who might need or benefit from the products advertised, they focus “on convincing people that they may be at risk for a wide array of health conditions” rather than genuinely educating consumers, concluded a 2007 study in the Annals of Family Medicine. Drug manufacturers maintain that their ads are not misleading. The Pharmaceutical Research and Manufacturers of America, an industry group, says: “Consistent with recently updated guidelines, PhRMA is committed to a fair balance of risk and benefit information in all direct-to-consumer advertising.” The “brief” summaries in direct-to-consumer drug ads can take up a whole magazine page, and make it very difficult for a consumer to understand and weigh the risks and benefits of the medication.

Drug fact boxes are a possible solution that could eliminate a lot of the ambiguity. These user friendly boxes, similar to the nutrition labels found on food, include facts not found in the so-called brief summaries.They are the brainchild of a husband-and-wife team: physician-researchers Lisa Schwartz and Steven Woloshin from the Dartmouth Institute for Health Policy and Clinical Practice. In a study published by the Annals of Internal Medicine, they tested how much consumers could benefit from understandable information on drug products. The data were collected via mailed surveys measuring respondents’ reactions to two drug fact boxes versus traditional direct-to-consumer advertisements. One trial compared two potential treatments for heartburn and the other compared two potential preventive medications for cardiovascular events. Of respondents who received the drug fact box on heartburn, 70 percent were able to correctly identify the most effective treatment, as compared to 8 percent of the control group. It’s obvious that drug fact boxes have the potential to improve consumers’ knowledge of the potential benefits and side effects of medications. The FDA is considering requiring factboxes so there is hope but the wheels of government grind slowly.

The book lays out three fundamental principles for making intelligent health choices. Number one is early is better than late. By learning genetic predispositions we can treat disease, sometimes before it even happens. Number two is let the data do the work. Using evidence-based medicine and monitoring the continuous stream of data we create, whether it be our diet, exercise, moods or DNA, gives us a baseline from which we can evaluate our future health. And number three is openness is a powerful thing. The more accurate information available to researchers, care providers, and consumers, the better decisions we can all make and the more options we will have for successful outcomes.

By making the most of the new science and technologies available and using the best practices from genetics, behavioral science, and information technology we have the opportunity to sculpt a process to better manage our health. Putting the patient at the center of healthcare and creating a strategy to process all of health data available today is a great start towards meaningful healthcare reform. While Congress debates payment methodologies, health insurance, and all of the political considerations which crowd into the discussion, we the people can take more control of our health today. After all, they are our bodies…

tags: , ,
  • Jerry

    Thanks for the great article, Brian. You present an overview of not only where we’re headed with data and the internet, but how drug development is starting to fail. As we move forward with advances, it’s going to become important for drug companies to collaborate on drug development rather than keep information to themselves. The technology is available, unfortunately this goes against the grain of the profit driven model used by big drug companies.

    Google and Microsoft continue to creep into the healthcare market, which I see as a good thing. Both companies offer years of experience as well as unique ideas that will help bring the healthcare industry in line with the consumer market. In other words, we’ll gain some ground.

    Thanks again for the thoughtful commentary.

  • Paul Roemer

    Nicely written and very relevant Brian.

  • MedicalQuack

    This is the reality of data and how it can be used to create drugs in pharmacogenomics.

    Back in April of 2008 I had the opportunity to interview Dr. Patrice Milos, Chief Science Officer of Helicos, one of the companies who develops and sells the genomic sequencing machines, theirs is used at Stanford University and she explains the importance of Single molecular sequencing and the role it plays.

    Since this time there have been huge advancements with sequencing as well relative to cancer and other diseases, but one thing Dr. Milo and I agree upon back in April of 2008, it’s all about the software.

    The link above was the initial interview and I have subsequent posts after this as well. The big issues that continue to hinder some of these processes are money and cost, nothing new there.

  • Jeff Brandt

    Good article and thanks. Most consumer AKA patients what to be told what to do or only what the maximum of 2 choices. But things are slowly changing. The era of the informed consumer is here.

    One of the biggest problems of the Internet is the deluge of information, much that is incorrect.

    The other double edge sword is the individual opinions and review. Many people only give reviews when they are unhappy, many do not have the understanding or knowledge about the subject to make a good review. Any review site of any subject is pretty much worthless without a system that allows vendors, doctors, manufactures responses.

    Our company Communication Software Inc., hosted a EHR roadshow several years ago and one of our speakers was a Pharmacist from a EHR company that provide distilled Pharma and FDA testing information to the provider. He told the crowd that many of the drugs that are entering the market today, especially the highly advertised ones are no better than the previously available products that sell for much lower price.

    Jeff Brandt

  • thomas goetz

    Thanks for the considerate and thorough read of the book, Brian.

    One thing I should clarify is the idea of “where to start” – I think personal genomics are a wonderfully promising arrival of genetic data for regular people, but I don’t expect everybody, at this point, to sign up for these still-limited services, and I don’t expect everyone to understand what the services provide. For some people, however, it may be a fascinating and even useful way to start engaging with their health. For others, though, they may be much further along the spectrum and actually be facing serious decisions about treatments or care. These people, too, can successfully engage in relevant health information and take a role in making their healthcare decisions. The basic idea I’m trying to communicate is that engagement is essential to the process, whereever it begins – when people engage with their health, and do the hard work of using relevant information, they are much more likely to make better decisions and have better health.

    I appreciate how thoroughly you understood this, as demonstrated by your post – it makes me hopeful that the idea will resonate to a larger and responsive audience.

  • The EHR Guy

    I believe that Bing relies on MedStory (Microsoft’s specialized health search engine). This is why Bing gives much better results.

    I don’t think Google has a similar counterpart.

  • Brian Ahier

    The Robert Woods Johnson Foundation Pioneer Portfolio produced a video, “The Coming Revolution Data-Driven, Patient-Centered Health Care,” featuring Thomas Goetz along with Patricia Flatley Brennan, R.N., Ph.D., Dr. Tom Delbanco, M.D., Stephen Downs, David Eddy, M.D., Ph.D., James Ralston, M.D., M.P.H.,Paul Tarini, M.A. and Steve Woloshin, M.D., M.S.

    There was also a Pioneer Portfolio-sponsored luncheon at TED, which I look forward to hearing about. Check out the blog for updates:

  • Jen McCabe

    Brian –

    This is a great review and thanks for including Contagion Health’s #geutpandmove application.

    I hope you’ll join me for a challenge!

    Your review of data and behavior is fascinating. Data can, we’re finding, help inoculate and treat, but not often by itself.

    Your health isn’t about the data – it’s about what you do with it.

    As Thomas’ book illustrates, we must find tools and decision-pathways that make it easy for us to understand where each piece of data fits.

    Thomas’ point about ‘where to start’ with your own individual decision tree is a great one.

    Each person’s decision tree is different – and even more interesting- each person’s health decision tree is different for them EACH TIME they make a health-related decision.

    Only when we begin to examine each person’s individual history of making microchoices – or small, everyday decision that add up over time to better health – will be be able to establish a person’s pattern of decision-making, and thus incent him or her to adopt a ‘healthier’ decision-pattern over time.

    We should consider the delta, or change over time, within a highly time-, situation-, and individually contextual or “me” centric atmosphere when trying to reward healthier behavior(s) at the individual level.

    This ‘n=1’ or personalized focus must drive the design for tracking, data entry, and all attendant challenges related to how people make decisions.

    At the simplest level, for every person and every decision, there is the pre-decision stage, the ‘during’ decision stage, and the ‘after’ or post-decision stage.

    Thomas’ work centers on the ‘during’ decision stage, where you’ve gathered information and now have the option to take action. was built for the ‘pre’ decision phase, but emergent behaviors about decision-making we’re now able to track surprised us.

    We originally just wanted to incentivize people to *move* – a little bit at a time – and get them to quickly and simply accept that movement is the default choice.

    Even though it seems like sending or accepting a challenge is relatively simple, we’re motivating users on a very complex, communal level-by integrating social contagion and networking theory (getting friends and family in your existing social networks like Twitter and Facebook to challenge you to a small, healthy ‘microchoice’).

    In essence, is working in the ‘pre’ decision-phase, even though the choice a user has to accept or reject a challenge are very definitely an ‘active’ decision state. It’s a two for one deal, in terms of decision-making.

    Interestingly enough, we found the social tendency to want to share what we’re doing (and with or for/whom) means many will also ‘actively’ decide to notify the person they’ve challenged that they’ve completed an activity (which represents both an active decision and a ‘post’ decision choice: do the challenge, then afterwards tell someone you’ve done it). Another two for one deal.

    A challenge is really a beautiful decision-tree with many potential branches for both the person who initiates the challenge and the person who accepts.

    The really amazing thing is that people using are moving quickly from ‘talk’ (“I will dance for two songs if you will dance for two songs”) to action (“Completed the challenge! How about another?”).

    In other words, they’re making the decisions we want them to – healthy ones!

  • Brian Ahier

    You can listen to the Pioneer Portfolio-sponsored luncheon at TED:

  • Pieter Peach

    Looking forward to reading the book.

    Personalised medicine (whether it involves pharmacogenetics or not) is an interesting area. Explicit decision-making techniques around health/wellbeing will no doubt be the area with the greatest utility.

    We currently gather evidence based on studies from well defined population groups and generalise this, often innappropriately, to our patients and ourselves, when more often than not, the devil is in the detail.

    From a public health perspective, the ability to deploy a generic tool which delivers tailored information to individuals appropriate to their stage of behaviour change is a great opportunity. Combine this with the web’s new social dynamics (as Jen has done) and you have a potentially powerful tool. – Exploring decision-making

  • Heather Leslie

    Hi Brian,

    Thanks for this post. It started me pondering on a couple of phrases that leapt out at me…

    Your comment: “The more accurate information available to researchers, care providers, and consumers, the better decisions we can all make and the more options we will have for successful outcomes” and Goetz’s comment: “…let the data do the work…“.

    “The data”.”Better decisions”.

    We make a lot of assumptions about quality of our data, without necessarily understanding the implications. It is a bit of a two-edged sword, with a potential for enormous good and the potential for… well, bad decisions can happen with poor quality data.

    We, both clinicians and consumers, need to understand our data better, make assessments about the data completeness and quality, and factor this into our decision-making process to ensure that we make wise and safe health choices.

    The combination of quality health data, models such as Jen McCabe’s microchoices and Goetz’s decision trees, personalised medicine, evolving social networks, personal health records and clinician/consumer decision support gives huge potential to influence long-term health outcomes.

    We live in exciting times!

    I have posted a more detailed response on my blog –


  • Brian Ahier

    Exciting times indeed! You raise some critical points on data quality. GIGO is true in computer programming, but equally true in human decision making. Faulty, imprecise, or incomplete data will invariably lead to faulty results and decisions.

    Thanks for sharing your excellent blog with us.

  • The Health Dude

    “Health is wealth” is well known to all and everyone wants good health.So, Let us build a food habit discipline, keep pace with work, rest and exercise to Achieve good health, The ultimate wealth.