As we begin a new year and turn over a new leaf, I am thinking of new ideas and a fresh start to the year. It seems a bit arbitrary to have the turning of a calendar page be the impetus for new things, but New Year's resolutions are a longstanding annual tradition.

So, I'll start with an old idea: a resolution to exercise more. I bet more than half of the United States has made this one—or one of the related resolutions like losing weight and eating better—and for good reason. There are so many studies supporting the benefits of exercise. My favorite is the observational study that demonstrated that increased exercise is associated with better survival, independent of losing weight.

Of course, these are just observational data, but there are randomized studies in patients with diabetes that also show benefit of exercise, albeit on a surrogate endpoint of HbA1c. Interestingly, this study, from a year or two ago, showed benefit of aerobic exercise and weight resistance training, thus lending support to my exercise routine of about four times (five, in a good week) in the gym on the stationary bike, elliptical, or weight machines, with abs to start the workout. I have to admit, though, I do exactly 30 minutes of exercise, so as to "get with the guidelines," and not that often more. I also have to admit, when I was on service, I was down to just twice a week—but I did exercise both days this weekend! So, a resolution to re-engage my "ideally 7 days a week" (to quote the guidelines) exercise program. When I look at others my age and older, I'm encouraged by the fact that the guys in shape are always into exercise and sports, so getting exercising is an ideal way to keep healthy. Plus, I will be joining my college buddies for a ski trip to Jackson Hole later this month, so with the steep slopes there, I have to get in shape!

On to some new ideas: One of my long-term goals with my research is to try performing clinical trials using electronic health record (EHR) data and/or registry data. I've had the great fortune to work with the Thrombolysis in Myocardial Infarction (TIMI) Study Group for 23 years doing many types of randomized clinical trials in acute coronary syndromes and prevention. Many of these trials have helped shape clinical practice and guidelines by bringing new medications to use and helping sort out optimal strategies for management.

The downside of these trials is their cost—often hundreds of millions of dollars each. One trial we are doing now will cost more than a half billion dollars! Each of these trials is for one drug in one indication, yielding a high "level of evidence" for that particular drug and indication.

This, however, does limit the total number of things that can be studied. It is also important to note that these trials have been mostly funded by the pharmaceutical company that makes the drug in question (including times when the competing drug was found to be better). We thus have a lot of positive information from the drug companies' commitment to research.

The number of questions yet to be answered in clinical medicine is overwhelming, and many don't involve new drugs or the pharmaceutical companies who provide funding for trials. So, how can we get answers to these pending questions? Also, how can we continue to fund the larger and larger trials that are needed to get answers, when even the big pharmaceutical companies can't afford to fund them in the traditional format? The costs are simply too high.

Many people are now talking about using data from EHRs collected during routine clinical care as a solution to this problem. Why ask research coordinators to re-enter data from one EHR into an electronic case report form, and then have monitors double-check that they transcribed it correctly? Why not just get the data directly?

Another possible solution: using registry data in a trial and adding randomization to a treatment strategy. For example, in the new PINNACLE-AF Registry platform, patients could be randomized to one of the different new anticoagulants, which might give researchers a direct randomized comparison of how these drugs perform in the real world.

There are many challenges to doing this, but using existing data is clearly a viable way to simplify the process and reduce costs. This would only work for approved medications or treatment strategies, but there are many trials that are needed. So, one goal I've set for myself for the New Year is to try to see if I could help with doing such a trial.

What's your idea or resolution for the new year?

Christopher P. Cannon, MD, is a professor of medicine at Harvard Medical School in Boston, Massachusetts. He is also the Editor-in-Chief of CardioSource Science and Quality.

Keywords: Electronic Health Records, Registries, Weight Loss, Exercise, Diabetes Mellitus

< Back to Listings