An announcement yesterday from the Food and Drug Administration highlights a couple of items that I think are important in terms of not only how we understand medical issues and medical risk, but also how technology will help us get better and quicker answers in the future.
On the medical side, the question is when does a risk become a risk? On the information side, why aren’t we able to harness the power of data to answer questions about risk more quickly and accurately? And, if I have a device implanted in me, why can’t someone get in contact with me? After all, if my car has problem they send me a letter. If something is put in my body–for the most part–forget about it. [more]
Basically, the FDA issued an announcement that breast implants–both silicone and saline–may be associated with a significant increased risk of a rare type of lymphoma in the breast called anaplastic large cell lymphoma.
The number of cases they can document is very small–on the order of 60 or so–and the increased risk sounds high in one published report at 18 times. But the numbers are minuscule, and what we call the “confidence intervals” in their analysis are wide, meaning the real risk may be anywhere from twice the risk up to 157 times the risk.
I don’t mean to minimize this risk in any way shape or form, but in reality there are millions of women worldwide who have had implants and only a very small number of them would develop this lymphoma. Yes, I know if you are one of the women it is your life at stake, but the announcement–which was perfectly appropriate from a medical standpoint–will probably scare many more women than it will actually affect.
The FDA issued this guidance because they saw an association that needs further analysis and study, not because they concluded the risk was real or definite. They want doctors and patients to report to them any cases of this lymphoma in women with breast implants, and they want doctors and patients to be aware that any unusual swelling or change around the implant could be related to this lymphoma. I suspect that many more women are going to undergo diagnostic procedures such as biopsies than will be diagnosed with this lymphoma.
Which leads me to my next question: Isn’t there a better way to do this?
In fact, in the future, there will be.
I believe we are–after several decades of hope–actually arriving at a “tipping point” where we will be able to appreciate and expand the value of electronic medical records and other health information technologies to help us improve the quality of our medical care and answer some of the questions as to whether what we do makes a difference in our health. And, as our systems progress, we will be able to accumulate medical information on literally millions of people to determine whether there are unusual side effects from medications and medical devices.
In fact, that future is taking form right now.
I have been part of several meetings over the past two weeks where the theme of quality improvement and improved health care based on electronic information has been at the heart of the discussion.
One such project is in fact being funded and run by the FDA. It is called the Sentinel Initiative, and is intended to accumulate appropriately protected medical information (privacy is important to everyone, and protecting that privacy is at the heart of these efforts) to determine if any significant side effects are seen with new–and even existing-medications, biologics and medical devices once those medications and devices are widely available.
The ideal circumstance would be for a background surveillance system to be in place which would scan millions of records in real time to find those needle-in-the-haystack problems that can prove so serious. But we don’t yet live in an ideal world, so the FDA is working with other partners to develop alternative processes to answer some of the questions as to how we can begin to move down the road towards this new vision.
As a result, there is a program called “mini-sentinel” which is looking at what we can accomplish today, including how to accumulate the information and how to start asking some of the questions. There is much more information about mini-sentinel available on their website at www.mini-sentinel.org. This program currently has access to medical information and/or insurance claims data on approximately 60 million people, with a goal in several years of 100 million. (There is an excellent review of the mini-sentinel program in a recent edition of the New England Journal of Medicine.)
Which brings me to the relationship between today’s announcement and our ability to answer the question raised about the breast implants.
Truth be told, even if we had the sentinel program in effect today it would be difficult to answer this question.
Medical devices don’t meet the same criteria that medications do. Medications are entered into most electronic medical records in a specific format, as are diagnoses. Devices have no such requirement.
Let me give you an example:
A relative of mine had heart valve implanted a couple of weeks ago. The family (for their own reasons) wanted to know what kind of heart valve had been placed. There was no way to find that information until they went back to the doctor (who did a great job, by the way) for a post-operative visit many weeks after the surgery. She eventually found the answer to her question, but never received any document with that information. And if there was a recall of the valve at some time in the future, she would have no quick way to find out if she was at risk. To our knowledge, there is no scannable electronic system of patients such as my relative that contains information about their valve, its serial number and her vital information.
Think about it: at least with my car they can send me a letter. With my relative, if there is a problem she (or her family) will probably hear the news and then dial, dial, dial their doctor until they get an answer. And the odds are currently that the doctor will probably have to find the slip of paper that came with the valve in a paper medical chart instead of an electronic format.
I have talked to many people who have had the same experience with a medical device. One in particular was someone I met a couple of weeks ago who I assure you is an informed medical consumer. She told me the story that she had a hip replacement, and when a recall was announced she had to call her doctor’s office repeatedly to finally get the information whether or not her hip was the type recalled. The only place that information existed was in her doctor’s paper medical record. She was not happy, and I don’t blame her.
So here we are, with an announcement of a possible problem that affects a very small number of women and no simple way to get the answer whether or not the problem is real. I will leave it to you to decide whether the odds are too great or too little for you to influence your decision if you are planning on getting a breast implant. But once again we have a situation where there is going to be the inevitable media alarm, with the reality much different than the headlines will suggest.
We need to be an informed public, and we also need to understand the real risk of what we are talking about. And we deserve quick answers to these questions.
We are making a start in getting to that point, but we still have a long way to go.