Aug 22 2016

What Is Precision Medicine?

precision_medicineI feel that a lot of what I do here is separate marketing or ideological hype from reality. It’s a never-ending task, I think because humans have a tendency to view the world through narratives. We build a story about how we think the world works, and then that story drives our perception of the evidence (rather than the other way around).

Not only do we create our own narratives, but we communicate with each other through narratives, especially when we are trying to be persuasive. Marketing is an excellent example, because it has evolved over decades, trying multiple strategies, and preferring those strategies which work best. You may have noticed that many commercials and ads do not try to persuade you with fact, but rather try to convey a feeling, an image – a narrative. The same is also true of politics.

So deep is this tendency to view the world thematically rather than empirically, that even in science-based and highly technical areas like medicine we can encounter it. In the last few years I have been encountering the terms “personalized medicine” and “precision medicine” more often. Frequently this is in the context of blatant pseudoscience, but these terms are used even within respectable science-based medicine.

What do these terms actually mean and how is this different than what we are already doing?

The NIH has a precision medicine initiative. Here is their definition:

Precision medicine is an emerging approach for disease treatment and prevention that takes into account individual variability in genes, environment, and lifestyle for each person.

My initial response to this is – this is what we are already doing. Back in medical school I learned that we need to take an “individualized” approach to medicine. We are treating individual patients, and have to take into account the details of their medical history, social context, psychological state, and genetic background.

Precision medicine, therefore, is not an emerging approach, it is the same old approach we always had.

On the other hand, this is the same reaction I had when I first encountered the term, “evidence-based medicine.” Isn’t that what we are doing now? The answer turned out to be “sort of.” Modern medicine has always been based in scientific evidence, but evidence-based medicine was an initiative to formalize that approach, to quantify quality of clinical evidence, and to explore means of getting that evidence to the clinician when and where they need it.

As an aside I also think that EBM made a critical error in eliminating basic science research from their formal evidence evaluation, which focuses almost exclusively on clinical trials. This opened the door for highly implausible treatments to get a foothold using poor quality, preliminary, or biased clinical research. This deficiency is fixed with science-based medicine, which explicitly includes consideration of prior plausibility in evaluating the clinical evidence.

So what about precision medicine – even though this is what we have already been doing (like EBM) does the initiative formalize or prioritize anything useful? If I am being generous, then I think so. For the NIH initiative, for example, this is about dedicating researching dollars to studies that will provide information specifically useful for precision medicine.

I do think, however, that it is important to point out that precision medicine (and personalized medicine) do not represent, in my opinion, any discontinuity or fundamental change in approach to regular medicine. The only variable is the amount of information we have and therefore the degree of precision.

In a way, this is like high definition video. Video technology has been steadily progressing in terms of increased definition – more total pixels making up the image. We started at standard definition of 576i (“i” stands for interlaced). Then came the first “high definition” sets with 720p (“p” stands for progressive). Then 1080i, 1080p, quad HD and 4k. In other words, there is a continuum of increasing resolution, but you can call 720p “high definition.”

In the same way, we have always tried to used the best evidence available when treating individual patients. All evidence that we use to inform our medical decisions is based in group-level data – we test interventions in groups of people to derive statistical information, and then we apply those statistics to individuals.

Precision medicine is about doing this kind of research on smaller and smaller subgroups.┬áJust as with tv definition (although I know that technically the term “high definition” is regulated and refers to at least 720p), there is no point at which we are objectively doing precision medicine. There is just a continuum or more and more precise data applied to individual patients.

It has long been common practice in clinical trials to do subgroup analysis, separating out male, female, different age groups, and different ethnic backgrounds. This is not always possible, because you need large studies in order to have statistically meaningful data when you break it down into subgroups, but looking at subgroups to better target interventions is nothing new.

Further, we can break down subgroups by their medical history. Do we treat everyone, just those who already have disease, or do we base it on risk factors?

One thing that is relatively new is using genetic analysis to target treatments. This is increasingly becoming available because genetic analysis is getting quicker and less expensive, and we are gaining more and more knowledge about the association of specific genes and things like response to specific drugs.

We already use ethnic background and family history as a rough guide to probable genetic makeup. It is certainly more precise to simply test the individual for their specific genes. Genetic testing has been available for decades, and our knowledge base has been increasing steadily.

To emphasize one point – even precision medicine using specific genes to predict risk and response to intervention is still the application of group data to individuals. We know what those genes do because we have studied their statistical associations in large groups. We are just using more and more precisely defined groups.


Precision medicine and personalized medicine (which as far as I can tell are the same thing) are not different in any meaningful way from regular medicine. They do not represent a new approach, nor do they represent a discontinuity in the slow and steady progress of medicine.

The historical fact is that modern medicine has always had the goal of individualizing treatment as much as possible within the confines of our existing scientific evidence. This scientific evidence has been slowly progressing over the last century, and our ability to individualize has been getting more precise at the same pace.

Much like “high definition” is anything 720p and higher, “precision medicine” can be any time you treat a patient as part of any subgroup, rather than just a generic Homo sapiens. If I am treating my patient not just as a human but as a woman, or a middle-aged woman, or a middle-aged woman with a family history of Type II diabetes and hypertension, or with that profile plus certain genetic alleles that statistically predict how they will metabolize certain drugs – this is all precision medicine, just to varying degrees of precision.

At most precision medicine is a priority – prioritizing research so that studies are designed to drill down to the smallest subgroups possible. That is a reasonable and worthwhile priority.

I continue to have a problem, however, with the way such concepts are typically communicated to the public. There is a tendency to emphasize how new something is, and therefore to exaggerate how different it is from the past. This hype distorts reality, downplaying our prior knowledge and exaggerating innovation, often to the point of creating the impression of a discontinuity (a breakthrough or totally new approach) when in reality we only have incremental change.

Precision medicine is just a continued incremental improvement in scientific medical knowledge.

6 responses so far