Modern medicine has learned a great deal about how different variables effect medical outcomes. Modern statistical methods have produced a wealth of knowledge about how the human body interacts with its environment. For example, modern doctors now known, through hundreds of highly replicated studies, that cigarette smoking causes cancer, particularly lung cancer. Doctors have even quantified the relationship that smoking has on, for example, overall mortality. Today, it is known that people who smoke about two packs of cigarettes per day, on average, will live approximately a decade less than those who do not.
However, all of the studies that produce this sort of knowledge are bound by the constraints of statistical inquiry. To see how this can effect the quality of knowledge that doctors must use, consider the example above. While it is true that there is a very strong correlation between cigarette smoking and an early death, there are many statistical outliers. Everyone’s heard of the guy who smoked three packs per day and lived well into his 90s. Likewise, plenty of people who have never smoked a cigarette in their lives die of lung cancer in their 50s. From the statistical point of view, the degree of variation in outcomes for some populations away from the general trend is termed variance. Variance that occurs within the context of a known predictor variable is, by definition, assumed to be random and learn more about Eric.
However, let’s look at the above example again. Is the fact that the man who smoked 3 packs living into his 90s really random, or is there something more concrete going on there? Could it be that the man’s genetics somehow made him immune to the health consequences of chain smoking? This is where Eric Lefkofsky, founder of Tempus, believes that modern medicine has a long ways to go.
Tempus is creating a set of analytic tools that will allow doctors to use the entire human genome to draw conclusions about the likely health outcomes of almost any imaginable cohort. Instead of relying on crude population-level data, what’s going on the individual’s genetic level may lead to treatments more effective than anything ever before imagined and more information click here.