Defining personalized medicine only in the context of genomics for drug discovery, patient stratification, and biomarker discovery is thinking too small, says Hill, a former computational physicist who founded GNS Healthcare, a health data analytics company, in 2000. We need to consider the whole universe of clinical information generated by randomized clinical trials, claims, electronic health care records, payers and providers—what’s being called the “data exhaust” of the digital healthcare era.You have to read this article - it provides a small snippet of what big data sets can do to improve people's health.
Actually observing what happens to people in the real world will discover the life events or conditions predictive of more serious health care issues in the future. It's common knowledge that smoking regularly predisposes people to higher rates of lung cancer, and other pieces of information have yet to find predictive value; for example, there's building evidence that living near highways contributes to childhood asthma. Do do so, you need large numbers of people for each event you'd like to investigate, and therefore an even larger population if those events aren't frequent.
Luckily for health information analysts, there are thousands of people with almost any medical condition that comes to mind.
But the analysis of big data, as far as health information goes, is hampered by numerous privacy issues, for good reason. The drawback is that the exchange of health information between organizations, even between hospitals, incredibly problematic and fraught with liability issues, and aggregating a lot of health data (even if it's been gathered in a publicly funded medical system, as in Canada) is difficult. Small value studies become much more effort than their worth, and so they remain in someones files and are soon forgotten.
However, once big data is organized in a systematic way, some investigation can happen. Ironically enough, I think insurance companies better poised to use big data approaches to help improve people's health levels than the medical system is. Large insurance companies simply have enough of a population base to approach problems from a big data analysis perspective without having to deal with anyone else, so data sharing issues don't apply. Colin Hill's company provides a nice example:
GNS Healthcare focuses on automated, hypothesis-free discovery of cause-effect relationships in “mashups of different types of data,” said Hill, with applications in drug discovery, biomarker identification, comparative effectiveness, and risk assessment. For example, Hill’s company worked with Aetna to predict which of their members with metabolic syndrome will go on to develop type II diabetes. Using data from 36,000 individuals and 600,000 lab results, the GNS model predicts who is most at risk, suggesting where preventive efforts need to be focused.
Despite the web being littered with stories of insurance companies disallow subscriber claims, saying that an insurance company has absolutely no interest in spending anything on their clients isn't true either.
No one really wants to use insurance for what it's intended for - serious losses - but if insurers can use data analysis to get preventative care to the right people, they have the potential to avoid huge claims down the line, and also benefiting their customers with a greater level of health.
Read Exploiting Big Genomic Data: Key to Personalized Medicine