Healthcare: It’s the Data, Stupid

Fact: Today’s L.A. Times has a startling report: “A stunning number of people who work in healthcare settings lack paid sick time — as many as 75% of all home health aides, for example… Federal data indicate that as many as 29% of all workers in the ‘healthcare and social assistance’ job sector lack paid sick days. Healthcare employees who work while ill may end up hurting the people they are hired to help….”

Analysis: Mark Twain said the only two sure things in life were death and taxes.  So it’s no surprise that the two presidential campaigns are focusing on healthcare and the economy, since people are universally affected in personal ways.  Forget taxes for today, I’m interested in technology’s role in healthcare, which is growing, and there’s no more potentially game-changing facet of that than the role of data. 

That Los Angeles Times story has a number of examples where the data just pop out at you, bonking you on the head.  And there’s prescriptive (no pun) value when you find and understand good data; an example in the story is “A 1996 study in the American Journal of Epidemiology found that the incidence of disease in the workplace is reduced when workers have paid sick days.”  That’s evidence for a policy change with demonstrable ROI.

By the way, I wonder if James Carville gets residual credit for the newcomer cliche based loosely on his famous line from 1992.  The phrase, “It’s the data, stupid,” including the quotation marks for precision, now gets 1,900 hits on Google, and more than 4,400 on Live Search.  (Hey, notice that disparity?)

Data has to be aggregated and maintained as records, and I intend to write more about the subject of health records in future.  One area I’m particularly interested in is personal health records (often abbreviated as PHRs).  There’s a good story today on TMCnet (“Work Remains in Creating Access to Online Health Records“) giving a quick overview of two major partnering efforts – one by Kaiser Permanente using Microsoft’s HealthVault, the other by the Massachusetts state Blue Cross Blue Shield using Google Health. 

Most insurers can’t offer portable records to members unless they partner with a Microsoft or a Google,” according to Carlton Doty, a senior analyst with Forrester Research, quoted in TMCnet story.

I’ve been listening to an interesting podcast series by Perot Systems on “Healthcare Tech and the World,” and the most recent is a good interview with Dr. William Stead, Associate Vice Chancellor for Strategy/Transformation and the Chief Information Officer at Vanderbilt University Medical Center.  He makes some incontrovertible points:

  • “First, we’re actually going to shift from expert-based care to system-supported practice.”
  • “Second, we’re going to shift from medicine that is appropriate for mass delivery…we’re going to really shift to individualized medicine.”
  • “The third component is… that we… will embed support for health into a person’s living environment.”

We’ve got Microsoft efforts (some using current technologies, some in the R&D labs) which are addressing all three areas.  One of the most exciting new things we have going is called Amalga, as in amalgamating data streams from multiple mutually-incoherent sources into coherency.  We call it “a unified approach to healthcare intelligence,” and long before I joined Microsoft my group at DIA looked at the precursor prototype system (then known as Azyxxi, before Microsoft acquired the company) for use in the intelligence community.  That hasn’t happened yet, but the technology is impressively flexible and I think it could be used in almost any domain

And a final point back on the data front: a British National Health Service (NHS) doctor writes today in The Guardian defending the practice of data-mining in the healthcare arena (“All Patients Can Benefit from Our Collecting of Statistics“).  I recommend that article, as his point is valid across any measurable discipline:

You can measure outcomes in all branches of medicine, using various methodologies depending on the patient and the condition being treated. In surgery you might monitor how quickly a patient walks again after a hip replacement, or the number of avoidable errors a particular surgical team makes. In psychiatry you can measure the extreme and distressing outcomes like self-harm and suicide.  But you can also measure the number of patients who successfully return to work, or the number of addicted patients who live free of their dependency for a defined period or time. In recording and analysing those outcomes you can look for the commonalities that point to either success or failure and use this information to improve the quality and safety of the care you provide.”

It’s an exciting time if you like data, data-mining, and the unpredictable insights and value they bring to medicine and improved healthcare.  And, since I’m getting older like the rest of us, I’ll be paying more attention to healthcare anyway 🙂


Email this post to a friend

AddThis Social Bookmark Button

4 Responses

  1. This discussion reminds me of the concept of Combined Personal Time Off (CPTO) which human resources organizations introduced into my workplace in the 90s. If you haven’t heard of this one, it combines vacation and sick time into personal time thus either a) encouraging people to stay healthy and avoid malingering or b) screwing people who actually get sick by forcing them to use all their vacation time on being sick. Take your pick.

    I’d be interested to know if any of these articles distinguish between sick time and CPTO.


  2. A side note is that every enthusiastic discussion of data mining in health should, int these days of lightning-fast communication on the Interweb, be accompanied by a huge assertion of the paramount role of privacy in health care information. A lot of people have a nuance-free attitude to reading about corporations and government working together to mine citizen health records.


  3. Great points, Fred. On CPTO, I didn’t see that particular distinction in what I was reading, but your point makes sense.

    On privacy, you’re absolutely right on its criticality. Here’s a link to the Microsoft approach with HealthVault:


  4. […] Healthcare: It’s the Data, Stupid […]


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: