Is HealthVault missing the self quantification point?
If Microsoft wants HealthVault to take over from Google Health, plumbing in APIs for the tracking devices people already use to monitor health and fitness would be a great start.
When Google canned its energy monitoring service, Microsoft's charmingly named Hohm quickly followed suit. So when Google recently announced that Google Health was going away by January, the HealthVault team took the time to blog that they weren't quitting - and came up with a way for users to transfer data from Google Health into HealthVault. That made me go back and take a look at my HealthVault to see if I could plug in the stats from my Fitbit pedometer or the BodyMedia FIT and BodyTrack Scale I've tried out recently - both of which sync into Google Health.
Not in the UK. Not one of the devices that upload data to Google Health is listed as something you can get data from. HealthVault does support some tracking devices; but according to the directory on the site, it supports a grand total of three devices from two companies, two for blood pressure and one scale.
(UPDATE: as you can see in the comments, the situation is different for US HealthVault users, who can upload data from a lot more devices.)
RunKeeper - which started out as an app to track your walking and running and has evolved into a service for combining measurements from multiple tracking tools - supports more than that already, including heart rate monitors, scales, stride sensors, pedometers and more.
The increasingly popular self-quantification and tracking tools are crying out for one place to show you correlated data, but unless Microsoft does something quickly, it's not going to be HealthVault, and that's a huge shame because we're on the cusp of having personal tracking data be useful for professional health services. An acquaintance had a defibrillator implanted last year and they went from weekly to monthly to six monthly to annual checkups of the defibrillator; but they also send tracking data to a server in Belgium weekly, which notifies the London hospital if it looks like there might be an issue to look at sooner. If that was paired with the details of the runs he's been tracking with a chest strap heart monitor and a GPS stride sensor on his running shoes, the data might be even more useful to the researchers who subsidise the gadget that phones home to Belgium every week.
The government is opening up the NHS to competition for things like treating leg wounds more cheaply (announced today, in case you missed it), but digital health would save a lot more money. The Amalga system Microsoft has for health professionals mines information form multiple systems inside a hospital; two researchers at Microsoft Research New England used the data in Amalga to look for phrases in patient notes that suggest they're likely to be back in A&E within 30 days because their problem has been misdiagnosed - which happens to a fifth of all patients in the US Medicare system. It's expensive for the hospital and unpleasant (and dangerous) for patients, and it turns out to be surprisingly easy to predict (look for words like fluid that suggest an infection - it's often hard to diagnose the cause and running a few extra tests to make sure turns out to be cheaper than having someone come back and starting from scratch). What the researchers found is now a Microsoft product called Readmissions Manager.
That's all data gathered by health professionals, in the hospital system, but there's a lot of extra information that we're starting to gather ourselves. How about a project using HealthVault to correlate levels of activity and sleep with blood pressure measures in a wide population?
RunKeeper is going to do some of that with its HealthGraph API (think social graph plus health and fitness data - see whether you walk more and sleep less than your friends) which is getting a 'correlation engine' that might point out that people who sleep an hour longer than you tend to run a little faster and weigh a little less. That's great, and I'm sure it will have a good privacy policy, but I'd rather see a central repository that I can keep private if I want from a company I'm little more certain will be in business in 25 years when I want to look back at whether I was walking enough in my 40s to build up bone density as a protection against osteoporosis. Plus, RunKeeper doesn't have a hospital IT product to turn interesting information into a way of treating people better.
If the HealthVault team jumped on the APIs that device makers like BodyMedia and Fitbit and Withings and BodyTrace already have and pulled those into the service, they'd get a lot of engaged users, and they could jump start using all this user-generated tracking data for more than just working out who walked more last week.
The self-tracking revolution is currently as the same stage as the personal computer in the days of the Homebrew Club; you have to put a lot of effort in and what you get out is of more interest than utility unless you're already a fitness expert or have access to a doctor who will analyse the information for you. That's when Microsoft jumped into personal computers with Altair BASIC and look where it is now. I think HealthVault is missing a big opportunity to be at the heart of personal tracking here…
Mary