The mHealth Data Dilemma
I talk with people a lot about data collection. After all, when you break Mood 24/7 down to its core, you find a data collection service. When pitching the benefits of tracking a single daily health data point, I’ve found the general response to be quite telling of American culture…
Nine times out of ten I’ll get asked if we can get more data. This response is amazing to me, for two reasons:
The first reason being, we aren’t currently receiving even one health data point on a daily basis from over 95% of Americans (my stat)… And although I have no proof, yet, I personally feel that the level of care in America can be increased exponentially with just a single data point tracked per person per day. Regardless of the data point.
The second reason this response is amazing to me is the implication that it’s trivial to insert a new behavior into an individual’s routine for an extended period of time… let alone a new behavior multiple times per day. The assumption may sound logical, that if we can get one data point, then we should be able to get multiple data points from the same individual. But it is not realistic when looking at a value curve. From a user’s perspective, multiple data entries per day increases complexity, while providing marginal utility gains.
We also know that healthy behavior change just isn’t that easy. If we can use weight as a data point and obesity as a proxy for American interest in their own health over the past 25 years, then it’s easy to see that we have enough of a challenge on our hands getting people to track even one health data point daily.Which is a big reason why I push back every time I’m asked the question about getting more data.
The other reason is that we can’t afford to look past simple data sets and visualizations because they’re “not as useful” as all encompassing individual health data streams. The truth is there is a lot that can be trended and deduced from a single data point per day, and those health trends over weeks and months are far more valuable than the relationship between multiple daily data points.
That’s my take, anyway.
I’m very interested to read why I’m right or wrong on this one, as I see mHealth and data collection being synonymous moving forward.