Visualizing World Health
This weekend, I've been exploring ideas for making meaning out of data. I could wax philosophically about how we are all drowning in data, but I will spare you. You already know how much time you spend every day parsing through email, snail mail, news, tweets and "what not" to decide what is relevant or needs response. Earlier this week, I presented at the 2010 TXHIMA Conference and attended several talks by health service professionals struggling to support meaningful use of electronic health records and transition to ICD-10. On down the line, I know that the data gathering challenges we see today are just a warm-up for the challenge of creating a shared system of health data that can be trusted and beneficially mined.
Our future challenges include the following in roughly this order:
- Improving the quality of data transformed into an electronic format. We used to say "garbage in garbage out". Errors introduced into shared data will be difficult to catch and correct unless we clarify the rules for "good data" that will allow us to automate validation checks for data input into the system.
- Clarifying data context. Much has been written about issues of using data out of context; for example, intermixing clinical and claims data. The variations of the human body, it's maladies, the drug cornicopia and claims documentation requirements all contribute to some of the most complex data schemas one could imagine. The heart of deriving meaning from data is creating classifications and mapping data relationships in unambiguous ways. We have a ways to go to remove remaining ambiguities from our classification systems, agree upon a standard and transform vast amounts of unstructured data into structured data that we can access.
- Improving data access. I have a dream that the world will experience a renaissance in medicine within my lifetime driven by the ability to access results for treatments on populations at any level of granularity. For the first time in history, researchists will be able to access significant numbers of complete case studies cross-referenced to all the factors influencing their condition. Advanced statistics will reveal the most promising areas to study versus the arduous process of medical research and deadends we use today. To get there, we will need to transcend the motivations that is keeping our data siloed.
- Improving data presentation and personalization. My dream for utopia includes every person's ability to apply what has been learned toward cost-effectively improving his/her health. Interactive health portals, online communities and personal health records will become more engaging and less cumbersome. Participants will have ready access to the information and encouragement they need to make better choices and obtain timely care.
OK, so onto the good stuff...a list of the most interesting health-related data visualization sites I found this weekend:
