When I turned 40 I celebrated my birthday at The Library Hotel. They boast a collection of over 6000 books organized according to the Dewey Decimal System. Each of the floors of guest rooms are arranged by 10 different themes--social sciences, literature, languages, history, math & science, general knowledge, technology, philosophy, the arts, and religion. Reading and writing nooks abound and if you find the right window you can glimpse the NYC library. The regal lions that adorn the entrance, Patience and Fortitude were renamed in the 1930s by Mayor LaGuardia for inspiration during the Great Depression.
I envision the data arm of data & donuts, datamonger.health as a blend of social enterprise and social business. By design it will be profit-generating but also geared toward social impact. Learn-Make-Do are three categories of folders I rely on. Think about it. Most of what bleeds into your inbox should fit into these buckets or just be deleted.
The blog has received quite a few page views over the last year. I don't typically check stats regularly. We all know how to get more hits, visits, or views if that is the goal. Keep writing listicles, misleading headlines, attaboys, and ride SEO glory right to AdSense glory.
Or pretend that all is right with the world of healthcare and medicine. Don't challenge anyone--accept the dogma, and keep the factory open. Don't be surprised when the disease of familiarity becomes pandemic.
One of my earliest sustainer memberships was from a physician I admire and respect. He wrote a simple note, "Keep the site ad-free". More joined over the years and I feel accountable to how I spend resources. Most fuels travel to Washington DC and medical conferences throughout the country. Although press credentials gain access, I typically cover my travel and lodging.
I am thinking of 3 relevant insights from the last few weeks of meetings or travel. I will summarize in a later blog but for now here are a few in no particular order:
1. Datasets are useful for hypothesis generating. Especially if you are relying on real world evidence. But when you want to test your hypothesis? You need to use a different dataset from the one you originally used for your hypothesis.
2. I value 12000 responses in non-probability sampling more than 100 from a probabilistic model. What this means is we don't really have a guide for reaching people or survey respondents. Is their bias in both approaches? Certainly.
3. Survey responses are only episodic determinations. If you want to measure the patient journey, you will need measures that can be tracked longitudinally. I see so many data points that add value for that particular survey administration but have little use in future assessment.
*Bonus--passive data can also be informative. This will be the sweet spot for wearables. Think GPS or activity information regarding movement and activity and how this might correlate with mood, anxiety, or depression...