Undue reliance on or use of facts that can be quantified or analysed using mathematical or statistical methods; inappropriate application of such methods, especially in the fields of sociology and anthropology.
I work with data. Big messy data. I don't necessarily like the term big. I am using it here as a familiar. I think if you show up with a plan and a set of skills you don't need all of the data--you should be trying to curate or generate the right data. If anything, I like wide data. Lots of columns and variables to create the granularity we need in our over-aggregated data populations.
Importantly, we also need critical thinking and personal judgement. Providers are tired of regulation, false doomsday warnings of artificial intelligence and power algorithms. Treating to the mean in actual patient populations might seem like evidence-based medicine but what we need is practice-based evidence. Just ask Siddhartha Mukherjee, MD, DPh.
The Laws of Medicine should be required reading for a data career in healthcare. Maybe now that patient-centricity, patient-engagement, and shared decision making are all the rage perhaps we will starting to pay attention.
If you are aggregating data into large buckets you are losing the patient. The point of care is where the decisions are made. Decisions about risks and benefits and if the available data can guide the best care for THIS patient--the one in front of us. Patients are starting to wonder. Are you seeing the real me?
Here are the laws of medicine described by Siddhartha Mukherjee.
1. A strong intuition is much more powerful than a weak test
2. "Normals" teach us rules; "outliers teach us laws
3. For every perfect medical experiment, there is a perfect human bias
'laws of medicine' are really laws of uncertainty, imprecision, and incompleteness. They apply equally to all disciplines of knowledge where these forces come into play. They are laws of imperfection--Siddhartha Mukherjee
Read more here: Carpe Donut
An appreciation for chaos in the data world organically created an interest in Bayesian probabilities. We don't know anything for certain.
In fact probabilities are subjective--we seek a reasonable quantification of an idea or outcome. Naturally frequency data are brought to bear and combined with prior knowledge.
An important tenet of Bayesian thinking demands an update to probabilities when new information becomes available across all nodes.