"An approximate answer to the right problem is worth a good deal more than an exact answer to an approximate problem."
-John Tukey
A sigmoidal regression for determining high risk patients
A sigmoidal regression is a way of determining if a single observation stands as a failure or a succes in a probability analysis, for determining if a patient might be of high risk one or a normal one this kind of tools could be use. In this case a "success" is when a patient is a high risk one, and a failure when a patient is not.
Multiple Linear Regression
For determining the result of the sigmoidal, a buch of variables must be cinsider such as: weight, age, gender, if the patient has diabetes or another kind of comorbility. (For determining the most important comorbilities we could use a corraltion matrix).
​
After the multiple linear regression(MLR) is done we could mix the idea of the sigmoidal function and the MLR, in order to determine if a pacient with a specific sets of characteritics could be classified as a high risk patient or not.
What is next?
Find some repository such as Kaggle or GitHub in order to download real data and perform the sigmoidal analysis.