Emergency Department Admissions: Analysis of ECDS Dataset (part 4)

SMLR modelling of an ECDS extract of 237k admission records to the emergency departments of an undisclosed NHS Trust: the healthy vaccinee effect confirmed using decision to treat

I appreciate that getting one’s head around staged multivariate logistic regression modelling (SMLR) with propensity adjustment can be a bit too much for many folk, even with a large teapot on the go, several rounds of buttered toast, and the promise of fresh-baked mince pies.

But we’ve been here before, and those with sufficient stamina might want to flip back to this article to refresh the mind and illuminate the soul. Those with little time to spare might prefer a simple summary of what I’m doing and why.

SMLR: A Simple Summary

It’s pretty much a given that the older you get the more things go wrong with your body, so we might expect that older folk are more likely to get treated within the emergency department when they call for an ambulance. We might also expect them to be at greater risk of hospitalisation after that treatment. To account for this we can throw their age into an analysis.

But it isn’t just about age, for biological males are different from biological females (regardl…