How to generate statistically relevant group sizes for activity levels

#1
Hi, I hope you can help.

I am looking at allocating hospitals to the following groups, LHA( Low hospital activity) MHA(medium activity HHA(High hospital activity. For which I will then perform analysis on the costs associated with delivering that surgery, as we have costs associated with each hospital and procedure. What I am trying to prove that those hospitals with low activity either have high or low costs.

What I am not sure of is the allocation to groups. Could I use quartile Upper, Middle and Lower or is there a better formula to assign?

I have 137 suppliers of activity with volumes and spend per unit and I wish to review the amount (£) spent by each supplier to demonstrate that low activity increases the unit cost and those with high activity might have low costs per unit. However, I need to agree with what low medium and high is

Thanks for your help in advance
 

Karabiner

TS Contributor
#2
Why do you want to artificially group these units? You could directly analyse the association
between activity (the measure which you want to bas the grouping on) and the dependent
variable(s), without categorization. This would potentially permit, for example, to demonstrate
things like "an X% increase of activity is associated with Y% cost decrease".

With kind regards

Karabiner
 
#3
Why do you want to artificially group these units? You could directly analyse the association
between activity (the measure which you want to bas the grouping on) and the dependent
variable(s), without categorization. This would potentially permit, for example, to demonstrate
things like "an X% increase of activity is associted with Y% cost decrease".

With kind regards

Karabiner
The challenge is the 137 suppliers have a profile of activity of 1 through 47 and a cost associated from £412 to £20878. So i want make sure i treating each group fairly.

How would you do it?
 

hlsmith

Not a robit
#4
@Karabiner - this is a redundant post.

@Openview2 - please don't post the same thing twice - it wastes people's time - since @Karabiner just typed the same thing I did in another place. It's the flip of a patient having to explain the condition to schedule an appointment, then to the nurse, then to the learner (resident), and finally to the physician. The patient's time is wasted.
 
#6
@hlsmith sorry I thought I had posted it down a level in the forum, so I wanted to make sure I was not allocating just to one area. I am not wanting to waste time. Just trying to get additional ideas. So my mistake for posting may be in the wrong section. Thanks for the support. @Karabiner could you look at the other post as well then as you might not have seen this