Group 1: 
+Timely issue clinically, especially with prescription opidoids
-However opioid fraud/abuse is an issue much more common in younger populations (i.e., people in the age range 20-50 years old), which are not captured in Medicare
+There is quite a bit of variability in Medicare Part D prescription drug use (Reference: http://www.ncbi.nlm.nih.gov/pubmed/22316446; Reference-My NEJM Comment!! http://www.ncbi.nlm.nih.gov/pubmed/22571214)
+You note in Limitations that over-the-counter drugs won't be covered…but actually fraud is more likely occurring with the more novel therapeutics still on-patent (i.e., not over-the-counter meds)
+ Figure 1: NPs and PAs essentially function as family practioners/internal medicine doctors as providers of primary care; this wide variability between NPs/PAs vs FPs/IMs is very interesting (as drug companies target advertising dollars toward FPs/IMs but not NPs/PAs)
-You have to somehow account for inherent variability in patient populations (for instance, certain docs may see more pain patients than other docs…thus leading to appropriately differential amounts of pain medication prescribing). how would you account for this?
+ Interesting methodology using Personalized PageRank!
- I would be careful making the assumptions on Slide 13; for instance, spine surgeons, pain medicine doctors, physiotherapists, and internal medicine doctors may all prescribe the same drug(s) related to pain, but this is often appropriate as we tinker around with these meds collectively; although this system is probably more disjointed than it should be, it doesn't necessarily always lead to fraud (most of the time, I would say it probably doesn't…though it does point to an inefficient system of providing medical care).
-Agree with your first bullet point on Slide 14. You could focus on opioid use in ophthalmologists or cardiologists or kidney doctors for instance (these people should almost never have to prescribe pain medications for extended periods of time)

Group 2
+ Mostly the same feedback as Group 1 applies here.

Group 3
+ I'm sorry I wasn't able to be in class during this presentation. This is a great idea, depending on how the numbers and associations shake out, we could publish this in a good journal! Please keep in touch and let me know how I can help this progress along!
- We you able to find a solution to the 9- versus 5- digit zip code issue?
-As you note in your presentation, there doesn't seem to be a great relationship between serious complication score and deprivation index. Is there a relationship between deprivation index and readmissions?

Group 4
Seems like an interesting idea

Group 5
I think I maybe needed to be there in person to understand the full context, but nice start