Actuarial Outpost
 
Go Back   Actuarial Outpost > Actuarial Discussion Forum > Health - Disability
FlashChat Actuarial Discussion Preliminary Exams CAS/SOA Exams Cyberchat Around the World Suggestions

DW Simpson
Actuarial Jobs

Visit our site for the most up to date jobs for actuaries.

Actuarial Salary Surveys
Property & Casualty, Health, Life, Pension and Non-Tradtional Jobs.

Actuarial Meeting Schedule
Browse this year's meetings and which recruiters will attend.

Contact DW Simpson
Have a question?
Let's talk.
You'll be glad you did.


Reply
 
Thread Tools Search this Thread Display Modes
  #1  
Old 09-11-2018, 11:06 AM
WhosOnFirst WhosOnFirst is offline
Member
Non-Actuary
 
Join Date: Mar 2015
Posts: 137
Default ACO Savings

I read an article about ACO savings. There is a link to the actual study in the story.

https://www.forbes.com/sites/bruceja.../#7432f9702bbc

I quickly looked over the study and while I appreciate their methodology, there aren't enough details shown (in my opinion) to investigate if some selection bias exists between the ACO population and the non-ACO population. The information may be there but I haven't done a solid read of the study yet.

If the conclusions hold true, they are showing about a 1% decrease in annual claims cost for the ACO population. While not horrible, I don't think this is the panacea that some proponents make it out to be.
Reply With Quote
  #2  
Old 09-11-2018, 12:49 PM
sjb554's Avatar
sjb554 sjb554 is offline
Member
SOA
 
Join Date: Jun 2011
Studying for FAP
Favorite beer: Miller Lite
Posts: 1,188
Default

Quote:
Originally Posted by WhosOnFirst View Post
I read an article about ACO savings. There is a link to the actual study in the story.

https://www.forbes.com/sites/bruceja.../#7432f9702bbc

I quickly looked over the study and while I appreciate their methodology, there aren't enough details shown (in my opinion) to investigate if some selection bias exists between the ACO population and the non-ACO population. The information may be there but I haven't done a solid read of the study yet.

If the conclusions hold true, they are showing about a 1% decrease in annual claims cost for the ACO population. While not horrible, I don't think this is the panacea that some proponents make it out to be.

I agree with your conclusion that ACOs save money, and also that they aren't the single best thing to happen to healthcare. If the comparison point is sticking with FFS, ACOs are a good path forward as a piece of the solution.



It's important to consider the current political climate here. I heard a guy from the OMB speak a few months ago, and he said:


1) I am a libertarian
2) These programs don't save money
3) I am trying to shut them down
4) Mic drop


__________________
HTML Code:
hi
Reply With Quote
  #3  
Old 09-11-2018, 04:52 PM
kooky cookie kooky cookie is offline
Member
 
Join Date: Oct 2001
Posts: 117
Default

Thanks for sharing that - I'll check it out.

For your consideration:

http://pnhp.org/blog/2017/11/09/slug...organizations/

https://www.wsj.com/articles/SB10001...96902005944398

https://www.forbes.com/sites/robertp.../#3dae7c2b65f2

I think what I find most disturbing in these sorts of situations (sweeping changes promulgated from above) is how often there is little to no evidence that things will work out like policymakers want them to.

Last edited by kooky cookie; 09-11-2018 at 05:08 PM..
Reply With Quote
  #4  
Old 09-12-2018, 12:13 PM
WhosOnFirst WhosOnFirst is offline
Member
Non-Actuary
 
Join Date: Mar 2015
Posts: 137
Default

Quote:
Originally Posted by kooky cookie View Post
Thanks for sharing that - I'll check it out.

For your consideration:

http://pnhp.org/blog/2017/11/09/slug...organizations/

https://www.wsj.com/articles/SB10001...96902005944398

https://www.forbes.com/sites/robertp.../#3dae7c2b65f2

I think what I find most disturbing in these sorts of situations (sweeping changes promulgated from above) is how often there is little to no evidence that things will work out like policymakers want them to.
I haven't worked a lot with ACOs. The few situations I did work with were fairly narrow and the contracts with the provider paid the provider a portion of the PMPM difference from year 1 to year 2. What I saw in the results was that many of the high cost members who were managed by the physician had a decrease in cost because the high PMPM in year 1 was due to a one time event that was unlikely to repeat. The population that I was looking at also had about 15% of the members never even interacting with the physician group but that 15% was a significant portion of the payments to the physician group.

I agree that evidence is sketchy at this point. I also think that the administrative aspects of these programs are not well thought out at this point in time. I'm hopeful that this will improve over time.
Reply With Quote
  #5  
Old 09-12-2018, 01:43 PM
George Frankly's Avatar
George Frankly George Frankly is offline
Member
SOA AAA
 
Join Date: Aug 2011
Location: CO
Favorite beer: Sam Smith Oatmeal
Posts: 10,158
Default

Quote:
Originally Posted by WhosOnFirst View Post
I haven't worked a lot with ACOs. The few situations I did work with were fairly narrow and the contracts with the provider paid the provider a portion of the PMPM difference from year 1 to year 2. What I saw in the results was that many of the high cost members who were managed by the physician had a decrease in cost because the high PMPM in year 1 was due to a one time event that was unlikely to repeat. The population that I was looking at also had about 15% of the members never even interacting with the physician group but that 15% was a significant portion of the payments to the physician group.

I agree that evidence is sketchy at this point. I also think that the administrative aspects of these programs are not well thought out at this point in time. I'm hopeful that this will improve over time.
I'm not super close to ACOs, but I am close to value-based care (I work in oncology).

Just a comment on the methods related to your underlined bit, as I understand them (again, not an ACO expert). This is often called 'regression to the mean,' and it's quite common. You typically see large upfront expenses for many dx, both for diagnostics and tx, and then you see the pt get it under control and costs stabilize. I even see it in my cancer pts, you get a dx, a bunch of labs/imaging, then massive drug/radiation/surgery bills, and after six months it's tapered off a lot. Might come back later, of course.

But there is symmetry in the calculation, those pts also exist in the control group. So it happens in your study group, but also in the comparator group. So it get's baked into the trend factor for pre-post studies, or is present in the control group for concurrent control studies.

And you also have pts on the other side of the coin, who were cheap last year and have a stroke this year and rack out $100k in bills. But again, you have it on both sides. Please don't tell the folks in Political I'm making a 'both sides' argument.

Now, your comment about ACO selection bias is a good one. I think it's likely that the more motivated practices are much more likely to join these programs, and comparing more motivated physicians to less motivated physicians... could be a problem.

My two cents.
Reply With Quote
  #6  
Old 09-12-2018, 08:32 PM
kooky cookie kooky cookie is offline
Member
 
Join Date: Oct 2001
Posts: 117
Default

I believe that one of the biggest problems with the analyses that have been done is that the costs of setting up and maintaining the ACO are excluded. These costs, which I would think can be considerable, include technology/infrastructure and organizational costs, among others. ACOs are conglomerations of primary care practices, specialist practices and hospitals. These entities probably don't have EMRs that can talk to each other; if they can't, tests will be duplicated and care will not be well-coordinated (so savings will be missed out on). And an ACO would have to create organizational structure to lead the effort to improve care and lower spending, to measure everything and to figure out how to divvy up payments. This is to say nothing of the costs that CMS incurs in overseeing ACOs, promulgating rules, studying results, etc.

Another big problem is that ACOs are compared to entities that do not exist as ACOs and that do not have the same incentives to pay attention to the ACO measures. One person put this as, ACOs game the system while comparison entities don't.

While the article focused on ACOs in the MSPP program, there is also the Pioneer ACO program. Most of the Pioneer ACOs have dropped out of the program because they could not achieve savings and/or the costs of participating were too high.

Finally, it's hard for me to see primary care and hospitals as being on the same "side" - primary care is like maintenance on your house or car. The idea is to keep things in good working order and avoid expensive repairs. I believe that the proper role of primary care is to keep patients in "good working order" and keep them out of the hospital if possible. But hospitals, of course, want primary care (and specialists) to route patients to them. Seems like perverse incentives at work whenever hospitals and primary care are in the same organization.
Reply With Quote
  #7  
Old 09-13-2018, 11:15 AM
WhosOnFirst WhosOnFirst is offline
Member
Non-Actuary
 
Join Date: Mar 2015
Posts: 137
Default

Quote:
Originally Posted by George Frankly View Post
I'm not super close to ACOs, but I am close to value-based care (I work in oncology).

Just a comment on the methods related to your underlined bit, as I understand them (again, not an ACO expert). This is often called 'regression to the mean,' and it's quite common. You typically see large upfront expenses for many dx, both for diagnostics and tx, and then you see the pt get it under control and costs stabilize. I even see it in my cancer pts, you get a dx, a bunch of labs/imaging, then massive drug/radiation/surgery bills, and after six months it's tapered off a lot. Might come back later, of course.

But there is symmetry in the calculation, those pts also exist in the control group. So it happens in your study group, but also in the comparator group. So it get's baked into the trend factor for pre-post studies, or is present in the control group for concurrent control studies.

And you also have pts on the other side of the coin, who were cheap last year and have a stroke this year and rack out $100k in bills. But again, you have it on both sides. Please don't tell the folks in Political I'm making a 'both sides' argument.

Now, your comment about ACO selection bias is a good one. I think it's likely that the more motivated practices are much more likely to join these programs, and comparing more motivated physicians to less motivated physicians... could be a problem.

My two cents.
You nailed it exactly, what I was seeing was regression to the mean. The problem with the specific programs I was looking at is that there were only 20-40 enrolled and they were selected primarily based upon very high year 1 costs. I was seeing a lot of members who went from $100K in claims to $20K in claims that was likely due to regression to the mean as opposed to the medical management. The problem with the program payment mechanisms was that we paid on the $80K "savings". Meanwhile, if there was a patient that went from $10K in claims to $50K that wasn't considered a balancing cost against the payout. I was making an effort to get the powers that be to redesign the payment structure to account for the downside risk as well as the upside to balance out some of the regression to the mean. The short story is that the physician groups wouldn't sign those agreements and take the risk. I was going to try to push forward performing better selection of the managed members but I left that team to pursue less frustrating (and more exciting) work.

I want to be clear that this is not a condemnation of ACO or medical management in general. This was a failure of these specific programs due to a lack of understanding by the people who implemented them. That's why I found the actual study presented in the article to be worth investigating.
Reply With Quote
  #8  
Old 09-13-2018, 11:29 AM
George Frankly's Avatar
George Frankly George Frankly is offline
Member
SOA AAA
 
Join Date: Aug 2011
Location: CO
Favorite beer: Sam Smith Oatmeal
Posts: 10,158
Default

Quote:
Originally Posted by kooky cookie View Post
I believe that one of the biggest problems with the analyses that have been done is that the costs of setting up and maintaining the ACO are excluded. These costs, which I would think can be considerable, include technology/infrastructure and organizational costs, among others. ACOs are conglomerations of primary care practices, specialist practices and hospitals. These entities probably don't have EMRs that can talk to each other; if they can't, tests will be duplicated and care will not be well-coordinated (so savings will be missed out on). And an ACO would have to create organizational structure to lead the effort to improve care and lower spending, to measure everything and to figure out how to divvy up payments. This is to say nothing of the costs that CMS incurs in overseeing ACOs, promulgating rules, studying results, etc.

Another big problem is that ACOs are compared to entities that do not exist as ACOs and that do not have the same incentives to pay attention to the ACO measures. One person put this as, ACOs game the system while comparison entities don't.

While the article focused on ACOs in the MSPP program, there is also the Pioneer ACO program. Most of the Pioneer ACOs have dropped out of the program because they could not achieve savings and/or the costs of participating were too high.

Finally, it's hard for me to see primary care and hospitals as being on the same "side" - primary care is like maintenance on your house or car. The idea is to keep things in good working order and avoid expensive repairs. I believe that the proper role of primary care is to keep patients in "good working order" and keep them out of the hospital if possible. But hospitals, of course, want primary care (and specialists) to route patients to them. Seems like perverse incentives at work whenever hospitals and primary care are in the same organization.
Regarding the first point, I'm not sure how the programs you're looking at are structured. If the payers are making payments to the providers, in the form of a PMPM payment or some other form of cash, then those should absolutely be deducted from any savings paid out. If the providers are making investments without the payer chipping in, then you wouldn't deduct those. I'd consider those as practices investing with the hope of getting shared savings.

Agree with your second point.

The last point is a tough one. You're right, a lot of these programs are trying to encourage physicians to keep pts out of the hospital. But that means less revenue to the hospital. The only way you can make that work is to share revenue with the hospitals through the ACO-type programs. Conceptually, let's say you avoid a $20k admit, thereby saving the payer $20k. And let's say on that $20k bill, the hospital would have made a $5k profit. If the payer then paid $8k to the hospital as shared savings, then the hospital is up $3k, and the payer saves $12k net. Now, in practice, teasing that all out is impossible. No perfect way to do it, the question is whether or not you can come up with a reasonable way.

I hope that makes sense. Need more
Reply With Quote
  #9  
Old 09-13-2018, 11:31 AM
George Frankly's Avatar
George Frankly George Frankly is offline
Member
SOA AAA
 
Join Date: Aug 2011
Location: CO
Favorite beer: Sam Smith Oatmeal
Posts: 10,158
Default

Quote:
Originally Posted by WhosOnFirst View Post
You nailed it exactly, what I was seeing was regression to the mean. The problem with the specific programs I was looking at is that there were only 20-40 enrolled and they were selected primarily based upon very high year 1 costs. I was seeing a lot of members who went from $100K in claims to $20K in claims that was likely due to regression to the mean as opposed to the medical management. The problem with the program payment mechanisms was that we paid on the $80K "savings". Meanwhile, if there was a patient that went from $10K in claims to $50K that wasn't considered a balancing cost against the payout. I was making an effort to get the powers that be to redesign the payment structure to account for the downside risk as well as the upside to balance out some of the regression to the mean. The short story is that the physician groups wouldn't sign those agreements and take the risk. I was going to try to push forward performing better selection of the managed members but I left that team to pursue less frustrating (and more exciting) work.

I want to be clear that this is not a condemnation of ACO or medical management in general. This was a failure of these specific programs due to a lack of understanding by the people who implemented them. That's why I found the actual study presented in the article to be worth investigating.
Gotcha. I'm not all that surprised. I have seen some pretty appalling methods applied to measure savings over the years. And I've also seen studies that really looked pretty solid, but the results were just unbelievable, for reasons I couldn't quite unpack.
Reply With Quote
  #10  
Old 09-13-2018, 11:57 AM
exponentialpi's Avatar
exponentialpi exponentialpi is offline
Member
SOA AAA
 
Join Date: Nov 2008
Location: Po
Studying for ACA
College: Graduate of hard knocks
Favorite beer: Cold Smoke
Posts: 1,457
Default

Quote:
Originally Posted by George Frankly View Post
I have seen some pretty appalling methods applied to measure savings over the years.
I have seen this vary widely over my career. The craziest one was another department using a simple interest calculation for a multi-year study. I was a greenhorn way back then and did not speak up.
Reply With Quote
Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off


All times are GMT -4. The time now is 05:11 PM.


Powered by vBulletin®
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.
*PLEASE NOTE: Posts are not checked for accuracy, and do not
represent the views of the Actuarial Outpost or its sponsors.
Page generated in 0.20967 seconds with 11 queries