PBM Data – Apples and Oranges

When I first saw the data below from AIS’s quarterly pharmacy benefits survey (2Q 2010), I was excited. I hadn’t ever seen a breakdown of claims filled by tier across multiple PBMs. But, I was disappointed as I dug in.

Percentage of Claims Processed per Tier, per PBM, as of 2nd Quarter 2010 
           
  

% of Claims Processed

Company

1st Tier

2nd Tier

3rd Tier

4D Pharmacy Management Systems

78

16

7

ACS, Inc.

36

50

14

Aetna Pharmacy Management

64

20

16

BioScrip

50

39

11

Burman’s Specialty Pharmacy

90

8

2

CIGNA Pharmacy Management

67

26

7

Envision Pharmaceutical Services, Inc.

70

25

5

Factor Support Network Pharmacy

95

5

First Health Services Corporation

48

37

15

FutureScripts

66

18

16

HealthSmart RX

48

41

11

HealthTrans

69

16

16

Maxor National

40

55

5

National Pharmaceutical Services

78

17

5

Navitus Health Solutions

68

28

2

Northwest Pharmacy Services

54

37

9

OncoMed The Oncology Pharmacy

90

10

Partners Rx Management

72

19

9

PBM Plus, Inc.

60

35

5

Prescription Solutions

55

25

20

Prime Therapeutics

68

20

12

RegenceRx

73

15

12

RESTAT

71

17

12

SXC Health Solutions/informedRx

70

24

6

United Drugs

64

29

7

Walgreens Health Services Division

55

30

15

 

Where is the data from Express Scripts, Medco Health Solutions, and CVS Caremark? I’m sure someone chose not to participate, but I don’t think they would be hurt by these numbers.

Why are the numbers so different across companies?

  • 1st tier ranges from 36% to 95%
  • 2nd tier ranges from 5% to 55%
  • 3rd tier ranges from 0% to 20%

What do I do with this? Do they all define tiers the traditional way (i.e., 1st tier = generics, 2nd tier = brand formulary, 3rd tier = non-formulary brands)?

Are they all even in the same business (i.e., a specialty only company might not have any generics)?

Do they serve similar populations (i.e., Medicaid is very different than commercial)?

Now, I could certainly drill in one by one. For example, if I pull out the companies that I believe have similar diversified clients, I get a few more focused questions:

 

1st Tier

2nd Tier

3rd Tier

Aetna Pharmacy Management

64

20

16

CIGNA Pharmacy Management

67

26

7

Envision Pharmaceutical Services, Inc.

70

25

5

FutureScripts

66

18

16

Navitus Health Solutions

68

28

2

Prime Therapeutics

68

20

12

RegenceRx

73

15

12

SXC Health Solutions/informedRx

70

24

6

Walgreens Health Services Division

55

30

15

 

Now, I see some distinct clustering that maps closer to what I would expect. I would expect a PBM to have 65-70% generic (1st tier) utilization); 25-30% brand formulary (2nd tier) utilization; and 5-10% non-formulary brand (3rd tier) utilization.

The two things that jump out for me here are then:

  • Why does Walgreens have such low 1st tier utilization? My general perception is that they do a good job at driving generic utilization and have incentives in place for their pharmacists to do that.
  • Why does Navitus have only 2% third tier utilization? Do they have a closed formulary (i.e., no 3rd tier drugs are covered without some medical exception)?

In the same issue of Drug Benefit News (DBN 8/13/10) and the same survey, there were a few other data points:

  • The total number of claims processed by PBMs increased by 9.97% over the past year to 2.56B.
    • Does that mean that the remaining ~600M claims are cash?
    • Does that equate to a 9.97% increase in utilization or is that just more covered lives or more concentration of business among the PBMs that respond to the survey?
  • Average copays were:
    • $9.98 for the 1st tier
    • $23.69 for the 2nd tier
    • $36.93 for the 3rd tier

IMHO (in my humble opinion), I think these copays show a problem…not with the data but with plan design. Generics should definitely be below $10 so we’re alright here although I would probably shoot for $8. A difference of $14 between tiers 1 and 2 is too low. $15 should be the minimum and $20 is probably better. The same goes for the $13 difference between tiers 2 and 3.

My ideal plan design would be $8, $25, and $45 (or the equivalent average using percentage copays).

2 Responses to “PBM Data – Apples and Oranges”

  1. Thanks for your citation to the AIS’s Quarterly Pharmacy Benefit survey, chart originally published in 8/13/10 issue of Drug Benefit News. As the director of this survey, I want to add my insight on some of your points.

    I would like to clarify that Express Scripts, Medco, and CVS/Caremark do participate in the survey, although obviously not with regard to the “% of claims per tier” set of questions addressed in this particular chart. I believe benefit design issues such as tier definitions and copay levels are typically decided at the plan/client level, so a PBM with a very diverse client base may not have the information to track this across their book of business and may not feel that such a compilation would be valid. I invite all participants to answer only the questions that apply to their particular business model, and as you pointed out, there are many.

    The fine print that DBN published at the bottom of our original chart addressed some of the points you raised about whether the companies defined tiers in the same way (no), or whether they have similar business models (no) or whether they served similar populations (no). The intention of the survey is to compare pharmacy benefit metrics to the extent possible, while still keeping the survey open to all companies that offer competing or alternative approaches to those offered by the “big 3” PBMs. PBMs are incredibly disparate—apples and oranges as you observed, plus many stranger fruits! I like to think that by acknowledging and embracing this fact my survey provides an important tool for transcending the Medco/Caremark/Express Scripts narrative and furthering evolution in the field.

    Of course this often makes it very challenging to draw conclusions from the survey results, and I applaud your attempt to apply the data to your specific questions. For the most part, I think the conclusions you were able to draw from the data made a lot of sense. While my role is to simply research and share the raw data with all its caveats, you’ve provided a really nice example of how I hope end-users will take bits and pieces of the data and combine with other knowledge to further their own understanding of the pharmacy benefit.

    Susan Namovicz-Peat
    Director, Directories & Databases
    Atlantic Information Services, Inc.
    snpeat@aispub.com

    • Susan –

      Thanks for the detailed comment. Just to be clear…I love the research that you do and find it very helpful. I do think any data used without a deep understanding can be mis-used. PBM data is often very difficult to compare (often intentionally). I often find it funny that some people at the Big 3 don’t even know about the smaller PBMs. I brought up Catalyst to a sales person at one company the other day and they didn’t even know who they were. I was surprised, but I know life can be fairly myopic if you let it be.

      I would argue that the larger PBMs do have the data across tiers so I’m not sure why they wouldn’t share it. It’s always a proxy and does vary dramatically based on client plan design.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: