News And Updates


CVS’s 2019 Formulary Removals – Negligent Handling Diabetes Test Strip Change

Summary

Around the first week in August for the past four years, CVS has made an official announcement of its next year’s formulary changes. It posts these changes on its own websites.

But that changed in 2018. During an August 8th 3Q2018 Earning Conference Call,  CVS said that 2019 formulary details would be available around October 1, 2018.  

But, CVS waited until November 16, 2018 to post details on its own website. We believe that there has been a willful intent by CVS to minimize attention to upcoming formulary changes fearing bad publicity would delay government approval of their merger with the insurance company Aetna. The review included the Department of Justice, state Attorney Generals and Judge Richard Leon of the U.S. District Court of the District of Columbia.    

In particular, CVS has been negligent in giving ample warning to diabetes patients of an upcoming switch in blood glucose test strips from LifeScan’s OneTouch brand to Roche’s Accu-Chek brand.   We estimate that this change is going to affect up to 7.2% of CVS’s 94 Million covered lives,  or 6.8 Million people who have Type I or Type II diabetes, as they require daily blood glucose monitoring via a glucometer and test strips.

We conclude with a look at the impact of this event on state “frozen formulary” laws.

 

What is a formulary?

One of the most effective cost-controls employed by pharmacy benefit managers (PBMs) are their national formularies. Formularies are lookup tables embedded in software at the retail and mail order point of sale that alerts pharmacists as to which drugs are covered by a customer’s drug benefit plan.   

Via formularies,  PBMs’ have the power to affect the demand for patented, but therapeutically equivalent drugs.  This power enables them to negotiate rebates with pharmaceutical manufacturers (Pharma) in return for formulary placement.  Placement entails more that just inclusion vs exclusion, but also entails a whole array of other conditions including prior authorization, step therapy, and quantity limits.

The more a PBM shelters a given drug from competition via its array of formulary controls, the higher Pharma is willing to pay formulary rebates to avoid those controls.

 

PBM formulary switches and exclusions are growing

It has only been since 2012 that PBMs began to make significant formulary exclusions of FDA-approved prescriptions drugs. Before that, they would include non-preferred drugs in Tier III of a formulary which carried the highest copayments.

In our 2005 paper The Effect of Corporate Structure on Formulary Design, we looked at key therapeutic classes with small molecule brands having close therapeutic equivalents — proton pump inhibitors (PPI), COX-2 inhibitors, and 2nd generation antihistamines.  We found that  PBMs practiced what we called the “the Soprano approach” to negotiation — threaten harm to all, but negotiate payoffs from all to abstain from harm.

We called this negotiating strategy a “sin of omission” as there was no evidence of harm like formulary exclusion.  It was a sin of omission in that PBMs should have excluded high cost brands in cases where there were generics that were close therapeutic equivalents.  For example, in the mid-2000s, PBMs should have excluded the high cost brand PPI Nexium in favor of other near equivalent PPIs that had gone generic like omeprazole (Prilosec).

Below is a spreadsheet from a 2005 paper The Effect of Corporate Structure on Formulary Design  indicating a tendency in the mid-2000s for PBMs not to exclude brands that had close therapeutic equivalents.

Formulary design has changed significantly since the mid-2000s.  Now, there is a decided tendency among the Big 3 PBMs to grant exclusivity to drugs in formulary therapeutic classes where there are a number of therapeutic equivalents.  

In our 2017 paper Three Phases of the PBM Business Model, we argued that the trend of formulary exclusions stemmed from fundamental shift in the PBM business model after 2010.  PBMs shifted from a reliance on gross profits from mail order generics to a renewed reliance on retained rebates, this time from high list priced specialty drugs.

In addition to outright increases in net exclusions (see graph below), PBMs have become more aggressive in making one-for-one switches between two therapeutic equivalents resulting in no net change in the total number of exclusions.

 

CVS’s upcoming formulary switch of branded diabetes test strips

This paper focuses on CVS’s handling of one upcoming formulary switch effective January 1, 2019. It involves the replacement of LifeScan’s OneTouch branded diabetes glucometer and test strips with Roche’s Accu-Chek branded glucometer and test strips.  It is by no means the only upcoming switch as “leaked” (more on this later) by a PBM contract consultant and on websites of various plan sponsors using CVS as their PBM.

By far,  it is this test strip switch that will affect the most people.   We estimate that this change will affect up to 7.2% of CVS’s 94 Million covered lives,  or 6.8 Million people who have Type I or Type II diabetes, as they require daily blood sugar monitoring.

The switch in coverage for a medical diagnostic device requires patients to invest a lot more time and energy than a switch between therapeutic equivalent pills.  First, users will have to order the new glucometer and have it arrive before being able to use the new test strips. There is not telling how long that will take.

Then users will have to spend time learning how to calibrate and read their new glucometer.  Finally, if users have installed OneTouch software on their PCs with histories of readings, they will have to transfer that data to Accu-Chek software which will be no easy feat.

 

Why did CVS decide to switch coverage for test strips?

Our search of past formularies confirmed that CVS gave LifeScan’s OneTouch exclusive formulary placement between 2015-2018.  OneTouch likely enjoyed exclusive coverage going back even longer. Our cursory view of the formularies of the other two big PBMs — Express Scripts and OptumRx — indicated that for the past few years the other two PBMs also included OneTouch on their formularies, although not always exclusively. Finally only CVS decided to make a switch in 2019.

Obviously, gross rebates and related net costs to CVS were important factors in making this impactful switch.  Although we are a leading critic of the PBM practice of making formulary choices on the basis of gross rebates rather than net costs after rebates, we give CVS the benefit of the doubt here.  

Give the inconvenience that this switch will cause to users,  we just cannot fathom CVS making this switch based on small differences in net costs. CVS made the switch because Roche’s likely decided that 2019 was the year to go “all in” with rebates.  Note that it looks like Roche’s substantial rebate commitment to win CVS over in 2019 likely left Roche in no position to offer substantial rebates to the other two Big PBMs as they both continue their exclusive coverage for OneTouch in 2019.    

But, quality may have been another significant factor involved in CVS’s decision to switch.  While LifeScan was the pioneer in electrochemical glucometers and test strips, it has lost its lead to the likes of Accu-Chek, Contour, a division of Panasonic Health Products, and FreeStyle Lite, a division of Abbott.   

Evidence to support this comes from results of a Blood Glucose Monitor System Surveillance Program conducted by the respected Diabetes Technology Society. (see below). They found that only 6 out of 18 meters tested passed their 95% accuracy threshold.   Contour and Accu-Chek were at the top. Both LifeScan OneTouch brands failed the accuracy threshold test.

Another factor in CVS’s decision to switch might have been a change in ownership of LifeScan in 2018.  According to Wikipedia entries, after first being put up for sale in 2017,  LifeScan, a division of Johnson & Johnson, was bought finally by the private equity firm Platinum Equity in June 2018.  

The reason why this sale might have been a factor is that large Pharma companies have a developed a strategy of bundling rebates to lock in exclusive formulary placements for an array of their drugs.  The new private equity firm who bought LifeScan in 2018 was in no position to compete with Roche on basis of  bundled rebates.

We wish to reiterate that our problem with CVS is not its choice to replace OneTouch with Accu-Chek.  The switch looks good both from a net cost and well as from a quality standpoint. The problem we have is CVS’s handling of the switch.  

 

CVS delayed for three months an official announcement of its 2019 formulary removals

Around the first week in August for the past four years, CVS has made an official announcement of its next year’s formulary changes. It posts details of these changes on its own websites.  Plan sponsors using CVS as their PBM follow suit soon thereafter. This practice is similar to that of the other two big PBMs — Express Scripts and OptumRx

But that changed in 2018.

During a 3Q2018 Earning Conference Call on August 8, 2018,  CVS said that formulary change details would be available around October 1, 2018.

However, on October 1, 2018, there was no official announcement.  Instead, CVS posted a .pdf refresh of its 2018 formulary exclusions accessed via the following URL.   

https://www.caremark.com/portal/asset/Formulary_Exclusion_Drug_List.pdf

Then, on or after November 16, 2018, CVS deceptively posted the long delayed 2019 formulary change details using THE SAME URL previously used to access its October refresh of the 2018 formulary.  In other words, it deceptively replaced a 2018 formulary .pdf with a 2019 formulary .pdf  leaving no trail indicating there was a sequence of postings.

The document number at the bottom of 2019 .pdf was 106-25923A and the date used was 010119 v2. It also had a document date of November 16, 2018

 

Fortunately we downloaded the 2018 refresh which now can be accessed  from our website https://nu-retail.com/CVS_100118_refresh.pdf

The document number at the bottom of the 2018 refresh pdf was 106-25923A (same as 2019), but the date used was 100118.  

Here is a the timeline indicating that an official October 1st announcement would have come toward the end of CVS’s efforts to gain government approval for earlier announced merger with the insurance company Aetna.  

October 26, 2017     WSJ first reports merger talks between CVS and Aetna

December 3, 2017   Negotiations complete, submit to DOJ for approval

October 1, 2018   CVS scheduled to announce 2019 removals. Instead posts refresh of 2018 removals using the following URL               https://www.caremark.com/portal/asset/Formulary_Exclusion_Drug_List.pdf

October 10, 2018     DOJ approves merger

October 18, 2018     Multi-state coalition of Attorney Generals approves merger

November 16, 2018   CVS posts actual 2019 Formulary Exclusion List using same URL as the 2018 formulary refresh 

November 28, 2018  CVS announces close of merger agreement with Aetna

November 29, 2018  Federal Judge raises prospect of not approving merger

Any bad publicity from an October announcement of the 2019 formulary changes might have derailed the conclusion of the government approval process.  As it turned out, the merger was approved. The absence of an October 1st announcement might have helped.  

In our opinion, CVS’s failure to announce this major change is a case of negligence.   There has been a willful intent on CVS’s part to downplay this switch in order to minimize potential bad publicity during a period where CVS’s merger with the insurance company Aetna was being reviewed.

 

“Leaks” of CVS’s 2019 Formulary Switches

While CVS delayed posting the official announcement until November 16, 2018, there has been a series of  “leaks” in the form of PBM consultant blog posts and postings of .pdfs on websites of plan sponsors using CVS as their PBM.  Basically, a PBM consultant and plan sponsors “jumped the gun” in these announcements thinking that CVS would made it official by October 1, 2018.

We were first alerted to the test strip switch on September 27, 2018.  Knowing that CVS had promised to announce details by October 1st, our online search that day turned up the following  blog post  by the CEO of the largest pure PBM contract consultant Pharmaceutical Strategy Group (PSG).

“CVS is changing their only covered test strips, One-Touch to Accu-Check where members will need to obtain a new meter and learn any new or different features of the new preferred products.”

Further searches revealed other “leaks” of 2019 details in the form of .pdf files. Below is a listing of some of our findings.  Each file was stamped with CVS’s logos and marked with CVS-generated document numbers and date.

 

What should have CVS done?

With a few exceptions (see below), the following  footnote at the end of the each .pdf on plan sponsor sites is the only effort to date to inform and guide consumers through this switch.

“An ACCU-CHEK  blood glucose meter may be provided at no charge by the manufacturer for those members currently using a meter other than ACCU-CHEK. For more information on how to obtain a blood glucose meter, call 1-877-418-4746

ACCU-CHEK brand test strips are only preferred options.”

In our search of online websites,  we found only  Carefirst Blue Cross Blue Shield, one of CVS’s largest clients covering 3.2 million in the Mid-Atlantic region and the State Health Benefits Plan (SHBP) of Georgia making a real effort to inform and guide members.

On November 14, 2018 , Caremark published a letter that it would be sending via regular mail to every impacted member. Carefirst even labeled the test strip change as constituting “the majority of the negative disruption”

Below is a screenshot of the full letter Carefirst said it would be sending to each impacted member.

The following is a list of other measures  we have come up with that CVS should have undertaken to smooth the transition.

  • Like Carefirst, urge all client plan sponsors to send a letter via regular mail to all impacted members informing them of switch and instructions for ordering a new meter.
  • Upon request, offer December delivery of new Accu-Chek glucometer and one package of test strips. (done by the SHBP of Georgia)
  • Get retail pharmacies involved early by offering to build up inventory with new Accu-Chek meters and test-strips.
  • Post online videos showing how to use new meters. (done by the SHBP of Georgia)
  • Make sure history of readings stored on members’ PC can be transferred easily between OneTouch software programs to Accu-Chek software programs.

 

Will CVS handling of this switch hasten the adoption of “frozen formulary” laws?

Currently, there is a smattering of states that have passed legislation limiting how often PBMs can make formulary changes during any plan year and require prior notice before the changes takes effect. Due to the trend toward formulary exclusions, more and more  states are considering adopting their own “frozen formulary” laws.  

These so-called  “frozen formulary” laws generally focus on changes within any given plan year.  We could find no state law dealing with requirements for sufficient notification of  formulary changes for the new plan year. 

For example, the general rule has been for the Big 3 PBMs to announce new plan year formulary changes around the first week in August, a full 5 months before the new plan year.  CVS’s official announcement delay to November 16th occurred only 35 days before the beginning of the new plan year.  

As late as this was, it looks like CVS’s November official notification did not violate any existing state laws.  

Still, if CVS’s switch in diabetes test strip coverage causes as much disruption and negative publicity as we anticipate,  we could envision that this event becoming the spark for a whole slew of “frozen formulary” laws requiring a minimum of 3 months or 90 days prior notification of new plan year formulary changes, including individual letters mailed out to impacted consumers.  


The Problem with CVS’s “Guaranteed Net Cost” PBM Business Model

Summary:

Consider this meta:  CVS opaquely is substituting one opaque source of gross profits — guaranteed net cost markup — for another opaque source — retained rebates.

The pharmacy benefit manager (PBM) CVS Caremark has offered its self-insured corporate clients an alternative business model called “Guaranteed Net Cost”.  The pricing scheme features 100% pass-through of drug rebates and the end of rebate retention as an opaque source of PBM gross profits.

But, CVS has glossed over the fact that their “guaranteed net cost” price to plans is not the same as the net costs to them.  Until CVS tells us otherwise, the new business model allows for a opaque markup on top of PBM net cost. In graphs below, we demonstrate how a markup of guaranteed net costs serves as an opaque offset to foregone rebate retention.  

It is naive to think that CVS Caremark is about to give back a significant source of its annual gross profits without some sort of offset. In fact, CVS admits as much as their spokesperson is quoted as saying

CVS’ manager of corporate communications, Christina Beckerman, told Fierce Healthcare that the company does not expect CVS Health’s profitability to increase or decrease as a result of the shift to 100 percent pass-through rebates.

It is not even clear that CVS’s new business model lessens the incentives to Pharma to inflate list prices in order to compete on rebates for formulary placement.

The Problem With the Current PBM Business Model

The current PBM reseller business model features five major streams of revenue and gross profits.  Four of the five are opaque.

  1. Opaque rebate retention % on speciality (biotech) drugs in return for preferred or exclusive placement on formularies;
  2. Opaque rebate retention % on small molecule brand drugs in return for preferred or exclusive status on formularies;
  3. Opaque profit margins on 90-day generic Rx filled by captive mail order operations of the PBM;
  4. Opaque “spread margins” added by the PBM on top of reimbursements to retail pharmacies included in their networks;
  5. Transparent claims processing and data fees.

The opacity of drug rebates is magnified by the fact that reimbursements for brand drug Rx and related rebates come at different times.   It is impossible for plans match up these two streams and calculate a single net price its pays per drug.

Since the early 2000s, PBMs have continually come under attack for not acting in the best interest of their clients.  We have written a number of papers since 2004 pinpointing an opaque reseller business model as the source of this misalignment.

The PBM reseller business model is in stark contrast to two other transparent business models used by managed care companies:  

  1. a PMPY fee-for-service agency model where 100% of all reimbursements and rebates are passed through to plans.
  2. a risk-based insurance model with capitated premiums paid by plans.

Until the PBM Medco’s merger with Express Scripts in 2012, Medco’s financial 10-Q and 10-K reports to the SEC broke out gross rebates received — a credit to cost of sale — and rebates retained — a credit to sales.  We were able to calculate with certainty Medco’s “rebate retention rate”, a name we coined fifteen year ago in 2003.

We calculated that Medco’s rebate retention rate — the percentage of gross rebates retained — fell from 55% in 1Q03 to 28% in 2Q05.  This rapid decline was due to the sudden awareness by clients of the whole rebate retention scheme. To offset this loss, Medco began to push clients toward its captive mail order and fat margins it began to earn on mail order generic Rx fills.

The share of Medco’s overall gross profits coming from retained rebates reflected  outrageous rebate retention rates.  For 3Q04, we derived with certainty from Medco’s 10-Q that 71% of its gross profits came from retained rebates from small molecule brand drugs.  By 2Q05, we estimated with certainty that Medco’s retained rebate share of gross profits had dropped to 48% with the difference going to their newly found focus on mail order generics.

In our 2017 paper “Three Phases of the PBM Business Model”, we carried forward our mid-2000s work on disaggregating PBM gross profits by sources.  Below is a summary of that work.

Here is a graph of the above data:

CVS’s Guaranteed Net Cost Business Model

On December 5, 2018,  CVS Caremark introduced a new pharmacy benefit manager (PBM) business model option for self-insured corporate drug benefit plans.   

The core of this new business model is a simplified reimbursement price paid by plans to CVS that the company craftily describes as “Guaranteed Net Cost”.  Craftily, in that this so-called “cost” is really a “price” where the difference between “cost” and “price” is a markup.

The company touts the following distinguishing features of this simplified reimbursement price.

  • Drug cost predictability and simplicity
  • 100% of rebates are passed through to plan sponsors
  • Simpler payments flow — no retrospective rebates or inflation adjustments
  • Simpler way to compare different PBM contract proposals

Note this new pricing model is for brand drugs only dispensed at retail, mail order and specialty pharmacies.  The generic Rx drug reimbursement pricing scheme remains the same. That is to say, it preserves an opaque “spread margins” that PBMs like CVS add on top of CVS reimbursements to retail pharmacies for generic Rx drug fills.

This new CVS’s initiative clearly is in response to the tsunami of criticism by plan sponsors over an opaque PBM business model and the difficulty in matching initial Rx reimbursements at an inflated list prices with retrospective rebates occurring months later.

The Problem with CVS’s Guaranteed Net Cost Business Model

One: Opaque Markups

The problem with CVS’s new business model is that guaranteed net cost to plans is not necessarily the same as the net cost to PBMs. CVS never states unequivocally that its guaranteed net cost to plans = net cost to CVS.  In other words, CVS’s new business model allows for an opaque markup on top of its net cost.

Consider this meta:  CVS opaquely is substituting one opaque source of gross profits — guaranteed net cost markup — for another opaque source — retained rebates.

It is naive to think that CVS is about to give back some of its oligopolistic profits.  In fact, CVS admits as much as their spokesperson is quoted as saying

“ CVS’ manager of corporate communications, Christina Beckerman, told Fierce Healthcare that the company does not expect CVS Health’s profitability to increase or decrease as a result of the shift to 100 percent pass-through rebates”

The following is a numeral example of how the opaque markup can serve as a 1-for-1 substitute for retained rebates:

Here is a graphical depiction of our view that CVS is substituting an opaque markup for an opaque rebate retention:

 

To CVS’s credit, its new guaranteed net cost eliminates timing complexity. It does this by taking a risk and netting the current period Rx reimbursement with an estimated “expected” rebate rather than wait to credit plans with the actual rebate when it is paid by Pharma months later.  

CVS certainly is justified in including some markup as compensation for taking the risk that their estimated expected rebates turn out to be less than actual rebates.

Instead, CVS decided not mention markup at all,  let alone a justified markup as a compensation for assuming timing risk.

TWO: Doubtful Elimination of Incentive to Play the High List – High Rebate Game

Under the current retained rebate business model,  PBMs are incentivized to favor drugs with the highest gross rebates to the exclusion of therapeutically equivalent drugs with the lowest net cost.  To be in a position to win this rebate game, Pharma is driven by the PBM-created rebate game to inflate list prices for its brand drugs.  See our paper: Blame PBMs (Not Pharma) for Drive Drug Price Inflation.

The list price – net price bubble began around 2010 and reached its peak in 2017. It was in 2017 that AbbVie first broke the PBM rebate game  winning formulary placement by Express Scripts despite pricing its late entrant Hepatitis C Virus (HCV) drug Mavyret with an ultra low list price with no rebate potential.  However, this was an exception and the norm remains that the basis for formulary placement is gross rebates over net price (list price – gross rebates).

Below is a graphical depiction of how AbbVie broke the rebate game with its ultra low list = no rebate drug HCV drug Mavyret.

It is possible that the rebate game of high list – high gross rebate may be lessened under CVS’s new guaranteed net cost new business model. This is because the basis for PBM profits — markups — could be any number as opposed to being tethered to something like % of gross rebates or % of net cost.

Below is a depiction of CVS’s flexibility in choosing a markup that is independent of the list price or gross rebate. 

On the other hand, we can see the possibility that the new business model preserves the status quo. Here is our line of reasoning for this:

it is likely that brand drug list prices, which are publically available,  will serve as an upward bound for guaranteed net cost as it would look bad for CVS to set a guaranteed net cost that exceeded a drug’s list price.

To look good, CVS will want to show that guaranteed net costs is consistently 40% to 70% below the brand list prices.  

To achieve these percentages while still having room for oligopolistic markups, CVS will signal to Pharma that, while formulary placement is no longer based on gross rebates, high list – high rebate drugs afford CVS latitude in setting guaranteed net cost markups.

Below is a graphical depiction of why, under the new business model, CVS still would be incentivized to favor the high list – high rebate drug.