Wednesday, August 30, 2017

Which PBM Best Managed Drug Spending in 2016: CVS Health, Express Scripts, MedImpact, or Prime? (rerun)

This week, I’m rerunning some popular posts while I work on the forthcoming 2017-18 Economic Report on Pharmaceutical Wholesalers and Specialty Distributors. Click here to see the original post and comments from March 2017. 

ICYMI, I published a follow up analysis that includes OptumRx in Which PBM Best Managed Drug Spending in 2016: How Did OptumRx Compare?

Four of the largest pharmacy benefit managers (PBMs)—CVS Health, Express Scripts, MedImpact, and Prime Therapeutics—have raised their swords and released their 2016 drug trend reports. (Links below.) That means it’s time for some Drug Channels-style fun! Below, I review what the reports reveal about drug spending—and which questions remain unanswered.

Drug trend measures the year-over-year changes in a plan sponsor’s pharmacy benefit drug spending. As you will see, all four PBMs reported that their drug trend percentages for 2016 were in low single digits. The unweighted average among them was a mere 3.5%. About one-third of PBM clients experienced year-over-year declines in drug spending.

Put more simply: The U.S. doesn’t have a drug spending problem. It's fake news.

Specialty drugs have doubled as a share of plan sponsors’ benefit costs. Since these products account for only about 1% of claims, pharmacy benefit management is increasingly focused on handling these patients and their complex, hard-to-treat diseases. Unfortunately, payers seem set on discriminating against people who receive these therapies.

Keep in mind that the PBMs’ reports are primarily marketing documents, not peer-reviewed research studies. But as many PBM executives have said: What we do in life echoes in eternity!


We enter the arena with the four PBMs’ 2016 drug trend reports. As always, I encourage you to have the strength and honor to review the source materials for yourself.
  • Prime Therapeutics (Press release only. However, Prime gave me a sneak peek at the full report, which it will release in April.)
The larger PBMs that are owned by health insurers—Aetna, Humana, and OptumRx (UnitedHealth)—don’t publish drug trend reports. (UPDATE: See comment below.) As far as I know, none of the smaller PBMs releases any information on their management of drug trend.

Note that our comparison focuses only on commercial payers: employers and health plans. Those are the only data reported by CVS and Prime. Express Scripts provided separate trend figures for Medicare, Medicaid, and Exchanges. MedImpact provided separate trend data for Medicare and Medicaid.

FYI, Express Scripts provides the most complete reports in the colosseum. I delve into their data in Express Scripts Reveals What Really Drives Drug Spending—And Why the Government May Do No Better.


The chart below summarizes the reported 2016 commercial drug trend figures. As you can see, all four PBMs reported that plan sponsors experienced low single digit growth in pharmacy benefit spending. I know that complaining about “drug prices” is politically fashionable. Yet it appears that drug spending may be growing at a rate comparable to (or slower than) that of other parts of the healthcare system.

[Click to Enlarge]

Three of the four PBMs provided the share of its clients that experienced negative trend, i.e., those payers spent less on pharmacy benefit drugs in 2016 than they spent in 2015.
  • CVS Health: 38%
  • Express Scripts: 33%
  • Prime Therapeutics: 30%
Note that CVS did not report separate trend figures for traditional and specialty drugs, and would not provide the data when I asked. However, the company told me that it would publish a specialty trend report sometime in Q2. Such an exciting mystery!


Here are my observations on the 2016 results.

1) The components of drug spending changes vary among PBMs.

The total change in drug spending equals the sum of changes in unit cost and changes in utilization. Understanding the mix of price and quantity is crucial to understanding what drives drug spending. For a more complete discussion, see my post on the Express Scripts report.

As the chart below illustrates, the underlying components of trend differed among the four PBMs.

[Click to Enlarge]

Prime had the lowest increase in unit cost, but it tied with CVS for greatest increase in utilization. (As I discuss below, Prime also reported the lowest share of cost in specialty drugs.) MedImpact was the only PBM to report a drop in overall utilization. This drop was offset by the biggest increase in unit cost.

2) Specialty drugs’ share of costs has doubled over the past five years..

Each PBM develops its own definition of “specialty drugs.” They do this for internal management purposes and to describe benefit coverage to beneficiaries. These definitional differences partly explain the varying estimates of pharmacy benefit costs from specialty drugs. As the chart below illustrates, specialty drugs accounted for 30.4% to 37.8% of that share. For the three PBMs for which we have historical data, the share of spending from specialty drugs is about twice the 2011 figure.

[Click to Enlarge]

Note that for 2016, the chart shows specialty drugs’ share of pharmacy benefit costs after rebates. This figure is not comparable to that for prescription dispensing revenues of specialty drugs from retail, mail, long-term care, and specialty pharmacies. That’s because traditional drug have larger rebates than specialty drugs, but these rebates do not impact prescription dispensing revenues. Therefore, specialty drugs’ share of costs is higher than their share of prescription revenues. We estimate that for 2016, specialty drugs accounted for 28% of the pharmacy industry’s prescription dispensing revenues. (See Section 1.2.2. of our 2017 Economic Report on U.S. Pharmacies and Pharmacy Benefit Managers.)

MedImpact reported that specialty claims accounted for a mere 0.88% of commercial claim volume. I presume that the figures were comparable at the other PBMs, which didn’t report claims. The disparity between claim volume and costs represent a dangerous trend for patients. As I note in Employer Pharmacy Benefits in 2016: More Specialty Drug Cost-Shifting Means More Problems for Patients, third-party payers increasingly discriminate against the few patients taking specialty drugs.

3) Overall drug trends are converging.

In previous years, the PBMs’ trend figures were often vastly different. But for 2016, the traditional, specialty, and overall trend figures are surprisingly comparable. Express Scripts, Medimpact, and Prime all reported that spending on traditional drugs declined while spending on specialty drugs grew by 13% to 14%.

I attribute this convergence to measurement approaches that are more similar than those in my previous analyses. Here’s my table summarizing the methodology behind the 2016 numbers.

[Click to Enlarge]

Consider rebates. For 2016, the PBM figures show the change in drug spending net of rebates. Prior to 2015, rebates were excluded from some PBMs’ drug trend figures but included in others. The 2016 results therefore may not comparable to previous years’ reporting that excluded rebates. This more consistent treatment of rebates partly explains the similar drug trend figures. For more on rebates, see Section 5.3. of our 2017 pharmacy and PBM report.

4) Each PBM uses a different sampling approach.

Each PBM generated its 2015 and 2016 figures differently.
  • Express Scripts pooled member-level data across plan sponsors for the available months of eligibility. It excluded sponsors if “they were not Express Scripts clients in both 2015 and 2016, if they had less than 12 months of claims data in either year, if they had retail-only benefits, if they had 100% or 0% copayment benefits, if they had eligibility shifts exceeding 20% for commercial plans (eligibility shifts exceeding 50% for Medicare and Medicaid plans), or if they were contractually prohibited from inclusion.” It’s not clear whether these exclusions were material. An Express Scripts spokesperson told me that its sample included more than 30 million members.
  • CVS Health calculated trend using data from 1,100 clients with more than 23.5 million lives. It excluded an unspecified number of clients with “abnormally high membership shifts (up or down)” and those “with abnormally low or high trends.”
  • MedImpact included clients with 24 months of continuous claim and eligibility data. It did not report any exclusions.
  • Prime Therapeutics’ press release states that the company “represents nearly 20 million members nationally.” A Prime spokesperson told me that its 2016 drug trend figures were based on a subset of 14.6 million members from its commercial full service Blue Cross Blue Shield plans.

I know you come to Drug Channels for the best in PBM gladiatorial action. So, here’s something to crowe about: a private home video of me reading emails from readers. Click here if you can’t see the video.

Keep those comments coming!

No comments:

Post a Comment