I have always been critical about the use of prescription claim data in measuring adherence, though I understand that it is one of the only mechanisms readily available to estimate adherence. Discussions that emerged from “The Rewards of Performance” led to a closer examination of Mirixa SSI claims this quarter.
For those that don’t know, the Mirixa SSI Retail Stars Adherence Intervention Program alerts pharmacies of patients with poor compliance (percentage of days covered or PDC) taking one or more medications being emphasized thru the EQuIPP measures. These alerts are based entirely on claims data. The pharmacy is paid $12 for each case completed and $2 if the pharmacy attempts to complete the case and the patient declines to take part.
I asked my clinical director to summarize the SSI activity our pharmacy has seen during the first quarter of 2015. Here is what she reported:
-
We received 12 SSI Retail Stars Adherence Interventions Program notices
-
Each one of these required an average 20 minutes of pharmacist time
-
Many of these 12 , once the discussion started, required the pharmacist to do a medication reconciliation to address the issue.
-
All 12 (100%) of these SSI cases WERE FALSE POSITIVE adherence issues
It is quite clear, for at least the first 12 cases this year at our pharmacy, claims data does not tell a compelling story about compliance. Consider some examples:
Falsely Identified:
Patient fell and broke his hip in late Feb. He underwent surgery and was hospitalized for 3 weeks. Patient is also a part of pharmacy’s synchronization program. We will be re-syncing medication at upcoming fill.
Falsely Identified:
On 9/16, our pharmacist discussed the change from simvastatin to atorvastatin with the patient. He asked if he could finish simvastatin left at home (as suggested by his doctor). We agreed this was fine (as the change was not prompted by an ADR to simvastatin). The 9/12/14 fill of atorvastatin will obviously last >90 days since he finished simvastatin first. We continue to monitor his compliance regularly with all medications.
Falsely Identified:
Last 3 refills were appropriate: 12/6/14, 1/5/15, and 2/2/15. Reviewed patient profile as Mirixa Platform identified non-compliance for Simvastatin, Metformin, and Lisin/HCTZ. Compliance issue was falsely identified and last three fills/refill history looks appropriate. Intervention was documented into Mirixa Platform. Patient reported no issues with medications at pick-up
Falsely Identified:
Per Pharmacy records, compliance is appropriate and claims were billed through new processor. Refill history: 1/2/15, 2/9/15 Patient was recently hospitalized a couple of weeks ago.
Falsely Identified:
Refill dates include: 1/5/15,1/19/15,2/2/15,2/16/15,3/2/15. This non-adherence falsely-identified.
Falsely Identified:
Reviewed patient profile as Mirixa Platform identified non-compliance for Glipizide. Compliance issue was falsely identified and last three fills/refill history looks appropriate (11/10/14, 1/10/14, 3/2/15). Patient fills 60 days at a time. Intervention was documented into Mirixa Platform. Patient reported no issues with medications at pick-up
Falsely Identified:
Patient’s record shows adequate and timely compliance. All medications are lined up to fill around the 25th each month. Pharmacy refill history shows refill dates this year of: 12/26/14, 1/21/15, and 2/25/15.
Falsely Identified:
Pharmacy dispensing software indicates that medication is being filled appropriately with last 3 refills dates of: 12/4/15, 1/7/15, and 2/6/15. Additionally, patient is enrolled in pharmacy’s medication synchronization program and medications are reviewed by pharmacist monthly for ADRs, changes, and promotes adherence.
The use of claims data to alert a pharmacist to a potential problem is a benefit to both the PBM and the pharmacy. Unfortunately, the reliance of pharmacy quality measurements (i.e. EQuIPP) on PDC taken from claims data is seriously flawed. Some might consider our first quarter to be an aberration statistically. This is not the case. Our in-house compliance screen (see “Addressing Compliance“) routinely finds that a significant number of apparent compliance issues have legitimate explanations. The current mechanism for evaluating pharmacy is certainly flawed, and new measures that actually measure the impact that pharmacists can have on patient outcomes are desperately needed.
Side Note:
In “The Rewards of Performance” it was pointed out that the example plan withheld money paid for SSI cases by Mirixia from performance payments for overall EQuIPP based measures. This is essentially double jeopardy for the pharmacy. With a payment of a mere $12 per case (if completed), a pharmacy spends more on pharmacist salary than they are being paid for the case. To aggravate the matter, the pharmacy’s PDQ may already being hurt by the compliance in question, possibly reducing their performance rating and reward. Now it is becoming apparent that the claims based compliance may not have been an actual problem to start. Withholding the amount in the performance payment just adds insult to injury.