Wednesday, July 13, 2011

Co-Payments and Prescription Abandonment

In Wake-Up Call for Co-pay Cards, I mention that co-pay card programs can be valuable tools in helping employers ensure a healthy workforce by reducing prescription abandonment data.

A few people asked me about the data, so here’s a link to the original scientific paper: Shrank WH, et al, “The Epidemiology of Prescriptions Abandoned at the Pharmacy,” Annals of Internal Medicine, November 2010.

As far as I know, this article is the best publicly-available, peer-reviewed analysis of how co-payment levels influence whether a patient actually picks up a prescription. Lots-o'-Data in the original article, but read on for a few highlights along with a strawberry-scented chart.


The Annals study analyzes 10.3 million prescriptions filled by 5.2 million people over a 3-month period in 2008. All data came from CVS Caremark (NYSE:CVS) by matching CVS retail transactional activity to Caremark claims data. (I don’t consider this factor to be material in interpreting the results.) Full details of the methodology are described in the paper.

Prescriptions were classified into three mutually exclusive outcomes:
  1. Filled prescription, indicating that the patient purchased the prescription;
  2. Return-to-Stock (RTS), indicating that the patient abandoned the prescription; or
  3. RTS with fill, indicating that the patient abandoned the prescription and it was returned to stock, but the patient purchased a prescription for a medication in the same medication class at the same or another pharmacy.

The overall abandonment rate was 3.27%, comprised of 1.77% RTS (outcome #2) and 1.50% RTS with fill (outcome #3).

Table 3 (page 637) relates various factors to the relative risk of prescription abandonment. Here’s my visual summary of the bivariate relationship between the patient’s out-of-pocket costs and prescription abandonment rates (outcome #2 above).

As you can see, co-pays above $20 seem to be the tipping point for people who decide not to fill their scripts. Amazingly, even a co-pay of $0.00 (824K prescriptions) had an abandonment rate of 1.5%!

Table 2 (page 636) has useful data on abandonment by medication class.
  • Opiates and antiplatelet medications were least likely to be RTS prescriptions (1.0% and 0.9%, respectively) or RTS with fill (0.8% and 1.1%). Antihypertensives, oral diabetic medications, and statins also had comparatively low abandonment rates.
  • Among daily-use therapies, higher rates of RTS were seen for proton-pump inhibitors (2.6%), asthma medications (3.5%), and insulin (2.2%).
  • Medications used on an as-needed basis, such as dermatologic agents (RTS rate, 2.9%) and cough and cold medications (RTS rate, 3.6%) were also abandoned more commonly.
One other curious data point: Electronic prescription were almost 65% more likely to be abandoned than traditional scripts. More non-Sunnyside news for e-prescribing given the just-published study showing e-prescribing & handwritten error rates are similar.


  1. Al GodleyJuly 13, 2011

    Adam, as always I love your posts.  I've noticed one thing from my personal prescription habits:  Drug stores often auto refill my 'as needed" meds even when I don't need them.  I've had a significant issue getting them to get this corrected.  I know that I have RTS because of this.  As for opiates...well, if you get pain meds I suspect you fill those since they are hard to get and you might need them later.  I wonder how these practices impact the numbers?

    Interesting info on e-scripts!  They are my personal favorite and actually increase my adherence.  My doc sends them to the pharmacy and they are ready when I arrive.  No waiting!

    Hope you summer is not as hot as what we have in the south this week!

  2. Al,

    Interesting observation. I presume your "as needed" meds are maintenance drugs, which were no more likely to be abandoned than non-maintenance scripts in the data. However, new scripts were almost 3 times as likely to be abandoned than refills.

    Enjoy the summer!


  3. Jeff WagnerJuly 13, 2011

    Thanks for a very interesting article Adam.  I believe one of the implications that the editors cite (E-Rx's may result in increased abandonment) is only partially correct.  While it is true that the measured abandonment rate will rise, it does not consider the fact that there is tremendous leakage today due to handwritten prescriptions not even making it to the pharmacy to begin with. ("Leaky Bucket")  These aren't captured in any metric that I am aware of.

  4. Jeff Wagner makes a good point. The fact is, some prescriptions are not needed at all and others are written for complaints that are minor, self-limited or adequately treated with non-prescription therapies. Physicians write prescriptions for lots of reasons and sometimes it's just to terminate the encounter and get the patient out of the office. Patients know this, so some primary non-adherence (i.e., "leakage") is rational and even appropriate.

    As for the article in JAMIA that you mention above, Adam, it's regrettable that it has been interpreted as an indictment of electronic prescribing.  It's not.  The article reports a very dated (2008 data) analysis of a small (3,898) and unrepresentative sample of COMPUTER-GENERATED prescriptions orders that were received at one chain. The authors themselves state "we did not exclude any prescriptions based on the mode of transmission to the pharmacy (e.g., hand delivery, facsimile, or other electronic transmission)."  That, my friends, is NOT electronic prescribing.  Regrettably, few people in the lay media apparently read any further than the unfortunate title.  Shame on them.

  5. Hmmm, your comment makes me wonder about the "$290 billion in costs from prescription non-adherence" that I keep hearing about from the pharmacy associations.

    BTW, I interpreted the JAMA article to suggest that the biggest problem with e-prescribing was neither hardware or software, but human error (meatware). 61% of the errors were due to "omitted info."

  6. That's precisely my point, Adam.  True (i.e., computer-to-computer) electronic prescribing uses NCPDP's telecommunication standard that makes it very difficult to issue a prescription with essential information omitted.  The Rx will simply not be successfully transmitted without all required fields. Just because a prescription is generated on a computer does not mean that it's an electronic prescription.  I think it's important that we get our terms straight here.

  7. At a recent speaking engagement, the study's lead author did clarify that this is, in fact, their interpretation -- ePrescriptions will always make it to the pharmacy, while some portion of hand-written prescriptions are abandoned before the pharmacy or PBM has any visibility to them.

  8. SunpharmacyJuly 16, 2011

    1.  We've had e-prescriptions abandoned because the prescribers sent the same order to several pharmacies pennecontemporaneously.  (Incidentally, this adds to uncompensated time wasted in the pharmacy.)

    2.  "Meatware" (apt description) errors are not all omissions--we've caught numerous "list drop" errors where the prescriber selected the wrong item on the drug drop down list and default errors where days supply is always "30" regardless of sig or quantity to dispense.
    If anything, at this point at least, we're making more calls on errors on e-prescriptions than on traditional handwritten prescriptions.

    Tom Connelly, RPh
    Sun Pharmacy
    Rising Sun, MD

  9. Drug Channels:  I read it for the pictures.