Where Should Copays be Set for Specialty Drugs?

A continually popular question that I receive is how high can specialty copays be set before patients stop taking their medication? 

Unfortunately, empirical data to guide the answer to this question is fairly limited.


Four studies of elasticity of specialty medications have been published, most examining data from the early to mid-2000s.  The first published study by Goldman et al. examined price elasticity for specialty medications used to treat cancer, kidney disease, rheumatoid arthritis (RA), and multiple sclerosis (MS). 

                                              Summary of Studies of Specialty Pharmaceutical Price Elasticity

Author Year Population Conditions* Key Dependent Variable
Karaca-Mandic (2010) 2000-2005 35 large private employers RA Medication initiation and continuation
Gleason (2009) 2006-2008 13 million commercially insured IBD, MS, RA, Psoriasis Prescription abandonment, defined as a reversal of a previously adjudicated claim
Curkendall (2008) 2002-2004 45 large employers RA Medication adherence and persistency
Goldman (2006) 2003-2004 15 large employers Cancer, Kidney disease, MS, RA Medication initiation and continuation

For specialty meds used to treat RA, a doubling of patient cost-sharing was correlated with a 21 percent reduction in patient spending.  Elasticity for MS medications was -.07, and kidney disease and cancer did not demonstrate any statistically significant relationship between patient cost-sharing and prescription demand.  Karaca-Mandic et al. found that for newly diagnosed RA patients doubling the average annual OOP cost under the pharmacy benefit (from $400 to $800) reduced the probability of initiating a biologic for RA by 9.3 percent.  For current users, doubling the average OOP cost under the pharmacy benefit reduced the probability of continuation by 3.8 percent (from 80 to 77 percent).  Interestingly, this study found that higher family OOP burden for medical expenditures, not just specialty, was also associated with decreased likelihood of initiation of a biologic.

In one of the most widely quoted studies, Gleason et al. (2009) examined abandonment of specialty prescriptions at the pharmacy for patients newly initiating a biologic for the treatment of MS or tumor necrosis factor (TNF) blocker for treatment of RA or other conditions. Prescription abandonment was defined as “reversal of an adjudicated claim with no evidence of a subsequent adjudicated paid claim in the ensuing 90 days.”  For MS, the abandonment rate for patients with OOP less than $100 was 5.7% compared to more than 25% for OOP expense exceeding $200.  For TNF, the abandonment rate increased most significantly at copays exceeding $500.

                                                                                                                                                               Results from Gleason et al.

 Out-of-Pocket Expense

Multiple Sclerosis (N=2,791)

TNF Factor (N=7,313)

Unadjusted Abandonment Rate Odds Ratio from Logistic Regression Unadjusted Abandonment Rate Odds Ratio from Logistic Regression
$0-$100 5.7 Reference Group 4.7 Reference Group
$101-$150 5.3 0.9 10.5 2.3*
$151-$200 10.6 2.0 14.6 3.7*
$201-$250 26.8 7.3* 10.6 2.1
$251-$350 25.8 6.5* 13.4 3.3*
$351-$500 26.2 6.1* 16.3 4.4*
>$500 28.5 6.7* 26.4 7.0*


Finally, Curkendall et al. examined the relationship between cost-sharing and compliance with two TNF blockers, adalimumab and etanercept.  A weekly OOP expense of $0-$40, which represented 95% of patients, was associated with a medication possession (MPR) of 0.53, compared to a MPR of 0.35 among individuals with a weekly OOP expense exceeding $40 (5% of the patients).  A separate analysis of medication persistency suggested that the poor adherence was primarily due to therapy discontinuation rather than patients missing doses. 

While these studies have provided important early insights on price sensitivity for specialty medication, their primary focus on rheumatoid arthritis and their examination of data from the early to mid-2000’s limit their application (exception being Gleason et al).  The specialty landscape has changed dramatically in recent years—annualized spend per commercial member has grown 50%, patient cost-sharing has grown rapidly with a fourth tier becoming the mainstay, and the majority of spend is now managed under the pharmacy benefit rather than the medical benefit.    Clearly more research is needed in this area.  In the meantime, Gleason’s work perhaps provides the greatest insight to this question, keeping in mind that the copay threshold for significantly greater non-adherence was different for the two therapy classes studied.  Even at $100, they observed a doubling of the abandonment rate for TNF factor so a $100 copay would be a conservative design.

  1. #1 by Larry Gorkin on May 17, 2011 - 11:44 am

    Interesting analysis. When I looked at this from another angle, several years ago, I was struck with the hypothesis that payers that had established five or more “prior auth” conventional DMARDs before an RA patient could gain access to a TNF-alpha inhibitor used less TNF-alpha as compared to payers who had established “reasonable” co-pays (e.g., 20% co-pay), without the “prior auth” barriers, for these $15,000 to $25,000 biologic agents, in terms of annual cost.
    I suspect you would enjoy reading some of my reports at the above website.
    Kind regards, Larry

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 )

Google+ photo

You are commenting using your Google+ 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: