Tiered pharmacy benefits can be lethal

Annals of Internal Medicine
September 16, 2014
Comparative Effectiveness of Generic and Brand-Name Statins on Patient
Outcomes: A Cohort Study
By Joshua J. Gagne, PharmD, ScD; Niteesh K. Choudhry, MD, PhD; Aaron S.
Kesselheim, MD, JD, MPH; Jennifer M. Polinski, ScD, MPH; David Hutchins,
MBA, MHSA; Olga S. Matlin, PhD; Troyen A. Brennan, MD; Jerry Avorn, MD;
and William H. Shrank, MD, MSHS

Background: Statins are effective in preventing cardiovascular events,
but patients do not fully adhere to them.

Objective: To determine whether patients are more adherent to generic
statins versus brand-name statins (lovastatin, pravastatin, or
simvastatin) and whether greater adherence improves health outcomes.

Design: Observational, propensity score–matched, new-user cohort study.

Setting: Linked electronic data from medical and pharmacy claims.

Participants: Medicare beneficiaries aged 65 years or older with
prescription drug coverage between 2006 and 2008.

Intervention: Initiation of a generic or brand-name statin.

Measurements: Adherence to statin therapy (measured as the pro- portion
of days covered [PDC] up to 1 year) and a composite outcome comprising
hospitalization for an acute coronary syndrome or stroke and all-cause
mortality. Hazard ratios (HRs) and absolute rate differences were estimated.

Results: A total of 90 111 patients who initiated a statin during the
study was identified; 83 731 (93%) initiated a generic drug, and 6380
(7%) initiated a brand-name drug. The mean age of patients was 75.6
years, and most (61%) were female. The average PDC was 77% for patients
in the generic group and 71% for those in the brand-name group (P <
0.001). An 8% reduction in the rate of the clinical outcome was observed
among patients in the generic group versus those in the brand-name group
(HR, 0.92 [95% CI, 0.86 to 0.99]). The absolute difference was -1.53
events per 100 person-years (CI, -2.69 to -0.19 events per 100
person-years).

Limitation: Results may not be generalizable to other populations with
different incomes or drug benefit structures.

Conclusion: Compared with those initiating brand-name statins, patients
initiating generic statins were more likely to adhere and had a lower
rate of a composite clinical outcome.

Primary Funding Source: Teva Pharmaceuticals.

From the Discussion:

In a head-to-head comparison, we found that patients initiating generic
statins were more likely than those initiating brand-name statins to
adhere to their prescribed treatment and had an 8% lower rate of a
composite end point of cardiovascular events and death. Generic drug use
has been widely recognized to reduce patient out-of-pocket costs and
payer spending. Most persons in the United States are enrolled in
prescription drug insurance programs with tiered benefits that require
higher copayments for brand-name prescriptions than bioequivalent
generic versions. Among patients in our study, the mean copayment for
the index statin prescription was $10 for generic drug recipients and
$48 for brand-name drug recipients. Our finding that adherence is
greater with generic statins than with brand-name statins is therefore
not surprising and is consistent with other studies that have shown a
direct relation between higher copayments and lower adherence.

http://annals.org/article.aspx?articleid=1905128

****


Comment by Philip Verhoef

This study further dispels the notion that more skin in the game leads
to better outcomes.

The researchers prospectively examined 90,111 patients who received a
new prescription for statin therapy and divided them into receiving
generic (83,731 patients) versus name-brand (6380 patients) for the 3
statins that were generic at that time. They used propensity scores to
"match" patients since they weren't able to properly randomize patients
to generic versus name-brand. They then measured adherence and outcomes
and found the following:

* 77% adherence in patients receiving generics versus 71% adherence in
patients receiving name brand

* 8% reduction in the rate of pooled clinical outcomes (stroke, acute
coronary syndrome or all cause mortality) for patients receiving generics

* They note that the out of pocket costs to fill the generic was 10
dollars and the branded was 48 dollars

They basically conclude that patients who have to pay more for their
medications are less likely to be adherent, and that the difference in
clinical outcomes occurred as a direct result of the increased cost,
leading to poorer adherence and consequently more strokes/ACS/death.

There's lots in here:

* Generics work

* Generics are cheaper

* Relatively small differences in adherence can lead to substantial
clinical outcomes (ie, perhaps we should focus on improving adherence to
prescribed therapies)

But I think that the biggest take-home is that if patients have to pay
more for medications, they will not take them and then they will die.
Ergo, shifting costs to patients leads to more death.


(Philip Verhoef, MD, PhD is Clinical Instructor of Medicine at The
University of Chicago Medicine, a member of the PNHP Advisory Board, and
co-president for PNHP-Illinois.)