Transcript Iowa Medicaid Preferred Drug List Powerpoint

Iowa Medicaid
Preferred Drug List
Presented by:
Timothy Clifford, MD
September 28 & 29, 2004
1
Overview

Today and next month:

Review data and evidence to design 1st
PDL
This may seem like a daunting task
 Does the PDL need to be perfect?

No such thing
 Maintenance is fluid and evolutionary

2
The 1st Year


Together we will create an extensive PDL
affecting most drug categories
The 1st year will be relatively simple



“Relative” may not seem so until next year
Recommended “easy” approach (despite more
savings being available through PA approach)
Base year—subsequent versions will be more
complex and with the potential for fewer
choices in order to realize greater savings
3
Focus for Next Two Days
Need to keep an open but skeptical
mind
 Listen carefully and critically
 Concentrate on drugs with draft PDL
positions that concern you

4
Ranking Drugs
Categorize drugs as:
 Preferred without conditions, or
 Preferred with conditions not involving PA (eg.
age ranges Ortho-Evra preferred if under 21
years), or
 Preferred with conditions involving PA (eg.
Genotropin GH), or
 Non-preferred with all non-preferred drugs
equal, or
 Non-preferred with same non-preferred drugs
favored over others (eg. Protonix less nonpreferred than Prilosec)
5
PDL Basics
In many PDL categories:
 Although there may be many differences
in individual responsiveness to any one
given product, the majority who
eventually respond to any drug in the
category will respond to the first drug
tried
 Law of diminishing returns can be
validated with utilization data
6
Success Targets
Drug response averages:
 1st product from roughly equivalent
class works 60 – 65% of the time
 2nd increases to 75 – 85%
 3rd to 85 – 90%
 4th to near 95%
 No matter how many drugs are
available, it will never be 100%
7
The Initial Drug Selections
Unless there is a need for a particular
characteristic of one drug that is not
present in the others (or vice-versa
with side effects), then the initial
choice should be based on the
average probability of response in the
population (as per studies), since this
cannot be predicted with any greater
certainty at the patient level
8
How to Test Choices

PA is the best method for testing
and validating clinical arguments for
medical necessity based on relative
risks or relative differences in
efficacy between preferred and nonpreferred drugs

Must be an acceptable cost:benefit
ratio to curtailing access to initial
choices
9
PDL Engineering
There are many different ways to
create a PDL
 Two different groups can follow the
same process precisely and arrive at
a different result
 Two different group can follow
grossly different rules and reach the
same results

10
Sorting Through It All
There is much information and data
from manufacturers and other
presenters
 The efficacy or value of the
manufacturer products to the
practice of medicine is not in dispute
 Keep it simple and focused on the
following three key issues

11
Three Keys
1.
2.
3.
Does manufacturer have proof that their
product is clinically better/safer than preferred
choices for the majority of the Medicaid
members—not just subpopulations?
Can manufacturer demonstrate that their
product is as or more cost-effective than the
preferred choices?
If the above cannot be shown, then focus on
what PDL criteria should be in order to access
the product via the PA process.
12
PDL vs. PA


The decision to make a drug preferred or
non-preferred can be simplified by using
the three keys as precepts
The level of clinical complexity necessary
for a P&T Committee decision on
preferred status is much different and
markedly simpler than it is for that
involved in determining prior
authorization approval criteria because…
13
Prior Authorization Criteria


In a PDL, the objective is to designate as
preferred the most cost-effective drugs
that will work for the majority of the
Medicaid population as initial choices
The PA arm or component of the PDL
requires a greater level of purely clinical
reasoning (the issues are different):

Does this individual (yes/no?) need this
particular drug (y/n?) for this condition (y/n?)
at this time (y/n?)?
14
In Summary…



The PDL is all about creating an array of
cost effective drugs that will suffice for
most patients, most of the time
All other drugs are available via PA
Preferred drugs are a set of tools that can
be used freely and hopefully prudently
without permissions
15