The Challenge of Formulary Management in the DoD

1
The Challenge of Formulary Management in the DoD
UF 101

• Dr. Dave Bretzke, Dr. Angela Allerman, and Major
  Wade Tiller
• PEC Conference 2007

2
Objectives

• Describe the UF decision making process
• Differentiate the roles and responsibilities of
  the DoD PEC and the DoD PT
• Explain why UF decisions apply across both the
  direct care and purchased care points of service
• Discuss the key elements that must be considered
  in all Uniform Formulary decisions

3
Uniform Formulary

• Intended to provide uniform availability across
  DoD dispensing venues
• Legislated and published regulation
• Must do
• How to do it
• Introduce 3rd tier (pay for preference)
• Copays governed by law
• Limits ability to increase spread
• Difficult to change

4
Uniform Formulary

• Decisions apply across dispensing venues
• Defined role of DoD PT Committee
• Types of decisions
• Formulary status
• PA, QL, MN
• Moved decision from DoD PT Committee to TMA
  Executive Director (Dr. Winkenwerder)
• Introduced visibility through Beneficiary
  Advisory Panel (BAP)

5
Uniform Formulary

• Must review all drug classes
• Implement one at a time
• New drugs
• PT defines class complete freedom
• Class set as BCF or ECF
• Within every class, at least one agent must be
  BCF / ECF

6
Loading the hopper

• Mandate to review every drug class
• Identify drug class candidates
• DoD high spend, high use
• Competition
• New drugs / generics

7
Loading the hopper (cont)

• Feasibility
• Current situation
• Likely constituents
• Likely outcomes
• Cover BCF/ECF requirement
• Determine priority
• Staff workload
• Industry considerations
• Put on calendar

8
Uniform Formulary Decision Cycle

• Preliminary review (T-180 270)
• PT Committee meets (T-90 180)
• Defines class, BCF or ECF
• Request for Pricing Info (T-75 100)
• Price Quotes Received (T-20 30)
• PT Committee meets (T)
• Final review, recommendations
• BAP meets (T45)
• Comments to TMA
• TMA Director reviews signs minutes (T60)
• Implementation of decisions (no later than T240)

90-Day Decision Cycle
9
Everything Happens at Once
Aug 063 Classes
10
Why review Statin Class now?

• Statin contract set to expire
• Simvastatin available in a generic version
• Exclusivity?
• 1 MHS drug class
• 1M users
• 550M / year (9 of MHS total)
• How would you define the class?

11
Antilipidemic I Class
Generic name Brand name Generic availability?
Statins Statins Statins
Atorvastatin Lipitor No
Fluvastatin Lescol Lescol XL No No
Lovastatin IR Lovastatin ER Mevacor Altoprev (previously Altocor) Yes No
Pravastatin Pravachol Yes except Pravachol 80mg
Rosuvastatin Crestor No
Simvastatin Zocor Yes
Statin Combinations Statin Combinations Statin Combinations
Atorvastatin/Amlodipine Caduet No
Pravastatin/Aspirin Pravigard PAC Discontinued May 13, 2005
Lovastatin/Niacin Advicor No
Simvastatin/Ezetimibe Vytorin No
Add-on therapies Add-on therapies Add-on therapies
Niacin Niaspan, Niacin IR No, yes
Ezetimibe Zetia No
12
Antilipidemics-I Evaluation Strategy

• Developed by the PEC
• Factors considered
• Clinical needs of the DoD population
• Generic availability
• Market competition within the therapeutic class
• Approved by DoD PT Committee
• Develop bidding scenarios
• Put it on the street

13
PEC Process for Determining Relative Clinical
Effectiveness

• UF Rule DoD PT Committee considers information
  regarding safety, effectiveness, and clinical
  outcomes when comparing drugs in the class

14
Relative Clinical Effectiveness
ProcessAntilipidemic I agents

• Team approach
• Developed key questions to answer
• Class divided into intensive LDL lowering (gt45 ?
  in LDL) and low to moderate LDL lowering (lt45 ?
  in LDL) drugs
• Efficacy differences
• Surrogate outcomes (LDL ?, HDL ?)
• Clinical outcomes (myocardial infarction, acute
  coronary syndromes, mortality, revascularization,
  stroke)
• Safety / Tolerability differences
• Common adverse effects
• Serious elevated liver enzymes, proteinuria,
  myopathy, rhabdomyolysis
• Drug interactions, special populations
  (pediatric, pregnancy)
• Other Factors (pleiotropic effects, dosing
  initiation schedules)
• Initial assessment of DoD population needs

15
Intensive vs. low to moderate LDL lowering
HMG CoA Reductase Inhibitors (Statins) HMG CoA Reductase Inhibitors (Statins) HMG CoA Reductase Inhibitors (Statins) HMG CoA Reductase Inhibitors (Statins) HMG CoA Reductase Inhibitors (Statins) HMG CoA Reductase Inhibitors (Statins) HMG CoA Reductase Inhibitors (Statins) HMG CoA Reductase Inhibitors (Statins)
Statin LDL reduction Lovastatin (Mevacor) Lovastatin (Altoprev) Pravastatin (Pravachol) Simvastatin (Zocor) Fluvastatin (Lescol, Lescol XL) Atorvastatin (Lipitor) Rosuvastatin (Crestor)
25-30 20mg 20mg 20mg 10mg 40mg
30-40 40-80mg 40mg 40mg 20mg 80mg XL 10mg
40-45 80mg (40mg X 2) 60mg 80mg 40mg or Vytorin 10/10 20mg 5mg
45-50 Note Ezetimibe or Niacin generally decreases LDL up to an additional 15 Note Ezetimibe or Niacin generally decreases LDL up to an additional 15 Note Ezetimibe or Niacin generally decreases LDL up to an additional 15 80mg or Vytorin 10/20 40mg 10mg
50-55 Note Ezetimibe or Niacin generally decreases LDL up to an additional 15 Note Ezetimibe or Niacin generally decreases LDL up to an additional 15 Note Ezetimibe or Niacin generally decreases LDL up to an additional 15 Vytorin 10/40 80mg 20mg
gt55 Note Ezetimibe or Niacin generally decreases LDL up to an additional 15 Note Ezetimibe or Niacin generally decreases LDL up to an additional 15 Note Ezetimibe or Niacin generally decreases LDL up to an additional 15 Vytorin 10/80 40mg
Combination agents follow their parent statin
16
85 of the population
85 of the patients Will use simvastatin lt
40mg/day
45 LDL Reduction
Majority of DoD population can meet LDL goals
with simvastatin
17
Relative Clinical Effectiveness
ProcessAntilipidemic I agents

• Define search strategy for evidence
• FDA website
• MEDline search for approved and off-label uses
• Evidence Based Medicine (EBM) sites searched
• Existing guidelines (ACC, AHA, NCEP ATP III,
  VA/DoD)
• Database, retrospective cohort studies (safety)
• Solicit provider opinion
• Survey development
• SG Consultants teleconference
• Survey to providers and pharmacists
• Invited pharmaceutical manufacturer presentations
• Ongoing discussion with cost-effectiveness expert

18
Relative Clinical Effectiveness
ProcessAntilipidemic I agents

• Evidence Based Medicine (EBM)
• Integrating best research evidence with needs of
  patient/population
• EBM principles
• Helps to assess the quality of clinical evidence
• rigorous study design
• appropriate statistical tests
• inclusion of patients similar to DoD
• appropriate inclusion and exclusion criteria
• appropriate objective clinical endpoints
• appropriate definition of adverse events
• valid conclusions based on studied outcomes and
  statistical tests
• Goal for relative clinical effectiveness section
  is to use the best quality evidence when
  determining the differences between the class
  members

19
Relative Clinical Effectiveness
ProcessAntilipidemic I agents

• Systematic process to identify best evidence
  available
• Hierarchy of levels of evidence
• Systematic reviews
• Condense the results of several trials (10s-100s)
  into an understandable format
• Can help determine differences in efficacy,
  safety between drugs
• Sources Cochrane, UK, Canada, Oregon
• Some limitations different patient populations,
  different background medications
• Randomized, double-blind, controlled trials
• Head-to-head trials
• Placebo controlled trials
• Non-randomized trials unequal allocation of drug
  treatment
• Cohort studies, case control studies no
  intervention given
• Case reports, anecdotal reports lowest level of
  evidence small numbers of patients

20
Relative Clinical Effectiveness Process Sections
of class review

• Efficacy
• Discuss key issues in efficacy differences
  between class members or limitations in the
  available evidence
• Safety
• How likely will each drug in the class cause
  harm?
• Often need evaluation of sources other than
  clinical trials
• Well-designed retrospective health claims
  database study
• Published FDA data available
• Required lab monitoring
• Tolerability assessment How likely will patients
  take the drug?
• Drug discontinuations due to adverse effects
• Other Factors anything else we can think of
• Overall clinical effectiveness conclusion

21
Antilipidemics I Clinical Evidence Summary
CHD CHD CHD CHD CHD Stroke
1o Prevention 2o Prevention 2o Prevention 2o Prevention Mixed 1o 2o Prevention Stroke
  1o Prevention  General Acute Coronary Syndrome  Atherosclerotic Progression Mixed 1o 2o Prevention Stroke
Atorva ASCOT CARDS TNT IDEAL GREACE MIRACL PROVE-IT REVERSAL ASAP 4D ASCOT CARDS TNT SPARCL
Fluva LIPS FLORIDA Riegger et al ALERT
Lova AFCAPS Knatterud et al   MARS
Prava WOSCOPS LIPID CARE PROVE-IT PACT REVERSAL REGRESS L-CAD ALLHAT PROSPER CARE LIPID
Simva HPS 4S IDEAL A to (Z) MAAS ASAP HPS 4S HPS
Rosuva ASTERIOD  
Active control trial

• Most studies compared a statin to placebo
• Update to ATP III NCEP guidelines primarily based
  on 5 major clinical trials ALLHAT-LLT,
  ASCOT-LLA, HPS, PROSPER, and PROVE-IT

22
Antilipidemic I agents Efficacy Conclusion

• No head-to-head trials of equivalent doses of
  different statins for reducing coronary events or
  mortality
• Meta-analysis (Zhou et al) In low to moderate
  doses of atorva prava simva in long-term
  cardiovascular prevention using adjusted indirect
  comparison
• Evidence summary

Atorva Fluva Lova Prava Simva Rosuva
1o CHD 1o CHD 1o CHD ?10 mg ?20-40 mg ?40 mg ?40 mg
2o CHD General pop General pop ?10-80 mg ?80 mg(after PTCA) ?40 mg ?20-40 mg
2o CHD ACS ACS ?80 mg X X X
2o CHD Revasc Revasc ?80 mg ?80 mg(after PTCA) ? 40-80 mg(after CABG) ?40 mg ?20-40 mg

Stroke Stroke Stroke ?10-80 mg ?40 mg ?20-40 mg
? statistically significant RRR X available
trials, but no statistically significant
difference shown
23
Relative Clinical Effectiveness
ProcessAntilipidemic I agents

• First draft of written class review
• Assess military readiness issues if applicable
• Gather additional clinical information
• Evaluate feasibility of classifying one or more
  drugs as non-formulary under the UF
• Requires input from PEC management, clinical
  staff, economic experts, and procurement experts
• Initial consideration of medical necessity
  criteria
• Final written draft completed, edited by PEC
  staff
• Revisions completed, slides developed for meeting
• Ongoing process steps are not necessarily
  sequential

24
Challenges of Relative Clinical
EffectivenessAntilipidemic I agents

• Lack of comparative head-to-head trials overall,
  or problems when available
• Is the comparison drug inferior?
• Comparison of non-equivalent doses of statins
• PROVE-IT trial compared 80 mg atorvastatin with
  40 mg pravastatin in ACS patients
• What to do in absence of head-to-head trials
• Calculate NNT
• Compare hazard ratios, odds ratios, confidence
  intervals
• Similar point estimates? Overlapping confidence
  intervals
• Efficacy vs. Effectiveness
• Strict study conditions may not reflect use of
  drug in real world
• Co-morbid conditions not represented in RCTs

25
Challenges of Relative Clinical
EffectivenessAntilipidemic I agents

• Safety
• Safety data for new market entry is often
  incomplete and based on 1,000s of patients,
  rather than 1,000,000s of patients
• Length of time on the market and provider
  experience can determine provider confidence in a
  drugs safety profile
• Provider opinion
• Past experiences and perceptions play a role
• Degree of therapeutic interchangeability within
  the drug class
• Difficult to conclusively determine that Drug A
  is better than Drug B
• Can usually say Drug A is no worse than Drug B

26
Introduction Relative Cost-Effectiveness

• The UF final rule states
• If a pharmaceutical agent in a therapeutic class
  is not cost-effective relative to other
  pharmaceutical agents in that therapeutic class,
  it may be classified as a non-formulary agent
  (Federal Register / Vol. 69, No. 63 / Thursday,
  April 1, 2004 / Rules and Regulations)
• Applies to
• Therapeutic class reviews
• New agents in classes already reviewed

27
Relative Cost-effectiveness Process

• PEC performs two different types of economic
  analyses
• Pharmacoeconomic Analysis
• Identifies, measures, and compares the costs
  (i.e., resources consumed) and consequences
  (clinical, economic, and humanistic) of
  pharmaceutical products and services (Bootman,
  et al., Principles of Pharmacoeconomics 2nd Ed.,
  New York, St. Martins Press,1996)
• Budget Impact Analysis
• By definition, a BIA estimates the impact on
  annual healthcare use and cost for the first,
  second and subsequent years after the
  introduction of the new product for a national or
  health plan population (Mauskopf, JA et al.)

28
Pharmacoeconomic AnalysisProcess

• What type or types of analyses performed?
• Clinical team provides insight as to how the
  agents differ in terms of relative clinical
  effectiveness and to identify significant
  clinical outcomes
• The type of pharmacoeconomic analysis performed
  is dependent upon the how the outcomes are
  measured and the extent that these outcomes
  differ between the agents within a therapeutic
  class
• Scour the literature for every pharmacoeconomic
  analysis that has ever been performed on the
  agents within the class
• Industry for any published or unpublished
  pharmacoeconomic models they use to support their
  respective agents
• Synthesize the information and develop
  pharmacoeconomic models to critically evaluate
  the cost-effectiveness of the agents within the
  class.

29
Pharmacoeconomic Analysis ProcessAntilipidemic I
Agents

• Strong intermediate outcome evidence
• Mean LDL reduction by drug and strength
• Total cholesterol/HDL ratios
• Head-to-head RCT
• Long-term clinical outcomes evidence
• There are no head-to-head trials comparing
  equivalent doses of statins that evaluate
  clinical outcomes for reducing mortality or other
  clinical outcomes (e.g., myocardial infarction,
  stroke, need for revascularization)

30
Pharmacoeconomic AnalysisModels

• The Annual Cost per 1 LDL Decrease decision
  analytic model.
• The Annual Cost per Patient Treated to Goal Monte
  Carlo simulation model
• The Medical Cost Offset Model (industry)
• Compared the cost-effectiveness of these agents
  based on their predicted outcomes and total
  predicted health care expenditures for CHD and
  CHD risk-equivalent patients.
• The Cost per Event-Free Patient decision analytic
  model
• Based on the results of the IDEAL Trial, compared
  the cost-effectiveness of the agents included in
  that trialhigh-dose (80mg) atorvastatin
  (Lipitor) vs. low-dose (20-40 mg) simvastatin
  (Zocor, generics)

31
Pharmacoeconomic AnalysisResults

• Results of the first three cost-effectiveness
  analyses showed ezetimibe/simvastatin (Vytorin)
  to be the most cost effective high LDL lowering
  agent.
• Results of the fourth analysis revealed that
  high-dose (80 mg) atorvastatin (Lipitor) was more
  effective but considerably more costly compared
  to low dose (20-40mg) simvastatin (Zocor,
  generics).

32
Budget Impact AnalysisAffordability
Affordability
Cost-Effectiveness
Quality
Efficacy
Safety
33
Budget Impact Analysis Overview

• Models the MHS budget impact dynamics of a UF
  decision
• Factors included in the model
• Market growth
• Market share migration
• Change in utilization among therapeutic class
  agents as beneficiaries migrate from NF to UF
  agents
• Point of Service Migration
• Change in utilization among POS
• Costs associated with migration from NF to UF
  agents
• Medical necessity processing fees
• Provider costs associated with switching patients
  from NF to UF agents
• Lab costs
• Cost reduction associated with NF 22.00 copay

34
Budget Impact AnalysisMethodology Overview

• Dynamic Markov Model
• Differs from static model in that it incorporates
  the element of time
• Migration transitions from NF to UF occur
  gradually overtimeno longer assumes
  instantaneous migration
• Open not closedbeneficiaries enter and exit
  model over timeallows for market growth
• Allows for providers/beneficiaries to change
  agents over time (migrate back and forth between
  Markov states)

35
Budget Impact Analysis Results

• Interesting
• All scenarios that included Lipitor on the UF
  were more favorable compared to scenarios without
  Lipitor on the UF
• Ultimately
• The Uniform Formulary scenario that included
  atorvastatin (Lipitor), ezetimibe/simvastatin
  (Vytorin), and simvastatin (Zocor, generics) 80
  mg as the high LDL lowering agents on the UF
  was the most cost effective UF scenario.

36
PT Committee DecisionFactors Considered

• Relative Clinical Effectiveness
• Relative Cost Effectiveness
• Pharmacoeconomic Analysis
• Vytorin the most cost-effective agent intensive
  dose statin
• No economic benefit to exclude any of the low to
  moderate LDL lowering agents from the UF
• Budget Impact Analysis
• The UF scenario that included Vytorin, Lipitor,
  and Simvastatin 80mg along with the low to
  moderate LDL lowering agents was the most
  cost-effective UF scenario
• Market Conditions
• Potential Impact/Disruption to Beneficiaries

37
Distribution of Unique Users (UUs) by Agent and
POS
UUs
Source Utilization Data - PDTS BPA -
Manufacturers Current MTF (Prime Vendor),
Retail TMOP from PDTS FSS - VAPBM
38
After the PT Meeting

• BAP meeting
• Minutes from DoD PT BAP meetings presented to
  TMA Executive Director and signed
• Minutes posted
• MTF formulary management (pushout) documents
• Medical Necessity criteria
• Finalize forms and submit to ESI / post
• Real work begins implementation!

39
MTF Implementation

• Review decisions
• Develop implementation plan
• Provider education
• New vs. existing patients
• Confirm prices loaded
• Follow-up MUE / DUE
• Execute

40
PEC Monitoring

• Develop monitoring plan
• Ongoing
• Specific targets
• Monthly trend analysis
• Combined
• By Point of Service
• Monthly cost per ???? tracking
• Rx, 30-day Eq Rxs, Treatment day, Tab/cap
• Prime Vendor availability

41
PEC Monitoring

• Field alerts
• Add to PACER
• Return and Report
• DoD PT Committee
• BAP
• Dr. Winkenwerder
• Conferences
• ??????

42
PEC Monitoring - Example

• Statins
• Ongoing analysis
• Trend by drug by month
• Cost per day by month
• Specific targets
• Simvastatin cannibalization
• Equipotent doses (e.g. Lipitor 10 20mg)
• Up prescribing
• Brand over-marketing (e.g. Lipitor Vytorin vs.
  class)
• Cost effective
• High Potency Use (Crestor vs. Lipitor 40 80mg
  vs. Vytorin)

43
Summary - Statins

• Extremely important drug class for DoD
• Use Vytorin in patients who need gt 45 reduction
  in LDL
• Our efforts to sustainthe TRICARE Rx benefit
  require that MTF prescribers continue using
  simvastatin
• I strongly encourage MTF commanders, doctors and
  pharmacists to maximize the use of simvastatin
• Dr. Winkenwerder, ASD-HA

44
Summary - Overall

• The UF process and drug class reviews are
  comprehensive and (sometimes painfully) thorough
• Best real world application of EBM reviews and
  pharmacoeconomic analyses
• MTF input is extremely valuable
• Strives to find the best balance of access vs.
  costs
• Future offers depend on todays performance

45
Contact Info

• Major Wade Tiller, RPh, MHA, PhD
• Biomedical Sciences Corp
• Deputy Director, DoD Pharmacoeconomic Center
• kevin.tiller_at_amedd.army.mil
• David R. Bretzke, PharmD
• Procurement / Informatics Pharmacist
• DoD Pharmacoeconomic Center
• david.bretzke_at_amedd.army.mil
• Angela Allerman, PharmD, BCPS
• Clinical Pharmacy Specialist
• DoD Pharmacoeconomic Center
• angela.allerman_at_amedd.army.mil