ForecastingDemand for Critical Medical Technologies: Issues

1
Forecasting Demand for Critical Medical
TechnologiesIssues Options 

• Neelam Sekhri
• Global Health Forecasting Working Group

2
Outline

• What is Going Wrong and Why
• Recommendations What We Can Do Now
• Take Demand Forecasting Seriously
• Create Global Health Infomediary
• Share Risks and Align Incentives
• Recommendations Building a Foundation for Long
  Term Access
• Next Steps

3
What is Going Wrong and WhyThe Changing Global
Health Environment
4
The Big Picture Problem

• Despite the influx of substantial new monies
• The right products arent getting to the people
  who need them when they need them

5
What is Going Wrong?
Leading to Unreliable Demand Forecasts
Complex Rapidly Evolving Market..

• Funders
• More Money
• New Mechanisms

• Missing
• Accountability
• Institutions
• Policies

• Suppliers and Products
• New Players
• New Products
• Variable quality

The right products are not getting to the the
people who need them when they need them
Lack of credible data

• Uncertain
• Funding
• Uptake
• Demand/Price

• Buyers
• New Buyers
• Complex supply
• chains

• Inappropriate
• Methodologies
• Incentives

• Intermediaries
• New stakeholders

6
The ACT STORYCQ Resistance and popularity of
ACTs
Source The Economist
7
The ACT STORYACT Production Process (Based on
Coartem)
8
THE ACT STORYTracking Coartem Forecast
Performance
9
The ACT STORY New MoneyHistorical and Future
Funding for Malaria
10
The ACT STORY New StakeholdersACT Supply Chain
11
The ACT STORY New Products
12
Risk Map Current ACT Supply Chain
13
Incentive Map Current ACT Supply Chain
14

The ACT STORY Consequences
15
Recommendations
16
Opportunities for Short Term Action

• Technical constraints
• Information Gaps
• Methodologies in the face of market discontinuity
• Structural constraints
• Asymmetric distribution of risks lead to
  misaligned incentives

17
Recommendations Framework
Consolidates disparate information
sources Enables more accurate forecasts Establishe
s common baseline forecasts
Increases credibility and transparency of
forecasting process through adoption of sound
principles Expands forecasting expertise for
global health products
Shares suppliers risk Motivates all players to
take forecasting seriously and share
information Motivates funders to reduce risk
18
Why?
Take Forecasting Seriously

• Impact of Poor Forecasting on Access is Large
• Demand Forecasting Drives Supply Planning
• Forecasting is Unique Expertise

19
How ?
Take Forecasting Seriously

• Understand What Demand Forecasting Means for
  Stakeholders
• Adopt Principles Of Good Forecasting
• Strengthen Technical Forecasting Capability

20
Principles Why
Take Forecasting Seriously
Forecasting is a Process, not a Number

• Credibility and Trust In Process
• Increase Confidence and Reduce Market Risk
• Reduce Forecast Variation

21
Principles of Good Forecasting

• Customer-focused principles
• Identify principal customers/decision makers and
  understand their needs.
• Understand and clearly communicate purpose and
  decisions it will affect.
• Create forecasting process independent of
  planning and target setting.
• Protect forecasting process from political
  interference and ensure transparency.

22
Principles of Good Forecasting

• Process- context-focused
• 4. Embed forecast into broader environment
  taking account of market conditions, public
  policy, competitive forces, regulatory changes,
  health program guidelines.
• Create dynamic forecasting process that
  continually incorporates and reflects changes in
  the market, public policy and health program
  capabilities.

23
Principles of Good Forecasting

• Methodology- data-focused principles
• Choose the methodologies appropriate to the data
  and market environment.
• Keep methodologies simple and appropriate to the
  situation. Dont introduce too much complexity,
  but include sufficient detail to address the
  level of investment risk and accuracy required.
• Make forecast assumptions clear and explicit.
• Understand data and their limitations. Use
  creativity and intelligence in gathering and
  introducing data into forecasts.

24
Selection Tree for Forecasting Methods
25
Develop Technical Capacity

• Recruit students from specialized graduate
  programs
• Provide extensive training for existing personnel
• Recruit experienced forecasters from industry
• Create forecasting methodology resource base for
  developing countries

26
Why?
Create Global Health Infomediary

• Key stakeholders require similar information
  across variety of diseases products
• Some data exist but are not shared systematically
• More and better market research is needed!

27
From Needs To Demand
Effective Demand
28
Approaching the Information Gap
Create Global Health Infomediary
29
Developed Country Model

• Market data collected by Infomediaries (ex. IMS
  Health)
• Customized market information gathering/analysis
  by many private organizations

Independent organizations...
...with credible processes and well established
reputations...

• Quality of information and methodologies
• Credibility of information and methodologies

Provide Information and Analyses to Stakeholders

• Market Analyses
• Baseline Demand Forecasts

30
Global Health Infomediary
INFOMEDIARY
31
Key Functions
Create Global Health Infomediary

• Serve as a neutral party that maintains
  relationships with supply chain partners and has
  credibility with all stakeholders
• Create and Maintain Central Repository on demand
  and supply
• Provide Continuous Data Gathering
• Generate Transparent, Baseline Aggregate Demand
  Forecasts

32
Data Collection Points for Medical Products
PI Importer
Wholesaler
Retail Pharmacies/Dispensing Doctors
Manufacturer
Specialty distribution/ Homecare
Patient
Community Clinics/ Physician offices
Hospital
IMS Presentation by Peter Stephens 26/2/07
33
Retail pharmacy and medical the most common types
of data collected IMS
IMS Presentation by Peter Stephens 26/2/07
34
Retail Pharmacy Volume Precision across low and
lower middle income countries (52-100) IMS
packs within /- 22.5 of manufacturer
estimates, 2004 data
IMS Presentation by Peter Stephens 26/2/07
35
Why?
Share Risk and Align Incentives through
Contracting

• Funders and Buyers bear limited risk for poor
  forecasting
• Efficient risk sharing motivates all parties to
  perform better
• Contracts are important to share risks align
  incentives

36
Creative Contracting
Share Risk and Align Incentives through
Contracting

• Minimum Purchase Commitments
• Quantity Flexibility Contracts
• Buyback Contracts
• Revenue Sharing
• Real Options

37
Long-termAvoiding reducing risk

• More predictable, reliable donor resources
• Streamlined product approval regulatory
  processes
• Product development support that is attuned to
  market forces (more pull, some push)
• Strengthened health systems / in-country supply
  chains that feed information and orders to
  suppliers (more pull)

38
Recommendations Framework
Consolidates disparate information
sources Enables more accurate forecasts Establishe
s common baseline forecasts
Increases credibility and transparency of
forecasting process through adoption of sound
principles Expands forecasting expertise for
global health products
Shares suppliers risk Motivates all players to
take forecasting seriously and share
information Motivates funders to reduce risk
Building a foundation for long-term access
39
Consultation and Next Steps
http//www.cgdev.org/section/initiatives/_active/
demandforecasting.

• Soliciting comment and critique through March 23,
  2007
• Finalizing the work of the group by April 2007
  for May publication and launch

40
Appendix
41
Working Group Members

• Deborah Atherly, PATH
• Jorge Carrion, PAHO
• Robert Chisholm, Consultant
• Renia Coghlan, MMV
• Peter Evans, Consultant
• Gian Gandhi, IAVI
• John Hurvitz, Covington Burling
• Steve Jarrett, UNICEF Supply Division
• Andrew Jones, GAVI Alliance
• Steve Kinzett, PATH
• Ruth Levine, CGD
• Andrea Longhi, NHS
• Elisabetta Molari, Global Fund
• Morgan Musongole, Zambia Malaria Control Program

• Angeline Nanni, PneumoADIP
• Donné Newbury, Bristol-Myers Squibb
• Hans Rietveld, Novartis
• Mark Rilling, USAID
• Nina Schwalbe, TB Alliance
• Neelam Sekhri, The Healthcare Redesign Group Inc.
• Marcus Soalheiro Cruz, Nortec Quimica
• Anil Soni, Clinton Foundation
• Jeffrey Sturchio, Merck
• Krista Thompson, BD
• Christine Tonkin, IAPSO
• Saul Walker, International Partnership for
  Microbicides
• Edward Wilson, John Snow, Inc.

42
What Is Demand Forecasting?

• Demand Forecasting ?Needs Estimates
• Demand Forecasting ? Demand Creation/Advocacy
• Demand Forecasting ?Target Setting

Estimates Effective Market Demand Product needs
which have or will have purchasing power behind
them and will result in actual orders