Docstoc

What is Analysis and Advocacy The Philosophy and Tools of A2

Document Sample
What is Analysis and Advocacy The Philosophy and Tools of A2 Powered By Docstoc
					                                       1




What is Analysis and Advocacy?
   The philosophy & tools of A2

   40th Summer Seminar on Population
         June 2nd – July 2nd, 2009
        East-West Center, Honolulu
                                          2




Strengthening strategic information for
better HIV/AIDS programming in Asia
                                     3




A2 is:
         An outgrowth of a growing
          sense of frustration
                                                                                                                    4


We know what to do about HIV in Asia…
       • We understand                                                                        FSW
         Asian epidemic                Clients
                                 Low or no risk
                                                                                        MSM                 IDUs

         dynamics                   males
                                                                                        Low or no risk
                                                                                           females


       • We have




                                 HIV Infections in millions
                                                              10

         examples of                                          8

         effective                                            6

                                                              4
         responses                                            2

                                                              0
          Thailand – 6 million                                1985      1990     1995       2000     2005         2010
           infections averted                                        With behavior change    Without prevention
                                                                                                       5


           …but we’re not doing it
                                                     Prevention coverage in Asia




                         Percent covered
                                           100
• Coverage is low…                          80
                                            60
                                            40
                                            20
                                             0
                                                       IDU            FSW                 MSM
                                                       Southeast Asia      Western Pacific


                                                             HIV among sex workers
                                                30
                                                25

• …and HIV epidemics                            20




                                       % HIV+
                                                15

  continue to grow                              10
                                                 5
                                                 0
                                               93      94   95    96    97    98     99      '00 '01 '02
                                           Rangoon, Burma        Guangxi, China           Hanoi, Vietnam
                                           Vientiane, Lao PDR    Papua, Indonesia         Riau, Indonesia
                                                         6


              What’s Gone Wrong?
• Lots of data, but no one puts together the “big picture”
• Despite 2nd generation surveillance guidelines,
  existing information on
      - Past and present responses
      - Epidemiology (HIV and STI) and
      - Behavioral data
   is not being analyzed in integrated fashion
• Critical analysis of response alternatives missing
• Data quality improvement is not being done
• Data is not being used for effective advocacy
                                      7




In short, the data collected is not
     influencing decisions….
                                       8




…and the understanding and political
commitment needed to make the right
       decisions is lacking!
                                      9




So what can we do about it? A2

   Integrated Analysis and Advocacy
                  10




How is A2 done?
                                                          11


1. Local teams gather & synthesize existing data

           Current state of the epidemic and response

• Existing data collected & critically analyzed:
   –   Epidemiological/behavioral/biological (IBBS,BSS)
   –   Sizes of key populations
   –   Responses: programs and policies
   –   Program costs and coverage
• Key trends in HIV, behaviors and responses
• Identify gaps & quality issues in data systems
                                                                                                                                                                                        12

                                                               The process itself builds understanding
                                                                        of our data and its limitations
                                                                                                                                          • Bangladesh A2
                                                                                                                                            – How many are at risk?
          PANCHAGAR
                   LA
                      LM
   THAKURGAONNILPHAMARION
                                          IR
                                             H   AT
                                                 KURIGRAM

                    DINAJPUR RANGPUR



                                         GAIBANDHA                                                    Asam
                                                                                                                                            – Involved government, NGOs, communities
                                                    UR
                                                     ALP




                         JAIPURHAT                             SHERPUR
                                                                                                                           SYLHET
                                                         JAM




                   NAOGAON                                                              SUNAMGANJ
                                         BOGRA                                  NETROKONA
CHAPAI NAWABGANG                                                         MYMENSINGH

                  RAJSHAHI                                                                         MOULVIBAZAR

                                                                                                                                              in meetings on specific populations
                                                                                             ARIA




                             NATOR          SIRAJGANJ
                                                    TANGAIL                     KISHOREGANJ HABIGANJ
                                                                                                NB
    West Bengal




                                                                                         BRAHMA




                                                            GAZIPUR
                                         PABNA
                                                                  NARSINGDI
                        KUSHTIA                         DHAKA




                                                                                                                                            – Reviewed service statistics, scientific
                                                  MANIKGANJ
                  MEHERPUR
                                            RAJBARI            NARAYANGANJ
                                                           Dhaka Division
                                                                                                                Tripura
                    CHUADANGA
                                                                                                                  CHHARI
                                                                 UR




                         JHENAIDAH                                    MUNSHIGANG
                                MAGURAFARIDPUR
                                                                  ARIP




                                                                                      COMILLA
                                                                                                                KHAGRA
                                                                   MAD




                                                                         SHARIYATPUR
                               JESSORE                                           CHANDPUR
                                            NARAIL
                                                 GOPALGANJ



                                                                                                                                              studies, expert opinions
                                                                                                     FENI
                                                                 BARISAL
                                                                                 LAKSMIPUR
                                                                                                                           RANGAMATI
                                    RA




                                                                                        NOAKHALI
                              SATKHI




                                         KHULNA
                                                          JHALAKATI
                                                       PIROJPUR
                                                  BAGERHAT                                                  CHITTAGONG
                                                                                BHOLA




                                                                                                                                          • Outputs
                                                                     PATUAKHALI
                                                                BORGUNA



                                                                                                                             BANDARBAN

                                                                Bay of Bengal                                  COX'S BAZAR




                                                                                                                                            – Consensus on size of risk populations
                                                                                                                                     ar
                                                                                                                                 n  m
                                                                                                                              ya
                                                                                                                            M




                                                                                                                                            – New estimates of PHA
                                                                                                    13
 Careful review identifies issues in data
      systems & aids interpretation
• Vietnam national estimates and A2
                 – Inconsistent trends -> protocol issues in the field
                 – HCMC trends explained
                     100
 Percent IDUs HIV+




                      80

                      60

                      40

                      20

                       0
                         94

                                95


                                       96

                                              97

                                                     98


                                                            99

                                                                   00

                                                                          01

                                                                                 02


                                                                                        03

                                                                                               04
                       19

                              19


                                     19

                                            19

                                                   19


                                                          19

                                                                 20

                                                                        20

                                                                               20


                                                                                      20

                                                                                             20
                                                                                   14
           2. Develop a local model of the epidemic
                        the Asian Epidemic Model (AEM)
Sizes & behavioral
trends in
• Clients
• Sex workers
• IDUs
                                     Injecting drug users     Female sex workers
• MSM
• Population at large



                    AEM
                   Calculation           Adult males              Adult females
                     Engine



Probabilities of                                        Observed
transmission                                            HIV trends
and start years                                         (white lines)
                                                                         15

          Use the model to determine where
           new infections come from…
                               Proportion new infections by population
• Help in deciding                           in Thailand
                               1.0
  appropriate focus for
                               0.8
  effective programs
                               0.6
  – Clients and sex workers,
                               0.4
    IDU, MSM
  – At advanced stages         0.2


    husband to wife & MTCT     0.0
                                     1988 1990 1995 2000 2005 2010

                                 IDU                 MSM
                                 Client from FSW     Husband from wife
                                 FSW from client     Wife from husband
                                 Casual sex
                                                                                            16

        …and to tell us where the epidemic is going

• Determine levels of current,                                 Current and annual new HIV
                                                                        infections
  cumulative and new
  infections in the country                                120,000




                                 HIV infections (number)
   – Adult and child                                       100,000

• Make projections to assess                                80,000
  future needs and impacts                                  60,000
   – ART needs                                              40,000
   – Affected children (in                                  20,000
     development)
                                                                0
                                                                1996      2001       2006
                                                                        Cumulative   New
                                                                                               17


Dhaka – high risk rapidly growth epidemic
                                       100
IDU




                Percent HIV positive
Hotel SW                                80
Street/Res SW
                                        60
Clients
Gen Male
                                        40
Gen Female
MSM                                     20
MSW
                                          0
                                          00

                                                  05

                                                          10


                                                                  15

                                                                          20

                                                                                  25

                                                                                          30
                                       20

                                               20

                                                       20


                                                               20

                                                                       20

                                                                               20

                                                                                       20
                                                18
3. Evaluate the impact of different programs
   choices & resource allocation decisions…..
     Responses and resources needed for maximum
                       impact

• Through linked AEM/GOALS Modeling
• The GOALS model: 3 modules…
   − Resource needs module
   − Impact module
   − Capacity module
                                                                        19
         GOALS relates programs, costs & behaviors

  Program mix      Budget   Coverage
                                        Improved Policy
Policy Interventions   $                environment


                                       Behavior change
                                       -age at first sex        AEM for
                                       -number of partners      epi impact
Prevention             $      %
                                       -condom use
                                       -STI treatment
                                       -safe injections

Care and treatment     $      %        Increased care,
                                                                Treatment
                                       treatment & mitigation
Mitigation             $      %                                 coverage
Program support        $
                                                20

      GOALS Resource Needs Module
• Inputs:
  – Population size,
  – Unit costs and
  – Coverage (current and to be achieved).
• Answers crucial policy questions:
  – What are the costs to achieve a specific
    prevention goal?
  – What financial resources are required to
    implement a national/provincial HIV plan?
                                                                                                    21


           Allows costing of prevention plans
                  Total Resource Required(2006-2010)                      • Yunnan provincial plan
                                                                            based on targets in
                                                                            National 5 Year Action
¥600,000,000
                                                                            Plan (2006-2010)
¥500,000,000                                                              • 2006: US$25.3 m
¥400,000,000                                           Policy etc.        • 2010: US$64.2 m
                                                       Mitigation
¥300,000,000
                                                       Care and Support
¥200,000,000
                                                       Prevention
¥100,000,000
                                                                            Note: numbers are
         ¥0                                                                 not final and are
               2006   2007    2008    2009    2010                          for illustration only
                                                                                               22
                                         How much will care cost us?
   Cost of ART (in Millions of USD)
                                               ART costs in Thailand
                                      600
                                      500
                                      400
                                      300
                                      200
                                      100
                                        0
                                        00




                                               05




                                                           10




                                                                      15




                                                                                 20




                                                                                          25
                                      20




                                             20




                                                         20




                                                                    20




                                                                               20




                                                                                        20
                                             First line only    First and second line

Source: Thai MOPH, World Bank, East-West Center,
        Expanding Access to ART in Thailand, 2005
                                                                                                                 23

                              Allows determination of resource gaps

                                                                                   • Guangxi, China
                          Resource Gap Scenario 1 (2006-2010)
                                                                                   • By 2010: RMB 1.5b
                                                                                     (US$190m)
               60,000
                                                                                   • Gap between funds to
               40,000
                                                                                     implement the latest
                                                                                     national HIV/AIDS plan &
                                                                                     available funds
RMB (10,000)




               20,000
                                                                                   • Raises the issue: How to
                   -                                                                 increase resources and
                            2006        2007        2008       2009        2010      choose most effective
               (20,000)                                                              program combinations
               (40,000)

               (60,000)
                                                                                         Note: numbers are
                             Available Resources   Resource Needs   Resource Gap         not final and are
                                                                                         for illustration only
                                                                                        24
              GOALS Impact Module behaviors for AEM
              Based on program coverage & resource allocations
                                 Baseline Behaviors
                          Risk Behavior before Interventions

Goals, Objectives and
Intervention choices       % HIV+      %
                                                # sharing     Freq of
                           among     sharing
                                                partners     injecting
                            IDU      needles
                                                                         AEM
 Pre-Intervention: 40%
 of IDU reached with
 peer education &
 outreach
                                     IMPACT MATRIX
 Post-Intervention: 90%
 of IDU reached with
 peer education &
 outreach                                                                      Impact

                          % HIV+      %
                                               # sharing     Freq of
                          among     sharing
                                               partners     injecting
                           IDU      needles
                                                                         AEM
                          Risk Behavior after Interventions


   Impact matrix based on published studies of program effectiveness
                                                                                                  25

         HCMC explored impact of 4 policy alternatives:
                                Scenario A          Scenario B       Scenario C      Scenario D
             Scenarios                                                            Recommended
                             Clients and FSW       MSM & MSW            IDU
                                                                                    Scenario
Migrant workers                    60%                 30%              30%             50%
Workplace                          60%                 30%              30%             50%
High school students               30%                 30%              30%             30%
Out of school youth                50%                 30%              30%             40%
Women in community                 30%                 30%              30%             30%
Direct FSW                         70%                 60%              60%             65%
Indirect FSW                       70%                 60%              60%             65%
IDU counseling & testing           30%                 30%              80%             60%
Needle and Syringe Program         10%                 10%              30%             20%
Methadone Program                 12.8%               15.8%            16.8%           13.4%
Naltrexone Program                 5%                  5%               5%              5%
Peer outreach for MSM              30%                 80%              30%             50%
Peer outreach for MSW              30%                 80%              30%             50%
                                $4,008,976          $4,003,596       $4,005,579      $4,006,138
Prevention Costs in 2010                                             VND 63.69
                             VND 63.74 billion   VND 63.66 billion                VND 63.70 billion
                                                                      billion
                                                                      26

          Comparison in relation to the baseline:
New HIV infections under various policy scenario alternatives (A-D)

                       2006-2010
16,000
14,000
12,000
10,000
 8,000
 6,000
 4,000
 2,000
   -
      02

      03

      04

      05

      06

      07

      08

      09

      10

      11

      12

      13

      14

      15

      16

      17

      18

      19

      20
    20

    20

    20

    20

    20

    20

    20

    20

    20

    20

    20

    20

    20

    20

    20

    20

    20

    20

    20
                 Baseline                    A - Clients & FSW
                 B - MSM & MSW               C - IDU
                 Recommended Scenario
                                                            27

    4. Turn strategic information into action
        Implementation of the most effective policies and
        programs, and mobilization of adequate resources

• Interactively engage policymakers, communities,
  program managers, donors & others to:
   – Build a common understanding of epidemic
   – Identify realistic alternative program packages
   – Evaluate costs, effects, & benefits of packages
• Show them the consequences of their choices
Identify policy processes and                                               28

who to engage:
policy makers, program
                                      Policy mapping
managers, communities,
donors, other stakeholders

Meet decision makers                Interactive engagement
   (formal/informal)
•Increase understanding                           Analyze impacts of
                                  Reprioritize
•Identify issues                                  response alternatives
                                  & redefine
•Present impacts of
                                    response         GOALS/AEM
  response alternatives
•Solicit input on what
  to change

                                 Evidence for
                                decision making
                                                       Reevaluate who is
                    Policy briefs, meetings,           engaged in process
                    data for planning, M&E
                                29




How is this done in practice?

   Methodology in the field
                                                                        30
     Affected
   communities                                    Technical
                      A2 in-country             Working Group
  Policymakers
 Decision makers

                                                  Advocacy &
Program managers
                                                   Data Use
                                                    Group
                           Synthesis
   Researchers            specialist(s)       In-country technical
                                              support, review and
International NGOs         Advocacy                 advice
     & donors
                          specialist(s)
 The public and the     A2 core team at           Regional A2
      media                                       Team (EWC,
                        local institution
                                                   FHI, HPI)

Who to engage,                                 Regional technical
involve & inform                            support, review and tools
                                                 development
                                                31


           Features of the A2 process

•   Inclusive – engages those it targets
•   Leverages existing local data & capacity
•   Remains focused on responses
•   Aims for sustainability by engaging local
    agencies
                                                          32


                Where does A2 operate?
• The 5 formal A2 sites supported by USAID:
   –   Bangladesh (FHI Bangladesh)
   –   Guangxi, China (Guangxi Center for AIDS and STD)
   –   Ho Chi Minh City, Vietnam (HCMC PAC, NIHE)
   –   Thailand (PRI, Burapha University)
   –   Yunnan, China (Yunnan CDC)
   –   Started in late 2004
• Variants in
   – Hong Kong (Special Preventive Program)
   – Various Chinese provinces
• Future site
   – Malaysia (Malaysian AIDS Council)
                                33




What has A2 achieved to date?


        Mobilization
                                                                                        34


     Put prevention back on the map in Thailand
• Working with Director General DDC
  & MOPH                                      Proportion new infections by population
   – Thailand adopted and announced                         in Thailand

     goal of cutting new infections in half   1.0
     by 2010
                                              0.8
   – Office of Health Security allocated
     500M Baht                                0.6
• Organizing prevention consortia for         0.4
  key populations as input to National
  AIDS plan                                   0.2

• Supporting provincial planning in           0.0
  Chonburi                                          1988 1990 1995 2000 2005 2010
   – Responds to decentralization of            IDU                 MSM
     plans & budgets                            Client from FSW     Husband from wife
   – Mobilized local funds for AIDS             FSW from client     Wife from husband
                                                Casual sex
                                                                                                                        35


        Sized up the situation in Bangladesh
                                                                  Needle sharing by program exposure




                                 Percent shared passively last
• First size estimates for at-                                   70
                                                                 60
                                                                 50
  risk populations                                               40




                                             time
                                                                 30
• Provided donor inputs on                                       20
                                                                 10
  need for focus                                                  0
                                                                      Central-A      Northwest-A Northwest-B
   – USAID, DFID, World Bank                                     Exposed to NEP      Not exposed        Overall

• Promoted scale up of FSW                                        Last time FSW condom use by program
                                                                                exposure
  & MSM programs



                                 Percent used condom last
                                                                 60
                                                                 50
                                                                 40




                                           time
                                                                 30
                                                                 20
                                                                 10
                                                                  0
                                                                      Brothel-     Street        Street      Hotel
                                                                      National    Central-A   Southeast-A   Central-A

                                                           Intervention exposed         Not exposed         Overall
                                        36


        Gave prevention focus in HCMC
• Adapted National Action
  Plans for comprehensive
  focus on at-risk pops
• Formally recognized MSM
  as at-risk pop
• Expanded budget for at-risk
  pops
   – 1.4B VND in 2005
   – 21.9B VND in 2007
• Expanding to Haiphong
                                                  37


      Expanded local to national level in China

• Informed local planning in Yunnan
  & Guangxi
    – MSM focus added
    – Resource gaps assessed
• YN/GX as resource for other
  provinces & NCAIDS
• Adapted tools for China
    – Chinese GOALS
    – AEM now being applied for
      national projections
• Providing input to national
  planning
    – GFATM consolidation
                                                                                                                                                                 38

    Mobilized MSM responses in Hong Kong
                                                                                            Number of MSM Living with HIV in Hong Kong

                                                                               16,000



• Data showed HIV growing




                                           Number of men HIV+
                                                                               12,000



  rapidly among MSM                                                             8,000




• Prepared scenarios on what
                                                                                4,000


                                                                                        0

  was possible & costs of




                                                                                    85


                                                                                                90


                                                                                                        95


                                                                                                                  00


                                                                                                                         05


                                                                                                                                 10


                                                                                                                                          15


                                                                                                                                                  20
                                                                                 19


                                                                                               19


                                                                                                      19


                                                                                                                20


                                                                                                                       20


                                                                                                                               20


                                                                                                                                         20


                                                                                                                                                20
  inaction                                                                                           MSM Non-intervened        MSM Intervened



• Met with MSM community,               Impact of effective MSM prevention
  policy makers, AIDS Trust                                                    1400

                                                                               1200

  Fund




                                               Millions of Hong Kong dollars
                                                                               1000


• Results                                                                       800

                                                                                600

   – $9M HKD special fund                                                       400


   – Community mobilization                                                     200



   – Prioritization of MSM prevention                                               0




                                                                                 05




                                                                                                             10




                                                                                                                                    15




                                                                                                                                                            20
                                                                               20




                                                                                                           20




                                                                                                                                  20




                                                                                                                                                           20
                                                                                                     Annual additional cost   Cumulative additional cost



                                        Comparison of ART costs with and
                                        without effective MSM prevention
                                39




What has A2 achieved to date?


  Policy change & resource
   mobilization in real time
                                         40




What are some of the issues arising in
            the field?

              Mutations
                                                41


             Staffing and sustainability
• Process requires dedicated staff
  – Often difficult to get full time people
  – Sometimes responsibilities divided
• No natural locus in some places
  – Limited capacity of organizations
  – Requires engaging multiple skills sets
• Requires resources and funding
  – Staff salaries, travel and training, etc.
• Best analysts not always best advocates
  – Requires a strong team approach
                                                       42


       Devolution and decentralization

• In many countries, responsibility is or is
  becoming local (provincial or state)
  – Sometimes limited capacities for analysis
  – Less data available
  – Resource decisions made at this level
• A2 process in current form challenging
  – Efforts underway in Chonburi & Haiphong to learn
    to work at this level
                                                          43


                     Time and tools
• Process is time consuming
  – Data collection & analysis is the bulk
  – Only true the first time
• Tools and analysis require substantial training
  – Addressed through regional trainings
  – Need for more comprehensive guidelines and training
    materials
• Tools need simplification and automation
  – Will increase ability to use locally
  – Try to reduce data input needs
                                                       44


          Ownership and obfuscation

• Data sometimes cannot be used in public
  – “Data is power”
• Transparency can be a problem
  – Hard to justify conclusions if data inaccessible
  – Advocacy needs to be done “behind closed doors”
• Provincial-national linkages can be problematic
                                                                  45


      Partnerships, planning and prioritization

• Current M&E systems weak
   – Lots of indicator collection, little synthesis & analysis
   – Sustained data quality issues
• A2 can fill gap in analyzing/improving data systems and
  responses, but…
   – Need access to data being collected in such systems
   – Need to form bidirectional links to ongoing M&E
• Being holistic, A2 offers real evaluation of responses
• Need closer collaboration with..
   – National programs, strategic planners, communities, donors
                                                       46


       The benefits of the A2 process
• Information from multiple sources collated and
  analyzed for first time
• Data gaps identified and filled
• Better evaluations of program effectiveness: past,
  present and future
• More informed decision makers
  – Better decisions
  – Increased resource mobilization
• Empowered communities
                                            47




And ultimately, as understanding grows ….


         Stronger responses,
            fewer infections, and
               better care
          48




The end

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:9
posted:9/20/2011
language:English
pages:48