Date This publication was produced for review by the United States Agency for International Development. It was prepared by (First author’s First Name, Last Name),



Download 0.58 Mb.
Page18/21
Date conversion06.08.2017
Size0.58 Mb.
1   ...   13   14   15   16   17   18   19   20   21

6.4Using Demographic Data to Estimate Contraceptive Consumption


  • For these reasons, we have conducted the following forecast using a population (or demographic) based methodology and reconciling this with anectodal information from our site assessments and interviews. The 2007-2008 population based forecast was conducted using data from the 2001 Reproductive Health Survey for Azerbaijan, including CPR, method mix, rate of abortion, source mix, TFR and other variables.

  • This forecast also considered assumptions and factors provided through our key informant interviews.

6.5Methodology


  • We conducted this exercise both using a software program called Spectrum18 developed by the Futures Group International. We also conducted this same forecast using a basic manual approach with key family planning stakeholders. This stakeholder meeting provided us with the opportunity to: 1) confirm our reservations about using logistics data as the primary source for this forecast, 2) advocate for the importance of using this evidence based approach in developing forecasts and funding allocations, 3) build capacity in the forecasting methodologies, and 4) learn about many of the assumptions and factors that may influence such a forecast.

  • We developed the following two scenarios and forecasted each of them:

    • Scenario #1: constant indicators from 2001 RHS





    • 2001

    • Pills

    • IUDs

    • condoms

    • VFT

    • MMX

    • 1.80%

    • 11%

    • 5.80%

    • 1.00%

    • Source Mix (public sector)

    • 12.40%

    • 93.80%

    • 3.10%

    • 4%

    • Source: 2001 RHS









  • Given the current situation in the country with stock outs and service delivery limitations, we do not estimate that the CPR would have changed dramatically in the past 5 years. Therefore, our first scenario has kept all information constant. It is important to note that while the population growth rate stays the same, the population increases thereby being the only major dynamic indicator in this scenario.

  • Scenario # 2: increased modern MMX (pills, condoms, spermicides) and increased public sector source mix (pills, spermicides)

  • While we do not envision any changes in the past data, there may be opportunity for the current interventions (i.e., ACQUIRE) to affect demand and access in the immediate future. The likely result would be a very slight increase in modern method mix with a corresponding decrease in traditional. In addition, IF public sector contraceptives are made available, it is likely that there will be a slight uptake in methods from the public sector. For this scenario, we have assumed the following method and source mix changes:

  • Scenario # 2: increase in modern MMX and public sector source mix

  • 2001

  • Pills

  • IUDs

  • condoms

  • VFT

  • MMX

  • 5.40%

  • 11%

  • 7.20%

  • 3.60%

  • Source Mix (public sector)

  • 15.00%

  • 93.80%

  • 3.10%

  • 5%

  • Source: 2001 RHS








6.6Limitations


  • As indicated, for Scenario #1, the data was based on the 2001 RHS for Azerbaijan. This information is very old and much has likely occurred since then that will have affected the family planning program. Scenario #2 was based on 2001 RHS data in combination with key stakeholder conjecture. While changes were conservative, these were essentially guestimates.

  • Similarly, the numbers of users forecasted for some of the methods is extremely low. For example, for scenario #1, approximately 300 public sector spermicide acceptors were forecasted for each year. For condoms and pills, the number of public sector acceptors is between 1000-2000. These numbers are extremely low and therefore are highly sensitive to any kind of change (sensitivity).

  • Because of these limitations, we have only developed a two year forecast for each scenario and strongly recommend that a new forecast be conducted in six months using the 2006 DHS.
1   ...   13   14   15   16   17   18   19   20   21


The database is protected by copyright ©ininet.org 2016
send message

    Main page