Skip to main content

Changes

View changes from to


On August 19, 2025, 8:50:08 AM UTC, Gravatar Andreas Jahn:
  • Updated description of resource Spectrum File in Malawi HIV Estimates 2025 from

    _v08 updates (2025-05-21; Tiwonge Chimpandule): * Refit Shiny90 * Rerun uncertainty analysis Updated all program data with complete 2024 outputs. ART program data 2021-2024 were reduced by -2.5%, -4.9%, -5.7% and -7.0% in 2021, 2022, 2023 and 2024 to adjust for systematic overreporting that was discovered in 2023 and corroborated during a targeted DQA at a sample of 54 facilities in Aug 2024. The reduction was based on the difference between ARV actually issued to the clinic rooms at all facilities vs. estimated consumption based on the reported number of patients on ART. Note: some facilities in Blantyre with the most significant suspected overreporting already started to drop "ghost patients" in Q3 and Q4 2024. This is consistent with the declining ART reporting bias in Blantyre estimated by the 2025 Naomi model. Replaced VLS program data 2019-2024 with: Total routine VL samples collected (DBS and plasma) for the denominator. Number of patients with VL<1000 based on extrapolated VLS from ADJUSTED VLS rates using the Kuzama pa Kalondo model for the numerator. The model was developed in 2024/25 to remove the DBS bias and bias from over/underrepresentation of age/sex and district specific VL testing rates. For 2012-2018: used crude LIMS data (all routine VL results from plasma and DBS samples) as Kuzama analysis was not possible for these earlier years due to missing data elements. Changed Advanced options > Child / Adult transition parameters from "East Africa" to "Southern Africa" pattern (following suggestion from Eleanor, UNAIDS). This results in slightly higher AIDS mortality. We are unable to triangulate mortality estimates with empirical sources, so challenging to validate if this is more appropriate for Malawi. Changed Advanced options > HIV-related fertility reductions to the new default parameters (2024).
    to
    _v09 updates (2025-08-19; Tiwonge Chimpandule refitted Shiny90 and uncertainty analysis as these appeared incomplete in v08) Updated all program data with complete 2024 outputs. ART program data 2021-2024 were reduced by -2.5%, -4.9%, -5.7% and -7.0% in 2021, 2022, 2023 and 2024 to adjust for systematic overreporting that was discovered in 2023 and corroborated during a targeted DQA at a sample of 54 facilities in Aug 2024. The reduction was based on the difference between ARV actually issued to the clinic rooms at all facilities vs. estimated consumption based on the reported number of patients on ART. Note: some facilities in Blantyre with the most significant suspected overreporting already started to drop "ghost patients" in Q3 and Q4 2024. This is consistent with the declining ART reporting bias in Blantyre estimated by the 2025 Naomi model. Replaced VLS program data 2019-2024 with: Total routine VL samples collected (DBS and plasma) for the denominator. Number of patients with VL<1000 based on extrapolated VLS from ADJUSTED VLS rates using the Kuzama pa Kalondo model for the numerator. The model was developed in 2024/25 to remove the DBS bias and bias from over/underrepresentation of age/sex and district specific VL testing rates. For 2012-2018: used crude LIMS data (all routine VL results from plasma and DBS samples) as Kuzama analysis was not possible for these earlier years due to missing data elements. Changed Advanced options > Child / Adult transition parameters from "East Africa" to "Southern Africa" pattern (following suggestion from Eleanor, UNAIDS). This results in slightly higher AIDS mortality. We are unable to triangulate mortality estimates with empirical sources, so challenging to validate if this is more appropriate for Malawi. Changed Advanced options > HIV-related fertility reductions to the new default parameters (2024).


  • Uploaded a new file to resource Spectrum File in Malawi HIV Estimates 2025


  • Changed value of field sha256 of resource Spectrum File to e242c38e003b4c0466d532d122235c64987934498b3e3d4618c263e44499933e (previously 33c61e680e78c39f8dfdae1d12f7b8dbb7c154026a12e45fbc8434cb41a3d611) in Malawi HIV Estimates 2025