IMPROVING GLOBAL ESTIMATES: PROCESS AND MILESTONES IATT Webinar June 7 th 2016

Transcription

1 Where did we start? Where did we get to? Where do we want to go? IMPROVING GLOBAL ESTIMATES: PROCESS AND MILESTONES IATT Webinar June 7 th 2016 Mar6na Penazzato Paediatric advisor HIV Department WHO, Geneva- Switzerland

2 Where did we start? Paediatric ARV Drug Op:miza:on PADO mee:ng (Dakar, Senegal, October 2013) Ø Challenges in the development and uptake of ARVs for children Ø Mid and long- term priori:es Ø Research priori:es Mobilize communi:es Op:mize drug regimens TREATMENT 2.0 Provide point of care diagnos:cs Reduce costs Adapt delivery systems 2

3 Objectives Estimate the number of children living with HIV by 2020 globally, in countries with generalized epidemic and in the 21 priority countries Characterize the size and age of the population on a yearly basis Estimate the number of children in need of ART by 2020 Explore size and age of the paediatric population in need of ART

4 Assumptions on Coverage reached by 2020 MAX Adult ART coverage 95% PMTCT coverage 95% Paediatric ART coverage 100% Likely (Based on current performance) PMTCT coverage 95% if currently >75% 90% if % if % if <25 Paediatric ART coverage 100%if currently >75% 90% if % if % if <25 STABLE (No improvement) Adult ART coverage stable PMTCT coverage stable Paediatric ART coverage stable All values are mid-year values Projected values for Ethiopia ( ) and Kenya ( ) were calculated applying the percentage change between years for the total of all countries. This is because the country files stopped at ART need for children under 5 based on population living with HIV by single year age group

5 3,500,000 3,000,000 3,193,775 2,836,736 Children living with HIV 2,500,000 2,000,000 1,500,000 1,000, ,000 2,665,191 MAX Adult ART coverage 95% PMTCT coverage 95% Paediatric ART coverage 100% 2,302,510 1,960,718 1,829, LMIC 21 PMTCT Priority Countries Generalized epidemic countries

6 Roadmap to streamline access to Develop accurate forecas:ng of demand for paediatric ARVs Explore new drug delivery systems Accelerate approval of new drugs and new formula:ons Reach patent- sharing agreement for key agents ARVs for children Iden:fy innova:ve financial mechanisms to sustain a diminishing market CollecSve engagement in driving innovasons that meet unique needs of infants and children

7 June 17 th - 18 th 2014 October 28 th - 29 th 2015 Where did we get to?

8 Modelling approach 1. Proposed structure 2. Parameters 3. Data sources

9 Modelling structure Is this reasonable representa:on of the natural history? Does this reflect the programma:c reality Can this match Spectrum?

10 We discussed Do these parameters need to be stra:fied based on (ie mode of transmission, age, regimens,..)? How many strata do we need by parameter? Do we expect these parameters to change in the next 5 to 10 years? In which direc:on? Which value do we aaach to these parameters? Input data vs default value.

11 Decided to maintain and adjust Spectrum 5/15/17 11

12 New Infec:ons Fer:lity differen:al in HIV infected (and on ART) or changing paaerns over :me and concentrated epidemics BF in HIV infected vs general popula:on HIVDR impact transmission rate Adherence to interven:ons Length of first line for pregnant women that may impact transmission

13 Survival Iden:fica:on and ini:a:on by age Survival off ART Survival on ART by age at ini:a:on (adolescents) and :me since ini:a:on Adherence to the interven:on (reten:on/adherence) Failure by line of treatment and :me from ini:a:on

14 Steps required To adjust in SPECTRUM Default assump:on for aari:on on ARVs Add na:onal EID to Valida:on panel of Spectrum Age stra:fica:on at treatment ini:a:on Paeds compartmental model for CD4 CTX effect to be included from ARROW To gather data on BF dura:on Fer:lity on and off ART Early vs late ini:a:on of ART in PW Reten:on during BF Interac:on between reten:on and VL suppression Feasibility of collec:ng start of ARV by weeks of gesta:on. Age stra:fica:on for newly ini:ated children Mar:son dataset to look at CD4 To explore in Spectrum or other models Impact of :me of ARVs in PW HIVDR impact Effect of change in fer:lity on es:mates for concentrated epidemics Tes:ng strategies and impact on outcomes Explore role of BF and maternal health on mortality off ART and on ART.

15 Mean:me a few changes happened. More infec?ons are prevented Policies adopted in 144 LMIC 80% - OpSon B+ New HIV infec:ons among the 21 Global Plan countries dropped below 200,000 in 2014

16 Mean:me a few changes happened. More children are in need for ART Change in policy was uptaken very rapidly resul:ng in an increasing number children living with HIV being eligible for ART.

17 Mean:me a few changes happened. More children age into adolescence 2.1 million adolescents were living with HIV globally in HIV first cause of deaths among adolescents in Africa

18 Mean:me a few changes happened. More guidelines coming soon Treat All! (no more eligibility) Treat sooner (impact on incidence) Introduce beaer sequencing (more emphasis on 2 nd and 3 rd line) Enhance postnatal prophylaxis in high risk babies Breaspeed for longer Expand tes:ng net for children and scale up EID (as early as birth to minimise early mortality) Adapt service delivery models to specific popula:ons and type of pa:ents (ie adolescents) Support ac:vely adherence and reten:on to op:mise programme quality

19 2 nd edi:on: October 28th - 29th 2015 Con:nue improving Spectrum es:mates To measure progress To improve forecas:ng To inform programme planning Objec:ves: 1. Iden:fying relevant changes for Spectrum 2. Outline data gaps to be filled 3. Propose next steps to address crucial parameters or assump:ons. 4. Explore approaches for beaer forecas:ng

20 Steps required To adjust in SPECTRUM Default assump:on for aari:on on ARVs Add na:onal EID to Valida:on panel of Spectrum Age stra:fica:on at treatment ini:a:on Paeds compartmental model for CD4 CTX effect to be included from ARROW Reten:on by age, CD4 and :me on ART Reten:on paaern HIV prevalence by age Fer:lity by CD4 Transmission probabili:es To gather data on BF dura:on Fer:lity on and off ART Early vs late ini:a:on of ART in PW Reten:on during BF Interac:on between reten:on and VL suppression Feasibility of collec:ng start of ARV by weeks of gesta:on. Age stra:fica:on for newly ini:ated children Mar:son dataset to look at CD4 HEUs mortality To explore in Spectrum or other models Impact of :me of ARVs in PW HIVDR impact Effect of change in fer:lity on es:mates for concentrated epidemics Tes:ng strategies and impact on outcomes Explore role of BF and maternal health on mortality off ART and on ART.

21 Where do we want to go? Further refinement of Spectrum to capture beaer some of the programmasc reality (that will depend on availability of data in countries, par:cularly age- disaggregated data) Explore opportuni:es for further improvement of the model structure to promote a more granular approach to modelling treatment response Addi:on of a ForecasSng module to Spectrum: more discussion to explore detailed next steps needed Resource mobilisason to support research needed to collect primary data and to conduct analysis by observa:onal database

22 UNAIDS reference group mee:ng Fall 2016