Modelling HIV epidemics in the antiretroviral era: the UNAIDS Estimation and Projection package 2009

Abstract

Objective: The UNAIDS Estimation and Projection Package (EPP) is a tool for country-level estimation and short-term projection of HIV/AIDS epidemics based on fitting observed HIV surveillance data on prevalence. This paper describes the adaptations made in EPP 2009, the latest version of this tool, as new issues have arisen in the global response, in particular the global expansion of antiretroviral therapy (ART).

Results: By December 2008 over 4 million people globally were receiving ART, substantially improving their survival. EPP 2009 required modifications to correctly adjust for the effects of ART on incidence and the resulting increases in HIV prevalence in populations with high ART coverage. Because changing incidence is a better indicator of program impact, the 2009 series of UNAIDS tools also focuses on calculating incidence alongside prevalence. Other changes made in EPP 2009 include: an improved procedure, incremental mixture importance sampling, for efficiently generating more accurate uncertainty estimates; provisions to vary the urban/rural population ratios in generalised epidemics over time; introduction of a modified epidemic model that accommodates behaviour change in low incidence settings; and improved procedures for calibrating models. This paper describes these changes in detail, and discusses anticipated future changes in the next version of EPP.

Figure 1

The impact that scaling up to 100% of antiretroviral therapy (ART) need by 2015 would have if begun in 2005 in an advanced epidemic. (A) Prevalence increases as a result of prolonged survival of those on ART, while (B) incidence declines. Note that the UNAIDS Estimation and Projection Package (EPP) estimates incidence using the traditional epidemiological definition of percentage of susceptibles contracting HIV rather than on a whole-population basis.

Figure 2

The compartments in the modified Reference Group model with antiretroviral therapy (ART) used in the UNAIDS Estimation and Projection Package (EPP) 2009. Once they become eligible for ART, those living with HIV can move to untreated or first-line therapy compartments based on the number of treatment slots available. If they fail first-line therapy they can move to second-line therapy.

Figure 3

Example of the Weibull distributions for time (A) from infection to antiretroviral therapy (ART) eligibility and (B) from ART eligibility to death. When combined, they approximate (C) the original Weibull survival function used in previous versions of the UNAIDS Estimation and Projection Package (EPP) and Spectrum. This figure is for a CD4 eligibility threshold of 200 cells/mm³ and normal (11-year survival) progression pattern.

Figure 4

Incidence display on the Results page. The user can select to display new infections, that is, incidence as either (A) the percentage of those susceptible to infection who contracted HIV in the last year or (B) an absolute number of new infections in a year.

Figure 5

Incidence distribution displays. UNAIDS Estimation and Projection Package (EPP) 2009 allows the display of the distribution of incidence among the subpopulations comprising a national projection. The vertical axis shows the number of new infections in thousands.

Figure 6

A comparison of the outcomes of the Bayesian melding routines as implemented in the UNAIDS Estimation and Projection Package (EPP) 2007 (A) and EPP 2009 (B) with the same data set. This illustrates how the incremental mixture importance sampling (IMIS) procedure used in EPP 2009 provides many more unique curves and better uncertainty estimates in similar amounts of computational time.

Figure 7

An example of the impact of applying a ϕ-shift with an incidence constraint to Zimbabwe urban data. The left-hand figure shows the near-zero incidence resulting from a four-parameter Reference Group model fit, while the right-hand side shows the results of allowing ϕ to vary, providing a low, but stable incidence pattern.