Publications & Reports

HIV Mathematical Modelling to Support Swaziland's Development of its HIV Investment Case.

Kelly S, Shattock A, Kerr CC, Gama T, Nhlabatsi N, Zagatti G, Harimurti P, Wilson DP, Gorgens M


Background and context:

Swaziland is currently developing its HIV investment case and is thus focusing on which high impact interventions to prioritize. As a starting point to this process, the Government of Swaziland and UNAIDS requested the World Bank to estimate, through mathematical modelling, what the impact of the high impact (‘game changer’) interventions would be. The World Bank contracted the HIV modelling team at UNSW Australia to conduct this work. This report summarizes the outcomes of this modelling effort.

Purpose of the study:

At the request of the Government, this modelling study aimed to assess the impact of moderate or high-level scale-up of the following five interventions, separately and jointly:

  1. Antiretroviral therapy (ART) for people living with HIV;
  2. Voluntary medical male circumcision (VMMC);
  3. Implementation of conditional cash transfers (CCT) for girls and young women aged 15–24 years;
  4. Prevention of mother-to-child transmission (PMTCT) of HIV;
  5. Tuberculosis (TB)/HIV treatment.

Summary of results

a) By 2030 in Swaziland, compared to the situation had current coverage levels been maintained, it is possible to reduce new HIV infections by 27% and AIDS-related deaths by 12%. By jointly implementing all interventions, it was estimated that by 2030 an additional 27,400 new infections and an additional 7,400 AIDS-related deaths could be averted (Table 1 and Figure 1).

b) The highest marginal impact interventions are ART, VMMC, and CCT. Rapid scale-up of ART for people living with HIV, VMMC in males aged 10–49 years, and CCT for girls and young women aged 15–24 years showed the greatest impact on reducing more new HIV infections and AIDS-related deaths at population level by 2030. VMMC alone accounts for 49% of new HIV infections averted by 2030, ART for 27% of all infections averted, and CCT for 36% of new infections averted (Table 25 and Figure 9). ART alone for over half (57%) of AIDS-related deaths averted.

c) It is possible to reduce new HIV infections amongst girls and young women aged 15–24 years by 23% and at the national level by 10% by implementing the CCT program for girls and young women alone. Cash transfers have multiple benefits and have also been shown to positively impact school enrollment rates, preventive healthcare, and household consumption in poor and vulnerable populations (not measured in this modelling effort) [1]. If the CCT program were implemented among 60% of girls and young women aged 15–24 years in Swaziland by 2018, maintained until 2030, it is estimated that there would be 5,100 (23%) fewer new infections in this age group alone, and almost twice as many (9,900) new infections in the total population could be averted. d) PMTCT shows lower marginal impact as the impact of the other game changer interventions – due to already-high coverage of the program – is already highly effective. Scale-up of PMTCT to 90% coverage by 2018, maintained at 90% until 2030, would result in an estimated 1,200 (1.2%) fewer new HIV infections among infants born to HIV-positive mothers by 2030. As a secondary benefit, there would be 200 (0.3%) fewer deaths among HIV-positive mothers, and over time among their children (Table 19 and Figure 7). Since the current PMTCT coverage is already high at 84%, continued funding of this program is required to maintain high coverage and low-levels of new congenital infections. However, decreasing MTCT from 2% to 1%, as well as breastfeeding from 51% to 30% in 2018 and to 20% in 2030, as modeled by the moderate scale-up scenario, will obviously avert even more new infections (1.2%), which has important societal and downstream epidemiological benefits. e) Moderate scale-up of TB/HIV intervention accounts for 35% of AIDS-related deaths averted by 2020 By 2030, scale-up of ART coverage among people co-infected with TB/HIV to 75% by 2015, 85% by 2018, and 90% by 2030 would result in an estimated 1,200 (2%) fewer AIDS-related deaths. Moreover, with increased coverage of TB/HIV co-treatment it is estimated that an additional 3,800 people would be cured of TB by 2030 (Tables 22 and 23; Figure 8).


  • Journal: Report to the World Bank
  • Published: 08/10/2014