Table 1.

Study design and setting overview.

Reference	Study design	Setting	Population	Time horizon	HIV prevalencea	Perspectiveb	Intervention description
VMMC
Binagwaho et al. (2010) [15]	Deterministic compartmental simulation	Rwanda	0-49 yoc, male population	Lifetime	2.7%	Health care payer	Scale-up of VMMC to infants, adolescents, and adults
Njeuhmeli et al. (2011) [16]	Deterministic compartmental simulation	Sub-Saharan Africa	15-49 yo, general population	Lifetime	4.8%	Health care payer	Scale-up of VMMC
Uthman et al. (2011) [17]	Probabilistic decision analysis	Sub-Saharan Africa	15 + yo, male population	Lifetime	5.5%	Health care payer	Uptake of VMMC
Duffy et al. (2013) [18]	Cross-sectional descriptive cost-analysis	Uganda	18 yo and older, male population	Lifetime	5.9%	Health care payer	PrePex device for VMMC
Menon et al. (2014) [19]	Impact analysis	Tanzania	10-49 yo, male population	Lifetime	4.5%	Health care payer	Scale-up of VMMC
Awad et al. (2015) [20]	Deterministic compartmental simulation	Zimbabwe	10-49 yo, male population	15 years	13.3%	Health care payer	Prioritisation of VMMC subpopulations by age, geographic location, sexual risk profile
Awad et al. (2015) [21]	Deterministic compartmental simulation	Zambia	10-49 yo, male population	15 yearsc	11.5%	Health care payer	Prioritisation of VMMC subpopulations by age, geographic location, sexual risk profile
Haacker et al. (2016) [22]	Deterministic compartmental simulation	South Africa	15-59, male population	Lifetime	18.8%	Health care payer	Age prioritised VMMC scale up
Kripke et al. (2016) [23]	Deterministic compartmental simulation	Malawi	10 + yo; male population	15 years	9.6%	Health care payer	Age prioritised VMMC scale up
Kripke et al. (2016) [24]	Deterministic compartmental simulation	Zimbabwe	20-29 yo; male population	15 years	13.3%	Health care payer	Age prioritised VMMC scale up
Kripke et al. (2016) [25]	Deterministic compartmental simulation	Sub-Saharan Africa	10-49 yo; male population	15 years	4.8%	Health care payer	Age prioritised VMMC scale up
Kripke et al. (2016) [26]	Deterministic compartmental simulation	Eswatini	10-49 yo; male population	15 years	27.4%	Health care payer	Age prioritised VMMC scale up
Kripke et al. (2016) [27]	Deterministic compartmental simulation	Malawi, South Africa, Eswatini, Tanzania, Uganda	10-49 yo; male population	15 years	9.6% (Malawi) 18.8% (South Africa) 27.4% (Eswatini) 4.5% (Tanzania) 5.9% (Uganda)	Health care payer	Age prioritised VMMC scale up
Njeuhmeli et al. (2016) [28]	Deterministic compartmental simulation	Zimbabwe	Male infants	36 years	13.3%	Health care payer	Early infant male circumcision
PrEP
Pretorius et al. (2010) [29]	Deterministic compartmental simulation	South Africa	15-49 yo, general population	10 years	18.8%	Health care payer	PrEP is scaled up to recruit all uninfected individuals
Hallett et al. (2011) [30]	Microsimulation	South Africa	HIV serodiscordant couples	Lifetime	18.8%	Health care payer	PrEP for uninfected partner in serodiscordant relationships
Cremin et al. (2013) [31]	Deterministic compartmental simulation	KwaZulu-Natal, South Africa	15-54 yo, general population	10 years	27.0% (KZNc)ref	Program	Combination prevention strategies of VMMC, early ART, and PrEP
Nichols et al. (2013) [32]	Deterministic compartmental simulation	Macha, Zambia	12 + yo, general population	10 years	7.7% (Macha)	Health care payer	Prioritisation of PrEP
Verguet et al. (2013) [33]	Deterministic compartmental simulation	Sub-Saharan Africa	15-49 yo, general population	5 years	4.8%	Health care payer	PrEP intervention to pre-existing levels of MC, ART, and condom use
Alistar et al. (2014) [34]	Dynamic compartmental simulation	South Africa	15-49 yo, general population	20 years	18.8%	Health care payer	PrEP is scaled up to recruit all uninfected individuals
Nichols et al. (2014) [35]	Deterministic compartmental simulation	Macha, Zambia	12 + yo, general population	40 years	7.7% (Macha)	Health care payer	Uptake of PrEP and TasP in combination
Cremin et al. (2015) [36]	Deterministic compartmental simulation	Nyanza province, Kenya	General population	5 years	13.9% (Nyanza)	Health care payer	Dynamic interaction between key determinants of PrEP impact and cost-effectiveness
Cremin et al. (2015) [37]	Deterministic compartmental simulation	Gaza province, Mozambique	Adult male mine workers	5 years	30.0% (female) 17.0% (male)	Health care payer	Time-limited PrEP uptake among sexual partners of miners
Ying et al. (2015) [38]	Micro-costing analysis	Uganda	HIV serodiscordant couples	10 years	7.1%	Program	Targeted PrEP for serodiscordant couples
Glaubius et al. (2016) [39]	Deterministic compartmental simulation	South Africa	15-54 yo, general population	1) 10yrs 2) lifetime	18.8%	Societal	Long-acting injective antiretrovirals used for PrEP
Walensky et al. (2016) [40]	Deterministic compartmental simulation	South Africa	18-25 yo, high risk women	5 years	Incidence: 5.0% (high risk women)	Program	Long-acting PrEP
Cremin et al. (2017) [41]	Deterministic compartmental simulation	Nairobi, Kenya	Key populations	10 years	4.8%	Health care payer	PrEP provided to FSW
TasP
Barnighausen et al. (2012) [42]	Discrete time mathematical model	South Africa	15 + yo, general population	10 years	18.8%	Health care payer	Increased coverage of TasP, ART under the current WHO eligibility guidelines, and MMC
Granich et al. (2012) [43]	Deterministic compartmental simulation	South Africa	15 + yo, general population	1) 5 years 2) 40 years	18.8%	Program	Enhanced combination prevention strategy
Smith et al. (2015) [44]	Individual-based simulation modelling study	KwaZulu-Natal, South Africa	18 + yo, general population	10 years	27.0% (KZN)ref	Health care payer	Home HIV counselling and testing
Bershteyn et al. (2016) [45]	Individual-based simulation modelling study	South Africa	General population	20 years	18.8%	Health care payer	Age-targeting outreach with HIV treatment and prevention
Ying et al. (2016) [46]	Dynamic compartmental model	KwaZulu-Natal, South Africa	General population	10 years	27.0% (KZN)ref	Program	Home HIV testing and counselling
PMTCT
Halperin et al. (2009) [47]	Modelling analysis	Sub-Saharan Africa	Pregnant, HIV-infected women	1 year	4.8%	Service delivery	Antiretroviral prophylaxis programs and family planning programs
Nakakeeto et al. (2009) [48]	Forecasting model	Burkina Faso, Cameroon, Cote d’Ivoire, Malawi, Rwanda, Tanzania, and Zambia	HIV-infected women, HIV-exposed infants	8 years	0.8% (Burkina Faso) 3.7% (Cameroon) 2.8% (Cote d’Ivoire) 9.6% (Malawi) 2.7% (Rwanda) 4.5% (Tanzania) 11.5% (Zambia)	Health care payer	PMTCT package including: family planning, HIV testing and counselling, and provision of antiretroviral and cotrimoxazole prophylaxis
Orlando et al. (2010) [49]	Cost-effectiveness analysis	Malawi	Pregnant, HIV-infected women	42 months	16.9% (ANC)	Societal and Private	HAART-based intervention
Robberstad et al. (2010) [50]	Decision analysis	Tanzania	Pregnant, HIV-infected women	18 months	6.6% (ANC)	Health care payer	HAART-based intervention
Shah et al. (2011) [51]	Decision-based analytical model	Nigeria	Pregnant, HIV-infected women	1 year	2.8%	Health care payer	2009 WHO PMTCT guidelines (long-course ART)
Kuznik et al. (2012) [52]	Cost-effectiveness analysis	Uganda	Pregnant, HIV-infected women	19.3 years	7.1%	Health care payer	Combination ART
Binagwaho et al. (2013) [53]	Cost-effectiveness analysis	Rwanda	HIV-infected pregnant women and their infants	Lifetime	2.7%	Health care payer	Dual ARV and short course HAART prophylaxis with breastfeeding or replacement feeding
Fasawe et al. (2013) [54]	Decision analysis	Malawi	Pregnant, HIV-infected women	10 years	16.9% (ANC)	Health care payer	Implementation of Option B +
Maredza et al. (2013) [55]	Cost-effectiveness analysis	South Africa	Pregnant, HIV-infected women	24 months	28.0% (ANC)	Health care payer	HAART-based intervention
Gopalappa et al. (2014) [56]	Deterministic compartmental simulation	Kenya, South Africa, Zambia	15-49 yo, female population	Lifetime	5.9% (Kenya) 18.8% (South Africa) 11.5% (Zambia)	Program	Implementation of Option B +
Ishikawa et al. (2014) [57]	Decision analysis	Zambia	Pregnant, HIV-infected women	18 months	11.5%	Health care payer	Comparison between Option A, Option B, and Option B +
Yu et al. (2014) [58]	Decision analysis	South Africa	Pregnant, HIV-infected women	18 months	28.0% (ANC)	Health care payer	1) tested and treated promptly at any time during pregnancy (promptly treated cohort), 2) no testing or treatment until after delivery and appropriate standard treatments were offered (remedy treated cohort)
Zulliger et al. (2014) [59]	Cost-effectiveness analysis	South Africa	Pregnant, HIV-infected women	1 year	28.0% (ANC)	Health care payer	Expedited initiation onto lifelong ART in pregnant women who met South African ART eligibility criteria
Price et al. (2016) [60]	Decision analysis	Zambia	Pregnant women	Lifetime	11.5%	Health care payer	Daily oral PrEP during pregnancy and breastfeeding
Tweya et al. (2016) [61]	Individual-based simulation modelling study	Malawi	Primigravida women	50 years	16.9% (ANC)	Health care payer	Option B vs. Option B +
Other biomedical
Verguet et al. (2010) [62]	Cost-effectiveness analysis	South Africa	15-49 yo, female population	1 year	26.3% (Female)	Health care payer	Impact of microbicides distributed alongside condoms
Williams et al. (2011) [63]	Dynamic compartmental model	South Africa	General population	20 years	18.8%	Health care payer	Tenofovir gel uptake by sexually active women
Long et al. (2013) [64]	Dynamic compartmental simulation	South Africa	15-49 yo, general population	10 years	18.8%	Health care payer	HIV screening and counselling, ART, VMMC, microbicides
Mbah et al. (2013) [65]	Dynamic compartmental simulation	Zimbabwe	15-49 yo, female population	10 years	13.3%	Health care payer	Praziquantel as a preventive anthelminthic chemotherapy
Terris-Prestholt et al. (2014) [66]	Deterministic compartmental simulation	Gauteng Province, South Africa	15-49 yo, general population + FSW and their partners	15 years	17.6% (Gauteng)	Health care payer	Uptake of tenofovir gel by women
Mvundura et al. (2015) [67]	Impact analysis	Sub-Saharan Africa	15-49 yo, general population	1 year	4.8%	Health care payer	Distribution of 100,000 female condoms
Moodley et al. (2016) [68]	Semi-Markov simulation	South Africa	Adolescents enrolled in school	Lifetime	10.2% (females 15-24) 3.9% (males 15-24)	Health care payer	Hypothetical HIV vaccination provided to adolescent students
Moodley et al. (2016) [69]	Semi-Markov simulation	South Africa	Adolescents girls enrolled in school	Lifetime	10.2% (females 15-24) 3.9% (males 15-24)	Health care payer	National implementation of hypothetical HIV vaccination to adolescents
Wall et al. (2018) [70]	Cost-benefit analysis and cost-effectiveness analysis	Zambia	HIV serodiscordant couples	5 years	11.5%	Donor	Couples’ testing and counselling with TasP for seropositive partner
Behavior change
Enns et al. (2011) [71]	Stochastic network simulation	Eswatini, Tanzania, Uganda, Zambia	15-49 yo, general population	10 years	27.4% (Eswatini) 4.7% (Tanzania) 7.1% (Uganda) 11.5% (Zambia)	Program	Concurrency reduction campaigns focused on behaviour change scenario: 1) increased monogamy, 2) high-risk partnership reduction, 3) untargeted partnership reduction
Structural
Fieno et al. (2014) [72]	Cost simulation	South Africa	Women aged 15-20 yo, bottom quarter of income distribution	6 years	18.8%	Health care payer	Cash transfers
Remme et al. (2014) [73]	Cost-benefit analysis and cost-effectiveness analysis	Malawi	Adolescent girls attending school	18 months	9.6%	Health care payer	Cash transfers
Rutstein et al. (2014) [74]	Decision-tree model	Malawi	15-49 yo, partners of STI clinic indexes	1 year	9.6%	Health care payer	Partner notification

^aWorld Bank 2017 HIV prevalence estimates

^bHealth care payer perspective refers to costs incurred or saved by the governmental healthcare system; Donor perspective refers to costs incurred of saved by international donors; Program and service delivery perspective refers to costs incurred by a stakeholders implementing HIV program; Societal perspective refers to all of society regardless of the payer; Private perspective takes into account the costs incurred by service providers

^cAbbreviations: ANC = antenatal care clinic; ARV = antiretrovirals; ART = antiretroviral therapy; FSW = female sex worker; HAART = highly active antiretroviral therapy; KZN = KwaZulu-Natal, South Africa; MC = male circumcision; MMC = medical male circumcision; PMTCT = prevention of mother-to-child transmission; PrEP = pre-exposure prophylaxis; TasP = treatment as prevention; VMMC = voluntary medical male circumcision; WHO = World Health Organization; yo = years old.