The Joint United Nations Program on HIV/AIDS (UNAIDS) estimated that providing antiretroviral (ARV) prophylaxis to pregnant women living with HIV for the prevention of mother-to-child transmission (PMTCT) probably averted about 600 000 infant infections between 1995 and 2011 . In 2010, the WHO recommended two PMTCT prophylaxis options, Option A and Option B, for pregnant women not eligible to receive antiretroviral therapy (triple ARV regimen; ART) for their own health, that is, who do not have a CD4+ cell count of less than 350 cells/μl, active tuberculosis (TB) disease, or co-infection with hepatitis B virus . Option A recommends maternal AZT (zidovudine) prophylaxis beginning from 14 weeks of gestation continued through pregnancy and 7 days postpartum and daily NVP (nevirapine) for the infant until 1 week after the cessation of breastfeeding. Option B recommends triple ARVs for prophylaxis from 14 weeks of gestation continued through pregnancy and until 1 week after all infant exposure to breast milk has ended .
Although the role of ARVs in preventing mother-to-child transmission and in improving health benefits of HIV-infected persons has been well established, only recently, studies provide evidence of ART preventing sexual transmissions [4,5]. In light of this evidence, there have been proposals to increase ART eligibility, and, because limited resources constrain universal expansion, there have been suggestions for incremental expansion of eligibility, such as all pregnant women and discordant couples irrespective of CD4+ cell count, raising the CD4+ cell eligibility threshold to less than 500 cells/μl, or all persons with increased risk of HIV infection .
In Malawi, the lack of capacity to conduct CD4+ cell count testing for all HIV-positive pregnant women (to determine eligibility for treatment or prophylaxis) and concerns about health of women who frequently start and stop ARVs due to frequent pregnancies led to the decision to offer ART treatment for life to all HIV-positive pregnant women . This approach is called Option B+ . It has the advantage of being easier to implement and has potential benefits for the long-term health of the mother, but requires additional resources to maintain women with high CD4+ cell counts on treatment.
As national programs consider Option B+ and other treatment eligibility options, they will need to know the costs and benefits. Cost–effectiveness analysis of Option B compared with Option A or no intervention has been conducted for generalized epidemics in low-income and middle-income countries [9–11]. However, only two studies present cost–effectiveness of Option B+, one for Malawi and the other for Zimbabwe [12,13]. Both studies showed reduced child infections, improved health benefits to the mother, and higher costs under Option B+ and concluded that Option B+ is cost-effective in the long-term in reducing child infections and improving health benefits. The cost–effectiveness of Option B+ over Options A and B could be region-specific and depends on factors such as crude birth rate, median age at infection, and the number of births during HIV infection. In this study, we estimate costs and child infections averted in each of Options A, B, and B+ for four countries, Kenya, South Africa, Zambia, and Vietnam that differed in total fertility rate, median age at first birth, spacing of births, and breast feeding patterns, but with similar median age at infection, presenting different scenarios of ART discontinuation patterns in Options A and B. Unlike previous studies, we also include the effect of Option B and B+ on adult sexual transmission of HIV and estimate cost–effectiveness as the net costs per child and adult infection averted.
We developed a simple Excel-based deterministic model that simulates the experiences of a typical woman in each setting starting from age 15 through to age 49, assuming the woman will be infected at age equal to the median age at infection in the country and will survive through age 49 (Fig. 1). The model simulates the woman's status (HIV-infected or not, pregnant or not, breastfeeding or not) for each month, based on the median age at HIV infection, the total fertility rate, the median age at first birth, the median birth interval, and breastfeeding patterns (Table 1) [14–19]. A woman infected with HIV starts ART when her CD4+ cell count drops below the eligibility threshold (CD4+ cell <350 or <500 cells/μl). The median time from infection to a CD4+ cell count of less than 500 cells/μl varies from 19 to 69 months depending on age at infection and the time from infection to CD4+ cell count less than 350 cells/μl varies from 32 to 98 months . A pregnant woman who is infected with HIV, but not eligible for ART for her own health, initiates one of the three PMTCT treatment options, Option A, Option B, or Option B+. At every month of infection when a woman is pregnant or breastfeeding, there is a probability of transmission of HIV to the child depending on the prophylaxis regimen (Table 2) [12,20–23].
Assumptions on fertility and breastfeeding patterns
On the basis of the total fertility rates in year 2010, we assumed that, between the ages of 15 and 49, each woman in Kenya, South Africa, Zambia, and Vietnam would give birth to five, two, six, and two children, respectively (Fig. 2) . We assumed the time between pregnancies as 31, 40, 33, and 29 months in the four countries, respectively . We assumed breastfeeding for a maximum period of 3 years after birth, with the monthly probability of breastfeeding varying by country and decreasing with time [16–19]. The median durations of breastfeeding in the four countries are about 22, 16, 20, and 18 months, respectively. For mothers who are HIV-positive, because there is a chance of HIV transmission from the mother to the child, the monthly probabilities of breastfeeding were adjusted to include the monthly probabilities of child survival .
Assumptions on prevention of mother-to-child transmission treatment regimen and costs per pregnancy
Under all PMTCT options, we assumed that, at every pregnancy until HIV diagnosis, a woman would be first tested for HIV at an antenatal clinic at a cost of $3.5 per test. For an HIV-infected woman under Options A and B, every pregnancy when not on ART, we assumed one CD4+ cell test to determine eligibility for ART for her own health, at a cost of $20 . For Option A, for the mother, we assumed twice-daily AZT prophylaxis during pregnancy starting from 14 weeks gestation, single-dose NVP at onset of labor, and twice-daily AZT+3TC [AZT+lamivudine (3TC)] for 7 days postpartum at a total cost of $45. We also assumed once-daily NVP for the child starting from birth through 1 week after end of breastfeeding at a cost of $0.008 per day [8,21]. For Option B, we assumed once-daily triple ARV prophylaxis of AZT/3TC/efavirenz (EFV) for the mother, one of the regimens recommended by WHO, starting from 14 weeks gestation through 1 week after end of breastfeeding at a cost of $55 for the period of pregnancy and a daily cost of $0.313 for the duration of breastfeeding. We assumed once-daily NVP prophylaxis for the infant starting from birth through 6 weeks of age at a total cost of $0.336. Under Options A and B, we assumed a woman who became eligible for ART for her own health would use AZT/3TC/EFV at a cost of $0.313 per day. For Option B+, we assumed a once-daily ART regimen of AZT/3TC/EFV starting from diagnosis and continuing through age 49 at a cost of $0.313 per day or $114 per year. For the child, we assumed once-daily NVP prophylaxis starting from birth through 6 weeks of age at a total cost of $0.336. The costs have been summarized in Table 2.
Assumptions on mother-to-child transmissions and cost per infected child
Under Options A and B, we simulated transmissions from mother to child by assuming a 2% chance of perinatal transmission and 0.2% chance of transmission per month during breastfeeding under both infant prophylaxis (Option A) and maternal prophylaxis (Option B). Under Option B+, we used the same transmission rates if the mother started ART during the current pregnancy and lower rates (perinatal transmission of 0.5% and breastfeeding transmission of 0.16% per month) if the mother had started ART with a previous pregnancy (Table 2) . For children infected with HIV, WHO recommends an LPV/r (lopinavir/ritonavir) or NPV-based regimen, the type of regimen determined based on prior exposure to these drugs. In the simulation, we assumed a discounted cost per infected child of $4785, which was estimated by averaging costs over all regimens weighted by mortality for up to 30 years of life of the child, discounting at 3%, and assuming ART coverage in children to be 80% .
Assumptions on sexual transmissions to serodiscordant partner and related costs
ART regimens, in addition to averting mother-to-child transmissions, also reduce sexual transmissions by about 90–96% [4,5]. We included this benefit by estimating the additional infections averted from preventing sexual transmissions from infected mothers to serodiscordant partners due to the additional years of triple ART prophylaxis under Options B and B+. On the basis of simulation models of the epidemics in each country, we determined that each person-year of ART would avert about 0.17 infections in Kenya, 0.2 infections in Zambia, 0.11 infections in South Africa, and 0.1 infections in Vietnam . We also assumed an average discounted lifetime cost saving of $8400 per infection averted. (This figure is calculated using the 2009 prices of ARVs for lower-middle income countries of $141 for first line and $1378 for second line  and assuming that prices for second-line ARVs decline to $930 by 2015. The cost of diagnostics and monitoring tests is $180 per patient per year and the service delivery costs are $102 per patient per year . Need for treatment begins 8 years after infection and survival on first-line ART is assumed to be 86% for the first year on first-line therapy and 90% for each subsequent year . Those failing first-line treatment are assumed to get second line with a survival of 90% per year. With these assumptions, a typical patient survives for about 6 years on first-line treatment and 7 years on second line. All costs are discounted at 3% per year to the time of infection. For more information, see the interactive ART costs calculator, which can be accessed through the ‘Policy Tools’ link on the Futures Institute website, http://www.FuturesInstitute.org.)
Assuming an ART eligibility threshold of CD4+ cell count less than 350 cells/μl, we estimated the duration of time on prophylaxis for PMTCT before becoming eligible for ART, the total cost of the PMTCT option including HIV and CD4+ cell tests, prophylaxis drugs, and ART drugs for mother and child, and the number of children infected. We also estimated the number of children infected and the costs of ART for mother and child under a ‘No PMTCT’ option, that is, no intervention for PMTCT, but women initiate ART when they reach the ART eligibility threshold. We estimated incremental cost–effectiveness ratios (ICERs) as the incremental cost per infant infection averted in pair-wise comparisons of a PMTCT option with the next most effective option (in averting child infections), eliminated strongly dominated options, and re-evaluated ICERs among the remaining options. An option is strongly dominated if it costs more and averts fewer infections than another option. Further, we included the number of sexual transmissions and costs averted from the additional years of ART under Options B and B+ and re-estimated the ICERs. Under the assumption that ART eligibility could likely expand in the future, we re-evaluated the above measures for an eligibility threshold of CD4+ cell count less than 500 cells/μl. All results are presented as values per woman and costs are in 2011 US dollars.
If the ART eligibility threshold is CD4+ cell count less than 350 cells/μl, the durations of time a pregnant women is on prophylaxis in the four countries varies between 12 and 20 months in Option A, 42 and 83 months in Option B, and 90 and 98 months in Option B+ (Table 3). If ART eligibility is expanded to CD4+ cell count less than 500 cells/μl, the prophylaxis durations are between 12 and 14 months in Option A, 35 and 59 months in Option B, and 50 and 69 months in Option B+ (Table 4). Therefore, before becoming eligible for ART under CD4+ cell count less than 350 cells/μl criteria, compared with Option B wherein women discontinued ARVs between pregnancies, mothers on Option B+ are on ART for an additional duration of 15, 21, 48, and 43 months, in Kenya, Zambia, South Africa, and Vietnam, respectively. Equivalent numbers if ART eligibility is CD4+ cell count less than 500 cells/μl are 10, 14, 23, and 15 months, respectively.
On average, in the duration of time between becoming infected with HIV and CD4+ cell count dropping to less than 350 cells/μl, each woman in Kenya, Zambia, South Africa, and Vietnam has four, three, two, and two, pregnancies, respectively. In the duration of time between becoming infected and CD4+ cell count dropping to less than 500 cells/μl, each woman has three, two, one, and one pregnancies in the four countries, respectively. Considering the above number of births and thus spacing between two consecutive pregnancies, we estimate that, Option B, where women not eligible for ART for their own health are on ARVs only during pregnancy and breastfeeding, leads to women discontinuing ARVs for only about three, seven, twenty-three, and twenty months in Kenya, Zambia, South Africa, and Vietnam, respectively. The discontinuations are longer in South Africa and Vietnam because of lower fertility rate and wider birth intervals compared with those in Kenya and Zambia.
Option B+ averted the most number of transmissions, both mother-to-child and sexual transmissions to serodiscordant partners (Tables 3 and 4). If ART eligibility is set at CD4+ cell count less than 350 cells/μl, compared with Options A and B, Option B+ results in about 32% fewer infant infections in Kenya, 27% fewer in Zambia, 29% fewer in South Africa, and 26% fewer infant infections in Vietnam. If ART eligibility is expanded to CD4+ cell count less than 500 cells/μl, because the duration on ART increases in Options A and B, the percentage reductions in infant infections in Option B+ will be lower at about 24, 26, 22, and 23% in the four countries.
Option B+ averts the largest number of total transmissions (mother-to-child and sexual), between 1.13 and 2.24 if ART eligibility is at CD4+ cell count less than 350 cells/μl or between 0.76 and 1.57 if eligibility is expanded to CD4+ cell count less than 500 cells/μl (Tables 3 and 4). If ART eligibility is at CD4+ cell count less than 350 cells/μl, compared with Option B, Option B+ averts 16, 22, 65, and 52% more transmissions in Kenya, Zambia, South Africa, and Vietnam, respectively. The impact of Option B+ in comparison with Option B is the highest in women in South Africa and Vietnam because, compared with women in Kenya and Zambia, their fertility rates are lower and birth intervals are wider, both of which, as reported earlier, lead to longer periods of ART discontinuation in Option B. For the same reasons, if ART eligibility is expanded to CD4+ cell count less than 500 cells/μl, the percentage increase in total transmissions averted in Option B+ compared with Option B will be about the same in Kenya (16%) and Zambia (22%), but decrease to 35 and 24% in South Africa and Vietnam, respectively.
The total PMTCT-related costs per woman, including costs for HIV and CD4+ cell tests, prophylaxis drugs for mother and child, ART costs for mothers’ own health through age 49, and discounted life-time treatment costs for HIV-infected children, are higher in Option B compared with Option A. The costs are between $2700 and $3700 in Option B and $2400 and $3200 in Option A if ART eligibility is at CD4+ cell count less than 350 cells/μl and, if eligibility is expanded to CD4+ cell count less than 500 cells/μl, the costs are between $3000 and $3700 in Option B and $2700–$3300 in Option A (Tables 3 and 4). As expected, Option B+ costs the same under both ART eligibility thresholds, between $3000 and $3600. In Kenya and Zambia, Option B+ costs less than Option B (because the higher effectiveness avoids ART costs for infected children). In South Africa and Vietnam, Option B+ costs a little more than Option B if ART eligibility is at CD4+ cell count less than 350 cells/μl, but costs approximately the same if ART eligibility is expanded to CD4+ cell count less than 500 cells/μl.
The discounted life-time ART costs saved from averting sexual transmissions during the additional years of prophylaxis ART in Options B and B+ are between $2700 and $11 000 in the four countries, if ART eligibility is CD4+ cell count less than 350 cells/μl and between $2000 and $7700, if eligibility is expanded (Tables 3 and 4). When these discounted life-time ART costs saved in Options B and B+ are included, if ART eligibility is CD4+ cell count less than 350 cells/μl, the net cost of Option B is about $170 in South Africa, $60 in Vietnam, and negative in Zambia and Kenya. These Option B costs are much lower than the costs of Option A, whose net costs are the same as their total PMTCT costs because the prophylaxis drugs under Option A are assumed to have no effect on sexual transmissions. Thus, when treatment costs averted are included, Option B costs less than Option A. Option B+ had negative net costs in all four countries (Fig. 3), implying that Option B+ would cost less than Options A and B. Under expanded ART eligibility, these differences in costs across the three PMTCT options were impacted by the additional cost of ART drugs in Options A and B, and the reduced difference in sexual transmissions averted in Options B and B+ compared with Option A (Fig. 4).
Considering incremental PMTCT costs per child infection averted, the No PMTCT option has higher costs than Option A and averts zero infections in children, and Option B has higher PMTCT-related costs, but averts the same number of infant infections compared with Option A. Eliminating the strongly dominated ‘No PMTCT’ and Option B from subsequent analysis and re-estimating cost–effectiveness indicates that, if ART eligibility is CD4+ cell count less than 350 cells/μl, compared with Option A, Option B+ would incur an additional cost of about $6000, $6800, $23 000, and $21 500 per infant infection averted in Kenya, Zambia, South Africa, and Vietnam, respectively. If ART eligibility is expanded, compared with Option A, Option B+ would incur an additional cost of about $6300, $7200, $13 500, and $9800 per infant infection averted in the four countries.
Considering total costs per total infections (mother-to-child and sexual transmissions) averted, No PMTCT, Option A, and Option B are all strongly dominating, costing more but being less effective than another option. Option B+ is cost saving under both ART eligibility thresholds and in all four countries.
Estimates indicate that, in all four countries, transmissions from mother to child are lower in Option B+ compared with Option B because of lower risks of transmission when the mother is on ART before the initiation of pregnancy. In Kenya and Zambia, because of high fertility rates and short birth intervals, the additional time on maternal ART prophylaxis in Option B+ compared with Option B is very short. These additional costs of ART are outweighed by its benefits in averting infant infections and future costs. In South Africa and Vietnam, the PMTCT-related costs of Option B+ is only a little higher than that of Option B if ART eligibility threshold is CD4+ cell count less than 350 cells/μl, but is approximately the same amount if eligibility is expanded to CD4+ cell count less than 500 cells/μl. In all four countries, under both ART eligibility criteria, because the risk of infant transmissions is alike in both Options A and B, they result in the same number of infant infections. However, the PMTCT-related costs of Option B are higher than Option A because of higher drug costs. Option B+ is the most cost-effective strategy in all four countries, in agreement with literature estimates for Zimbabwe and Malawi, whose estimates indicate cost–effectiveness of Option B+ in preventing infant transmissions and increasing health benefits for the mother [12,13].
Options B and B+ averted a considerable number of sexual transmissions. Estimates of net costs indicate that, in the long-term, Option B is likely to be less costly compared with Option A, and Option B+ is likely to be less costly than Option B. Cost–effectiveness estimates indicate Option B+ will be cost saving in all four countries and under both ART eligibility thresholds.
The analysis is subject to limitations. We simulated the life of a woman only from age 15 to age 49 and assumed 100% survival through age 49 so that transmissions can be estimated over the same period of time under all three PMTCT options. We did not consider the possible change in mortalities or health benefits for the mother under the different PMTCT options or the challenges in taking lifelong ART under Option B+. We did not consider any possible differences in costs of ART from changes in survival due to ART under Option B. We assumed the duration of time from initiation of infection to CD4+ cell count dropping to ART eligibility threshold to be the same under both Options A and B. The analysis is based on a simple deterministic model that did not include detailed disease progression of HIV or transmissions. However, we believe, simulating women in four different countries with considerably different characteristics that create different patterns in durations of infection, pregnancy, and exposure to transmissions, provides insight into the additional costs and effectiveness of Option B+ under different settings. Although we costed AZT/3TC/EFV for the ART regimen, the analysis is essentially unchanged with the new preferred 1st-line regimen of TDF/3TC/EFV, recommended in the new 2013 guidelines, released after this analysis was done .
The benefits of Option B+ in averting infant infections have been well studied. This analysis confirms that Option B+ is the most cost-effective strategy. Support in favor of Option B+ is further reinforced from the consistency in results across countries with diverse characteristics sensitive to cost–effectiveness, including fertility rates ranging between 2 and 6, age at first birth between 20 and 27 years, and median interval between births between 30 and 40 months, which encompasses values of majority of countries. Apart from these benefits in averting infant and adult infections and being cost-effective, Option B + could be an effective strategy, for both implementation and intervention, as it would eliminate the need for timely monitoring to identify women becoming eligible for ART, especially in settings with limited access to CD4+ cell testing. Intervention through timely treatment improves the life of HIV-infected women.
The authors would like to thank Ellen McRobie from Imperial College London for coordinating the submission of their article to this supplement.
J.S. and N.S. conceived the concept and directed the study; C.G., and J.S. performed the analysis and modeling; C.G., J.S., N.S., and M.M. contributed to the interpretation of results and writing.
Conflicts of interest
The study was funded by a grant from the Bill and Melinda Gates Foundation. The funder had no input to the analysis or description of results.
4. Donnell D, Baeten JM, Kiarie J, Thomas KK, Stevens W, Cohen CR, et al. Heterosexual HIV-1 transmission after initiation of antiretroviral therapy: a prospective cohort analysis
5. Cohen MS, Chen YQ, McCauley M, Gamble T, Hosseinipour MC, Kumarasamy N, et al. Prevention of HIV-1 infection with early antiretroviral therapy
. N Engl J Med
6. Delva W, Eaton JW, Meng F, Fraser C, White RG, Vickerman P, et al. HIV treatment as prevention: optimising the impact of expanded HIV treatment programmes
. PLoS Med
7. Schouten EJ, Jahn A, Midiani D, Makombe SD, Mnthambala A, Chirwa Z, et al. Prevention of mother-to-child transmission of HIV and the health-related Millennium Development Goals: time for a public health approach
8. World Health Organization. Use of antiretroviral drugs for treating pregnant women and preventing HIV infection in infants Programmatic update, 2012. [Accessed 20 February 2013].
9. Shah M, Johns B, Abimiku A, Walker DG. Cost-effectiveness of new WHO recommendations for prevention of mother-to-child transmission of HIV in a resource-limited setting
10. Johri M, Ako-Arrey D. The cost-effectiveness of preventing mother-to-child transmission of HIV in low- and middle-income countries: systematic review
. Cost Eff Resour Alloc
11. Binagwaho A, Pegurri E, Drobac PC, Mugwaneza P, Stulac SN, Wagner CM, et al. Prevention of mother-to-child transmission of HIV: cost-effectiveness of antiretroviral regimens and feeding options in Rwanda
. PLoS One
12. Fasawe O, Avila C, Shaffer N, Schouten E, Chimbwandira F, Hoos D, et al. Cost-effectiveness analysis of Option B+ for HIV prevention and treatment of mothers and children in Malawi
. PLoS One
13. Ciaranello AL, Perez F, Engelsmann B, Walensky RP, Mushavi A, Rusibamayila A, et al. Cost-effectiveness of World Health Organization 2010 guidelines for prevention of mother-to-child HIV transmission in Zimbabwe
. Clin Infect Dis
15. Rutstein SO, Shea O. Trends in birth spacing. DHS Comparative Reports No. 28
. Calverton, Maryland, USA:ICF Macro; 2011.
16. Kenya National Bureau of Statistics (KNBS) and ICF MacroKenya Demographic and Health Survey 2008–09
. Calverton, Maryland:KNBS and ICF Macro; 2010.
17. Central Statistical Office (CSO), Ministry of Health (MOH), Tropical Diseases Research Centre (TDRC), University of Zambia, and Macro International Inc.Zambia Demographic and Health Survey 2007
. Calverton, Maryland, USA:CSO and Macro International Inc; 2009.
18. Department of Health, Medical Research Council, OrcMacroSouth Africa Demographic and Health Survey 2003
. Pretoria:Department of Health; 2007.
19. Committee for Population, Family and Children [Vietnam], and ORC MacroVietnam Demographic and Health Survey 2002
. Calverton, Maryland, USA:Committee for Population, Family and Children and ORC Macro; 2003.
20. Stover J, Brown T, Marston M. Updates to the Spectrum/Estimation and Projection Package (EPP) model to estimate HIV trends for adults and children
. Sex Transm Infect
2012; 88 (Suppl 2):i11–i16.
22. Rollins N, Mahy M, Becquet R, Kuhn L, Creek T, Mofenson L. Estimates of peripartum and postnatal mother-to-child transmission probabilities of HIV for use in Spectrum and other population-based models
. Sex Transm Infect
2012; 88 (Suppl 2):i44–i51.
23. CHAI, PMTCT and Pediatrics Impact & Cost Model. International AIDS Society Conference; July 2011, Rome, Italy.
24. Eaton JW, Menzies NA, Stover J, Cambiano V, Chindelevitch L, Cori A, et al. Health benefits, costs, and cost-effectiveness of earlier eligibility for adult antiretroviral therapy and expanded treatment coverage: a combined analysis of 12 mathematical models
. Lancet Glob Health
25. WHO, Towards Universal Access: Scaling up priority HIV/AIDS interventions in the health sector, Progress Report 2010, WHO, UNAIDS, UNICEF.
26. Stover J, Bollinger L, Avila C. Estimating the impact and cost of the WHO 2010 Guidelines for Antiretroviral Therapy, AIDS Research and Treatment Vol 2011, Article ID 738271. .
27. Mahy M, Lewden C, Brinkhof MWG, Dabis F, Tassie JM, Souteyrand Y, Stover J. Derivation of parameters used in Spectrum for eligibility for antiretroviral therapy and survival on antiretroviral therapy
. Sex Transm Infect
2010; 86:ii28–ii34. .
© 2014 Lippincott Williams & Wilkins, Inc.