Estimating the costs and perceived benefits of oral pre-exposure prophylaxis (PrEP) delivery in ten counties of Kenya: a costing and a contingent valuation study.

Background: Kenya included oral PrEP in the national guidelines as part of combination HIV prevention, and subsequently began providing PrEP to individuals who are at elevated risk of HIV infection in 2017. However, as scale-up continued, there was a recognized gap in knowledge on the cost of delivering oral PrEP. This gap limited the ability of the Government of Kenya to budget for its PrEP scale-up and to evaluate PrEP relative to other HIV prevention strategies. The following study calculated the actual costs of oral PrEP scale-up as it was being delivered in ten counties in Kenya. This costing also allowed for a comparison of various models of service delivery in different geographic regions from the perspective of service providers in Kenya. In addition, the analysis was also conducted to understand factors that indicate why some individuals place a greater value on PrEP than others, using a contingent valuation technique. Methods: Data collection was completed between November 2017 and September 2018. Costing data was collected from 44 Kenyan health facilities, consisting of 23 public facilities, 5 private facilities and 16 drop-in centers (DICEs) through a cross-sectional survey in ten counties. Financial and programmatic data were collected from financial and asset records and through interviewer administered questionnaires. The costs associated with PrEP provision were calculated using an ingredients-based costing approach which involved identification and costing of all the economic inputs (both direct and indirect) used in PrEP service delivery. In addition, a contingent valuation study was conducted at the same 44 facilities to understand factors that reveal why some individuals place a greater value on PrEP than others. Interviews were conducted with 2,258 individuals (1,940 current PrEP clients and 318 non-PrEP clients). A contingent valuation method using a "payment card approach" was used to determine the maximum willingness to pay (WTP) of respondents regarding obtaining access to oral PrEP services. Results: The weighted cost of providing PrEP was $253 per person year, ranging from $217 at health centers to $283 at dispensaries. Drop-in centers (DICEs), which served about two-thirds of the client volume at surveyed facilities, had a unit cost of $276. The unit cost was highest for facilities targeting MSM ($355), while it was lowest for those targeting FSW ($248). The unit cost for facilities targeting AGYW was $323 per person year. The largest percentage of costs were attributable to personnel (58.5%), followed by the cost of drugs, which represented 25% of all costs. The median WTP for PrEP was $2 per month (mean was $4.07 per month). This covers only one-third of the monthly cost of the medication (approximately $6 per month) and less than 10% of the full cost of delivering PrEP ($21 per month). A sizable proportion of current clients (27%) were unwilling to pay anything for PrEP. Certain populations put a higher value on PrEP services, including: FSW and MSM, Muslims, individuals with higher education, persons between the ages of 20 and 35, and households with a higher income and expenditures. Discussion: This is the most recent and comprehensive study on the cost of PrEP delivery in Kenya. These results will be used in determining resource requirements and for resource mobilization to facilitate sustainable PrEP scale-up in Kenya and beyond. This contingent valuation study does have important implications for Kenya's PrEP program. First, it indicates that some populations are more motivated to adopt oral PrEP, as indicated by their higher WTP for the service. MSM and FSW, for example, placed a higher value on PrEP than AGYW. Higher educated individuals, in turn, put a much higher value on PrEP than those with less education (which may also reflect the higher "ability to pay" among those with more education). This suggests that any attempt to increase demand or improve PrEP continuation should consider these differences in client populations. Cost recovery from existing PrEP clients would have potentially negative consequences for uptake and continuation.

Implementation and resource needs for long-acting PrEP in low- and middle-income countries: a scoping review.

INTRODUCTION: Several low- and middle-income countries (LMICs) are preparing to introduce long-acting pre-exposure prophylaxis (LAP). Amid multiple pre-exposure prophylaxis (PrEP) options and constrained funding, decision-makers could benefit from systematic implementation planning and aligned costs. We reviewed national costed implementation plans (CIPs) to describe relevant implementation inputs and activities (domains) for informing the costed rollout of LAP. We assessed how primary costing evidence aligned with those domains. METHODS: We conducted a rapid review of CIPs for oral PrEP and family planning (FP) to develop a consensus of implementation domains, and a scoping review across nine electronic databases for publications on PrEP costing in LMICs between January 2010 and June 2022. We extracted cost data and assessed alignment with the implementation domains and the Global Health Costing Consortium principles. RESULTS: We identified 15 implementation domains from four national PrEP plans and FP-CIP template; only six were in all sources. We included 66 full-text manuscripts, 10 reported LAP, 13 (20%) were primary cost studies-representing seven countries, and none of the 13 included LAP. The 13 primary cost studies included PrEP commodities (n = 12), human resources (n = 11), indirect costs (n = 11), other commodities (n = 10), demand creation (n = 9) and counselling (n = 9). Few studies costed integration into non-HIV services (n = 5), above site costs (n = 3), supply chains and logistics (n = 3) or policy and planning (n = 2), and none included the costs of target setting, health information system adaptations or implementation research. Cost units and outcomes were variable (e.g. average per person-year). DISCUSSION: LAP planning will require updating HIV prevention policies, technical assistance for logistical and clinical support, expanding beyond HIV platforms, setting PrEP achievement targets overall and disaggregated by method, extensive supply chain and logistics planning and support, as well as updating health information systems to monitor multiple PrEP methods with different visit schedules. The 15 implementation domains were variable in reviewed studies. PrEP primary cost and budget data are necessary for new product introduction and should match implementation plans with financing. CONCLUSIONS: As PrEP services expand to include LAP, decision-makers need a framework, tools and a process to support countries in planning the systematic rollout and costing for LAP.

Unit cost repositories for health program planning and evaluation: a report on research in practice with lessons learned.

BACKGROUND: Most low- and middle-income countries have limited access to cost data that meets the needs of health policy-makers and researchers in health intervention areas including HIV, tuberculosis, and immunization. Unit cost repositories (UCRs)-searchable databases that systematically codify evidence from costing studies-have been developed to reduce the effort required to access and use existing costing information. These repositories serve as public resources and standard references, which can improve the consistency and quality of resource needs projections used for strategic planning and resource mobilization. UCRs also enable analysis of cost determinants and more informed imputation of missing cost data. This report examines our experiences developing and using seven UCRs (two global, five country-level) for cost projection and research purposes. DISCUSSION: We identify advances, challenges, enablers, and lessons learned that might inform future work related to UCRs. Our lessons learned include: (1) UCRs do not replace the need for costing expertise; (2) tradeoffs are required between the degree of data complexity and the useability of the UCR; (3) streamlining data extraction makes populating the UCR with new data easier; (4) immediate reporting and planning needs often drive stakeholder interest in cost data; (5) developing and maintaining UCRs requires dedicated staff time; (6) matching decision-maker needs with appropriate cost data can be challenging; (7) UCRs must have data quality control systems; (8) data in UCRs can become obsolete; and (9) there is often a time lag between the identification of a cost and its inclusion in UCRs. CONCLUSIONS: UCRs have the potential to be a valuable public good if kept up-to-date with active quality control and adequate support available to end-users. Global UCR collaboration networks and greater control by local stakeholders over global UCRs may increase active, sustained use of global repositories and yield higher quality results for strategic planning and resource mobilization.

National Evaluation of HIV Service Resource Allocation in Tanzania.

Using time-driven activity-based costing (TDABC), we examined resource allocation and costs for HIV services throughout Tanzania at patient and facility levels. This national, cross-sectional analysis of 22 health facilities quantified costs and resources associated with 886 patients receiving care for five HIV services: antiretroviral therapy, prevention of mother-to-child transmission, HIV testing and counseling, voluntary medical male circumcision, and pre-exposure prophylaxis. We also documented total provider-patient interaction time, the cost of services with and without inclusion of consumables, and conducted fixed-effects multivariable regression analyses to examine patient- and facility-level correlates of costs and provider-patient time. Findings showed that resources and costs for HIV care varied significantly throughout Tanzania, including as a function of patient- and facility-level characteristics. While some variation may be preferable (e.g., needier patients received more resources), other areas suggested a lack of equity (e.g., wealthier patients received more provider time) and presented opportunities to optimize care delivery protocols.

Estimating the cost and efficiency gain of rolling out a multi-month dispensing programme for antiretroviral treatment in Tanzania.

Background: Globally, efforts to curtail the HIV pandemic are growing. The Joint United Nations Programme on HIV and AIDS (UNAIDS) and partners set the 95-95-95 targets to be achieved by 2025. Tanzania's ongoing transition from single-month ARV to longer multi-month dispensing (MMD) involves significant planning and shifts in existing resources, including health commodities, clinical staff and storage space. This study aimed at evaluating the costs and efficiency gains of rolling out MMD compared to the prior monthly dispending (MD) standard of care before the new guidelines.Methods: The analysis employed a health provider perspective utilising prior costing data collected to estimate cost of treatment for HIV/AIDS, including salaries, laboratory costs, antiretroviral drugs, other supplies and overhead costs. The projections were run from 2018 to 2030 using the Spectrum package for Tanzania.Results: Our model estimated that total treatment cost without MMD (including salaries, laboratory costs, antiretroviral drugs, other supplies, and overhead costs) is estimated to rise from USD 189 million in 2018 to USD 244 million in 2030. The introduction of a six-month MMD would lead to the total annual facility-based treatment costs being reduced to USD 205 million in 2030. When comparing MD to a six-month MMD, the total savings over the 13-year period would be USD 425 million. The introduction of six-month MMD for stable patients would reduce the average cost from USD 180 to USD 156 per patient per year if stable patients were only required to make six-monthly visit.Conclusions: The introduction of differentiated service delivery models (DSDMs) and MMD is already contributing to significant cost savings for Tanzania and will continue to do so as the country puts more stable patients on MMD. The potential gains from MMD implantation could further be harnessed if retention of treatment and viral suppression monitoring are prioritised.

The voluntary medical male circumcision Site Capacity and Productivity Assessment Tool (SCPT): An innovative visual management tool to optimize site service delivery.

Given constrained funding for HIV, achieving global goals on VMMC scale-up requires that providers improve service delivery operations and use labor and capital inputs as efficiently as possible to produce as many quality VMMCs as feasible. The Voluntary Medical Male Circumcision Site Capacity and Productivity Assessment Tool (SCPT) is an electronic visual management tool developed to help VMMC service providers to understand and improve their site's performance. The SCPT allows VMMC providers to: 1) track the most important human resources and capital inputs to VMMC service delivery, 2) strategically plan site capacity and targets, and 3) monitor key site-level VMMC service delivery performance indicators. To illustrate a real-world application of the SCPT, we present selected data from two provinces in Mozambique-Manica and Tete, where the SCPT was piloted We looked at the data prior to the introduction of SCPT (October 2014 to August 2016), and during the period when the tool began to be utilized (September 2016 to September 2017). The tool was implemented as part of a broader VMMC site optimization strategy that VMMC implementers in Mozambique put in place to maximize programmatic impact. Routine program data for Manica and Tete from October 2014 to September 2017 showcase the turnaround of the VMMC program that accompanied the implementation of the SCPT together with the other components of the VMMC site optimizatio strategy. From October 2016, there was a dramatic increase in the number of VMMCs performed. The number of fixed service delivery sites providing VMMC services was expanded, and each fixed site extended service delivery by performing VMMCs in outreach sites. Alignment between site targets and the number of VMMCs performed improved from October 2016. Utilization rates stabilized between October 2016 and September 2017, with VMMCs performed closely tracking VMMC site capacity in most sites. The SCPT is designed to address the need for site level data for programmatic decision-making during site planning, implementation, monitoring and evaluation. Deployment of the SCPT can help VMMC providers monitor the performance of VMMC service delivery sites and improve their performance. We recommend use of the customized version of this tool and model to the need of other programs.

Moving away from the "unit cost". Predicting country-specific average cost curves of VMMC services accounting for variations in service delivery platforms in sub-Saharan Africa.

BACKGROUND: One critical element to optimize funding decisions involves the cost and efficiency implications of implementing alternative program components and configurations. Program planners, policy makers and funders alike are in need of relevant, strategic data and analyses to help them plan and implement effective and efficient programs. Contrary to widely accepted conceptions in both policy and academic arenas, average costs per service (so-called "unit costs") vary considerably across implementation settings and facilities. The objective of this work is twofold: 1) to estimate the variation of VMMC unit costs across service delivery platforms (SDP) in Sub-Saharan countries, and 2) to develop and validate a strategy to extrapolate unit costs to settings for which no data exists. METHODS: We identified high-quality VMMC cost studies through a literature review. Authors were contacted to request the facility-level datasets (primary data) underlying their results. We standardized the disparate datasets into an aggregated database which included 228 facilities in eight countries. We estimated multivariate models to assess the correlation between VMMC unit costs and scale, while simultaneously accounting for the influence of the SDP (which we defined as all possible combinations of type of facility, ownership, urbanicity, and country), on the unit cost variation. We defined SDP as any combination of such four characteristics. Finally, we extrapolated VMMC unit costs for all SDPs in 13 countries, including those not contained in our dataset. RESULTS: The average unit cost was 73 USD (IQR: 28.3, 100.7). South Africa showed the highest within-country cost variation, as well as the highest mean unit cost (135 USD). Uganda and Namibia had minimal within-country cost variation, and Uganda had the lowest mean VMMC unit cost (22 USD). Our results showed evidence consistent with economies of scale. Private ownership and Hospitals were significant determinants of higher unit costs. By identifying key cost drivers, including country- and facility-level characteristics, as well as the effects of scale we developed econometric models to estimate unit cost curves for VMMC services in a variety of clinical and geographical settings. CONCLUSION: While our study did not produce new empirical data, our results did increase by a tenfold the availability of unit costs estimates for 128 SDPs in 14 priority countries for VMMC. It is to our knowledge, the most comprehensive analysis of VMMC unit costs to date. Furthermore, we provide a proof of concept of the ability to generate predictive cost estimates for settings where empirical data does not exist.

Costs and Cost Drivers of Providing Option B+ Services to Mother-Baby Pairs for PMTCT of HIV in Health Centre IV Facilities in Jinja District, Uganda.

BACKGROUND: In 2013, the World Health Organization (WHO) revised the 2012 guidelines on use of antiretroviral drugs (ARVs) for the prevention of mother-to-child transmission (PMTCT) of human immunodeficiency virus (HIV). The new guidelines recommended lifelong antiretroviral therapy (ART) for all HIV-positive pregnant and breastfeeding women irrespective of CD4 count or clinical stage (also referred to as Option B+). Uganda started implementing Option B+ in 2012 basing on the 2012 WHO guidelines. Despite the impressive benefits of the Option B+ strategy, implementation challenges, including cost burden and mother-baby pairs lost to follow-up, threatened its overall effectiveness. The researchers were unable to identify any studies conducted to assess costs and cost drivers associated with provision of Option B+ services to mother-baby pairs in HIV care in Uganda. Therefore, this study determined costs and cost drivers of providing Option B+ services to mother-baby pairs over a two-year period (2014-2015) in selected health facilities in Jinja district, Uganda. METHODS: The estimated costs of providing Option B+ to mother-baby pairs derived from the provider perspective were evaluated at four health centres (HC) in Jinja district. A retrospective, ingredient-based costing approach was used to collect data for 2014 as base year using a standardized cost data capture tool. All costs were valued in United States dollars (USD) using the 2014 midyear exchange rate. Costs incurred in the second year (2015) were obtained by inflating the 2014 costs by the ratio of 2015 and 2014 USA Gross Domestic Product (GDP) implicit price deflator. RESULTS: The average total cost of Option B+ services per HC was 66,512.7 (range: 32,168.2-102,831.1) USD over the 2-year period. The average unit cost of Option B+ services per mother-baby pair was USD 441.9 (range: 422.5-502.6). ART for mothers was the biggest driver of total mean costs (percent contribution: 62.6%; range: 56.0%-65.5%) followed by facility personnel (percent contribution: 8.2%; range: 7.7%-11.6%), and facility-level monitoring and quality improvement (percent contribution: 6.0%; range: 3.2%-12.3%). Conclusions and Recommendations. ART for mothers was the major cost driver. Efforts to lower the cost of ART for PMTCT would make delivery of Option B+ affordable and sustainable.

The short-term effects of the implementation of the "Treat All" guidelines on ART service delivery costs in Namibia.

The introduction of "Treat All" (TA) has been promoted to increase the effectiveness of HIV/AIDS treatment by having patients initiate antiretroviral therapy at an earlier stage of their illness. The impact of introducing TA on the unit cost of treatment has been less clear. The following study evaluated how costs changed after Namibia's introduction of TA in April 2017. A two-year analysis assessed the costs of antiretroviral therapy (ART) during the 12 months before TA (Phase I-April 1, 2016 to March 31, 2017) and the 12 months following (Phase II-April 1, 2017 to March 31, 2018). The analysis involved interviewing staff at ten facilities throughout Namibia, collecting data on resources utilized in the treatment of ART patients and analyzing how costs changed before and after the introduction of TA. An analysis of treatment costs indicated that the unit cost of treatment declined from USD360 per patient per year in Phase I to USD301 per patient per year in Phase II, a reduction of 16%. This decline in unit costs was driven by 3 factors: 1) shifts in antiretroviral (ARV) regimens that resulted in lower costs for drugs and consumables, 2) negotiated reductions in the cost of viral load tests and 3) declines in personnel costs. It is unlikely that the first two of these factors were significantly influenced by the introduction of TA. It is unclear if TA might have had an influence on personnel costs. The reduction in personnel costs may have either represented a positive development (fewer personnel costs associated with increased numbers of healthier patients and fewer visits required) or alternatively may reflect constraints in Namibia's staffing. Prior to this study, it was expected that the introduction of TA would lead to a significant increase in the number of ART patients. However, there was less than a 4% increase in the number of adult patients at the 10 studied facilities. From a financial point of view, TA did not significantly increase the resources required in the ten sampled facilities, either by raising unit costs or significantly increasing the number of ART patients.

Meta-analysis of average costs of HIV testing and counselling and voluntary medical male circumcision across thirteen countries.

Objective: Explore facility-level average costs per client of HIV testing and counselling (HTC) and voluntary medical male circumcision (VMMC) services in 13 countries.Methods: Through a literature search we identified studies that reported facility-level costs of HTC or VMMC programmes. We requested the primary data from authors and standardised the disparate data sources to make them comparable. We then conducted descriptive statistics and a meta-analysis to assess the cost variation among facilities. All costs were converted to 2017 US dollars ($).Results: We gathered data from 14 studies across 13 countries and 772 facilities (552 HTC, 220 VMMC). The weighted average unit cost per client served was $15 (95% CI 12, 18) for HTC and $59 (95% CI 45, 74) for VMMC. On average, 38% of the mean unit cost for HTC corresponded to recurrent costs, 56% to personnel costs, and 6% to capital costs. For VMMC, 41% of the average unit cost corresponded to recurrent costs, 55% to personnel costs, and 4% to capital costs. We observed unit cost variation within and between countries, and lower costs in higher scale categories in all interventions.

HIV treatment support services in Tanzania: a cost and efficiency analysis at facility and community levels.

Differentiated service delivery models provide an opportunity to reduce the cost of HIV treatment while increasing the focus on advanced patients. Also, stable patients who need less attention can visit facilities less frequently, saving time and money and reducing overcrowding. Tanzania needs treatment support services that can improve the quality of life, ensure adherence and result in viral suppression. Treatment support services can be provided through a variety of models, including at the community level, through the facility or through some combination of the two. Understanding the cost of each model is essential for policy-makers who must allocate resources. Data from neighbouring countries suggests that community-level support services can be beneficial to patients, especially in reducing loss to follow-up and death. Though community-based HIV services are available in Tanzania, uncertainty about the costs of these models remain. This study assessed the costs of treatment support services at 27 sites in Tanzania. The cost analysis found that the average unit cost of treatment support services is US$39 per patient per year. The analysis found that community-based models have clear advantages with regard to the number of patients that can be reached with support services. Costing data indicated that community-based models are also less expensive. The lack of data on retention limited any conclusions about whether community-based models are the most effective within the Tanzanian context.

Twenty Years Of Antiretroviral Therapy For People Living With HIV: Global Costs, Health Achievements, Economic Benefits.

Since the introduction of azidothymidine in 1987, significant improvements in treatment for people living with HIV have yielded substantial improvements in global health as a result of the unique benefits of antiretroviral therapy (ART). ART averted 9.5 million deaths worldwide in 1995-2015, with global economic benefits of $1.05 trillion. For every $1 spent on ART, $3.50 in benefits accrued globally. If treatment scale-up achieves the global 90-90-90 targets of the Joint United Nations Programme on HIV/AIDS, a total of 34.9 million deaths are projected to be averted between 1995 and 2030. Approximately 40.2 million new HIV infections could also be averted by ART, and economic gains could reach $4.02 trillion in 2030. Having provided ART to 19.5 million people represents a major human achievement. However, 15.2 million infected people are currently not receiving treatment, which represents a significant lost opportunity. Further treatment scale-up could yield even greater health and economic benefits.

The role of costing in the introduction and scale-up of HIV pre-exposure prophylaxis: evidence from integrating PrEP into routine maternal and child health and family planning clinics in western Kenya.

INTRODUCTION: Understanding the cost of strategies to reach and deliver pre-exposure prophylaxis (PrEP) to priority populations is essential to assess the cost-effectiveness and budget impact of HIV prevention programmes. Providing PrEP through maternal and child health and family planning clinics offers a promising strategy to reach women in high HIV burden settings. We estimated incremental costs and explored the cost drivers of integrating PrEP delivery into routine maternal and child health and family planning services in Kenya. METHODS: We conducted a costing study from the provider perspective within the PrEP Implementation for Young Women and Adolescents programme in western Kenya. We identified all within- and above-facility activities supporting PrEP delivery and measured clinical service time using time-and-motion studies. We obtained input costs from programme budgets, expenditure records and staff interviews. We estimated changes in costs if creatinine testing were postponed from initiation to first follow-up visit and if PrEP were prioritized to clients at high HIV risk using a behavioural risk assessment tool. We also projected costs under Ministry of Health (MOH) implementation assuming MOH salaries and programme supervision. We estimated annual numbers of PrEP visits from programme data abstracted from 16 facilities between November 2017 and June 2018. We report the cost per client-month of PrEP dispensed in 2017 USD. RESULTS: For an annual programme output of 24,005 screenings, 4198 PrEP initiations and 4427 follow-up visits, the average cost per client-month of PrEP dispensed in the study was $26.52. Personnel, drugs and laboratory tests comprised 43%, 25% and 14% of programme costs respectively. Postponing creatinine testing and prioritizing PrEP delivery to clients at high HIV risk reduced total programme costs by 8% and 14% respectively. In the MOH scenario assuming no changes in outputs, the projected cost per client-month of PrEP dispensed decreased to $16.54 and total programme costs decreased by 38%. CONCLUSIONS: Incremental PrEP costs are sensitive to the service delivery strategy used to engage priority populations. Postponing creatinine testing and prioritizing PrEP delivery to clients at high HIV risk may reduce costs. Context-specific cost data are crucial to assess the cost-effectiveness and affordability of PrEP delivery models.

Using mathematical modeling to inform health policy: A case study from voluntary medical male circumcision scale-up in eastern and southern Africa and proposed framework for success.

BACKGROUND: Modeling contributes to health program planning by allowing users to estimate future outcomes that are otherwise difficult to evaluate. However, modeling results are often not easily translated into practical policies. This paper examines the barriers and enabling factors that can allow models to better inform health decision-making. DESCRIPTION: The Decision Makers' Program Planning Tool (DMPPT) and its successor, DMPPT 2, are illustrative examples of modeling tools that have been used to inform health policy. Their use underpinned Voluntary Medical Male Circumcision (VMMC) scale-up for HIV prevention in southern and eastern Africa. Both examine the impact and cost-effectiveness of VMMC scale-up, with DMPPT used initially in global advocacy and DMPPT 2 then providing VMMC coverage estimates by client age and subnational region for use in country-specific program planning. Their application involved three essential steps: identifying and engaging a wide array of stakeholders from the outset, reaching consensus on key assumptions and analysis plans, and convening data validation meetings with critical stakeholders. The subsequent DMPPT 2 Online is a user-friendly tool for in-country modeling analyses and continuous program planning and monitoring. LESSONS LEARNED: Through three iterations of the DMPPT applied to VMMC, a comprehensive framework with six steps was identified: (1) identify a champion, (2) engage stakeholders early and often, (3) encourage consensus, (4) customize analyses, (5), build capacity, and (6) establish a plan for sustainability. This framework could be successfully adapted to other HIV prevention programs to translate modeling results to policy and programming. CONCLUSIONS: Models can be used to mobilize support, strategically plan, and monitor key programmatic elements, but they can also help inform policy environments in which programs are conceptualized and implemented to achieve results. The ways in which modeling has informed VMMC programs and policy may be applicable to an array of other health interventions.

Voluntary medical male circumcision service delivery in South Africa: The economic costs and potential opportunity for private sector involvement.

BACKGROUND: In 2010, the South African Government initiated a voluntary medical male circumcision (VMMC) program as a part of the country's HIV prevention strategy based on compelling evidence that VMMC reduces men's risk of becoming HIV infected by approximately 60%. A previous VMMC costing study at Government and PEPFAR-supported facilities noted that the lack of sufficient data from the private sector represented a gap in knowledge concerning the overall cost of scaling up VMMC services. This study, conducted in mid-2016, focused on surgical circumcision and aims to address this limitation. METHODS: VMMC service delivery cost data were collected at 13 private facilities in three provinces in South Africa: Gauteng, KwaZulu-Natal, and Mpumalanga. Unit costs were calculated using a bottom-up approach by cost components, and then disaggregated by facility type and urbanization level. VMMC demand creation, and higher-level management and program support costs were not collected. The unit cost of VMMC service delivery at private facilities in South Africa was calculated as a weighted average of the unit costs at the 13 facilities. KEY FINDINGS: At the average annual exchange rate of R10.83 = $1, the unit cost including training and cost of continuous quality improvement (CQI) to provide VMMC at private facilities was $137. The largest cost components were consumables (40%) and direct labor (35%). Eleven out of the 13 surveyed private sector facilities were fixed sites (with a unit cost of $142), while one was a fixed site with outreach services (with a unit cost of $156), and the last one provided services at a combination of fixed, outreach and mobile sites (with a unit cost per circumcision performed of $123). The unit cost was not substantially different based on the level of urbanization: $141, $129, and $143 at urban, peri-urban, and rural facilities, respectively. CONCLUSIONS: The private sector VMMC unit cost ($137) did not differ substantially from that at government and PEPFAR-supported facilities ($132 based on results from a similar study conducted in 2014 in South Africa at 33 sites across eight of the countries nine provinces). The two largest cost drivers, consumables and direct labor, were comparable across the two studies (75% in private facilities and 67% in public/PEPFAR-supported facilities). Results from this study provide VMMC unit cost data that had been missing and makes an important contribution to a better understanding of the costs of VMMC service delivery, enabling VMMC programs to make informed decisions regarding funding levels and scale-up strategies for VMMC in South Africa.

Correction: The Cost of Voluntary Medical Male Circumcision in South Africa.

[This corrects the article DOI: 10.1371/journal.pone.0160207.].

Correction: Estimating Client Out-of-Pocket Costs for Accessing Voluntary Medical Male Circumcision in South Africa.

[This corrects the article DOI: 10.1371/journal.pone.0164147.].

The Cost of Voluntary Medical Male Circumcision in South Africa.

Given compelling evidence associating voluntary medical male circumcision (VMMC) with men's reduced HIV acquisition through heterosexual intercourse, South Africa in 2010 began scaling up VMMC. To project the resources needed to complete 4.3 million circumcisions between 2010 and 2016, we (1) estimated the unit cost to provide VMMC; (2) assessed cost drivers and cost variances across eight provinces and VMMC service delivery modes; and (3) evaluated the costs associated with mobilize and motivate men and boys to access VMMC services. Cost data were systematically collected and analyzed using a provider's perspective from 33 Government and PEPFAR-supported (U.S. President's Emergency Plan for AIDS Relief) urban, rural, and peri-urban VMMC facilities. The cost per circumcision performed in 2014 was US$132 (R1,431): higher in public hospitals (US$158 [R1,710]) than in health centers and clinics (US$121 [R1,309]). There was no substantial difference between the cost at fixed circumcision sites and fixed sites that also offer outreach services. Direct labor costs could be reduced by 17% with task shifting from doctors to professional nurses; this could have saved as much as $15 million (R163.20 million) in 2015, when the goal was 1.6 million circumcisions. About $14.2 million (R154 million) was spent on medical male circumcision demand creation in South Africa in 2014-primarily on personnel, including community mobilizers (36%), and on small and mass media promotions (35%). Calculating the unit cost of VMMC demand creation was daunting, because data on the denominator (number of people reached with demand creation messages or number of people seeking VMMC as a result of demand creation) were not available. Because there are no "dose-response" data on demand creation ($X in demand creation will result in an additional Z% increase in VMMC clients), research is needed to determine the appropriate amount and allocation of demand creation resources.