Novel tuberculosis vaccines are in varying stages of pre-clinical and clinical development. This study seeks to estimate the potential cost-effectiveness of a BCG booster vaccine, while accounting for costs of large-scale clinical trials, using the MVA85A vaccine as a case study for estimating potential costs. We conducted a decision analysis from the societal perspective, using a 10-year time frame and a 3% discount rate. We predicted active tuberculosis cases and tuberculosis-related costs for a hypothetical cohort of 960,763 South African newborns (total born in 2009). We compared neonatal vaccination with bacille Calmette-Guérin alone to vaccination with bacille Calmette-Guérin plus a booster vaccine at 4 months. We considered booster efficacy estimates ranging from 40% to 70%, relative to bacille Calmette-Guérin alone. We accounted for the costs of Phase III clinical trials. The booster vaccine was assumed to prevent progression to active tuberculosis after childhood infection, with protection decreasing linearly over 10 years. Trial costs were prorated to South Africa's global share of bacille Calmette-Guérin vaccination. Vaccination with bacille Calmette-Guérin alone resulted in estimated tuberculosis-related costs of $89.91 million 2012 USD, and 13,610 tuberculosis cases in the birth cohort, over the 10 years. Addition of the booster resulted in estimated cost savings of $7.69–$16.68 million USD, and 2,800–4,160 cases averted, for assumed efficacy values ranging from 40%–70%. A booster tuberculosis vaccine in infancy may result in net societal cost savings as well as fewer active tuberculosis cases, even if efficacy is relatively modest and large scale Phase III studies are required.
Citation: Ditkowsky JB, Schwartzman K (2014) Potential Cost-Effectiveness of a New Infant Tuberculosis Vaccine in South Africa - Implications for Clinical Trials: A Decision Analysis. PLoS ONE9(1): e83526. https://doi.org/10.1371/journal.pone.0083526
Editor: Pere-Joan Cardona, Archivel Farma; Fundació Institut d'Investigació en Ciències de la Salut Germans Trias i Pujol. Universitat Autònoma de Barcelona. CIBERES, Spain
Received: April 16, 2013; Accepted: November 5, 2013; Published: January 15, 2014
Copyright: © 2014 Ditkowsky, Schwartzman. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Mr. Ditkowsky was supported by a studentship award from the Montreal Chest Institute Research Centre. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Nearly one-third of the world's population is infected with Mycobacterium tuberculosis (M. tuberculosis), with an estimated 9.4 million incident cases and 1.3 million deaths in 2009 , . Progress in reducing morbidity and mortality has been severely hampered by several challenges, including HIV co-infection, antibiotic resistance, and limited diagnostic and treatment capacity in many high-burden settings. As a consequence, tuberculosis (TB) control strategies are evolving to address novel diagnostic tools, treatment regimens for multi-drug resistance, TB-HIV program integration, and other potential solutions. For TB control in the longer term, modeling studies have underlined the importance of enhanced diagnostic capacity, expanded treatment of latent TB infection, new anti-tuberculosis drugs, and improved TB vaccines . In principle, an effective TB vaccine can circumvent some of the challenges posed by drug resistance, treatment adherence, and potentially HIV-TB .
The bacille Calmette-Guérin (BCG) vaccine is the only vaccine currently licensed for TB prevention. While it is considered modestly efficacious in preventing tuberculosis meningitis and disseminated TB in children, estimates of its true efficacy vary widely . Furthermore, it appears to provide limited or no protection against adult pulmonary TB, so that it cannot directly reduce transmission , . Because of these gaps, the last decade has witnessed substantial interest and investment in TB vaccine development. Select candidates, including the Aeras 402/Crucell Ad35 and MVA85A vaccines have entered Phase IIB clinical trials for safety and efficacy, based on promising Phase I safety and immunogenicity data. The MVA85A vaccine has recently undergone evaluation in BCG-vaccinated infants in South Africa, and is now being studied in HIV-infected adults in Senegal and South Africa , , , in trials conducted by the Oxford-Emergent Tuberculosis Consortium (OETC).
MVA85A vaccination of persons with previous exposure to M. tuberculosis or BCG vaccination appears to result in increased immunogenicity, compared to MVA85A vaccination of BCG/M. tuberculosis-naïve individuals , . However, a phase IIB randomized, controlled trial among 2,797 South African infants demonstrated no additional efficacy beyond BCG alone . Hence while immunological data for some new vaccines are promising, the substantial resources required for clinical development and rollout must be evaluated in the context of the potential downstream health benefits and cost savings. This is particularly relevant for funders and decision makers, who must consider further investments in vaccine development, versus investments in other promising approaches to TB control. Using a potential infant booster vaccination program as a case study, we examined the balance between vaccine development and administration costs, including those of further clinical trials, and later gains in TB morbidity, mortality, and the related cost savings.
We developed a Markov model, using TreeAge ProSuite 2009 (TreeAge Software, Williamstown, MA.) We compared two scenarios: 1) current neonatal BCG vaccination and DOTS coverage, without booster vaccination, and 2) current neonatal BCG vaccination and DOTS coverage, plus a new infant booster vaccine administered at age 4 months. (Figures 1–2). The model was calibrated to characteristics of the South African population, and examined a cohort of 960, 763 newborns entering the population (total born in 2009). We predicted active TB cases, TB deaths and related costs. The analysis was conducted from the South African societal perspective. We used a 10-year time frame, and a 3% discount rate .
Population and TB control parameters are listed in Table 1; these were assumed to remain constant over the 10-year simulation period. BCG vaccination was assumed to provide an initial 50% reduction in the rate of primary progression to TB disease, decreasing linearly to zero protection at the end of the 10 years , .
We assumed 90% BCG coverage of newborns , and that among BCG-vaccinated infants, 90% would receive the booster if offered. Booster vaccine protection waned linearly to zero over its 10-year duration of action. In the base case analysis, the booster vaccine was first assumed to confer 60% protection over BCG alone . We then varied the assumed efficacy from 40% to 70% over BCG alone. Neither vaccine was considered to provide any protection to children with clinical AIDS.
Infants could acquire TB infection at any stage of the simulation. After acquiring TB infection, they could progress immediately to active TB disease, or remain with latent infection. Initial infection with HIV led to an “early HIV” status. Progression to clinical AIDS could occur at any stage after initial infection. Based on South African data, we assumed that anti-retroviral therapy (ART) was offered to 58% of infected infants, and was associated with a 75% decrease in HIV-related annual mortality for those with early HIV, and a 9.8% decrease for those who progressed to clinical AIDS , , .
In addition, we assumed that isoniazid prophylaxis was provided only to children under 5 years of age who had HIV, and/or family contacts with active TB . We assumed that 50% of eligible infants with latent tuberculosis would be screened. Of those screened, 34% would be provided treatment . Treatment had an efficacy of 78%, with a 20% probability of treatment completion . Incomplete treatment was assumed to be ineffective. Other clinical parameters are described in Table 2.
Costs were expressed in 2012 US dollars. We included all TB-related health care costs, as well as family expenditures related to children's illness, and productivity losses by family members. South African gross national income (GNI) per capita was used to calculate income and productivity losses based on a 40-hour work week. Where possible, costs were obtained from previous cost-effectiveness analyses addressing TB treatment in South Africa , . Drug costs reflected the Global Drug Facility price list, and South African treatment recommendations , . DOTS program administrative costs were based on earlier evaluations of DOTS implementation in South Africa (Tables 3 and 4) .
Further research and development costs for a booster vaccine reflected estimated costs of Phase IIB and potential Phase III clinical trials of the MVA85A vaccine. Initial pre-clinical development and early phase clinical study costs were considered to be sunk i.e. already spent, as is also the case for other current leading vaccine candidates, so they were not included. Sample size and cost information was based on interviews with the OETC, which has not otherwise contributed to or reviewed this paper. The initial sample size calculation for a phase III trial involves a hypothesized efficacy of 60%, and is predicated on a lower limit of 30% vaccine efficacy for the 95% confidence interval. The estimated baseline risk of active TB is 1% per year. With these parameters, the target sample size is a total of 12,000 infants, with a total cost of $120 million, including $30 million in start-up/infrastructure costs, and $90 million for recruitment and follow-up ($7,500 per subject). Phase IIB trial costs, estimated at $30 million, plus an additional $10 million for infrastructure, were also added. [Of note, Phase I and II trials of the final commercial formulation of any vaccine will also be required and additional lot-to-lot Phase III consistency trials may be required depending upon the final design of the Phase III trial]. OETC reported that Aeras has previously publicly stated that the eventual target cost of vaccine production for use in the developing world is $2/dose. In fact, OETC anticipates that vaccine production costs would not reach this level without considerable economies of scale and purpose-built manufacturing facilities.
For the purposes of the present analysis, the total estimated Phase III trial cost was divided by the number of vaccine doses to be administered worldwide over 10 years after its introduction (based on 106.4 million BCG doses administered annually ), so as to attribute a share of the trial cost to every dose. This assumes that the trial cost would be recouped over the first 10 years of vaccine use. Phase III trial parameters are listed in Table 5. The Phase IIB trial costs (total $40 million) were also incorporated into the final research and development cost, and were attributed to each dose in the same way as Phase III costs. We examined alternate scenarios for true vaccine efficacy - while keeping 30% as the lower limit of the 95% confidence interval from any trial - and the resulting changes in sample size and cost. The target cost of vaccine production was initially set at $2/dose. In the base case analysis, we assumed a profit margin of $4/dose.
We conducted extensive sensitivity analyses for all assumed parameters, with epidemiologic parameters varied across published ranges. Sensitivity analyses included a multiway probabilistic analysis, with simultaneous variation of the parameters with the largest impact on predicted outcomes. We used a triangular distribution, where the base case value was most likely, and the low and high extremes of the distribution were half and double the base case assumption respectively. The probabilistic analysis consisted of 100 runs and 100,000 microsimulations.
For the birth cohort of 960,763 infants, we projected that the current BCG/DOTS strategy would cost $89.91 million over 10 years, with 13,607 active TB cases and 3,243 TB-related deaths over the same period. Assuming that the protective efficacy of the combined BCG-booster vaccination was 60% relative to BCG alone, we projected savings of $14.82 million, with 3,772 TB cases and 703 deaths averted. As expected, higher efficacy estimates resulted in greater cost savings as well as further improvements in morbidity and mortality. However, with an efficacy of 40% for the combined vaccinations, relative to BCG alone, there were still projected cost savings—even after accounting for the attendant increase in clinical trial sample size requirements (Table 6 and Figure 3).
When parameters that directly affected vaccine costs were varied, the booster vaccination strategy remained cost saving in most cases. For example, when the per-dose cost of the booster vaccine was doubled, there were still cost savings of $6.08 million for the booster strategy. With the assumption that the duration of vaccine action was halved, to only 5 years, the booster strategy resulted in a net cost of $1.24 million. The threshold value for cost savings was ≥5.72 years of vaccine activity.
In most other one-way sensitivity analyses, the booster strategy remained cost-saving relative to BCG alone, in addition to reducing TB morbidity and mortality. With the base case assumptions, any profit margin less than $17.13 per vaccine dose will still result in net cost savings for South African society (Table 7).
In “best” and “worst” case scenarios, key parameters were varied simultaneously: cost of booster vaccine, prevalence of initial single and multi-drug resistance, TB annual risk of infection, prevalence of HIV at birth, the probability of TB diagnosis, the cost of a DOTS visit, ART protective efficacy against HIV progression, and booster vaccine efficacy.
In the “best case” scenario, values for all these parameters were doubled, with the exception of the cost of booster vaccine, ART protection against HIV progression, and the probability of TB diagnosis, which were both halved, and the protective efficacy of combined vaccination compared to BCG alone, which was set to 70%. With the best case scenario, we predicted cost savings of $38.56 million, with prevention of 6,038 active TB cases and 4,854 TB-related deaths over 10 years.
In the “worst case” scenario, all key parameters were halved, except that the unit cost of the booster vaccine was doubled, the probability of TB diagnosis was set to 90%, ART reduced the probability of progression from asymptomatic HIV infection to AIDS by 90%, and the additional protective efficacy of combined vaccination compared to BCG alone was set to 40%. With this combination of assumptions, estimated costs were $7.32 million greater than for BCG alone, with 941 active TB cases and 137 TB-related deaths averted. A change in the proposed mechanism of vaccine action, with the booster assumed to protect against acquisition of M. tuberculosis infection, resulted in substantial cost savings, as well as further reductions in TB cases and related deaths (Table 7).
In the probabilistic sensitivity analysis we varied six key parameters: annual risk of TB infection, discount rate, HIV prevalence at birth, DOTS program cost, cost of lost work time for family members, and the booster vaccine's efficacy. The most likely cost for the booster strategy was consistently lower than that for the BCG alone strategy, with mean associated cost savings of $11.21 million; median savings were $11.53 million, with interquartile range $6.3–16.8 million (Figure 4).
From a societal perspective, infant booster vaccination with a new tuberculosis vaccine appears less costly than BCG vaccine alone, if the new vaccine has at least modest protective efficacy - even after clinical trial costs and various unfavorable assumptions are considered. These results are concordant with previous analyses addressing the potential impact of a pre-exposure TB vaccine. Findings were generally similar in a previous analysis examining neonatal BCG replacement with or without a booster . In that analysis, only at assumed protective efficacy values below the 40% lower limit used in the present analysis, was the replacement vaccine strategy more costly than BCG alone. In the recent infant Phase IIB study of MVA85A, there was no additional efficacy beyond BCG alone . Clearly, without large-scale clinical trial data supporting a more effective candidate vaccine, any predictions of cost-effectiveness remain hypothetical. Nonetheless, extensive sensitivity analyses suggest that a modestly effective booster vaccine will likely prove cost-effective. However, if mass vaccination is to occur, pricing must be within reach of low and middle-income countries, which is unlikely to be the case with the high-end estimate ($15/dose) we used in sensitivity analysis.
Abu-Raddad and colleagues estimated that an improved neonatal vaccine could reduce TB incidence by 39%–55%. They further concluded that a catch-up vaccination campaign, using a vaccine that provides pre-exposure protection, could reduce incidence by as much as 67%, in the absence of HIV infection . Another analysis suggested that a pre-exposure vaccine might be associated with a one-third drop in TB incidence, over the longer term .
A strength of the present analysis is that all costs were derived from previously published surveys, or interviews with OETC. Previous cost data for patients and families allowed us to adopt a societal perspective. Epidemiologic parameters were taken primarily from South African data. The remainder were taken from previously published models or statistics from similar settings. Our analysis is relevant to high TB incidence settings, including those with a high prevalence of HIV infection.
Our model assumed that major TB epidemiologic parameters remained constant over the course of the simulation, apart from vaccine effects. Hence we did not consider other emerging strategies to improve TB control, such as improved diagnostic capacity or expanded treatment of latent TB infection , . Our analysis used a static Markov model; as it focused on childhood TB over a short time frame, we did not consider secondary transmission (which is rarely the result of TB in young children), or herd immunity by vaccination.
The ultimate costs of vaccine research and development remain uncertain. We were able to incorporate some initial estimates from OETC, and we prorated these costs to South Africa's global share of BCG vaccination. It is premature to accurately estimate research and development costs, and several very large and costly clinical trials may ultimately be required before any candidate vaccine is suitable for licensure and distribution. The extent to which these costs will ultimately be included in the commercial vaccine purchase price remains uncertain, as do eventual production costs and profit margins. However, our sensitivity analysis did address the potential impact of varying research, development, and production costs—and of varying profit margins–on the vaccine purchase price.
At present, the absence of validated biomarkers for protective immunity against M. tuberculosis infection and disease means that clinical vaccine trials must use active TB as their primary endpoint. Even in very high TB incidence settings, such as sub-Saharan Africa, this entails very large sample sizes, and the attendant costs. However, the present analysis suggests that from a societal standpoint, the necessary investments of resources, time and money are likely to pay off in terms of cost savings as well as improved health.
The authors acknowledge the input of the Oxford-Emergent Tuberculosis Consortium with respect to clinical trial cost estimates used in this analysis.
Conceived and designed the experiments: KS JBD. Performed the experiments: JBD KS. Analyzed the data: KS JBD. Contributed reagents/materials/analysis tools: KS JBD. Wrote the paper: JBD KS.
- 1. World Health Organization (2011) Global Tuberculosis Report 2011. WHO/HTM/TB/2011.16. Available: http://www.who.int/tb/publications/global_report/en/index.html. Accessed: 30 Nov 2012.
- 2. World Health Organization (2011) Top 10 causes of death: fact sheet no. 310: WHO 2011. Available: http://www.who.int/mediacentre/factsheets/fs310/en/index.html. Accessed: 20 Jun 2011.
- 3. Fine PEM (1995) Variation in protection by BCG: implications of and for heterologous immunity. Lancet 346: 1339–1345.
- 4. Colditz GA, Berkey CS, Mosteller F, Brewer TF, Wilson ME, et al. (1995) The efficacy of bacillus Calmette-Guerin vaccination of newborns and infants in the prevention of tuberculosis: meta-analysis of the published literature. Pediatrics 96: 29–35.
- 5. Rodrigues LC, Diwan VK, Wheeler JG (1993) Protective effect of BCG against tuberculosis meningitis and military tuberculosis: a meta-analysis. Int J Epidemiol 22: 1154–1158.
- 6. Aeras (2012) Vaccine Development Process. Available: http://www.aeras.org/portfolio. Accessed: 26 Oct 2012.
- 7. Foundation AGTV (2011) Protective efficacy against tuberculosis (TB) disease, safety, and immunogenicity of MVA85A/AERAS-485 in HIV-infected adults. Available: http://clinicaltrials.gov/ct2/show/nct01151189?intr=“mva85a%2faeras-485”&rank = 1. Accessed: 5 Jun 2011.
- 8. Scriba TJ, Tameris M, Smit E, van der Merwe L, Hughes EJ, et al. (2012) A Phase IIA Trial of the new tuberculosis vaccine, MVA85A, in HIV- and/or Mycobacterium tuberculosis- infected adults. Am J Respir Crit Care 185: 769–778.
- 9. McShane H, Pathan AA, Sander CR, Keating SM, Gilbert SC, et al. (2004) Recombinant modified vaccinia virus Ankara expressing antigen 85A boosts BCG-primed and naturally acquired antimycobacterial immunity in humans. Nat Med 10: 1240–1244.
- 10. Beveridge NER, Price DA, Casazza JP, Pathan AA, Sander CR, et al. (2007) Immunisation with BCG and recombinant MVA85A induces long-lasting, polyfunctional Mycobacterium tuberculosis-specific CD4+ memory T lymphocyte populations. Eur J Immunol 37: 3089–3100.
- 11. Tameris MD, Hatherill M, Bernard LS, Scriba TJ, Snowden MA, et al. (2013) Safety and efficacy of MVA85A, a new tuberculosis vaccine, in infants previously vaccinated with BCG: a randomised, placebo-controlled phase 2b trial. Lancet 381: 1021–1028.
- 12. Weinstein MC, Siegel JE, Gold MR, Kamlet MS, Russell LB (1996) Recommendations of the Panel of cost -effectiveness in health and medicine. JAMA 276: 1253–1258.
- 13. United Nations International Children's Emergency Fund (2010) UNICEF South Africa Statistics 2010. Available: http://www.unicef.org/infobycountry/southafrica_statistics.html. Accessed: 7 Jun 2011.
- 14. McShane H (2013) Global progress in TB vaccine development. In Gillespie, S. editor. Respiratory Infection. The Biomedical & Life Sciences Collection, Henry Stewart Talks Ltd, London.
- 15. Violari A, Paed FC, Cotton MF, Med M, Gibb D, et al. (2008) Early antiretroviral therapy and mortality among HIV Infected infants. N Engl J Med 359: 2233–2244.
- 16. Johnson LF, Davies M, Moultrie H, Sherman GG, Bland RM, et al. (2012) The effect of early initiation of antiretroviral treatment in infants on pediatric AIDS mortality in South Africa: a model-based analysis. Pediatr Infect Dis J 5: 474–480.
- 17. Jaspan HB, Berrisford AE, Boulle AM (2008) Two-year outcomes of children on a non-nucleoside reverse transcriptase inhibitor and protease inhibitor regimens in a South African pediatric antiretroviral program. Pediatr Infect Dis J 27: 993–998.
- 18. Health Department: Republic of South Africa (2009) The National Tuberculosis Management Guidlelines 2009. Available: [http://familymedicine.ukzn.ac.za/Libraries/Guidelines_Protocols/TB_Guidelines_2009.sflb.ashx]. Accessed: 12 Jun 2011.
- 19. Chehab JC, Vilakazi-Nhlapo K, Vranken P, Peters A, Klausner JD (2012) Survey of isoniazid preventive therapy in South Africa, 2011. Int J Tuberc Lung Dis 16: 903–907.
- 20. Marais BJ, van Zyl S, Schaaf HS, van Aardt M, Gie RP, et al. (2006) Adherence to isoniazid preventive chemotherapy: a prospective community based study. Arch Dis Child 91: 762–765.
- 21. Sinanovic E, Floyd K, Dudley L, Azevedo V, Grant R, et al. (2003) Cost and cost-effectiveness of community-based care for tuberculosis in Cape Town, South Africa. Int J Tuberc Lung Dis 7: 556–562.
- 22. Wilkinson D, Floyd K, Gilks CF (1997) Cost and cost-effectiveness of alternative tuberculosis management strategies in South Africa- implications for policy. S Afr Med J 87: 451–455.
- 23. The South African National Tuberculosis Control Programme Practical Guidelines 2004. Available: http://www.kznhealth.gov.za/chrp/documents/Guidelines/Guidelines%20National/Tuberculosis/SA%20TB%20Guidelines%202004.pdf. Accessed: 2 Jul 2011.
- 24. International Drug Price Indicator Guide 2010. Available: http://erc.msh.org/dmpguide. Accessed: 7 Jun 2010.
- 25. Floyd K, Wilkinson D, Gilkds C (1997) Comparison of cost effectiveness of directly observed treatment (DOT) and conventionally delivered treatment for tuberculosis: experience from rural South Africa. BMJ 315: 1407.
- 26. UN Department of Economic and Social Affairs, Population Division (2010) World Population Prospects: the 2010 Revision Available: http://esa.un.org/wpp/Documentation/pdf/WPP2010_Volume-I_Comprehensive-Tables.pdf. Accessed: 18 Oct 2011.
- 27. Tseng C-L, Oxalade O, Menzies D, Aspler A, Schwartzman K (2011) Cost-effectiveness of novel vaccines for tuberculosis control: a decision analysis study. BMC Public Health 11: 55.
- 28. Abu-Raddad LJ, Sabatelli L, Achterberg JT, Sugimoto JD, Longini IM Jr, et al. (2009) Epidemiological benefits of more-effective tuberculosis vaccines, drugs, and diagnostics. Proc Natl Acad Sci U S A 106: 13980–13985.
- 29. Ziv E, Daley C, Blower S (2004) Potential public health impact of new tuberculosis vaccines. Emerg Infect Dis 10: 1529–1535.
- 30. Stop TB Partnership and World Health Organization (2006) Global Plan to Stop TB 2006–2015 (WHO/HTM/STB/2006.35). Available: http://www.stoptb.org/resources/publications/plans_strategies.asp. Accessed: 12 Jul 2011.
- 31. Boehme CCMD, Nabeta PMD, Hillemann D (2010) Rapid molecular detection of tuberculosis and rifampin resistance. N Engl J Med 363: 1005–1015.
- 32. World Health Organization (2011) South Africa: Health Profile 2011. Available: http://www.who.int/gho/countries/zaf.pdf. Accessed: 10 Jun 2011.
- 33. Central Intelligence Agency (2011) The World Factbook: South Africa. 2011. Available: https://www.cia.gov/library/publications/the-world-factbook/geos/sf.html. Accessed: 10 Jun 2011.
- 34. World Health Organization (2011) Immunization Profile-South Africa. 2011. Available: http://apps.who.int/immunization_monitoring/en/globalsummary/countryprofileresult.cfm?C=zaf. Accessed: 20 Jun 2011.
- 35. World Bank (2012) Gross national income per capita 2011, Atlas method and PPP. World Development Indicators database 2012. Available: http://databank.worldbank.org/databank/download/GNIPC.pdf. Accessed: 28 Nov 2012.
- 36. CIA (2011) Life Expectancy at Birth, CIA World Factbook 2011. Available: https://www.cia.gov/library/publications/the-world-factbook/rankorder/2102rank.html. Accessed: 7 Jul 2011.
- 37. Shanaube K, Sismanidis C, Ayles H, Beyers N, Schaap A, et al. (2009) Annual risk of tuberculosis infection using different methods in communities with a high prevalence of TB and HIV in Zambia and South Africa. PloS ONE 4: e7749.
- 38. Johnson L (2010) A model of paediatric HIV in South Africa. Cape Town, South Africa: University of Cape Town.
- 39. USAID (2008) South Africa Tuberculosis Profile 2008. Available: http://transition.usaid.gov/our_work/global_health/id/tuberculosis/countries/africa/southafrica.pdf. Accessed: 13 Jul 2011.
- 40. World Health Organization (2010) South Africa Tuberculosis Profile 2010. Available: http://www.doh.gov.za/docs/stats/2011/SouthAfricanTuberculosisProfile2011WHO.pdf. Accessed: 6 Jun 2011.
- 41. Surveillance TWIGPoA-TDR (2008) Anti-Tuberculosis Drug Resistance In The World: Report No. 4, 2008. Available: http://www.who.int/tb/publications/2008/drs_report4_26feb08.pdf. Accessed: 18 Jun 2011.
- 42. Oeltmann JE, Chengeta B, Mboya JJ, Wells CD, Kilmarx PH, et al. (2008) Reported childhood tuberculosis treatment outcomes, Gaborone and Francistown, Botswana, 1998–2002. Int J Tuberc Lung Dis 12: 186–192.
- 43. World Vision (2009) Overview of child health in South Africa 2009. Available: http://www.worldvision.co.za/child-health-now/overview-of-child-health-in-south-africa. Accessed: 2 Jul 2011.
- 44. Little K, Thorne C, Luo C, Bunders M, Ngongo N, et al. (2007) Disease progression in children with vertically-acquired HIV infection in sub-Saharan Africa: reviewing the need for HIV treatment. Cur HIV Res 5: 139–153.
- 45. Newell M-L, Coovadia H, Cortina-Borja M, Rollins N, Gaillard P, et al. (2004) Mortality of infected and uninfected infants born to HIV-infected mothers in Africa: a pooled analysis. Lancet 364: 1236–1243.
- 46. Department of Health, Republic of South Africa (2013) The South African antiretroviral treatment guidelines 2013. Available: http://www.sahivsoc.org/upload/documents/2013%20ART%20Guidelines-Short%20Combined%20FINAL%20draft%20guidelines%2014%20March%202013.pdf. Accessed: 26 Jul 2013.
- 47. Sutherland I (1966) The evolution of clinical tuberculosis in adolescents. Tuberculosis 47: 308.
- 48. Grzybowski S, Barnett G, Styblo K (1974) Contacts of cases of active pulmonary tuberculosis. Bull Int Union Tuberc 50: 90–106.
- 49. Menzies D (1997) Issues in the management of contacts of patients with active pulmonary tuberculosis. Can J Public Health 88: 197–201.
- 50. Stead W (1995) Management of health care workers after inadvertent exposure to tuberculosis: a guide for the use of preventive therapy. Ann Intern Med 122: 906–912.
- 51. Nolan CM, Elarth AM (1988) Tuberculosis in a cohort of Southeast Asian refugees. A five-year surveillance study. Am Rev Respir Dis 137: 805–809.
- 52. Comstock GW, Edwards LB, Livesay VT (1974) Tuberculosis morbidity in the US Navy: its distribution and decline. Am Rev Respir Dis 110: 572–580.
- 53. Whalen C, Jonson J, Okwera A, Hom DL, Huebner R, et al. (1997) A trial of three regimens to prevent tuberculosis in Ugandan adults infected with the human immunodeficiency virus. Uganda-Case Western Reserve University Research Collaboration. N Engl J Med 337: 801–808.
- 54. Guelar A, Gatell J, Verdejo J, Podzamczer D, Lozano L, et al. (1993) A prospective study of the risk of tuberculosis among HIV-infected patients. AIDS 7: 1345–1349.
- 55. Wood R, Maartens G, Lombard CJ (2000) Risk factors for developing tuberculosis in HIV-1-infected adults from communities with low or very high incidence of tuberculosis. JAMA 23: 75–80.
- 56. Edlin BR, Tokars JI, Grieco MH, Crawford JT, William J, et al. (1992) An outbreak of multi-drug resistant tuberculosis among hospitalized patients with the acquired immunodeficiency syndrome. N Engl J Med 326: 1514–1521.
- 57. Beck-Sague C, Dooley SW, Hutton MD, Otten J, Breeden A, et al. (1993) Hospital outbreak of multidrug-resistant Mycobaterium tuberculosis in patients with advanced HIV infection. N Engl J Med 328: 1137–1144.
- 58. Fischl MA, Uttamchandani RB, Daikos GL, Problete RB, Moreno JN, et al. (1992) An outbreak of tuberculosis caused by multiple-drug-resistant tubercle bacilli among patients with HIV infection. Ann Intern Med 117: 177–183.
- 59. Small PM, Shafer RW, Hopewell PC, Singh SP, Murphy MJ, et al. (1993) Exogenous reinfection with multidrug-resistant Mycobacterium tuberculosis in patients with advanced HIV infection. N Engl J Med 328: 1137–1144.
- 60. Daley C, Small P, Schecter G, Schoolnik GK, McAdam RA, et al. (1992) An outbreak of tuberculosis with accelerated progression among persons infected with the human immunodeficiency virus: An analysis using restriction-fragment-length polymorphisms. N Engl J Med 326: 231–235.
- 61. Lincoln EM (1950) Course and prognosis of tuberculosis in children. Am J Med 9: 623–632.
- 62. Grzybowski S, Enarson D (1978) The fate of cases of pulmonary tuberculosis under various treatment programmes. Bull Int Union against Tuberc 53: 70–74.
- 63. Horwitz O (1969) Public health aspects of relapsing tuberculosis. Am Rev Respir Dis 99: 183–193.
- 64. Rieder H (1999) Epidemiologic basis of tuberculosis control. First Edition ed. Paris, France: International Union Against Tuberculosis and Lung Disease.
- 65. Cohn D, Catlin B, Peterson K, Judson F, Sbarbaro J (1990) A 62-dose 6-month therapy for pulmonary and extrapulmonary tuberculosis. A twice-weekly, directly observed, and cost-effective regimen. Ann Intern Med 112: 407–415.
- 66. East African Tuberculosis Investigation Centre BMRC (1977) Results at 5 years of a controlled comparison of a 6-month and a standard 18-month regimen of chemotherapy for pulmonary tuberculosis. Am Rev Respir Dis 116: 3–8.
- 67. Somner A (1980) Short-course chemotherapy in pulmonary tuberculosis. A controlled trial by the British Thoracic Association (third report). Lancet 1: 1182–1183.
- 68. Algerian Working Group BMRC (1984) Controlled clinical trial comparing a 6-month and a 12-month regimen in the treatment of pulmonary tuberculosis in the Algerian Sahara. Am Rev Respir Dis 129: 921–928.
- 69. Benator D, Bhattacharya M, Bozeman L, Burman W, Cantazaro A, et al. (2002) Rifapentine and isoniazid once a week versus rifampicin and isoniazid twice a week for treatment of drug-susceptible pulmonary tuberculosis in HIV-negative patients: a randomised clinical trial. Lancet 360: 528–534.
- 70. Chee C, Boudville I, Chan S, Zee Y, Wang Y (2000) Patient and disease characteristics, and outcome of treatment defaulters from the Singapore TB control unit - a one-year retrospective survey. Int J Tuber Lung Dis 4: 496–503.
- 71. Parthasarathy R, Prabhakar R, Somasundaram P (1986) A controlled clinical trial of 3- and 5- month regimens in the treatment of sputum-positive pulmonary tuberculosis in South India. Am Rev Respir Dis 134: 27–33.
- 72. East African Tuberculosis Investigation Centre BMRC (1981) Controlled clinical trial of five short-course (4-month) chemotherapy regimens in pulmonary tuberculosis: Second report of the 4th study. Am Rev Respir Dis 123: 165–170.
- 73. Singapore Tuberculosis Service BMRC (1986) Long-term follow-up of a clinical trial of six-month and four-month regimens of chemotherapy in the treatment of pulmonary tuberculosis. Am Rev Respir Dis 133: 779–783.
- 74. Espinal MA, Kim SJ, Suarez PG, Kam KM, Khomenko AG, et al. (2000) Standard short-course chemotherapy for drug-resistant tuberculosis: treatment outcomes in 6 countries. JAMA 283: 2537–2545.
- 75. Nathanson E, Lambregts-van Weezenbeek C, Rich ML, Gupta R, Bayona J, et al. (2006) Multidrug-resistant tuberculosis management in resource-limited settings. Emerg Infect Dis 12: 1389.
- 76. Hessseling AC, Westra AE, Werschkull H, Donald PR, Beyers N, et al. (2005) Outcome of HIV infected children with culture confirmed tuberculosis. Arch Dis Child 90: 1171–1774.
- 77. Sonnenberg P, Murray J, Glynn JR, Shearer S, Kambashi B, et al. (2001) HIV-1 and recurrence, relapse, and reinfection of tuberculosis after cure: a cohort study in South African mineworkers. Lancet 358: 1687–1693.
- 78. Johnson JL, Okwera A, Vjecha MJ, Byekwaso F, Nakibali J, et al. (1997) Risk factors for relapse in human immunodeficiency virus type 1 infected adults with pulmonary tuberculosis. Int J Tuberc Lung Dis 1: 446–453.
- 79. Pulido F, Pena JM, Rubio R, Moreno S, Gonzalez J, et al. (1997) Relapse of tuberculosis after treatment in human immunodeficiency virus-infected patients. Arch Intern Med 157: 227–232.
- 80. Statistics South Africa (2010) Monthly Earnings of South Africa 2010. Available: http://www.statssa.gov.za/publications/P02112/P021122010.pdf. Accessed: 4 Jul 2011.
- 81. UNICEF Supplies and Logistics (2011) Vaccine Price Data 2011. Available: http://www.unicef.org/supply/index_57476.html. Accessed: 4 Dec 2011.