SYPHILIS INFECTION IN PREGNANT women is an important public health problem in sub-Saharan Africa with prevalence ranging from 3% to 17%.1–5 Untreated infection in a woman can lead to severe neurologic and cardiovascular complications,6–8 and can facilitate HIV transmission.9 Infection of the fetus can result in miscarriage, stillbirth, neonatal death, and low birth weight. Surviving infants with congenital syphilis suffer morbidity and premature mortality.8,10–13 More than 50% of children born with congenital syphilis are initially asymptomatic, making prenatal diagnosis of maternal infection vital to improving mother-child pregnancy outcomes.8,10
The devastating consequences of syphilis in both mother and child are preventable with effective screening followed by treatment with a single intramuscular dose of benzathine penicillin.6,7,11 Conventional testing for syphilis involves screening with an initial rapid plasma reagin (RPR) test followed by a confirmatory Treponema pallidum hemagglutination assay (TPHA) test of all RPR positives. However, this screening method has been difficult to implement in resource-poor settings. Staffing and operational problems pose challenges to initiation of the screening process where overtaxed and low-skill health workers may not have the time to see the benefits of screening in pregnancy. In addition, the complexities of the standard screening method require a return visit, which increases loss to follow-up, leaving many infected pregnant women undiagnosed and untreated.14–18
New rapid RPR and immunochromographic based strip (ICS) tests are easier to administer and provide same-day results.19,20 Rapid on-site, same-day testing and treatment seem promising in health clinics in sub-Saharan Africa using these tests.3–5,21,22 However, the standard screening method or, unfortunately no testing, remains the norm in many settings.17,23 Motivated by the availability of new rapid testing technology, we assess the cost-effectiveness of implementing a universal rapid point-of-care prenatal syphilis screening and treatment program within the setting of prenatal health clinics in South Africa with the potential to be generalizable to settings delivering prenatal care throughout sub-Saharan Africa.
Data from the literature were used to model the acquisition and subsequent natural history of syphilis in pregnant sub-Saharan African women over the course of their lifetime. We assessed the health and economic outcomes associated with screening strategies that varied by initial coverage, number of visits required, specific test (RPR and ICS), and need for confirmation with TPHA. We elected to use 2 comparators, no screening and conventional 2-step screening using a RPR test followed by a TPHA confirmatory test, because both of these serve as the baseline level of care depending on specific setting. Model outcomes include adverse pregnancy outcomes (miscarriage, low birth weight, congenital syphilis, stillbirth, and neonatal death), maternal life expectancy, and lifetime costs (2004 US dollars). Life-expectancy gains for children were estimated using a separate model simulating children born with and without congenital syphilis. We used the reduction in total discounted life years lost due to deaths of both mothers and their children as the primary measure of clinical benefit.
We conducted a cost-effectiveness analysis following methodological recommendations in published guidelines.24,25 The performance of alternative screening strategies is expressed using the incremental cost-effectiveness ratio, defined as the additional cost of a specific screening strategy, divided by its additional clinical benefit, compared with the next most expensive strategy. Final incremental cost-effectiveness ratios are calculated after eliminating both less effective strategies that have lower discounted life expectancy and higher discounted lifetime costs than an alternative strategy (i.e., “strongly dominated” strategies) and strategies with a higher incremental cost-effectiveness ratio than a more effective alternate strategy (i.e., “weakly dominated” strategies). The one exception is for strategies that provide greater health benefits, but have lower discounted lifetime costs than the baseline comparator (no screening or standard screening); these strategies are referred to informally as “cost saving” and the absolute difference in lifetime costs relative to the baseline comparator is provided. Sensitivity analyses assess the effect of varying baseline estimates and assumptions on our results.
Our computer-based model simulates the natural history of pregnancy, the risk of maternal syphilis, and the effect of syphilis infection on maternal and neonatal outcomes using a sequence of transitions between health states (Fig. 1). Health states, descriptive of each individual’s true underlying health, are defined to include stage of pregnancy (first, second, or third trimester), infection status, and if infected, distinguish the stage of syphilis (primary, secondary, early latent, and late latent with severe late infection referred to as tertiary syphilis).
Girls enter the model at age 15 and are followed throughout their lifetime. We assume an average of 6 pregnancies per person, derived from reported fertility rates for sub-Saharan Africa, and an average age at first pregnancy of approximately 16 years old.26,27 The incidence of syphilis infection is conditional on age with the highest rates of infection occurring under the age of 25.28 Among infected women, transmission from mother to child can occur throughout the pregnancy, and is conditional on disease stage with the highest rates of transmission occurring with primary, secondary, or early latent syphilis.10,15,29 Adverse pregnancy outcomes include miscarriage, low birth weight, congenital syphilis, stillbirth, and neonatal death. Miscarriages that occur during the third trimester are classified as stillbirths. Women can die of a pregnancy-related illness, syphilis-related complications, or other causes.
We assessed the comparative costs and outcomes associated with no screening; conventional 2-step screening using a RPR test followed by a TPHA confirmatory test; single-visit rapid RPR; and single-visit rapid ICS. The primary objective of our analysis was to assess the incremental cost-effectiveness of new single-visit rapid tests compared with the status quo—accordingly, we included as comparators no screening and conventional 2-step screening (RPR followed by TPHA in RPR-positives). Conventional 2-step screening differs from the rapid test strategies in that the confirmatory TPHA test is usually performed in a central laboratory facility. This requires women to return to the clinic approximately 2 weeks after the initial test to receive test results and, when necessary, receive treatment. In contrast, rapid tests are designed for use in rural and urban settings with minimal laboratory facility capabilities, allowing for same-day, on-site testing and treatment, thereby minimizing loss to follow-up and additional laboratory costs. The 2 rapid tests have different attributes. ICS is simpler to perform, requiring a finger stick blood sample as opposed to a venous blood draw, and provides more easily interpretable results than RPR.5,20 Although both rapid tests identify antibodies in the blood, the RPR test is nontreponemal specific, using cardiolipin-cholesterol-lecithin antigen as opposed to a syphilis specific antigen.8,30 Although several studies report a range of 93% to 98% for RPR specificity,20 there are reports of high false-positive rates of up to 28% of RPR positive women.1,20 The ICS test, on the other hand, is treponemal pecific, detecting antibodies specific to syphilis antigens. Treponemal specific tests have a better sensitivity profile when all stages of syphilis are considered, but remain positive for life which has potential implications for a higher false-positive rate in women who are repeatedly screened.20,30
We assumed screening uptake varies with stage of pregnancy because women are more likely to make their first antenatal visit during their second or third trimester.3,14,18,22,31,32 Although 3 weekly injections are recommended by the Centers for Disease Control and Prevention for late latent syphilis (or latent syphilis of unknown duration), we assumed treatment would be a single injection of 2.4 MU benzathine penicillin for all stages of syphilis, because this would be most feasible in a rural or resource-poor setting,1–3,23,33 and treatment failure rate approximates only 2%.12,34,35 We explored the potential impact of toxicity to penicillin in sensitivity analysis.
Selected clinical variables used in the model are shown in Table 1. Based on published data from sub-Saharan Africa, we assumed an average prevalence of 6%, with 2% primary or secondary syphilis and 4% with latent syphilis.1–5,8,14,15,17,18,22,28,31,32,36,37
Selected cost variables are shown in Table 2. Direct medical costs were based on secondary data (largely from South Africa) and reflected labor, counseling, laboratory equipment, testing and treatment supplies including the cost of penicillin, swabs, and syringes.19,21,37 Costs of hospitalization for treatment of adverse pregnancy outcomes were derived from published data on hospital cost of care for severely ill neonates and costs associated with congenital syphilis.21,38–41 Nonmedical costs were derived from the published literature.42 Costs are expressed in 2004 US dollars.43,44
Model-projected adverse pregnancy outcomes were compared with those from 5 published studies assessing the cost-effectiveness of screening strategies.3,5,21,31,45,46 Using base case values from comparator studies, the model produced similar values for congenital syphilis, low birth weight, stillbirth, and neonatal death (Table A1).
Over the lifetime of a population of 1000 women, or approximately 6000 lifetime pregnancies, single-visit rapid ICS averted 178 cases of congenital syphilis, 43 cases of low birth weight, 8 neonatal deaths, and 29 stillbirths in comparison to no screening (Table 3). A total gain of 303 healthy births was achieved using this strategy due to a combination of averted miscarriages in early pregnancy and prevented full-term pregnancy complications. Although single-visit rapid RPR was less effective than ICS, it was more effective than the standard two-visit RPR-TPHA strategy.
Table 4 shows the discounted and undiscounted life-expectancy gains and average lifetime costs per 1000 women and respective children associated with each strategy compared with no screening. Single-visit screening with ICS provided a gain of 2297 years of discounted life expectancy (6881 years undiscounted) with more than 95% of these years attributable to children with averted congenital syphilis. This strategy allowed for a cost savings of $170,030 per 1000 women over their lifetime compared with no screening. Although single-visit RPR was more costly and less effective compared with no screening, it provided a gain of 2118 discounted years (6364 undiscounted years) and saved $161,310.
Variations in test performance, test costs, and costs associated with congenital syphilis had the greatest effect on results. The choice between the 2 single-visit strategies was influenced by their comparative test sensitivity, and the cost of the ICS test kit, labor, and supplies. A screening strategy was considered the preferred strategy if it achieved greater benefits at lower cost than all other strategies.
Figure 2 shows selected results of sensitivity analyses. If the costs associated with congenital syphilis treatment were less than $200, screening with single-visit tests were no longer cost saving but had cost-effectiveness ratios less than per capita GDP for South Africa, a suggested threshold for very cost-effective strategies.47 If ICS test-related costs exceeded 2 times their base case value, the incremental cost-effectiveness ratio for single-visit rapid ICS testing was $14/LY compared with single-visit rapid RPR testing.
If overall test sensitivity of ICS fell below 88% (base case of 97%), single-visit rapid ICS was no longer preferred to single-visit rapid RPR. Similarly, if the test sensitivity of single-visit rapid RPR exceeded 92% (base case of 80%), then it was preferred to single-visit rapid ICS (Fig. 3). The overall policy results were not altered by plausible changes in the base case values of RPR or ICS specificity (96.4% and 94.1%, respectively) or by varying the very small potential mortality risk from fatal reactions to penicillin from 0.0001 to 0.001.
Unless the probability of returning for a second visit exceeded 87% and the test sensitivity of the RPR-TPHA test system was greater than 97%, single-visit rapid testing with ICS was always preferred.
We assessed the health and economic outcomes associated with new screening strategies that use same-visit rapid point-of-care tests. Screening with either ICS or RPR, in a single-visit screen and treat strategy, prevented substantial morbidity and mortality; over the lifetime of a cohort of women, considerable resources were saved by averting costly health consequences in the mother and child. Conventional 2-step screening using the RPR test followed by the TPHA confirmatory test was not as effective or cost-effective as either of the rapid point-of-care tests.
Compared with no screening, single-visit rapid ICS testing substantially decreased negative pregnancy outcomes, averting 178 cases of congenital syphilis, 43 cases of low birth weight, and 37 perinatal deaths while achieving an increase of 303 healthy births per 1000 women over the course of their lifetime. These gains in positive pregnancy outcomes were associated with discounted life expectancy gains of 60 and 2237 years for women and children, respectively, in addition to cost savings that exceeded $150,000 per 1000 women. Rapid RPR screening produced benefits that were slightly lower, but comparable in magnitude.
Although the expected values of rapid ICS and RPR were similar, relative to standard RPR-TPHA screening, the specific choice between rapid ICS and RPR testing was influenced by the relative cost and sensitivity of rapid RPR and ICS tests. Compared with RPR screening, ICS screening was no longer preferred when the ICS test-related costs increased more than 2-fold, when ICS test sensitivity fell below 88%, and/or when RPR test sensitivity increased to above 92%. Standard RPR-TPHA screening remained significantly disadvantaged with respect to either rapid screening method because of lower test system sensitivity, loss to follow-up, and additional testing, transportation, and time costs associated with the multiple visits required. Rapid on-site screening does not require sample transport and eliminates the need for women to make a second clinic visit since test results and treatment are given the same day, circumventing many of the problems associated with standard RPR-TPHA testing.
Resource-poor regions face significant demands on limited resources. Information on cost-effectiveness can be one useful component to priority setting, although other concerns such as cultural preferences, acceptability, equity, and feasibility are equally important to consider.48 Syphilis screening and treatment is highly cost-effective compared with other well-accepted public health interventions such as insecticide treated bed nets to prevent malaria and diphtheria-pertussis-tetanus, polio, and measles vaccinations.48 Although there is no consensus on a cost-effectiveness threshold above which an intervention would not be cost-effective, heuristics have been suggested. For example, the World Bank has described interventions costing $100 or less per life year saved as highly cost-effective.48 More recently, the Commission on Macroeconomics and Health has suggested that interventions with ratios below per capita GDP should be considered “very cost-effective.”47 Using this heuristic, strategies that have lower discounted lifetime costs and higher life-expectancy gains (i.e., cost saving) are considered very cost-effective. Per capita GDP in sub-Saharan Africa ranges between $600 and $13,700; South Africa’s per capita GDP is approximately $13,300, whereas Malawi has a very low per capita GDP of $600.49 We found single-visit screening and treatment using rapid tests would be cost-effective in even the poorest countries in sub-Saharan Africa.47,48
One must be cautious to not treat the “discounted lifetime” costs associated with the numerator of the cost-effectiveness ratio as “short-term financial costs” associated with, for example, an annual budget analysis. Although a cost-effectiveness analysis provides information on value for money, it is not equivalent to providing budgetary information on affordability and up-front financial costs required of the payor. Considerations of affordability may require one to choose one strategy over another given similar cost-effectiveness ratios. As such, there may be a rationale in some circumstances for choosing the rapid RPR test ($1.97) over the rapid ICS test ($3.18) if those who pay for screening are not the same as those who pay for negative neonatal outcomes. However, because the costs associated with congenital syphilis occur in close proximity to the pregnancy during which a woman would be screened, the cost savings associated with a screening program would occur over a relatively short time horizon. The implication of this fact, provided the parties who pay for screening are the same as those who pay for negative neonatal outcomes, would be a substantial savings in “financial costs.” For example, for 5000 women screened in a 1-year period, the difference in screening costs between RPR and ICS would approximate $6000, but ICS would save more than $67,000. This net of $61,000 could be invested in other health programs in need of immediate financial resources. Of course, other attributes of these 2 rapid tests (e.g., provider preferences) may also influence the choice between them.
In addition, specific attributes of ICS and RPR tests may potentially disadvantage their use. The RPR test is nontreponemal specific, which may result in substantial numbers of biologic false positives.1 In contrast, the ICS test is treponemal specific, detecting antibodies specific to syphilis antigens; while treponemal specific tests have a better test sensitivity profile, they remain positive for life which implies that among women who are repeatedly screened, a proportion who are not infected would have a positive test result.20,30 In both cases, false positive test results will result in the unnecessary treatment of uninfected women, which in turn will increase costs, the probability of treatment of side effects or reactions, and potentially contribute to resistance. That being said, the net impact of overtreatment in this scenario is expected to be small relative to the magnitude of the benefits, given that many women in sub-Saharan Africa are not screened, the number of previously screened and treated women is low, and rates of reinfection are high. In populations at lower risk of syphilis infection, however, these considerations may be more influential and our assumptions revisited.
Our analysis has several important limitations. There are formidable data gaps and the quality of data are heterogeneous. We restricted our analysis to pregnant women, and did not include the potential benefits from reduced transmission to partners, or from preventing syphilis-enhanced HIV transmission. We assumed injectable penicillin would be available, but there is the potential for supply shortages or a lack of trained personnel; although oral azithromycin is a possible alternative treatment, which would eliminate the need for injections, it is more expensive and there are concerns regarding macrolide resistance and the associated risk of treatment failure.33,50–53 Cost data were limited for tertiary syphilis sequelae as was information regarding life expectancy of syphilis-affected infants stratified by neonatal outcomes. Although we included the costs and potential mortality risk associated with adverse drug reactions to penicillin in a sensitivity analysis, and these had little influence on the overall policy results, we did not include the potential negative consequences because of resistance. Although these are likely to be minimal relative to the overall benefits of treatment, this is an area that should be revisited as data become available. In addition, in settings where the overall prevalence of syphilis is very low, and the benefits of screening are expected to be far lower, the tradeoffs associated with unnecessary treatment of women with false-positive results in terms of costs, toxicity, and resistance may be different. Our results may not be generalizable to specific scenarios and settings; as such, they are intended to primarily provide qualitative insight into the health and economic outcomes associated with syphilis screening and the components of the prenatal visit for pregnant women.
Our findings are generally consistent with those reported in other studies examining syphilis screening during pregnancy in similar populations.3,5,21,31,45,46,54 However, our results do differ from many studies in that we found rapid screening to be cost saving. This is likely a consequence of adopting a lifetime perspective and taking into consideration multiple adverse outcomes to mother and infant as well as patient costs. The most comparable syphilis screening analysis was collaborative work conducted by Bronzan et al. and Blandford et al., which examined the costs and benefits of RPR, ICS, and standard screening for preventing maternal and congenital syphilis over a single pregnancy as opposed to over a lifetime.5,21 Blandford found that screening with RPR or standard screening both resulted in lower total program costs but averted substantially fewer cases of congenital syphilis than ICS.21 These findings likely differ from our own because of assumptions about screening test performance (e.g., Blandford et al. assume a sensitivity and specificity of 100% with RPR-TPHA for all stages of syphilis compared with a sensitivity of 39%–71% for RPR alone) and the omission of patient time and transportation costs associated with returning for RPR-TPHA test results. However, sensitivity analyses demonstrated that ICS remained the most effective of the screening strategies in all but the most extreme scenarios.21 Bronzan also found the ICS test was preferred by nurses because it was quicker to perform and easier to interpret than the RPR test while also offering higher sensitivity.5 User acceptability is critical for facilitating uptake of screening programs by healthcare professionals who have many competing demands placed on their time.
Syphilis infection is a preventable cause of significant maternal and child mortality and morbidity. Timely diagnosis and treatment of syphilis can substantially reduce if not eliminate disease sequelae in pregnancy, childhood and adulthood, facilitating safe motherhood, and improving newborn health. The rapid ICS test is user-friendly, test kits are economically priced and results can be obtained the same day, making it possible to test and treat patients in a single prenatal visit. In situations where ICS tests are not available, the rapid RPR test is a good alternative. Treatment with penicillin is simple and inexpensive with a high probability of cure. Application of rapid point-of-care testing technologies and treatment in resource-poor settings where the burden of disease is greatest is as cost-effective as childhood vaccination. As efforts to improve maternal health take on a renewed priority, this analysis provides unequivocal support for prenatal care that includes syphilis testing.
1. Watson-Jones D, Changalucha J, Gumodoka B, et al. Syphilis in pregnancy in Tanzania. I. Impact of maternal syphilis on outcome of pregnancy. J Infect Dis 2002; 186:940–947.
2. Watson-Jones D, Gumodoka B, Weiss H, et al. Syphilis in pregnancy in Tanzania. II. The effectiveness of antenatal syphilis screening and single-dose benzathine penicillin treatment for the prevention of adverse pregnancy outcomes. J Infect Dis 2002; 186:948–957.
3. Jenniskens F, Obwaka E, Kirisuah S, et al. Syphilis control in pregnancy: Decentralization of screening facilities to primary care level, a demonstration project in Nairobi, Kenya. Int J Gynaecol Obstet 1995; 48(suppl):S121–S128.
4. Temmerman M, Gichangi P, Fonck K, et al. Effect of a syphilis control programme on pregnancy outcome in Nairobi, Kenya. Sex Transm Infect 2000; 76:117–121.
5. Bronzan RN, Mwesigwa-Kayongo DC, Narkunas D, et al. On-site rapid antenatal syphilis screening with an immunochromatographic strip improves case detection and treatment in rural South African clinics. Sex Transm Dis 2007; 34(7 suppl):S55–S60.
6. Hook EW III, Marra CM. Acquired syphilis in adults. N Engl J Med 1992; 326:1060–1069.
7. Garnett GP, Aral SO, Hoyle DV, et al. The natural history of syphilis. Implications for the transmission dynamics and control of infection. Sex Transm Dis 1997; 24:185–200.
8. Holmes KK, Sparling PF, Mardh P, et al., eds. Sexually Transmitted Diseases, 3rd ed. New York: McGraw-Hill Companies, 1999:chap 33–36, 84, 92 and 103.
9. Reynolds SJ, Risbud AR, Shepherd ME, et al. High rates of syphilis among STI patients are contributing to the spread of HIV-1 in India. Sex Transm Infect 2006; 82:121–126.
10. Wicher V, Wicher K. Pathogenesis of maternal-fetal syphilis revisited. Clin Infect Dis 2001; 33:354–363.
11. Darville T. Syphilis. Pediatr Rev 1999; 20:160–164; quiz 165.
12. Genc M, Ledger WJ. Syphilis in pregnancy. Sex Transm Infect 2000; 76:73–79.
13. Clark EG, Danbolt N. The Oslo study of the natural history of untreated syphilis; an epidemiologic investigation based on a restudy of the Boeck-Bruusgaard material; a review and appraisal. J Chronic Dis 1955; 2:311–344.
14. Bique Osman N, Challis K, Folgosa E, et al. An intervention study to reduce adverse pregnancy outcomes as a result of syphilis in Mozambique. Sex Transm Infect 2000; 76:203–207.
15. Donders GG, Desmyter J, Hooft P, et al. Apparent failure of one injection of benzathine penicillin G for syphilis during pregnancy in human immunodeficiency virus-seronegative African women. Sex Transm Dis 1997; 24:94–101.
16. Mullick S, Beksinksa M, Msomi S. Treatment for syphilis in antenatal care: Compliance with the three dose standard treatment regimen. Sex Transm Infect 2005; 81:220–222.
17. Gloyd S, Chai S, Mercer MA. Antenatal syphilis in sub-Saharan Africa: Missed opportunities for mortality reduction. Health Policy Plan 2001; 16:29–34.
18. Myer L, Wilkinson D, Lombard C, et al. Impact of on-site testing for maternal syphilis on treatment delays, treatment rates, and perinatal mortality in rural South Africa: A randomised controlled trial. Sex Transm Infect 2003; 79:208–213.
19. Peeling R. The Sexually Transmitted Diseases Diagnostics Initiative: Laboratory Based Evaluation of Rapid Syphilis Diagnostics. Geneva: WHO, 2003.
20. Peeling RW, Ye H. Diagnostic tools for preventing and managing maternal and congenital syphilis: An overview. Bull World Health Organ 2004; 82:439–446.
21. Blandford JM, Gift TL, Vasaikar S, et al. Cost-effectiveness of on-site antenatal screening to prevent congenital syphilis in rural eastern Cape Province, Republic of South Africa. Sex Transm Dis 2007; 34(7 suppl):S61–S66.
22. Beksinska ME, Mullick S, Kunene B, et al. A case study of antenatal syphilis screening in South Africa: Successes and challenges. Sex Transm Dis 2002; 29:32–37.
23. Watson-Jones D, Oliff M, Terris-Prestholt F, et al. Antenatal syphilis screening in sub-Saharan Africa: Lessons learned from Tanzania. Trop Med Int Health 2005; 10:934–943.
24. Siegel JE, Weinstein MC, Russell LB, et al. Recommendations for reporting cost-effectiveness analyses. Panel on Cost-Effectiveness in Health and Medicine. JAMA 1996; 276:1339–1341.
25. Weinstein MC, Siegel JE, Gold MR, et al. Recommendations of the Panel on Cost-effectiveness in Health and Medicine. JAMA 1996; 276:1253–1258.
27. 2004 World Population Data Sheet of the Population Reference Bureau: Population Reference Bureau, 2004.
28. Todd J, Munguti K, Grosskurth H, et al. Risk factors for active syphilis and TPHA seroconversion in a rural African population. Sex Transm Infect 2001; 77:37–45.
29. Hollier LM, Harstad TW, Sanchez PJ, et al. Fetal syphilis: Clinical and laboratory characteristics. Obstet Gynecol 2001; 97:947–953.
30. Hicks CB. Serologic testing for syphilis. UpToDate. December 8, 2006. Available at: http://uptodate.com
. Accessed April, 2007.
31. Hira SK, Bhat GJ, Chikamata DM, et al. Syphilis intervention in pregnancy: Zambian demonstration project. Genitourin Med 1990; 66:159–164.
32. Rotchford K, Lombard C, Zuma K, et al. Impact on perinatal mortality of missed opportunities to treat maternal syphilis in rural South Africa: Baseline results from a clinic randomized controlled trial. Trop Med Int Health 2000; 5:800–804.
33. Centers for Disease Control and Prevention. Sexually transmitted diseases treatment guidelines, 2006. MMWR Recomm Rep 2006; 55(RR-11):1–94.
34. Sheffield JS, Sanchez PJ, Morris G, et al. Congenital syphilis after maternal treatment for syphilis during pregnancy. Am J Obstet Gynecol 2002; 186:569–573.
35. Wendel GD Jr, Sheffield JS, Hollier LM, et al. Treatment of syphilis in pregnancy and prevention of congenital syphilis. Clin Infect Dis 2002; 35(suppl 2):S200–S209.
36. West B, Walraven G, Morison L, et al. Performance of the rapid plasma reagin and the rapid syphilis screening tests in the diagnosis of syphilis in field conditions in rural Africa. Sex Transm Infect 2002; 78:282–285.
37. Montoya PJ, Lukehart SA, Brentlinger PE, et al. Comparison of the diagnostic accuracy of a rapid immunochromatographic test and the rapid plasma reagin test for antenatal syphilis screening in Mozambique. Bull World Health Organ 2006; 84:97–104.
38. Bateman DA, Phibbs CS, Joyce T, et al. The hospital cost of congenital syphilis. J Pediatr 1997; 130:752–758.
39. Malan AF, Ryan E, van der Elst CW, et al. The cost of neonatal care. S Afr Med J 1992; 82:417–419.
40. Koontz SL, Molina de Perez O, Leon K, et al. Treating incomplete abortion in El Salvador: Cost savings with manual vacuum aspiration. Contraception 2003; 68:345–351.
41. Johnson BR, Benson J, Bradley J, et al. Costs and resource utilization for the treatment of incomplete abortion in Kenya and Mexico. Soc Sci Med 1993; 36:1443–1453.
42. Goldie SJ, Kuhn L, Denny L, et al. Policy analysis of cervical cancer screening strategies in low-resource settings: Clinical benefits and cost-effectiveness. JAMA 2001; 285:3107–3115.
44. U.S. Department of Labor Bureau of Labor Statistics: Inflation Calculator. Available at: http://www.bls.gov/cpi/
. Accessed January 29, 2007.
45. Terris-Prestholt F, Watson-Jones D, Mugeye K, et al. Is antenatal syphilis screening still cost effective in sub-Saharan Africa. Sex Transm Infect 2003; 79:375–381.
46. Fonck K, Claeys P, Bashir F, et al. Syphilis control during pregnancy: Effectiveness and sustainability of a decentralized program. Am J Public Health 2001; 91:705–707.
47. WHO. Commission on Macroeconomics and Health. Macroeconomics and health: Investing in health for economic development. Report of the Commission on Macroeconomics and Health. Geneva: WHO, 2001.
48. Disease Control Priorities in Developing Countries. 2nd ed. New York: Oxford University Press and The World Bank, 2006.
50. Hook EW III, Martin DH, Stephens J, et al. A randomized, comparative pilot study of azithromycin versus benzathine penicillin G for treatment of early syphilis. Sex Transm Dis 2002; 29:486–490.
51. Lukehart SA, Godornes C, Molini BJ, et al. Macrolide resistance in Treponema pallidum
in the United States and Ireland. N Engl J Med 2004; 351:154–158.
52. Mitchell SJ, Engelman J, Kent CK, et al. Azithromycin-resistant syphilis infection: San Francisco, California, 2000–2004. Clin Infect Dis 2006; 42:337–345.
53. Riedner G, Rusizoka M, Todd J, et al. Single-dose azithromycin versus penicillin G benzathine for the treatment of early syphilis. N Engl J Med 2005; 353:1236–1244.
54. Schackman BR, Neukermans CP, Fontain SN, et al. Cost-effectiveness of rapid syphilis screening in prenatal HIV testing programs in Haiti. PLoS Med 2007; 4:e183.
55. Simpson JL. Chapter 22—Fetal wastage. In: Gabbe: Obstetrics—Normal and Problem Pregnancies, 4th ed. Churchill Livingstone, 2002:729–730.
56. Zarakolu P, Buchanan I, Tam M, et al. Preliminary evaluation of an immunochromatographic strip test for specific Treponema pallidum
antibodies. J Clin Microbiol 2002; 40:3064–3065.
57. National Center for Health Statistics. December 16, 2004. Available at: http://www.cdc.gov/nchs/
. Accessed August 15, 2005.
Supplementary Model Information
A Markov model consists of a set of mutually exclusive and collectively exhaustive health states. A person entering the model can reside in only 1 health state at a given time. When an individual is in a specific health state they are indistinguishable from other individuals in that state. Transitions from 1 state to the next occur at defined recurring intervals (Markov cycles) of equal length (e.g., monthly or yearly) and are based on natural history data and probabilities from the literature. Our model cycle length is 3 months and our transition probabilities from 1 health state to another represent the probability of transitioning to that state within a 3-month time frame. Transition probabilities are based on population characteristics, including age, sex, and disease status. They may be constant or time-dependent as necessitated by the problem being modeled. This framework of transition probabilities and health states is used to allocate the population to specific health states and, at each cycle, reallocate these individuals into different health states over time. Costs and utilities are assigned to the respective health states to reflect costs and benefits associated with being in that state for 1 Markov cycle. Figure 4 illustrates how the average weighted cost of RPR-TPHA testing is calculated within the model. At the end of each cycle, these costs and benefits accrue and the total contribution of these values, including life expectancy, quality-adjusted life expectancy, and lifetime costs, are dependent on the length of time spent in the various health states. Synthesis of these model outputs allows for the comparison of different outcomes across strategies.42
We assume the duration of early latent syphilis is 2 years. Because of the extended duration of early latent and late latent disease, we assumed that if a woman became pregnant during either of these stages, she would incur the respective risks associated with that stage for her entire pregnancy. If a woman went on to develop tertiary syphilis, she no longer had the chance of becoming pregnant either because of advanced age or severity of disease. The majority of signs of tertiary syphilis would be apparent by 25 years after infection at which point it was assumed that all individuals remaining in the late latent stage of disease were asymptomatic and would not suffer the severe complications of infection.6–8,11
In terms of screening, if a woman was tested (and treated) during an earlier trimester of pregnancy, this did not protect her from acquiring syphilis later in the pregnancy or prevent her from being retested at a later date. Thus, our model includes the possibility of being rescreened throughout the pregnancy.
To place our results in the context of other published studies we performed analyses using the specific parameter values described in each of the studies identified (Table A1). Some studies provided more detailed descriptions of their unique setting, whereas others did not. No study specified all the parameters used in our model. As a result of incomplete specification and natural variation, our results are comparable but not identical to those reported by other researchers.