Halpern, Michael T.a; Read, Jennifer S.b; Ganoczy, Dara A.a; Harris, D. Robertc
Maternal-child transmission of human immunodeficiency virus type 1 (HIV) infection is the most common etiology of pediatric HIV infection in the US . In the US and several other countries, interventions aimed at decreasing mother–child transmission of HIV have consisted of zidovudine (ZDV) prophylaxis  and avoidance of breastfeeding [3,4]. Strong evidence has emerged that HIV-infected women with elective cesarean section (ECS) delivery – cesarean section performed before rupture of membranes and before onset of labor – are at significantly lower risk of transmitting HIV infection to their children [5,6]. The American College of Obstetricians and Gynecologists' Committee on Obstetric Practice recently has recommended that HIV-infected women be offered elective, or scheduled, cesarean section delivery to reduce the risk of vertical transmission of HIV . Thus, ECS may represent an additional intervention to decrease mother–child transmission of HIV.
Estimation of the potential overall costs and the incremental cost-effectiveness (cost per case of pediatric HIV infection prevented with ECS) would assist in the development of clinical and public health guidelines regarding the role of ECS in preventing vertical transmission of HIV. A recent cost-effectiveness analysis from the United Kingdom evaluated ECS . However, this analysis was performed prior to the completion of recent studies more precisely quantifying the impact of ECS, and was based on different practice patterns and costs than those existing in the US. The objective of the present study was to develop a decision analysis model to evaluate the cost-effectiveness and cost–benefit of ECS in comparison with vaginal delivery to prevent vertical transmission of HIV in the US.
Materials and methods
The decision analysis model was developed using Microsoft Excel (Microsoft Corporation, Redmond, Washington, USA). ECS was defined as cesarean section performed before onset of labor and rupture of membranes. There were two parts to the analysis: an individual-based analysis and a population-based analysis. While the individual-based analysis presents the per-patient costs and outcomes of ECS, the population-based analysis provides broader information on group cost and outcome changes for policy decisions and other applications.
An individual-based analysis was conducted to assess the costs and outcomes of different strategies to prevent vertical transmission of HIV on an individual level (i.e., the cost and outcome per HIV-infected pregnant woman). The modeling was based on pregnant women known to be HIV-infected. Each woman was assumed to have received HIV counseling and testing according to current US Public Health Service (USPHS) recommendations .
The costs and outcomes of ECS versus vaginal delivery were examined in three antiretroviral therapy (ART) scenarios: no ART, ZDV only, or combination ART. According to current USPHS guidelines , it is recommended that HIV-infected pregnant women and their infants receive ZDV prophylaxis. Although significant progress has been made in implementing these recommendations [10–13], some HIV-infected women and their infants do not receive ZDV prophylaxis or any other ART. Finally, an increasing proportion of HIV-infected women are receiving combination ART during pregnancy [10,14] for their own health, according to current USPHS guidelines .
Delivery of the infant was assumed to occur either via ECS or vaginal delivery. We assumed that all women not receiving ECS delivered vaginally.
Refraining from breastfeeding is an important component of preventing vertical HIV transmission . Since breastfeeding by HIV-infected women is extremely uncommon in the US [11,16], the model assumed HIV-infected women refrain from breastfeeding following delivery. As all children were assumed to receive formula, the cost of formula was not included in the model.
The model was developed from a health-care system perspective; therefore, only direct costs associated with ART and delivery, as well as discounted lifetime medical care for the HIV-infected child, were included. Indirect costs such as lost wages and direct non-medical costs such as transportation to medical care facilities were not incorporated. For HIV-infected women, the model time frame includes pregnancy, delivery, and the postpartum period. The model did not include costs of treatment of maternal HIV disease (except for combination ART received during pregnancy) or costs of routine pregnancy care not related to prevention of vertical transmission of HIV (such as ultrasonography for dating of pregnancy). Remote costs related to performance of ECS, such as need for subsequent cesarean delivery, also were not included. Further, as the model incorporated use of ART for only a relatively short period of time (3.2 months), no subsequent impact of this therapy on maternal health was assumed.
Vertical transmission rates used in the base model and in sensitivity analyses are presented in Table 1. HIV transmission rates according to mode of delivery for mother–child pairs not receiving ART or receiving ART during the antepartum, intrapartum, and postnatal periods (likely representing ZDV prophylaxis) were obtained from a published study  for the base model. Sensitivity analyses explored the impact of lower  or higher transmission rates with ECS. For the ZDV prophylaxis arm, ZDV was assumed to have been administered according to USPHS recommendations . Few data exist regarding vertical transmission rates among women receiving combination ART, including highly active ART. The reported vertical transmission rate among women receiving ZDV and lamivudine  was used to estimate the rate of vertical transmission among women receiving combination ART and delivering vaginally. Due to the paucity of data, transmission rates for women receiving combination ART and ECS were estimated proportional to the corresponding rates for women receiving ZDV only. Vertical transmission rates of 0.071 for women receiving ZDV and vaginal delivery, and 0.034 for those receiving ZDV and ECS, have been reported . The ratio of these rates is 2.088. A vertical transmission rate of 0.03 among women receiving combination ART (ZDV and lamivudine) and vaginal delivery also has been reported . Therefore, maintaining the same ratio, the transmission rate among women receiving combination ART and ECS was calculated as 0.03/2.088, or approximately 0.01 (rounded to two decimal places). Postpartum morbidity rates by mode of delivery (8.9% among women with ECS and 4.9% among women with vaginal delivery) were obtained from a published study  for the base model. Sensitivity analyses explored the impact of higher  or lower postpartum morbidity rates according to mode of delivery. Postpartum morbidity may include such events as: endometritis, surgical wound infections, urinary tract infection, thromboembolic disorders, anemia, and anesthesia-associated complications.
Estimated costs for ART (ZDV prophylaxis or combination ART), mode of delivery (ECS or vaginal delivery) with and without postpartum morbidity, and pediatric care (neonatal care and treatment of perinatally-acquired HIV infection) were obtained from the medical literature and inflated to 1998 dollars (Table 2). Costs for ZDV prophylaxis include a total of 3.2 months of therapy during pregnancy and during the intrapartum and postnatal periods . Costs for combination ART were estimated for a corresponding time period during pregnancy [20,21]. As the transmission rates in the combination ART model arm were based on data for women receiving ZDV plus lamivudine, the cost for combination ART was based on the average wholesale price for these two medications. For mothers who received combination ART during pregnancy, ZDV prophylaxis was assumed to have been administered in the intrapartum and postnatal periods also, as per current USPHS guidelines . For a majority of the analyses in this study, ECS was compared to vaginal delivery among women receiving identical ART. Thus, the cost of ART was the same in each comparison group and did not affect the results. The cost of ART was important only when women receiving one type of therapy were compared to those receiving a different therapy; this analysis is presented at the end of the Results section for the individual-based analysis.
Costs associated with vaginal delivery and cesarean section, with and without postpartum morbidity, as well as costs of neonatal care according to mode of delivery, were derived from a recently published study . As cost data are not available by specific types of postpartum morbidity, the model included associated costs only for the presence or absence of any postpartum morbidity. A published estimate of treatment costs for pediatric HIV disease (discounted at 5%) was used for the base model . Sensitivity analyses explored the impact of higher (non-discounted costs)  of treatment of pediatric HIV disease.
Total costs (including costs for ART, mode of delivery, postpartum morbidity, and pediatric care) were determined for each ART regimen/mode of delivery arm of the model, and the costs relative to benefits for each comparison then were assessed using cost–benefit and cost-effectiveness analyses. In these analyses, comparisons of costs relative to benefits (health outcomes) of different health care interventions are made.
The general formulation for a benefit–cost ratio is: (BENEFITa –BENEFITb)/(COSTa –COSTb), where:BENEFITa and BENEFITb represent the economic benefit resulting from intervention `a' (e.g., the decrease in expenditures for HIV-infected children) and from intervention `b';COSTa and COSTb represent the total cost of interventions `a' and `b', respectively. Benefit–cost ratios are expressed as a single number. Values greater than 1.0 indicate the intervention of interest (relative to the comparison intervention) saves money; for example, a benefit–cost ratio of 2.0 indicates that for every dollar spent on intervention `a', two dollars are saved in health-care expenditures in comparison with intervention `b'. Ratios less than 1.0 indicate the intervention costs more than it saves, relative to the comparison intervention. Most medical care interventions are not cost saving (i.e.,, benefit–cost ratios < 1.0). However, interventions that are not cost saving may be cost-effective.
The general formulation of a cost-effectiveness ratio is: (COSTa –COSTb)/(OUTCOMEa –OUTCOMEb), where:COSTa and COSTb represent the net cost of intervention `a' (e.g., ZDV prophylaxis plus ECS) and intervention `b' (e.g., ZDV prophylaxis alone); and OUTCOMEa and OUTCOMEb are the outcomes (e.g., years of life saved) from interventions `a' and `b', respectively. Cost-effectiveness analysis compares the net cost of a health-care intervention to the incremental improvement in effectiveness or patient outcomes. Cost-effectiveness ratios are generally expressed as a dollar amount expended on an intervention per case of disease avoided or per year of life saved. A cost-effectiveness ratio is termed `dominant' if it results in a negative value (the intervention both costs less and has better outcomes than a comparator intervention). It is commonly accepted that interventions with a ratio of less than US$20 000 per year of life saved are cost-effective .
For the cost–benefit analysis, outcomes were measured in monetary terms, yielding a benefit–cost ratio, i.e.,, the dollar amount saved by avoiding vertical transmission of HIV for every additional dollar spent on the ECS intervention. For the cost-effectiveness analysis, two outcome measures were used: an intermediate outcome of cases of mother-to-child transmission of HIV avoided, and a long-term outcome of years of the child's life saved. Calculations of years of life saved by avoiding vertical transmission of HIV were based on the difference between the average US life expectancy of 75.8 years  and the estimated life expectancy of 9.4 years for HIV-infected children .
Sensitivity analyses were performed in the individual-based analysis to evaluate the impact of changing key model parameters on projected costs and outcomes. Sensitivity analyses were performed on: vertical transmission rates according to maternal HIV disease status (advanced or not advanced disease; advanced disease was defined as: a maternal diagnosis of AIDS, or, in the absence of such a diagnosis, a CD4+ cell count of 200 cells/μl or a CD4+ percentage of < 14%)  and to mode of delivery (ECS), discounted lifetime treatment costs of pediatric HIV disease, and postpartum morbidity rates.
The costs and outcomes of different strategies to prevent vertical transmission of HIV also were evaluated on a population basis. For these analyses, we determined the overall annual costs (for both interventions and pediatric HIV treatment) and outcomes (number of cases of vertical transmission) in the US using the same population as the individual-based models (pregnant, HIV-infected women who refrain from breastfeeding).
We based model projections on a total of 4958 HIV-infected women delivering annually . The other model parameters used in the individual-based models (vertical transmission rates, costs) also were used for the population models. Data on population characteristics, including the HIV seroprevalence among pregnant women in the US and the percentage of women who receive ART during pregnancy, were obtained from published literature [14,19]. Among women receiving ART, we assumed that 29% received ZDV prophylaxis only and the remaining 71% received combination ART . Treatment for side effects of ZDV prophylaxis or combination ART and associated costs were not included in the model. An estimated 6% of women receiving combination ART were assumed to discontinue this therapy during pregnancy due to adverse events [28–33]. As the side effects of ZDV are less severe, discontinuation of this drug was not included in the model. Potential infant toxicities associated with maternal ART during pregnancy also were not included in the model.
Two types of sensitivity analyses were performed. First, the impact of different maternal HIV seroprevalence rates  on population costs and outcomes was evaluated. Second, to assess the impact of increased acceptance of ART, the percentage of women receiving ART was increased from the base case of 68% to 95%, an estimate chosen arbitrarily. The distribution of ZDV prophylaxis versus combination ART among the 95% of women receiving ART was based on published data .
Benefit–cost and cost-effectiveness ratios from the individual-based analyses are presented in Table 3. Among women receiving no ART during pregnancy, ECS is cost saving in comparison with vaginal delivery; for every dollar spent on ECS, more than two dollars are saved in avoided medical care costs for the child. In cost-effectiveness analyses, ECS results in both better outcomes and decreased costs in comparison with vaginal delivery. For mother–child pairs receiving ZDV prophylaxis, ECS increases total costs slightly as compared to vaginal delivery (as evidenced by a benefit–cost ratio of less than 1.0). However, use of ECS in this population is extremely cost-effective; the additional cost per year of life saved (as compared to ZDV prophylaxis with vaginal delivery) is only US$17. Despite an increase in overall cost associated with ECS among women receiving combination ART during pregnancy, this intervention improved outcomes and remained cost-effective .
The model is sensitive to variations in vertical transmission rates according to mode of delivery and to pediatric care costs, but not to variations in postpartum morbidity rates. ECS is less cost-effective (compared to vaginal delivery) when vertical transmission rates with vaginal delivery are already low, and more cost-effective when these transmission rates are higher. For example, the vertical transmission rates for women with advanced HIV disease are greater than those for all women combined. Thus, in comparison with vaginal delivery, ECS is cost saving among women with advanced HIV disease receiving either no ART or ZDV prophylaxis and is more cost-effective among those women receiving combination ART. Greater effectiveness of ECS (resulting in lower transmission rates) increased the benefit–cost ratios; decreased effectiveness produced decreased benefit–cost and cost-effectiveness ratios. The benefit–cost ratios increased substantially with higher pediatric treatment costs, indicating savings of US$2.20 to US$9.92 for every dollar spent on ECS; the benefit–cost ratios are almost 4.5 times greater using higher estimated costs as compared to the base model.
An additional cost-effectiveness analysis evaluated the costs and outcomes of utilizing combination ART for the woman's health without ECS versus following current recommendations for prevention of vertical transmission (ZDV prophylaxis) and utilizing ECS (data not shown). In the base model, vertical transmission rates for these two strategies are assumed to be equal (Table 1); thus, the benefit–cost ratio is zero and the cost-effectiveness ratio is undefined. With equal transmission rates, the estimated cost for ECS with ZDV prophylaxis is greater than the cost for vaginal delivery with combination ART (US$11 693 versus US$7903, respectively). However, if the vertical transmission rate with ECS is lower than in the base model, ZDV prophylaxis with ECS still costs more than combination ART with vaginal delivery but also results in fewer pediatric HIV cases and is cost-effective (US$310 000 per case avoided, US$4678 per year of life saved). If the transmission rate with ECS is higher than in the base model, combination ART with vaginal delivery results in both decreased costs and better outcomes.
The estimated total population costs and number of cases of vertical transmission associated with each prevention strategy, if all HIV-infected pregnant women delivering in a single year received the indicated strategy, are shown in Table 4. An HIV seroprevalence of 1.7 per 1000 pregnant women  was used for the base model. With no ART received during pregnancy, ECS would save US$27 million and prevent 446 cases of HIV transmission that otherwise would have occurred with vaginal delivery. With ZDV prophylaxis, delivery by ECS would cost US$224 297 more than vaginal delivery but would prevent 198 cases of HIV transmission. With combination ART, delivery by ECS would cost US$8.9 million more than vaginal delivery and prevent 120 cases of HIV. Changes in seroprevalence rates resulted in a proportional change in the total cost and outcomes associated with each mode of delivery and ART scenario. With a seroprevalence rate approximately three times that of the base model, the population costs and the number of perinatal HIV cases are approximately three times higher. Similarly, with a lower seroprevalence, the costs and annual number of cases prevented are proportionally lower than with the base model.
The population costs and outcomes, calculated as a weighted average across the three ART options (no ART, ZDV only, or combination ART) by mode of delivery, are presented in Table 5. An estimated 68% of women receive ART during pregnancy (19.7% ZDV prophylaxis, 48.3% combination ART) [14,19]. Using a weighted average of receipt of ART, ECS would prevent 239 cases and saves US$4.4 million. Assuming that a higher percentage (arbitrarily chosen to be 95%) of women receive ART, but with the same proportional distribution between ZDV prophylaxis and combination ART as in the base model, ECS would prevent 157 cases of vertical transmission of HIV but cost US$4.7 million more than vaginal delivery.
ECS is a cost-effective intervention to prevent vertical transmission of HIV among women receiving various antiretroviral regimens, who refrain from breastfeeding. The costs per case avoided also are less than a published threshold for HIV prevention programs . For women who do not receive any ART during pregnancy, ECS is cost-saving. Sensitivity analyses to evaluate the impact of model assumptions on the projected costs and outcomes indicated the model is sensitive to vertical transmission rates according to mode of delivery and costs associated with treatment for pediatric HIV disease. Results from the population-based analyses mirror those of the individual-based analyses; ECS is cost saving compared to vaginal delivery when no ART is used during pregnancy and results in increased costs but decreased vertical transmission when ZDV prophylaxis or combination ART is received.
The cost-effectiveness ratios in the present study are similar to or more favorable than values reported for other strategies to prevent vertical transmission of HIV, including voluntary testing and counseling, and ZDV prophylaxis [19,35]. The range of results are also similar to or more favorable than those of other interventions to prevent HIV transmission, including ZDV and lamivudine prophylaxis following sexual exposure (US$70 000 per case avoided) , blood donor screening (US$124 089 per case avoided) , and screening of health care workers to prevent transmission to patients (US$9 million per case avoided) . Further, the cost-effectiveness ratios for ECS are similar to or more favorable than values from other types of well-accepted health-care interventions [costs per year of life saved: beta blockers for myocardial infarction survivors (US$360–850); influenza vaccination for high risk individuals (US$490); mammography for women aged 50 (US$810); and cervical cancer screening every 5 years for women aged 65 years or older (US$1900); smoking cessation (men: US$705–988; women: US$1204–2058)][39,40].
A cost-effectiveness analysis from the United Kingdom evaluated the economic impact of ECS, with or without ZDV prophylaxis and avoidance of breastfeeding, for prevention of vertical transmission of HIV . Although the estimated costs and outcomes described in that study differ from those of our study, the findings are consistent in demonstrating the cost-effectiveness of ECS. Although Ratcliffe et al. argued that for women receiving no intervention to reduce vertical transmission of HIV (i.e.,, no receipt of ZDV prophylaxis, no ECS, and no avoidance of breastfeeding), use of ZDV prophylaxis alone was more cost-effective than ECS alone , this conclusion was based on less precise estimates of the efficacy of ECS in preventing vertical transmission than available for our analysis as well as a different health-care cost structure. Also, neither the impact of combination ART received during pregnancy nor cost savings from avoided cases of vertical transmission of HIV were included in their analysis.
This analysis adhered to published analytic principles for cost-effectiveness and cost–benefit literature  as well as suggested guidelines for the development of health economic models , including transparency of model parameters, description of model assumptions, and performance of sensitivity analysis for areas of uncertainty. Furthermore, we consistently used conservative assumptions in the analysis to prevent bias in favor of ECS. For example, deliveries other than via ECS were assumed to be all vaginal, instead of a mix of vaginal and non-ECS deliveries. The costs and postpartum morbidity rates resulting from such a mix would be higher than those associated with vaginal delivery alone, thereby reducing the difference in costs between different modes of delivery.
This analysis was limited by the extent of available published data regarding the parameters incorporated into the individual- and population-based models. Few published data exist regarding the costs of postpartum morbidity associated with mode of delivery. Therefore, cost estimates were derived from studies in the general US population of pregnant women, rather than a population of HIV-infected women, probably underestimating costs of postpartum morbidity. These studies evaluated costs for complications associated with vaginal delivery or cesarean section (both elective and non-elective); the cost of complications for ECS is likely to be less than that for all cesarean sections combined. Further, only an overall cost for the presence or absence of postpartum morbidity could be obtained from the literature. Finally, it is possible that the relative incidence of specific types of postpartum morbidity experienced by HIV-infected women differ from those of the general population. However, the model was not sensitive to postpartum morbidity rates. This is probably because the cost of treating postpartum morbidity is relatively minor in comparison with the cost of treating pediatric HIV disease, and therefore changes in the rates of postpartum morbidity have little impact on the overall cost-effectiveness.
Due to the paucity of available data regarding the risk of transmission of HIV to health-care workers involved in the delivery, we were unable to include such risk estimates in the model. As cesarean sections are surgical procedures and more invasive than vaginal delivery, there may be a higher risk of HIV infection for health-care providers performing ECS. Although this risk would be expected to be extremely small, it is probably not zero.
Only limited data exist regarding vertical transmission rates with combination ART received during pregnancy. Other studies [10,43] have suggested greater effectiveness for ZDV prophylaxis than was used in this model. Although sensitivity analysis on increased ZDV effectiveness was not performed, results would be similar to those observed with receipt of combination ART; with a lower vertical transmission rate, ECS is less cost-effective. Only very limited data are available regarding vertical transmission rates among HIV- infected women receiving combination ART regimens during pregnancy, especially those containing protease inhibitors. Data regarding the vertical transmission rate with one combination ART regimen were obtained from a recent, relatively large study . Use of these data is not meant to imply that other combination ART regimens would have the same vertical transmission rates. No data are available evaluating the joint impact of combination ART and ECS; transmission rates for this combination were estimated using the proportional change in rates for ECS versus vaginal delivery among women receiving ZDV.
Relatively little is known about the short- and long-term safety of combination ART during pregnancy, both for the mother and the fetus or infant [17,44,45]. Adverse events rates could vary according to the specific combination ART regimen received and the occurrence of adverse events could increase the costs associated with combination ART during pregnancy. Costs for different combination ART regimens were not explored in this study, as the analysis compared ECS with vaginal delivery when both groups received the same ART regimen (and thus incurred the same costs). If different combination ART regimens vary in terms of their impact on vertical transmission and/or adverse events experienced by the mother or fetus, then the differential costs of the regimens will be important.
Thus, to refine and improve upon the economic estimates presented in this analysis, further research needs to be performed. Most importantly, evaluation of the effectiveness of combination ART in preventing vertical transmission and of the possible maternal and pediatric adverse events related to exposure to such therapy is essential. Different combination ART regimens could have different effects on vertical transmission as well as types and rates of complications and treatment costs. If vertical transmission rates with combination ART were to reach 0%, ECS would only increase costs and not improve outcomes. As more precise data are collected, threshold analyses should be performed to evaluate the conditions under which ECS remains cost-effective. However, based on the findings of this study, ECS is likely to remain a cost-effective intervention under a wide range of possible clinical and economic scenarios.
1. Centers for Disease Control and Prevention. HIV/AIDS Surveillance Report.
2. Centers for Disease Control and Prevention. Recommendations of the public health service task force on the use of zidovudine to reduce perinatal transmission of human immunodeficiency virus. MMWR 1994, 43: (No.RR–11).
3. Centers for Disease Control. Recommendations for assisting in the prevention of perinatal transmission of human T-lymphotropic virus type III/lymphadenopathy-associated virus and acquired immunodeficiency syndrome. MMWR 1985, 34: 721–726.
4. American Academy of Pediatrics Committee on Pediatric AIDS. Human milk, breastfeeding, and transmission of human immunodeficiency virus in the United States. Pediatrics 1995, 96: 977–979.
5. The International Perinatal HIV Group. The mode of delivery and the risk of vertical transmission of human immunodeficiency virus type 1: a meta-analysis of 15 prospective cohort studies. N Engl J Med 1999, 340: 977–987.
6. The European Mode of Delivery Collaboration. Elective caesarean-section versus vaginal delivery in prevention of vertical HIV-1 transmission: a randomised clinical trial. Lancet 1999, 353: 1035–1039.
7. American College of Obstetricians and Gynecologists. Scheduled cesarean delivery and the prevention of vertical transmission of HIV infection. ACOG Committee Opinion Number 219.
Washington, DC: ACOG; August 1999.
8. Ratcliffe J, Ades AE, Gibb D, Sculpher MJ, Briggs AH. Prevention of mother-to-child transmission of HIV-1 infection: alternative strategies and their cost-effectiveness. AIDS 1998, 12: 1381–1388.
9. Centers for Disease Control and Prevention. US Public Health Service recommendations for human immunodeficiency virus counseling and voluntary testing for pregnant women. MMWR 1995, 44(RR–7): 1–15.
10. Fiscus SA, Adimora AA, Schoenbach VJ. et al
. Trends in human immunodeficiency virus (HIV) counseling, testing, and antiretroviral treatment of HIV-infected women and perinatal transmission in North Carolina. J Infect Dis 1999, 180: 99–105.
11. Simonds RJ, Steketee R, Nesheim S. et al
. Impact of zidovudine use on risk and risk factors for perinatal transmission of HIV. AIDS 1998, 12: 301–308.
12. Bertolli J, Simonds RJ, Thomas P, et al
. Implementation of recommendations for the medical care of HIV-exposed infants in the first year of life, USA. XIIth World AIDS Conference
. Geneva, June–July 1998 [abstract 23269].
13. Centers for Disease Control and Prevention. Success in implementing public health service guidelines to reduce perinatal transmission of HIV–Louisiana, Michigan, New Jersey, and South Carolina, 1995, and 1996. MMWR 1998, 1993, 37: 688–691.
14. Dorenbaum-Kracer A, Sullivan J, Gelber R, et al
. Antiretroviral use in pregnancy in PACTG 316: a phase III randomized, blinded study of single-dose intrapartum/neonatal nevirapine to reduce mother to infant HIV transmission. XIIth World AIDS Conference.
Geneva, June–July 1998 [abstract 23281].
15. Centers for Disease Control and Prevention. Public Health Service Task Force recommendations for the use of antiretroviral drugs in pregnant women infected with HIV-1 for maternal health and for reducing perinatal HIV-1 transmission in the United States. MMWR 1998, 47(No. RR–2): 1–30.
16. Bertolli JM, Hsu H, Frederick T, et al
. Breastfeeding among HIV-infected women, Los Angeles and Massachusetts, 1988–1993.XIth World AIDS Conference.
Vancouver, July 1996 [abstract WeC3583].
17. Blanche S, Rouzioux C, Mandelbrot L, Delfraissy JF, Mayaux MJ. Zidovudine-lamivudine for prevention of mother to child HIV-1 transmission. Sixth Conference on Retroviruses and Opportunistic Infections.
Chicago, January–February 1999 [abstract 267].
18. Read J, Kpamegan E, Tuomala R, et al
. Mode of delivery and postpartum morbidity among HIV-infected women: The Women and Infants Transmission Study (WITS). Sixth Conference on Retroviruses and Opportunistic Infections.
Chicago, January–February 1999 [abstract 683].
19. Gorsky RD, Farnham PG, Straus WL. et al
. Preventing perinatal transmission of HIV-costs and effectiveness of a recommended intervention. Public Health Rep 1996, 111: 335–341.
20. Physician's Desk Reference
. Montvale, NJ: Medical Economics Company, Inc.; 1997.
21. Cardinale V, ed. 1998 Drug Topics Red Book.
Montvale, NJ: Medical Economics Company, Inc., 1998.
22. Scott LL, Alexander J. Cost-effectiveness of acyclovir suppression to prevent recurrent genital herpes in term pregnancy. Am J Perinatol 1998, 15: 57–62.
23. Myers ER, Thompson JW, Simpson K. Cost-effectiveness of mandatory compared with voluntary screening for human immunodeficiency virus in pregnancy. Obstet Gynecol 1998, 91: 174–181.
24. Havens PL, Cuene BE, Holtgrave DR. Lifetime cost of care for children with human immunodeficiency virus infection. Pediatr Infect Dis J 1997, 16: 607–610.
25. Holtgrave DR, Qualls NL. Threshold analysis and programs for prevention of HIV infection. Med Decision Making 1995, 15: 311–317.
26. Vital and health statistics: Russian Federation and United States, selected years 1980–93. Vital Health Stat
27. Barnhart HX, Caldwell MB, Thomas P. et al
. Natural history of human immunodeficiency virus disease in perinatally infected children: an analysis from the Pediatric Spectrum of Disease Project. Pediatrics 1996, 97: 710–716.
28. Baruch A, Mastrodonato-Delora P, Schnipper E, Salgo M. Efficacy and safety of triple combination therapy with Invirase (saquinavir/SQV/HIV protease inhibitor), Epivir (3TC/lamivudine) and Retrovir (ZDV/zidovudine) in HIV-infected patients. XIth World AIDS Conference.
Vancouver, July 1996 [abstract MoB172].
29. Eriksen N, Helfgott A, Doyle M. Combination antiretroviral therapy for the treatment of HIV infection in pregnant women: safety profiles in women and newborns. Fifth Conference on Retroviruses and Opportunistic Infections.
Chicago, February 1998 [abstract 235].
30. Katlama C, Ingrand D, Loveday C. et al
. Safety and efficacy of lamivudine-zidovudine combination therapy in antiretroviral-naïve patients. A randomized controlled comparison with zidovudine monotherapy. Lamivudine European HIV Working Group.
JAMA 1996, 276: 118–125.
31. Raffi F, Auger S, Billaud E, et al
. Antiviral effect and safety of didanosine-stavudine combination therapy in HIV-infected subjects: interim results of a pilot trial. Fourth Conference on Retroviruses and Opportunistic Infections.
Washington, DC, January 1997 [abstract 554].
32. Skowron G, Leonung G, Dusek A, et al
. Stavudine (d4T), nelfinavir (NFV), and nevirapine (NVP): preliminary safety, activity and pharmacokinetic (PK) interactions. Fifth Conference on Retroviruses and Opportunistic Infections.
Chicago, February 1998 [abstract 350].
33. Staszewski S, Loveday C, Picazo JJ. et al
. Safety and efficacy of lamivudine-zidovudine combination therapy in zidovudine- experienced patients. A randomized controlled comparison with zidovudine monotherapy. Lamivudine European HIV Working Group.
JAMA 1996, 276: 111–117.
34. Laupacis A, Feeny D, Detsky A, Tugwell PX. How attractive does a technology have to be to warrant adoption and utilisation. Can Med Assoc J 1992, 146: 473–481.
35. Ecker JL. The cost-effectiveness of human immunodeficiency virus screening in pregnancy. Am J Obstet Gynecol 1996, 174: 716–721.
36. Pinkerton SD, Holtgrave DR, Bloom FR. Cost-effectiveness of post-exposure prophylaxis following sexual exposure to HIV. AIDS 1998, 12: 1067–1078.
37. Eisenstaedt RS, Getzen TE. Screening blood donors for human immunodeficiency virus antibody: cost-benefit analysis. Am J Public Health 1988, 78: 450–454.
38. Chavey WE, Cantor SB, Clover RD, Reinarz JA, Spann SJ. Cost-effectiveness analysis of screening health care workers for HIV. J Fam Pract 1994, 38: 249–257.
39. Tengs TO, Adams ME, Pliskin JS. et al
. Five-hundred life-saving interventions and their cost-effectiveness. Risk Anal 1995, 15: 369–390.
40. Cummings SR, Rubin SM, Oster G. The cost-effectiveness of counseling smokers to quit. JAMA 1989, 261: 75–79.
41. Udvarhelyi S, Colditz GA, Rai A, Epstine AM. Cost-effectiveness and cost-benefit analyses in the medical literature. Are the methods being used correctly?
Ann Intern Med 1992, 116: 238–244.
42. Halpern MT, Luce BR, Brown RE, Geneste B. Health and economic outcomes modeling practices: a suggested framework. Value in Health 1998, 1: 131–147.
43. Stiehm ER, Lambert JS, Mofenson LM, et al.
for the Pediatric AIDS Clinical Trials Group Protocol 185 Team. Efficacy of zidovudine and human immunodeficiency virus (HIV) hyperimmune immunoglobulin for reducing perinatal HIV transmission from HIV-infected women with advanced disease: results of Pediatric AIDS Clinical Trials Group Protocol 185. J Infect Dis
44. Blanche S, Tardieu M, Rustin P. et al
. Persistent mitochondrial dysfunction and perinatal exposure to antiretroviral nucleoside analogues. Lancet 1999, 354: 1084–1089.
45. Lorenzi P, Massarey Spicher V, Laubereau B. et al
. Antiretroviral therapies in pregnancy: maternal, fetal and neonatal effects. AIDS 1998, 12: F241–247.
© 2000 Lippincott Williams & Wilkins, Inc.