Microbicides are substances capable of reducing the transmission of HIV and other sexually transmitted pathogens when applied vaginally. Currently about 65 compounds are in some stage of development. Of these, 17 are in phase I or II clinical trials and four are expected to enter phase III trials later this year [1,2 (personal communication with Polly Harrison; March 2003)]. The new generation of products being evaluated have different mechanisms of action than nonoxynol-9 (N-9), the spermicidal product that has proven ineffective against HIV [2–4].
At present it is unknown what efficacy we can expect microbicides to have. The first generation of microbicides are likely to be less efficacious than the male-condom . Because of this, microbicides will be delivered using hierarchical messages that promote the concurrent use of microbicide with a condom as ‘best choice', but suggest microbicide use alone as a fallback option when condom use is not possible. This parallels other ‘harm reduction’ approaches to HIV prevention used with injecting drug users . Studies suggest that women find hierarchical counselling more acceptable than other counselling methods [7,8].
Evidence from many settings strongly suggests that many women will find microbicides easier to use than condoms and that they can be used when condom use is not possible [9–17]. This has raised the concern that women might abandon condoms in favour of microbicides. Indeed, the fear of ‘condom migration’ has been a major barrier to the widespread endorsement of microbicides as a potential method of HIV prevention .
This article presents a mathematical model of HIV transmission developed to estimate the level of protection provided by different combinations of condom and microbicide use. The model is used to identify the ‘break-even point’ at which any increase in risk associated with condom migration is counter-balanced by the increased protection afforded with microbicides, and to explore how different levels of microbicide use affect an individual's HIV-risk.
Condom efficacy and consistency of use
A meta-analysis of studies evaluating rates of sero-conversion among discordant couples who always use condoms estimated that condom HIV-efficacy is approximately 87%, with a potential range from 60% to 96% . Another meta-analysis estimates that consistent condom use reduces HIV incidence by 80% . A review by the US National Institutes of Health concludes that there is insufficient available data to estimate the efficacy of condoms against many STDs, and notes that efficacy is likely to vary by STD pathogen .
Despite such high HIV-efficacy, there is substantial evidence that in many settings levels of consistent condom use is low, especially within primary partnerships . Studies show that even after directed intervention, the percentage of couples that achieve consistent condom use seldom exceeds 20–30%, except where individuals know that one partner is HIV infected . Success in increasing the consistency of condom use has been greatest among sex workers and other vulnerable groups [8,23], but even here, many sex workers do not use condoms consistently with their non-commercial, primary partners . Even when condoms are used consistently they may be used incorrectly . The immense gap between the numbers requiring protection and those using condoms consistently and correctly could potentially be filled by microbicides.
Migration from the condom: current evidence
To date nine studies have examined how the availability of microbicidal-like products such as spermicides affect male-condom use (Table 1). These studies are limited because they focus on nonoxynol-9 spermicides, and tend to involve vulnerable groups of women. Nonetheless, six found that the availability of additional protection options along with counselling resulted in increased condom use [7,8,26–30]. Of the three studies that observed a lower consistency of condom use when both spermicides and condoms were provided [31–33], one was conducted among Kenyan sex workers , another among Colombian sex workers , and the third among high-risk women in Cameroon . Only two of these [31,33] measured the consistency of condom use, the greatest difference being between the ‘male-condom only’ and the ‘spermicides as a fallback’ study arms in the Colombian study (95% versus 78%), illustrating that migration may be a concern under certain circumstances . The Kenyan study used different measures of condom use at baseline and follow-up and so it is hard to accurately assess the amount of migration that occurred since it is difficult to determine if ‘consistent condom use’ is the same as ‘100% condom use'.
This study builds upon previous studies by Sokal et al. and Watts et al. that used mathematical modelling to identify factors affecting the potential impact of microbicides [34,35]. For this analysis we use an established mathematical equation that describes the probability of HIV transmission between susceptible and infected individuals [36–38]. As in the study by Watts et al., we refine the equation to consider the level of protection provided by different methods, incorporating both how the presence of an STD and the initial high HIV viraemia phase may facilitate HIV transmission . When just one method of protection is used, the probability (π) that a susceptible person becomes HIV infected over a fixed time period is:
where E is the average probability that an individual is protected for one sex act, in a partnership. We call this the ‘use-effectiveness’ against HIV of a method used in a particular partnership. It is given by the product of the clinical efficacy of the method against HIV transmission and the average consistency that it is used in the partnership. Also, p is the probability that a selected sexual partner is HIV infected; h is the probability that an HIV infected partner has high viraemia; s is the probability that within a partnership at least one person has an STD; m is the average number of sexual partnerships the person has over the fixed time-period; n is the average number of sex acts per partnership during the fixed time-period; β is the probability of HIV transmission per sex act; α is the multiplicative increase in the per sex act probability of HIV transmission during the high viraemia phase; and δ is the multiplicative increase in the per sex act probability of HIV transmission in the presence of another STD.
An analogous equation (not including the high viraemia and STD multiplicative co-factors) can also be used to describe the probability of STD transmission. In both cases, the main assumption is that the probability of transmission per sex act is independent of previous sex acts. It is a static model considering only one-way transmission of HIV over a fixed time-period and so is most reliable over short timeframes.
In this paper we use the equation in three ways. Firstly, we compare the protection provided against HIV and STDs by using either condoms only or microbicides only, with various consistencies, to identify when microbicides alone would provide more protection than condoms alone. Secondly, we compare the protection provided by condoms used alone at various consistencies with the protection provided by condoms and microbicides, allowing for the possibility that condom-consistency may decline following microbicide introduction. This is used to identify the ‘break-even point’ at which an individual is equally protected after microbicide introduction as before, assuming a microbicide HIV- and STD-efficacy of 50%. We choose an efficacy of 50% for illustrative purposes to look at the impact of a low-efficacy microbicide and the effect of migration from a high-efficacy method, namely condoms. We assume that condoms are 95% HIV- and STD-efficacious. This is a high estimate [19,20,39], that will produce conservative estimates of the amount of migration that can be tolerated without increasing risk. Thirdly, we develop a case-example from Cotonou, Benin, to explore how different patterns of condom and microbicide use in a specific setting interact to affect HIV-risk in four different groups: sex workers, the clients of sex workers, women with multiple sexual partners, and women in monogamous HIV-discordant partnerships. For this exercise, we use equation (1) with epidemiological and behavioural data from Cotonou [40–42] and the scientific literature [43–48], and consider the introduction of a 50% HIV- and STD-efficacious microbicide, assuming 10% condom migration. Cotonou was chosen for the case-example because comparable data for different groups were available, and Cotonou has already been a site for a phase II/III microbicide trial and is likely to be involved in future microbicide trials .
The estimates of microbicide-consistency are based on those of spermicide-consistency from the studies in Table 1. Two of these studies do not have the required available data. Five of the remaining seven studies report 50–90% of non-condom-protected sex acts being protected by spermicide. The other two studies indicate a low consistency of spermicide use. The reasons for this are uncertain but may be linked to the study populations in these studies being women with relatively low levels of sexual activity attending a family planning clinic in Texas or drug abusers at a medical centre in Miami. Overall, the studies in Table 1 indicate that high microbicide-consistency may be attained in many settings.
E is a measure of how a method of protection alters an individual's HIV-risk. The probability (π) that a susceptible person, using a particular method of protection, becomes HIV infected over a fixed time-period decreases as the use-effectiveness (E) increases. Any comparison of the risk of HIV infection associated with the use of different methods reduces mathematically to a comparison of the use-effectiveness of the different methods. Consequently, microbicides provide more protection than condoms if the use-effectiveness of microbicides is greater than the use-effectiveness of condoms. Within a partnership, for example, the protection provided by a microbicide of 50% HIV- and STD-efficacy used with 40% consistency is greater than that provided by condoms (95% HIV- and STD-efficacy) used with 20% consistency.
Similarly, we can compare an individual's HIV-risk before and after microbicide introduction. The use-effectiveness against HIV of condoms used with average consistency (percentage of sex acts in which a condom is used) fo in each partnership is Ec = foe, where e is the per sex act HIV-efficacy of condoms. The use-effectiveness against HIV of condoms, used with average consistency fR in each partnership after microbicide introduction, and microbicides, used in a proportion M of sex acts in which a condom is not used, is Em = fRe + (1 – fR)Mb. Here b is the per sex act HIV-efficacy of microbicides. The level of HIV protection provided by condoms and microbicides will be greater than the protection provided by condoms alone if the use-effectiveness of condoms and microbicides is greater than the use-effectiveness of condoms alone (Em > Ec), which can be written:
This inequality can be used to identify the maximum absolute drop in condom-consistency (condom migration) that could be tolerated following microbicide introduction without increasing HIV-risk.
From equation (2) we can identify the minimum consistency of condom use that can be tolerated after microbicide introduction (fR*) without increasing an individual's HIV-risk:
where λ = b/e is the ratio of the clinical efficacies against HIV of microbicides to condoms, and f0 and M are as defined in the Methods. We will refer to this threshold (fR*) as the ‘break-even point'.
Likewise, an analogous condition applies for STDs. To ensure that microbicide introduction results in the reduction of an individual's risk of both HIV and STDs we need λ to be the smallest of the relative clinical HIV-efficacy of microbicides to condoms, and their relative clinical STD-efficacy. For the rest of this analysis we assume that the λs are the same and by efficacy we mean efficacy against both HIV and STDs.
It can be seen from equation (3) that while the underlying probability of infection (π) is affected by the number of sexual partners and acts, and the underlying HIV/STD prevalence, these do not affect the break-even point. This suggests that the break-even point will be the same irrespective of the epidemiological setting in which microbicides are introduced. Although HIV-risk varies substantially between settings, the break-even point results from comparing solely the protection provided by different methods within the same risk context; thus, these factors do not determine when microbicide introduction will be beneficial. The break-even point depends only on the relative efficacy of microbicides to condoms, and the consistency with which they are used.
From equation (3), the maximum amount of migration that can be tolerated without increasing HIV-risk is greater for a higher relative clinical efficacy of microbicides to condoms (λ), for a higher consistency of microbicide use when condoms are not used (M), or for a lower initial condom-consistency (f0).
Finally, equation (3) can be used to identify the consistency of microbicide use required to ensure that total migration (condoms not used) will not increase risk. For this, the consistency of microbicide use must be greater than the ratio of the initial condom-consistency and the relative efficacy of microbicides to condoms M > f0/λ.
Further insights can be gained by looking at the relationship between the initial condom-consistency and the break-even point, for a microbicide that is 50% efficacious, for different levels of microbicide use (Fig. 1).
The point where each plot line meets the f0-axis indicates the initial condom-consistency at which, for a specific microbicide-consistency, total migration can occur without increasing risk. For example, if microbicides are used in 75% of non-condom-protected sex acts, then total migration can be tolerated if before microbicide introduction condoms are used less than 39% of the time.
The case study from Cotonou, Benin, illustrates how microbicide introduction may affect HIV-risk in four different groups. The most prevalent STD in Cotonou, for which there is available data, is Neisseria gonorrhoeae [24,41,42]. Table 2 compares an individual's monthly risk of HIV infection (π) before and after the introduction of a 50% efficacious microbicide, assuming a 10% absolute reduction in the consistency of condom use afterwards. The results show that the percentage reduction in the probability of HIV transmission increases for a lower initial condom-consistency and as microbicide use increases.
Table 2 illustrates that, under most conditions, the introduction of microbicides in this setting would be beneficial. The only scenarios of potential concern are where condom use is initially very high (90%) and microbicides are used in 50% or 75% of non-condom-protected sex acts. In these cases there is an increase in π, indicating an increased risk. However, if a microbicide is used in all acts not protected by a condom, this increased risk disappears.
We conducted a sensitivity analysis on these projections for different microbicide efficacies and found that if the efficacy is 30% then the scenarios that cause an increase in HIV-risk include those where initial condom-consistency is moderate (∼ 50%), and not just when it is very high (90%) which was the case for a microbicide efficacy of 50%. Similarly, if the efficacy is 70% then the initial condom-consistency has to be very high (90%) and microbicide use has to be moderate or lower (≤ 50% of non-condom-protected sex acts) for there to be an increase in HIV-risk.
Although the negative numbers in Table 2 highlight that care is needed when counselling consistent condom-users about microbicides, the results do not imply that introducing microbicides at a population level will not be beneficial. Using figures from Table 2, we can estimate the number of ‘new recruits’ (non-condom-users who start to use microbicides) that are required to balance any shifts away from condom use among high-consistency condom-users. Assuming microbicides are used in 50% of non-condom-protected sex acts, the static model suggests the following. For every four high-consistency condom-users who reduce from 90% to 80% consistency, only one new recruit is required in order to have a positive impact on HIV at a population level.
Equation (1) and the context specific data from Benin can be used to identify, for this setting, what level of condom- and microbicide-consistency must be achieved after microbicide introduction to reduce HIV-risk by various degrees. Table 3 shows the condom- and microbicide-consistency required to break-even or reduce the risk of HIV infection by 20%.
The results provide several important insights about whether migration should be a concern amongst sub-populations with different levels of condom use. Under most circumstances, there will be no increase in HIV-risk if moderate microbicide use is achieved. For example, if a 50% efficacious microbicide is used in 50% of sex acts not protected by condoms then for all partnerships:
low-consistency condom-users (30%) could reduce condom-consistency to 5% without increasing HIV-risk;
medium-consistency condom-users (50%) could reduce to 32% consistency and still break-even;
high-consistency condom-users (70%) could reduce condom use to 59%; and
very high-consistency condom-users (90%) could reduce condom use to 86%.
If microbicides are used in all non-condom-protected sex acts then for all partnerships:
low- and medium-consistency condom-users could cease condom use and still break-even;
high-consistency condom-users could reduce condom use to 37%; and
very high-consistency condom-users could reduce condom use to 79%.
For the partnerships considered, if condoms are initially used in 70% or less of sex acts, and if there is no reduction in condom-consistency following microbicide introduction, there will be a 20% or greater reduction in HIV-risk as long as microbicides are used in 50% of non-condom-protected sex acts. If microbicides are used more consistently, a 20% reduction in HIV-risk can be achieved even with substantial condom migration, particularly if condoms are used with 50% consistency or less before microbicide introduction.
First, the analysis illustrates how the level of protection provided by a product is determined by its use-effectiveness. Therefore, if an ‘easy to use’ method is used twice as often as a method that is twice as HIV- and STD-efficacious, then both methods will provide the same amount of protection against HIV and STD infection. This highlights that both efficacy and factors affecting consistency need to be considered and given equal weighting in the process of microbicide development. For example, a vaginal ring which gradually releases microbicide over many sex acts may have an advantage over a higher efficacy microbicide that needs to be applied before every sex act .
Second, the analysis shows that the maximum amount of condom migration that can be tolerated is greater for higher levels of microbicide efficacy and consistency of use but also for lower initial levels of condom-consistency. If few sex acts were previously protected by condoms, microbicides will have a greater potential to reduce HIV transmission and so it is worse for a high condom-consistency-user than for a low condom-consistency-user to migrate by 10%.
The results indicate that there are likely to be many situations in which the benefits of microbicide use outweigh the negative impact of condom migration. For a 50% efficacious microbicide, if the initial condom-consistency is moderate (50%), and microbicides are used in 50–100% of non-condom-protected sex acts, 18–50% fewer sex acts can be protected by condoms with no increase in risk. Migration from the condom may, however, be a potential problem among groups who initially use condoms with high levels of consistency (> 70%), and use microbicides with low consistency (< 50% of non-condom-protected acts). This highlights the importance of microbicides for women who use condoms inconsistently but could use microbicides consistently, such as women whose partners do not object to methods of protection being used but dislike condoms.
The break-even point is exclusively determined by the relative efficacy of microbicides to condoms, and the consistency with which they are used. Therefore, the decision whether to introduce microbicides into a population can be made independently of the prevalence of HIV/STDs and the number of sexual partners and acts people have in that setting. This is an important finding since accurate sexual behaviour data is hard to obtain due to its sensitive nature and the collection of epidemiological data is a costly process.
We were concerned in this analysis with whether migration from the condom would lead to an increased HIV-risk. These results illustrate that there are many situations where we would break-even. However, we do not want to just break-even, but to reduce HIV-risk. For the scenarios considered in the Benin case-study, if condoms are used in 70% or less of sex acts before microbicide introduction and their consistency does not decrease afterwards, then the introduction of a 50% efficacious microbicide will result in at least a 20% reduction in HIV-risk if it is used in half of the non-condom-protected sex acts. These results give insights into the likely impact of microbicides in different settings, which can be used to guide trial design.
The analysis provides insights into the extent to which migration from the condom could potentially be a concern at an individual level. Using a static model, however, means we cannot estimate how a microbicide may impact on the HIV epidemic. For this, dynamic modelling is required. Dynamic modelling could also be used to estimate the break-even point for a microbicide that is only efficacious against HIV and not other STDs, and to explore the potential impact and cost-effectiveness of microbicide introduction in different settings [50,51]. Future work will examine these issues.
We would like to thank our colleagues Lorna Guinness, Fern Terris-Prestholt and Lilani Kumaranayake, HIVTools Research Group, LSHTM, for their comments on the work. We would also like to thank Professor Michel Alary, Epidemiology Research Group, Centre Hospitalier Affilie Universitaire de Quebec (CHA) and Universite Laval, Quebec, Canada, for providing the Benin data. The work is funded by the Global Campaign for Microbicides/Program for Appropriate Technology in Health. A.F., P.V. and C.W. are members of the Health Economics & Financing Programme and the AIDS Programme, which are funded by the UK Department for International Development. P.V. and C.W. are also members of the MRC/Imperial Microbicide Development Programme, which is funded by the UK Department for International Development. The views and opinions expressed are those of the authors alone.
2.Harrison PF. The microbicide research and development ‘‘pipeline’': a status report
. Microbicides 2002
. Antwerp, 12–15 May 2002 [Abstract B/C-326]. http://www.itg.be/micro2002/
downloads/presentations/2Monday_May_13_2002/Track_B_C_ session/Polly_Harrison.pdf. Site accessed: 6 Dec 2002.
4.Wilkinson D. Nonoxynol-9 spermicide for prevention of vaginally acquired HIV and other sexually transmitted infections: systematic review and meta-analysis of randomised controlled trials including more than 5000 women
. Lancet Infect Dis
7.Gollub EL. Achieving safer sex with choice: Studying a women's sexual risk reduction hierarchy in an STD clinic
. J Womens Health Gend Based Med
8.Latka M, Gollub EL, French P, Stein Z. Male-condom and female-condom use among women after counseling in a risk-reduction hierarchy for STD prevention
. Sex Transm Dis
10.Boonstra H. Campaign to Accelerate Microbicide Development for STD Prevention Gets Under Way: The Guttmacher Report on Public Policy
. Guttmacher Institute; 2000. http://www
. guttmacher.org/pubs/journals/gr030103.html. Site accessed: 6 Dec 2002.
11.Green G, Pool R, Harrison S, Hart GJ, Wilkinson J, Nyanzi S, et al
. Female control of sexuality: illusion or reality? Use of vaginal products in south west Uganda
. Soc Sci Med
12.Pool R, Whitworth J, Green G, Mbonye A, Harrison S, Hart G, et al
. Ambiguity, sexual pleasure, and the acceptability of microbicidal products in Southwest Uganda
. Microbicides 2000
. Washington DC, 13–16 March 2000 [Abstract C24].
13.Ramjee G, Gouws E, Andrews A, Myer L, Weber AE. The acceptability of a vaginal microbicide among South African men
. Int Fam Plann Perspect
14.Sri Krishnan AK, Morrow K, Johnson S, Thamburaj E, Rosen R, Mayer KH, et al
. Same risks, disparate perceptions: Challenges to microbicide acceptability in the Chennai community
. Sixth International Congress on AIDS in Asia and the Pacific
. Melbourne, 5–10 October 2001 [Abstract 0777].
15.Srirak N, Sirirojn B, Wichajarn M, Rugpao S, Nelson K, Celentano DD, et al
. Acceptability of Thai males toward Buffergel as a vaginal microbicide in Chiang Mai, Thailand
. Microbicides 2000
. Washington DC, 13–16 March 2000 [Abstract C34].
16.Van der Wijgert JHHM K-SG, Coggins C, Dube SE, Nyamapfeni P, Mwale M, Padian NS. Men's attitudes towards vaginal microbicides and microbicide trials in Zimbabwe
. Int Fam Plann Perspect
17.Weiss E, Gupta GR. Bridging the Gap: Addressing Gender and Sexuality in HIV Prevention
. International Center for Research on Women; 1998.
18.Heise L. Topical Microbicides
. Takoma Park, MD 20912: Center for Health and Gender Equity (CHANGE); 1999. http://www
. genderhealth.org/. Site accessed: 6 Dec 2002.
19.Davis KR, Weller SC. The effectiveness of condoms in reducing heterosexual transmission of HIV
. Fam Plann Perspect
20.Weller S, Davis K. Condom effectiveness in reducing heterosexual HIV transmission
. The Cochrane Library. Oxford: Update Software: [Cochrane Review]; Issue 3, 2002.
21.Workshop Summary: Scientific Evidence on Condom Effectiveness for Sexually Transmitted Disease (STD) Prevention
. Hyatt Dulles Airport, Herndon, Virginia: National Institute of Allergy and Infectious Diseases, National Institutes of Health, Department of Health and Human Services; Reported prepared: July 20, 2001. http://www.niaid.nih.gov/dmid/stds/condomreport.pdf
. Site accessed: 6 Dec 2002.
22.Gardner R, Blackburn RD, Upadhyay UD. Population Reports: Closing the Condom Gap
. Vol. XXVII, No. 1, Series H, No. 9: Population Information Program, Center for Communication Programs, The Johns Hopkins School of Public Health, Baltimore; April, 1999. http://www.jhuccp.org/pr/h9edsum.stm
. Site accessed: 6 Dec 2002.
23.Fontanet AL, Saba J, Chandeying V, Sakondhavat C, Bhiraleus P, Rugpao S, et al
. Protection against sexually transmitted diseases by granting sex workers in Thailand the choice of using the male or female condom: results from a randomized controlled trial
24.Alary M, Mukenge-Tshibaka L, Bernier F, Geraldo N, Lowndes M, Meda H, et al
. Decline in the prevalence of HIV and sexually transmitted diseases among female sex workers in Cotonou, Benin, 1993-1999
25.Fishbein M, Pequegnat W. Evaluating AIDS prevention interventions using behavioral and biological outcome measures
. Sex Transm Dis
26.Hadden BR. An HIV/AIDS Prevention Intervention with Female and Male STD Patients in a peri-Urban Settlement in KwaZulu Natal, South Africa
. International Center for Research on Women: Women and AIDS Research Program; Project Summary [Conference Version]; June, 1997.
27.Marlow RM, Ziskind D, Jones DL. Use of female controlled microbicidal products for HIV risk reduction
. AIDS Care
28.Roddy RE, Zekeng L, Ryan KA, Tamoufe U, Weir SS, Wong EL. A controlled trial of nonoxynol 9 film to reduce male-to-female transmission of sexually transmitted diseases
. N Engl J Med
29.Steiner MJ, Glover LH, Bou-Saada I, Piedrahita C. Increasing barrier method use among oral contraceptive users at risk of STDs: what approach is best?Sex Transm Dis
30.Van Damme L, Ramjee G, Alary M, Vuylsteke B, Chandeying V, Rees H, et al
. Effectiveness of COL-1492, a nonoxynol-9 vaginal gel, on HIV-1 transmission in female sex workers: a randomised controlled trial
31.Farr G, Castro LAA, Disantostenfano R, Claassen E, Olguin F. Use of spermicide and impact of prophylactic condom use among sex workers in Santa Fe de Bogota, Colombia
. Sex Transm Dis
32.Richardson BA, Lavreys L, Martin HL, Jr., Stevens CE, Ngugi E, Mandaliya K, et al
. Evaluation of a low-dose nonoxynol-9 gel for the prevention of sexually transmitted diseases: a randomized clinical trial
. Sex Transm Dis
33.Roddy RE, Zekeng L, Ryan KA, Tamoufe U, Tweedy KG. Effect of nonoxynol-9 gel on urogenital gonorrhea and chlamydial infection: a randomized controlled trial
34.Sokal DC, King TDN, Crane S, Potts M. Modeling the efficacy of vaginal virucides: a simple projection
. In: Feldblum P. (presenter): Developing Virucidal Compounds for Intravaginal Use in Preventing the Sexual Transmission of HIV
. New York: Population Council; 29 June, 1992.
35.Watts CH, Thompson WA, Heise LL. The impact of microbicides for HIV prevention: Results of a mathematical modeling exercise
. XII World AIDS Conference
. Geneva, June–July 1998 [Abstract 354].
36.Garnett GP, Anderson RM. Strategies for limiting the spread of HIV in developing countries: conclusions based on studies of the transmission dynamics of the virus
. J Acquir Immune Defic Syndr Hum Retrovirol
37.Rehle TM, Saidel TJ, Hassig SE, Bouey PD, Gaillard EM, Sokal DC. AVERT: a user-friendly model to estimate the impact of HIV/sexually transmitted disease prevention interventions on HIV transmission
1998; 12(Suppl. 2)
38.Weinstein MC, Graham JD, Siegel JE, Fineberg HV. Cost-effectiveness analysis of AIDS prevention programs: concepts, complications and illustrations
. In: Turner CF, Miller HG, Moses LE (editors): Confronting AIDS: Sexual behaviour and intravenous drug use
. Washington DC: National Academy Press; 1989. pp. 471–499.
39.Pinkerton SD, Abramson PR. Effectiveness of condoms in preventing HIV transmission
. Soc Sci Med
40.Buve A, Weiss HA, Laga M, Van Dyck E, Musonda R, Zekeng L, et al
. The epidemiology of gonorrhoea, chlamydial infection and syphilis in four African cities
2001; 15(Suppl. 4)
41.Lowndes CM, Alary M, Gnintoungbe CA, Bedard E, Mukenge L, Geraldo N, et al
. Management of sexually transmitted diseases and HIV prevention in men at high risk: targeting clients and non-paying sexual partners of female sex workers in Benin
42.Lowndes CM, Alary M, Meda H, Gnintoungbe CAB, Mukenge Tshibaka L, Adjovi C, et al
. Role of core and bridging groups in the transmission dynamics of HIV and STIs in Cotonou, Benin, West Africa
. Sex Transm Infect
2002; 78(Suppl. 1)
43.European Study Group on Heterosexual Transmission of HIV. Comparison of female to male and male to female transmission of HIV in 563 stable couples
44.Chesson HW, Pinkerton SD. Sexually transmitted diseases and the increased risk for HIV transmission: implications for cost-effectiveness analyses of sexually transmitted disease prevention interventions
. J Acquir Immune Defic Syndr
45.Garnett G, Anderson RM. Sexually transmitted diseases and sexual behavior: insights from mathematical models
. J Infect Dis
1996; 174(Suppl. 2)
46.Korenromp EL, Van Vlient C, Grosskurth H, Gavyole A, Van der Ploeg PB, Fransen L, et al
. Model-based evaluation of single-round mass treatment of sexually transmitted diseases for HIV control in a rural African population
47.Quinn TC, Wawer MJ, Sewankambo N, Serwadda D, Li C, Wabwire-Mangen F, et al
. Viral load and heterosexual transmission of human immunodeficiency virus type 1
. N Engl J Med
48.Royce RA, Sena A, Cates W, Cohen MS. Current concepts: sexual transmission of HIV
. N Engl J Med
51.Watts C, Vickerman P. The impact of microbicides on HIV and STD transmission: model projections
2001; 15(Suppl. 1)