Trichomonas vaginalis is 6 to 20 times more common among HIV-positive women; reported prevalence of infection ranges from 12% to 63%.1 Given the significant burden of disease, the Centers for Disease Control and Prevention (CDC) 2010 sexually transmitted diseases treatment guidelines recommend annual Trichomonas screening in HIV-positive women.2 Because of high recurrence rates of Trichomonas infection among these women (18%–36%), repeat testing 3 months after treatment is recommended.3–5
Trichomonas vaginitis may increase the risk of HIV transmission to susceptible partners by a factor of 1.5 to 3.6,7 This elevated risk of transmission likely results from increased HIV viral shedding in the presence of Trichomonas infection.4,8,9 Treatment decreases HIV genital shedding.10,11 Mathematical modeling suggests that Trichomonas infection among HIV-infected men in the United States leads to an estimated 6% of new HIV infections among HIV-susceptible female partners each year. The cost of these new HIV infections secondary to Trichomonas infection could account for US$46 to US$382 million dollars annually.9
Metronidazole, the preferred antimicrobial treatment of T. vaginalis, is relatively inexpensive and eradicates approximately 95% to 97% of infections after a single treatment.2,12 Traditionally, saline microscopy has been used to diagnose Trichomonas, but the sensitivity is fairly poor (60%–70%).12 Until nucleic acid amplification tests (NAATs) became widely available for the diagnosis of Trichomonas, the gold standard for the diagnosis of Trichomonas infection was culture (InPouch TV test; BioMed Diagnostics, San Jose, CA). The sensitivity of Trichomonas culture ranges from 70% to 89%.13 Culture is less expensive than other more sensitive diagnostic methods, such as NAATs (APTIMA Trichomonas assay [Gen-Probe Inc, San Diego, CA] and AFFIRM VPIII [BD Diagnostics, Sparks, MD]) and rapid antigen tests (OSOM Trichomonas rapid test; Genzyme Diagnostics, Cambridge, MA),.4,12
Given the relative availability of screening techniques and affordability of effective treatment, we hypothesized that annual Trichomonas screening, treatment, and follow-up of HIV-positive women as recommended by the CDC would be cost-effective in decreasing the lifetime costs associated with new HIV infections in heterosexual susceptible male partners.
MATERIALS AND METHODS
We conducted a decision tree analysis to test the hypothesis that Trichomonas screening strategies are cost-effective due to prevention of transmission of HIV infections to male partners of HIV-positive women with untreated Trichomonas infection. This study was institutional review board exempt. Using Microsoft Excel 2010, we created a decision tree to compare cost-effectiveness of Trichomonas screening and treatment according to the CDC 2010 STD treatment guidelines versus no screening.2 The model represented a 12-month period. Two hundred women were entered into the model simultaneously: 100 women were screened for Trichomonas and 100 women unscreened. Women diagnosed as having Trichomonas infection at the initial screening were rescreened in 3 months.
Nucleic acid amplification tests are the most sensitive and specific modality for Trichomonas diagnosis, but because of cost, they are not readily available in most resource-limited settings. We chose to use culture in this model because it has increased sensitivity over microscopy, is generally available, and is affordable compared with NAATs. The sensitivity and specificity of culture were calculated using the means of current estimates. We assumed a sensitivity of 0.79 (0.7–0.89) and a specificity of 1.0.13–15 Using a mean prevalence from previous studies in HIV-positive women, we assumed 23% of women in the model would have a Trichomonas infection (range, 0.12–0.63).1,3,4,16 The reported range of Trichomonas prevalence among HIV-positive women varies by geographical location and study population, with the highest prevalence reported among urban HIV-positive women with a history of substance abuse.4
Because of high rates of recurrent Trichomonas infection among HIV-positive women (9%–36%), repeat screening in 3 months after treatment is recommended.3,5,17 In prior studies, it is difficult to discern if recurrent infections are due to treatment failure or reinfection. Rates of treatment failure after first-line therapy with metronidazole are reported as high as 5% (2%–5%) in HIV-negative women.2,4,12 HIV-positive women with Trichomonas infection treated with metronidazole have failure rates as high as 16.8% after single-dose therapy and 8.5% after 7 days of therapy.17 It is unclear as to whether these high rates of recurrence among HIV-positive women are secondary to reinfection or treatment failure. The reported rates of metronidazole efficacy are 90% to 95%.2 To avoid over estimating treatment failure, we chose to use a treatment failure rate of 5%. We calculated the probability of treatment failures after therapy and accounted for the proportion of women who would require repeat screening and treatment. We also accounted for the effect of screening failures or false-negative Trichomonas cultures.
Prolonged therapy for Trichomonas infection with metronidazole 500 mg by mouth twice daily for 7 days has been shown to be more effective than a single dose of 2 g in HIV-positive women.17 The costs of Trichomonas screening using InPouch TV culture ($29.63) and the costs of 7-day therapy with metronidazole (US$3.35) were calculated from published costs for medical fees in the United States as of 2008.18 These are mean cost estimates and account for labor costs. The costs were inflated to 2012 dollars using a medical inflation factor (1.1398) calculated from the United States Bureau of Labor Statistics (http://data.bls.gov/pdq/SurveyOutputServlet, accessed December 20, 2011).
The cost for repeat screening at 3 months in women with Trichomonas infection and cost of treatment for those with continued infection were accounted for. There were no screening or treatment costs for the 100 women in the no-screen group. The costs of no screening were attributed to the cost of new HIV infections in the susceptible male partners associated with untreated Trichomonas in HIV-positive women.
We calculated the likelihood of transmission of HIV to susceptible male partners based on the calculations for the risk of HIV transmission through vaginal intercourse. We assumed a 0.04% risk of HIV transmission from female to male per act of vaginal intercourse in the absence of antiretroviral therapy (ARV). Based on the previously reported frequency of sexual acts among HIV-positive women, we estimated the rate of sexual acts to be 1 per week (52 per year).19,20
For 100 HIV-positive women without Trichomonas infection, the expected rate of transmissions in a year would be 2.06 based on the formula [1 − (1 − 0.0004)52]. Using a conservative estimate based on a previously published risk of HIV acquisition/transmission, we assumed a 1.5 (1–3)-fold increased risk of HIV transmission secondary to Trichomonas infection.9,21,22 Assuming all 100 HIV-positive women had Trichomonas, approximately 3.09 cases of HIV transmission would occur using a simple approximation of 2.06 × 1.5. Using a Trichomonas infection prevalence 23% and assuming all partners are HIV susceptible, we calculated 2.3 cases [(3.09 × 0.23) + (2.06 × 0.77)] of HIV transmission to partners would occur annually among 100 HIV-positive women.
Multiple factors are associated with an increased risk of heterosexual HIV transmission. These factors may include but are not limited to the following: genital and plasma HIV RNA viral loads, HIV subtype, use of ARV therapy for patient treatment or partner prophylaxis,23 frequency and types of unprotected sex,2 male circumcision,24 and concomitant lower genital tract infections in either partner.10 It was difficult to estimate the proportion of HIV-positive women who are sexually active with susceptible male partners. Because of the multifactorial nature of HIV acquisition in susceptible male partners, estimating a risk of transmission is also difficult. In this model, we did not attempt to account for all potential factors associated with HIV transmission. From our basic model, which assumes no ARV use among HIV-positive women and all male partners are HIV susceptible, we adjusted parameters in an attempt to account for the effect of ARV therapy on HIV transmission.
From a recent meta-analysis of HIV infectiousness among discordant couples, we assumed a 91% (76%–96%) reduction in heterosexual HIV transmission to susceptible male partners of HIV-positive women on ARV therapy.23 Assuming a 91% reduction in HIV transmission, the transmission rate among 100 women not treated for Trichomonas infection would be 0.21 (2.3 × 0.09) and 0.19 (2.06 × 0.09) in Trichomonas-negative women.
To account for the number of male partners per year at risk of HIV acquisition, we extrapolated data from studies of partner notification and treatment of Trichomonas infection in HIV-positive women.5,17,20 In these studies, most women were heterosexual (98%), more than half were taking ART (58%), and 54% reported consistent condom use during at least 75% of sexual acts. Approximately one-third of all partners (33%; 95% confidence interval, 21%–44%) were reportedly unaware of the female partner’s HIV infection. More than half of HIV-positive women (52%) reported having sex with more than 1 partner in a 3-month period. Women with multiple sexual partners had a mean (SD) of 4 (1) partners per year.5,17,20
From these data, we assumed that 52% of HIV-positive women would have multiple sexual partners with approximately 4 partners per year.17,20 We selected a conservative estimate of the number of male partners unaware of their partner’s HIV infection (21%) based on previous reports that 21% to 44% of male partners were aware of their partner’s HIV positive status. Taking into account these factors, the number of male partners susceptible to HIV infection would be 54 per 100 women ([48 + (52 × 4)] × 0.21). When accounting for the number of HIV-susceptible male partners (21%) and the potential effect of ART (transmission reduction by 91%) in patients taking ARVs (36%),25 the transmission rates of HIV would range from 0.83 in women not treated for Trichomonas (23% prevalence) to 0.74 in those without Trichomonas infection.
The lifetime cost of each new HIV infection was assumed to be US$224,000. This amount is based on 2002 dollar estimates.9 The inflation factor (1.498) was applied to estimate the 2012 lifetime cost of HIV. The estimated lifetime cost of each case of HIV was US$335,552 (US $224,000 × 1.5; Table 1).
Sensitivity analyses were performed on multiple factors impacting the cost of Trichomonas screening within the original model to account for variability of our assumed values. The factors adjusted in the sensitivity analyses included the following: costs (screening, treatment, and new HIV infections), prevalence of HIV, and the sensitivity and specificity of Trichomonas culture. Factors associated with cost of screening, Trichomonas treatment, expedited partner therapy (EPT), and lifetime cost of HIV infection per case were varied by 50%. The sensitivity and specificity of the culture were varied according to published ranges.26 We also adjusted the assumptions of the model to account for variation in costs associated with the different Trichomonas screening methods. These included microscopy, NAATs, and the rapid antigen reagent test (Table 2).
Given the assumptions of our model, annual Trichomonas screening in HIV-positive women may be cost-effective in reducing the lifetime cost of new HIV infections among susceptible partners. Because of a decrease in new HIV infections associated with effective Trichomonas treatment, this approach could be associated with significant lifetime cost savings. From our calculations, untreated Trichomonas infection is associated with an HIV transmission rate of 2.3 per 100 HIV-positive women annually, assuming a 23% prevalence of Trichomonas. When screening with culture and treating Trichomonas infections for 7 days with metronidazole for the purpose of decreasing new HIV infections among all male partners, US$553 (US$475– US$645) is saved per woman in the lifetime costs of new HIV infections. This estimate represents the money saved per woman in the costs of new HIV cases avoided in susceptible male partners (Table 3).
The rate of HIV transmission used in this model was adjusted for multiple factors that could potentially impact the infectiousness of the female HIV positive partner. When varying these factors over published ranges, our model continued to demonstrate cost-effectiveness of screening and treatment of Trichomonas vaginitis to reduce HIV transmission. The reduction in HIV transmission associated with varying partner susceptibility had the greatest impact on cost-effectiveness of Trichomonas screening and treatment to prevent HIV transmission. However, partner susceptibility to HIV is not a clinically modifiable factor. The modifiable factor most associated with the greatest reduction in transmission is ARV use among HIV-positive women (Table 4).
Sensitivity analyses were performed on the original model. First, we adjusted the multiple assumptions within the basic model in which no women were taking ARV therapy and all male partners were susceptible to HIV infection. Trichomonas screening to prevent transmission of HIV continued to be cost-effective when changing all assumptions in this decision tree model. The factor that affected cost-effectiveness the most was the lifetime cost of HIV infection (Fig. 1).
Cost-effectiveness was maintained when accounting for the costs, sensitivity, and specificity of the different Trichomonas screening methods. We varied the sensitivity of these methods according to previously published studies.15,26 Nucleic acid amplification tests were associated with the greatest costs savings when adjusting the decision tree model for the cost, sensitivity, and specificity of NAATs (Table 2).
Based on 2009 estimates, 288,000 US women were living with HIV/AIDS.27 Using this estimate and the assumptions of our decision tree analysis, we calculated the potential number of HIV transmissions in 2009 that could have been avoided by treating Trichomonas infection. Assuming 2.06 heterosexual HIV transmissions to male partners per 100 women occur annually among women without Trichomonas infection, the number of female-to-male heterosexual transmissions in 2009 would have been approximately 5933 [(288,000 × 2.06)/100]. If 23% of HIV-positive women in 2009 were infected with Trichomonas, 66,240 (288,000 × 0.23) were 1.5-fold more likely to transmit HIV to susceptible male partners. When accounting for an increased rate of transmission due to Trichomonas infection, the number of probable female-to-male transmissions increases from 5933 to 6615. This represents 682 additional cases of HIV transmission per year. When considering the lifetime cost of HIV infection (US$335,552), preventing 682 new cases of HIV by treating Trichomonas infection in women would save US$228,846,464 (US$335,552 × 682) in lifetime costs of new HIV infections.
Assuming 21% of male partners are HIV susceptible and 52% of HIV-positive women have approximately 4 partners per year, 154,828 male partners would be susceptible [((288,000 × 0.52) × 4 + (288,000 × 0.48)) × 0.21]. Among women not screened and treated for Trichomonas, 3561(154,828 × 2.3) cases of HIV transmission would occur per year. When accounting for Trichomonas screening, treatment, and treatment failure among HIV-positive women, the transmission rate is 2.12 and the number of HIV infections would decrease to 3282 (154,828 × 2.12). Treatment of Trichomonas infection among HIV-positive women in the United States could potentially prevent 186 to 346 new cases of HIV per year. We assumed 36% of women would be taking ARV therapy.25 When adjusting for the effect of ART on HIV transmission, Trichomonas screening and treatment maintains cost-effectiveness by averting new HIV infections (Table 5).
According to current CDC HIV statistics, approximately 24% of all people living with HIV/AIDS in the United States are women.27 If our baseline cost-effectiveness model is correct, annual screening and treatment for HIV-positive women for Trichomonas infection is associated with a lifetime savings of US$553 (US$475– US$645) per woman in the prevention of new HIV infections to susceptible male partners. Based on HIV prevalence among women in 2009, if all 288,000 HIV-positive American women were screened according to current guidelines, the amount saved annually in the prevention of new HIV infections would approximate US$159,264,000 (US288,000 × US$553)]2 (Table 5).
Our decision tree model for the assessment of cost-effectiveness of Trichomonas screening and treatment in HIV-positive women to prevent HIV transmission suggests that this approach is cost-effective. The strength of the model is illustrated in the small variation in cost-effectiveness when performing a sensitivity analysis on the various assumptions. We attempted to account for many factors that may lead to dramatic shifts in costs. The most influential factor was found to be the lifetime cost of HIV infection (Fig. 1).
One of the most affirming findings in the model to support the cost-effectiveness of this approach is the relative in expense of EPT. The total cost of patient and EPT was only US$4.91 per infection episode. Patient and partner therapy were associated with an annual cost savings of US$553 per woman, the same as expected for patient treatment alone. Second, the diagnostic approach to Trichomonas screening had very little effect on cost-effectiveness. The cost savings per woman varied from US$337 (liquid-based cytology with a sensitivity of 57%) to US$656 (NAATs with a sensitivity of 95%; Table 2).
Our model accounted for a 5% failure rate of metronidazole. This rate of failure is based on a single dose of metronidazole 2000 mg.2 The annual cost savings per woman using single-dose therapy is US$554. This value is unchanged from the cost associated with the recommended therapy in HIV-positive women, metronidazole 500 mg twice daily for 7 days. This is the same regimen recommended as first-line therapy in suspected cases of decreased Trichomonas susceptibility to metronidazole. We did not account for the costs of repeat testing in the face of resistance, dose escalation, or the use of tinidazole.2
Cost-effectiveness models are theoretical and only as strong as the weakest assumption. Our model could not account for all factors associated with heterosexual HIV transmission and the effect of Trichomonas infection on transmission risks. In our initial model, we assumed that all women in this study were sexually active and that all partners were HIV susceptible. To determine a more realistic risk of HIV transmission, we estimated the number of susceptible male partners based on the proportion unaware of their partner’s HIV positive status (21%).5 We accounted for the proportion of women reporting more than 1 sexual partner per year (52%) and assumed a mean number of partners per year among these women to be 4 ± 1.17,20
In addition to estimating susceptible male partners, we adjusted the baseline model to demonstrate the effect of ARV therapy on HIV transmission. When reducing transmission rates by 91%, we assumed that 36% of these women were taking ARV therapy. We used this proportion based on the 2011 CDC estimate that 36% of HIV-infected Americans are taking ARV therapy.25 Antiretroviral therapy therapy was reported to be slightly higher (58%–65%) in prior studies of HIV-positive women with Trichomonas infection.5,17,20 Daily single-dose ARV therapy costs approximately US$12,000 per woman annually.28 Because of the complexity of our model, we could not account for the associated costs of ARVs.
Our model calculated HIV transmission rate according to the risk associated with vaginal intercourse. We did not account for the increased risk of HIV transmission in patients having receptive anal intercourse, which is associated with a higher rate of transmission (0.1%–0.2% vs. 0.5–1.0%).2 We were unable to account for new Trichomonas infections, including reinfection from untreated partners. Previous studies report a reinfection rate among HIV-positive women to be 27% (18%–36%).3,29 In our study of 200 women, this would have added an additional 6 cases in the screened group that would require additional treatment and 3-month follow-up.
Another important limitation of our study is that we were unable to account for the additive effect of concomitant sexually transmitted infections on HIV transmission. Ulcerative genital diseases such as herpes simplex virus and syphilis and gonorrhea cervicitis are associated with an increased risk of HIV transmission.30 Screening and treatment of other sexually transmitted infections may have a significant impact on the effects of Trichomonas and HIV transmission, but these potential effects are multifactorial.7
In conclusion, we have demonstrated that following current CDC STD treatment guidelines for the screening and treatment of Trichomonas infection in HIV-positive women would prove cost-effective in the prevention of new HIV cases. This model is based on the assumption that Trichomonas infection alone is associated with a 1.5-fold increased risk of HIV sexual transmission in heterosexual couples. Future studies in cost-effectiveness modeling could account for HIV transmission among men who have sex with men and the limitations of our model we have described. From these data presented, the United States could save a substantial amount of money and prevent approximately 186 to 346 cases of HIV per year by following the CDC 2010 STD treatment guidelines for the screening and treatment of T. vaginalis.
1. Cu-Uvin S, Ko H, Jamieson DJ, et al. Prevalence, incidence, and persistence or recurrence of trichomoniasis among human immunodeficiency virus (HIV)-positive women and among HIV-negative women at high risk for HIV infection. Clin Infect Dis 2002; 34: 1406–1411.
2. Workowski KA, Berman S. Sexually transmitted diseases treatment guidelines, 2010. MMWR Recommend Rep 2010; 59: 1–110.
3. Niccolai LM, Kopicko JJ, Kassie A, et al. Incidence and predictors of reinfection with Trichomonas vaginalis
in HIV-infected women. Sex Transm Dis 2000; 27: 284–288.
4. Miller MR, Nyirjesy P. Refractory trichomoniasis in HIV-positive and HIV-negative subjects. Curr infect Dis Rep 2011; 13: 595–603.
5. Gatski M, Mena L, Levison J, et al. Patient-delivered partner treatment and Trichomonas vaginalis
repeat infection among human immunodeficiency virus–infected women. Sex Transm Dis 2010; 37: 502–505.
6. McClelland RS, Sangare L, Hassan WM, et al. Infection with Trichomonas vaginalis
increases the risk of HIV-1 acquisition. J Infect Dis 2007; 195: 698–702.
7. Laga M, Nzila N, Goeman J. The interrelationship of sexually transmitted diseases and HIV infection: implications for the control of both epidemics in Africa. AIDS 1991; 5 (suppl 1): S55–S63.
8. Kovacs A, Wasserman SS, Burns D, et al. Determinants of HIV-1 shedding in the genital tract of women. Lancet 2001; 358: 1593–1601.
9. Chesson HW, Blandford JM, Pinkerton SD. Estimates of the annual number and cost of new HIV infections among women attributable to trichomoniasis in the United States. Sex Transm Dis 2004; 31: 547–551.
10. Wang CC, McClelland RS, Reilly M, et al. The effect of treatment of vaginal infections on shedding of human immunodeficiency virus type 1. J Infect Dis 2001; 183: 1017–1022.
11. Kissinger P, Amedee A, Clark RA, et al. Trichomonas vaginalis
treatment reduces vaginal HIV-1 shedding. Sex Transm Dis 2009; 36: 11–16.
12. Schwebke JR, Burgess D. Trichomoniasis. Clin Microbiol Rev 2004; 17: 794–803. Table of contents.
13. Nye MB, Schwebke JR, Body BA. Comparison of APTIMA Trichomonas vaginalis
transcription-mediated amplification to wet mount microscopy, culture, and polymerase chain reaction for diagnosis of trichomoniasis in men and women. Am J Obstet Gynecol 2009; 200: 188 e1–7.
14. Wendel KA, Erbelding EJ, Gaydos CA, et al. Trichomonas vaginalis
polymerase chain reaction compared with standard diagnostic and therapeutic protocols for detection and treatment of vaginal trichomoniasis. Clin Infect Dis 2002; 35: 576–580.
15. Huppert JS, Batteiger BE, Braslins P, et al. Use of an immunochromatographic assay for rapid detection of Trichomonas vaginalis
in vaginal specimens. J Clin Microbiol 2005; 43: 684–687.
16. Cu-Uvin S, Hogan JW, Warren D, et al. Prevalence of lower genital tract infections among human immunodeficiency virus (HIV)–seropositive and high-risk HIV-seronegative women. HIV Epidemiology Research Study Group. Clin Infect Dis 1999; 29: 1145–1150.
17. Kissinger P, Mena L, Levison J, et al. A randomized treatment trial: single versus 7-day dose of metronidazole for the treatment of Trichomonas vaginalis
among HIV-infected women. J Acquir Immune Defic Syndr 2010; 55: 565–571.
18. Staff PE. Medical Fees in the United States: Nationwide Charges for Medicine, Surgery, Laboratory, Radiology, and Allied Health Services. Los Angeles, CA: Practice Management Information Corporation; 2008; 1.
19. Boily MC, Baggaley RF, Wang L, et al. Heterosexual risk of HIV-1 infection per sexual act: Systematic review and meta-analysis of observational studies. Lancet Infect Dis 2009; 9: 118–129.
20. Schwebke JR, Desmond RA. A randomized controlled trial of partner notification methods for prevention of trichomoniasis in women. Sex Transm Dis 2010; 37: 392–396.
21. Sorvillo F, Smith L, Kerndt P, et al. Trichomonas vaginalis
, HIV, and African-Americans. Emerg Infect Dis 2001; 7: 927–932.
22. Van der Pol B. Trichomonas vaginalis
infection: The most prevalent nonviral sexually transmitted infection receives the least public health attention. Clin Infect Dis 2007; 44: 23–25.
23. Baggaley RF, White RG, Hollingsworth TD, et al. Heterosexual HIV-1 infectiousness and antiretroviral use: Systematic review of prospective studies of discordant couples. Epidemiology 2013; 24: 110–121.
24. Weiss HA, Quigley MA, Hayes RJ. Male circumcision and risk of HIV infection in sub-Saharan Africa: A systematic review and meta-analysis. AIDS 2000; 14: 2361–2370.
25. Vital signs: HIV prevention through care and treatment—United States. MMWR Morb Mortal Wkly Rep 2011; 60: 1618–1623.
26. Radonjic IV, Dzamic AM, Mitrovic SM, et al. Diagnosis of Trichomonas vaginalis
infection: The sensitivities and specificities of microscopy, culture and PCR assay. Eur J Obstet Gynecol Reprod Biol 2006; 126: 116–120.
27. Prejean J, Song R, Hernandez A, et al. Estimated HIV incidence in the United States, 2006–2009. PloS One 2011; 6: e17502.
28. Gebo KA, Fleishman JA, Conviser R, et al. Contemporary costs of HIV healthcare in the HAART era. AIDS 2010; 24: 2705–2715.
29. Kissinger P, Secor WE, Leichliter JS, et al. Early repeated infections with Trichomonas vaginalis
among HIV-positive and HIV-negative women. Clin Infect Dis 2008; 46: 994–999.
30. Laga M, Manoka A, Kivuvu M, et al. Non-ulcerative sexually transmitted diseases as risk factors for HIV-1 transmission in women: Results from a cohort study. AIDS 1993; 7: 95–102.