Tuli, Karunesh MD, PhD; Kerndt, Peter R. MD, MPH
Men who have sex with men (MSM) have a high risk for acquiring and spreading sexually transmitted infections (STIs) such as chlamydia, gonorrhea, hepatitis B, HIV, and syphilis.1 In recent years, outbreaks of syphilis have occurred among MSM in a number of US cities.2–6 Of special concern are the high rates of HIV infection found among those affected by the outbreaks. The prevalence of STIs is high among incarcerated individuals.7 Compounding the burden of illness suffered by inmates is the risk of transmission from released individuals to their sexual partners outside correctional facilities. At the same time, detention presents an opportunity to public health workers to intervene and reduce levels of infection among inmates and the potential for further transmission to others.8,9
Economic evaluations of STI screening among incarcerated MSM are not available in the published literature. The cost-effectiveness of screening for STIs among inmates in US prisons and jails has been assessed but without explicit attention to sexual orientation.10,11 Syphilis screening and HIV counseling and testing among incarcerated individuals were found to be cost-saving. Screening for chlamydia and gonorrhea was judged to be cost-effective among women but not among men. However, the effect of screening and treatment on future STI transmission was not examined.
In response to an outbreak of syphilis among MSM, the Los Angeles County Sexually Transmitted Disease Program initiated a screening, treatment, and condom provision intervention in the segregated MSM unit of the Los Angeles County Men's Jail (average census 300, average stay 46 days). The intervention started with screening for syphilis and HIV in March 2000.12–14 Screening for chlamydia and gonorrhea was initiated in April 2000. This article presents the results of an analysis that was conducted to determine the cost-effectiveness of the intervention in averting STIs among inmates of the MSM unit and partners with whom they have sex after leaving the jail.
MATERIALS AND METHODS
The cost-effectiveness analysis was conducted from a perspective in which all costs and benefits are included, no matter who pays or benefits. However, because the intervention is jail-based, productivity losses and inmate time were not included.
Cost estimates and other parameters were obtained from the Los Angeles County Department of Public Health and the literature and are summarized in Table 1.
Figure 1 provides a schematic overview of the model used for estimating the effect of the intervention on STI transmission. The model was adapted from the one developed by Hethcote and Yorke for gonorrhea transmission.16 MSM are categorized into those who are highly active sexually (represented by subscript 1 in the equations below) and those who are less active (represented by subscript 2). Hethcote and Yorke have shown that transmission dynamics of STIs such as gonorrhea depend critically on the existence of a small highly sexually active (core) group and on sexual interaction between this group and the less sexually active (noncore) group.
The daily rate of change in the size of the infected population is given by the following differential equation.
where Ii is the prevalence of infection and Ni is the size of the group.
Equation (Uncited)Image Tools
The number of infected MSM who recover per day is
where D is the duration of infectivity in days.
Equation (Uncited)Image Tools
The number of susceptible MSM who become infected per day is
where Ai is the number of new sexual partners per day, Q is the transmission probability per sexual partner, and Bi is as explained in Eq. 4 below
Note that the model assumes that (a) individuals do not display a preference for selecting sexual partners according to group; (b) individuals in the MSM unit do not have sex with individuals outside the unit; and (c) MSM do not have sex with women. As shown in the figure, transmission of infection is modeled within and outside the MSM unit. The rate at which MSM enter and leave the unit is defined by parameters presented in Table 1. The ratio of core to noncore individuals inside the unit is assumed to be the same as the ratio outside the unit.
The prevalence of infection among individuals entering the MSM unit is the result of a complex dynamic that includes duration of infectivity, transmission probability, and sexual interaction. The sexual transmission parameters listed in Table 1 are used in the model to generate prevalence levels similar to those found among individuals entering the MSM unit; they reflect existing treatment and control patterns. It is assumed that these patterns are similar inside and outside the MSM unit in the absence of the intervention. The introduction of the intervention alters the dynamic by reducing the duration of infectivity and if accompanied by a condom promotion activity, by reducing the probability of transmission.
The model was run twice for each infection. In the first run, no screening occurred. Infections that occurred over a 13-year-period were cumulated; a 3% annual discount rate was applied. In the second run, the intervention was initiated in the fourth year (that is, after prevalence of infection approximately equaled the value given in Table 1). Cost, intervention coverage, screening accuracy, and treatment parameters were as shown in Table 1. Infected individuals who were successfully identified and treated at intake were moved to the susceptible group in the MSM unit. It was assumed that the process of screening and treatment was completed during the intake process. Infected individuals who were not screened, detected, or successfully treated at intake joined the infected group in the unit. Again, infections occurring over a 13-year-period and intervention costs were cumulated. As in the first run, a 3% annual discount rate was applied to costs and number of infections. The 2 runs were compared to obtain the number of infections averted by the 10-year intervention.
The official policy in many correctional facilities is to prohibit sexual activity among inmates.1 However, it is likely that such activity occurs even if prohibition is in place. A study in Tennessee found that 18% of male prisoners reported same-sex intercourse while in prison; 7% reported such sex while not incarcerated.1
Three scenarios were developed to reflect uncertainty about sexual behavior in the MSM unit. In the first scenario, it is assumed that sex does not occur in the MSM unit. In this setting, some transmission is averted even if no screening occurs because infected inmates are not able to infect susceptible inmates; individuals who are still infected when they leave the unit can infect their partners outside the unit. If a screening and treatment intervention is implemented, additional infections are averted because fewer individuals are still infected when they leave the unit. However, this increment is smaller than the number of infections averted due to screening in a setting where inmates' sexual behavior is unchanged after incarceration; in the latter setting, the intervention alters transmission dynamics both inside and outside the unit. Thus, the first scenario provides a lower-bound estimate of intervention effectiveness and is the most conservative from a cost-effectiveness standpoint. The second scenario assumes unchanged sexual behavior.
The Los Angeles County Men's Jail started providing condoms to inmates in November 2001. In addition to the assumption about sexual behavior made in the second scenario, the third scenario assumes that 20% of screened inmates use condoms. This assumption is consistent with data from studies of condom use among individuals who are counseled regarding safe sex when they undergo HIV testing.11 In this scenario, transmission is reduced both because of condom use and the transfer of individuals from the infected group to the susceptible group (as a result of screening and treatment). Thus, this scenario provides a favorable (least conservative) estimate of intervention effectiveness. Additionally, since antiretroviral therapy is not included in the analysis, the third scenario is the only one in which HIV transmission is affected by the intervention.
All costs (Table 1) are in US dollars and have been adjusted to March 2000 using the medical care component of the consumer price index.36
Screening costs include laboratory costs (Public Health Laboratory, Los Angeles County Department of Public Health), cost of supplies used during specimen collection, and the cost of personnel time for collection and processing of specimens. For chlamydia and gonorrhea, amplification tests are performed on urine. For syphilis, blood tests are performed; a positive rapid plasma reagin test is confirmed by a specific treponemal test (Treponema pallidum particle agglutination assay). For HIV too, blood tests are performed; positive enzyme immunoassay tests are confirmed by Western Blot tests. Screening costs for HIV seropositive inmates are higher than those for seronegative individuals because of the need for confirmatory testing and additional counseling.
Treatment costs include medication costs (Los Angeles County Department of Public Health), cost of supplies and personnel time, and the cost of treating reactions to drugs. The treatment regime for chlamydia is azithromycin (1 g orally in a single dose), for gonorrhea ceftriaxone (125 mg intramuscularly in a single dose), and for syphilis benzathine penicillin G (2.4 million units intramuscularly in a single dose). Levofloxacin, a fluoroquinolone commonly used orally for gonorrhea treatment, is no longer recommended for use in California because of the increasing prevalence of resistance.37 Cefixime, also an orally administered medication for gonorrhea, is not being marketed in the United States anymore.38 Doxycycline is a cheaper alternative to azithromycin for the treatment of chlamydial infections. However, it requires week-long therapy and compliance may be less than optimal, especially in a jail setting.10,25
Cost saving from averted chlamydial and gonococcal infections includes the cost of outpatient treatment for urethritis and epididymitis and hospitalization costs for epididymitis. For syphilis, costs saved per averted case include the costs of treatment of primary, secondary, and early latent syphilis as well as the cost of long term sequelae (late latent syphilis, late benign syphilis, cardiovascular syphilis, and neurosyphilis). Each averted HIV case results in savings of lifetime treatment costs. Estimates provided by Holtgrave and Pinkerton for intermediate level care are used.19
The cost of condoms (to Los Angeles County Department of Public Health) is added to intervention costs in the third scenario (in which 20% of screened inmates use condoms). Usage is estimated—conservatively from the cost-effectiveness standpoint—at 1 condom per user per day.
Calculating Cost-Effectiveness Ratios
For chlamydia and gonorrhea, the cost of a symptomatic episode (given in Table 1) was multiplied by the number of infections averted to calculate treatment costs averted by the intervention. For HIV and syphilis, cost per case was similarly multiplied by the number of infections averted. Net cost was calculated by subtracting averted treatment costs from intervention cost. Cost-effectiveness ratios were obtained by dividing net costs by the number of infections averted. Ratios were not calculated if averted treatment costs exceeded intervention costs (that is, the intervention was found to be cost-saving).
Sensitivity analyses were conducted to study the effect of changes in key parameters on the cost-effectiveness of the intervention.
Table 2 summarizes the findings of the base case assessment of a 10-year intervention in which 40% of MSM are screened at entry into the unit and 63% of those who test positive are treated. The net discounted cost of the intervention is $179,121 in the most conservative scenario. Modeling indicates that the intervention could avert 339 chlamydia infections, 276 gonorrhea infections, and 241 syphilis infections. In the intermediate scenario, the intervention could avert 593 chlamydia, 586 gonorrhea, and 367 syphilis infections (with cost saving). In the least conservative scenario, the intervention generates additional cost savings and could avert 746 chlamydia infections, 791 gonorrhea infections, 3 HIV infections, and 443 syphilis infections.
Screening and treatment for syphilis is cost-saving in all 3 scenarios. For HIV, the intervention is cost saving in the only scenario in which an effort to reduce transmission (through condom promotion) was modeled. Screening and treatment for chlamydia and gonorrhea are not cost saving in any of the scenarios. Net cost per infection averted ranges from $42 to $215 for chlamydia and from $6 to $228 for gonorrhea.
The number of inmates is small in comparison with the total size of the MSM population from which they are drawn (Table 1). Consequently, the impact of the intervention on prevalence outside the unit (as well as overall prevalence) is small. For example, in the most conservative scenario, overall chlamydia prevalence at the end of the 13-year model run is 2.92% in the absence of the intervention and 2.91% with the intervention. However, even small changes in prevalence translate into sizeable numbers of infections averted.
Cost-effectiveness ratios are not significantly affected by changes in the proportion of MSM screened at entry (data not shown). However, the results are sensitive to changes in the proportion of MSM testing positive who receive treatment (Table 3). Data on HIV are not presented in the table because antiretroviral therapy was not included in the analysis. Increasing rates of treatment result in more favorable cost-effectiveness estimates for all STIs. Syphilis screening and treatment is cost saving even at very low treatment rates in all 3 scenarios. Screening and treatment for gonorrhea results in cost savings with a 70% treatment rate in the least conservative scenario. Increasing prevalence also results in more favorable cost-effectiveness estimates for all STIs (Table 4). Screening and treatment for syphilis remains cost-saving when prevalence is as low as half the base case value and also when cost per case is as low as a third of the base case value of $1244.
The study shows that the cost-effectiveness of the intervention depends on the sexual behavior of inmates. Even in a setting unfavorable to screening (no sexual activity among inmates) and with an intervention that screens 40% of inmates at intake and treats only 63% of identified infections, modeling indicates that large numbers of STIs could be averted at a low cost. The intervention could lead to cost-savings if incarcerated MSM continue to engage in sexual activity as they do outside jail. Higher levels of treatment and higher prevalence of infection result in more favorable cost-effectiveness estimates.
It should be noted that the scenarios presented here reflect uncertainty about sexual behavior in the MSM unit. Thus, the first (unfavorable) scenario provides a lower-bound estimate of intervention effectiveness and the third scenario the least conservative (most favorable) estimate; base case cost-effectiveness values should be interpreted as being located in the range delimited by the scenarios (Table 2).
Modeling indicates that screening and treatment for syphilis in the MSM unit is more cost-effective than similar interventions for chlamydia and gonorrhea. This study may have underestimated the cost-effectiveness of screening for HIV by not considering efforts by inmates who test positive to reduce the risk of transmitting infection to their partners after they leave the MSM unit. Even so, the intervention against HIV is cost saving in the third scenario.
In an economic evaluation of HIV counseling and testing in US prisons, Varghese and Peterman also concluded that the intervention is cost saving (even though they did not examine the effect of the intervention on future transmission).11 Similarly, Kraut et al. found screening of incarcerated men for syphilis to be cost saving.10 In their study, screening for chlamydia and gonorrhea was not cost-effective in preventing epididymitis (at a cost per case averted of $39,800 for chlamydial epididymitis and $421,579 for gonococcal epididymitis). However, they did not account for reduction in future STI transmission resulting from the intervention.
The benefits of the intervention are largely societal while the costs are incurred in the jail. Jail authorities are unlikely to see significant changes in STI care expenditures due to averted future infections since much of the decrease in the burden of infections occurs outside the MSM unit (especially in the scenario in which no sex occurs in the unit). In this context, continuation of screening activity in the future will probably depend on continuing personnel and financial support from the Los Angeles County Sexually Transmitted Disease Program.
The model used to study the effect of the intervention on STI transmission is based on the gonorrhea transmission model developed by Hethcote and Yorke.16 STI transmission dynamics are dependent on the existence of a highly sexually active (core) group that interacts with a less sexually active (noncore) group. Hethcote and Yorke have shown that interventions directed at identification and treatment of infected core group members can result in a much bigger impact on transmission than interventions delivered without regard to group membership. The present analysis does not address this question because it is unlikely that staff will be able to differentiate core group members from others before administering screening tests. However, if core group members enter the correctional system at a higher rate than noncore group members, the cost-effectiveness of a jail-based screening initiative is likely to be more favorable than found in this study.
We assumed that screening and treatment is completed during intake into the MSM unit. Delays in identification and successful treatment will reduce the number of infections averted in the unit. However, in the scenario in which no sex occurs in the unit, such delays do not alter transmission dynamics and the conservative cost-effectiveness findings of the study remain valid. A related issue is the tracking of individuals who test positive but leave the unit before receiving treatment. The base case treatment rate of 63% (among those who are screened and test positive) in the unit is conservative. The assumption is that the remaining individuals do not benefit from screening (even though costs are incurred in testing them). The cost-effectiveness of efforts to trace these individuals and treat them (in prisons or outside the correctional system) needs to be determined.
The present analysis underestimates the benefits of screening because transmission from MSM to women is not considered. If MSM engage in sexual activity with women after leaving the MSM unit, the intervention is likely to yield additional savings of treatment costs through prevention of pelvic inflammatory disease among female sex partners. The analysis also excludes productivity gains and intangible benefits such as pain and suffering averted by preventing future cases.
Additionally, the benefits of screening for chlamydia, gonorrhea, and syphilis are understated in the analysis for one other reason. STIs, especially syphilis, are known to facilitate HIV transmission.39,40 This mechanism is not considered in the analysis. Costs and benefits of antiretroviral therapy directed at inmates who test positive for HIV have also not been assessed in this study. Finally, the effect of continued condom use (after inmates leave the MSM unit) on future STI transmission has not been evaluated.
Future analyses should use model structures that account for differences between the STIs and between symptomatic and asymptomatic individuals and consider the impact of STIs on HIV transmission. Alternative analytical approaches to assessment of cost-effectiveness are also worth considering (e.g., comparing the intervention with another that attempts to identify highly sexually active individuals before screening). Finally, future evaluations should assess the cost-effectiveness of the intervention in an additional scenario in which MSM engage in a reduced level of sex after entering the unit.
In conclusion, using conservative assumptions, this study has found that the intervention could avert large numbers of STIs at a reasonable cost and can save costs in a scenario in which inmates continue to engage in sexual activity as they do outside jail. Modest success in efforts to promote condom use among inmates results in additional cost saving. Thus, allocation of funds to this intervention is a good public health investment.
1. Institute of Medicine. The Hidden Epidemic: Confronting Sexually Transmitted Disease. In: Eng TR, Butler WT, eds. Washington, DC: National Academy Press; 1997.
2. Centers for Disease Control and Prevention. Outbreak of syphilis among men who have sex with men – Southern California. MMWR Morb Mortal Wkly Rep 2001; 50:117–120.
3. Centers for Disease Control and Prevention. Primary and secondary syphilis–United States, 1999. MMWR Morb Mortal Wkly Rep 2001; 50:113–117.
4. Centers for Disease Control and Prevention. Primary and secondary syphilis among men who have sex with men – New York City, 2001. MMWR Morb Mortal Wkly Rep 2002; 51:853–856.
5. Centers for Disease Control and Prevention. Primary and secondary syphilis – United States, 2000–2001. MMWR Morb Mortal Wkly Rep 2002; 51:971–973.
6. Peterman TA, Heffelfinger JD, Swint EB, et al. The changing epidemiology of syphilis Sex Transm Dis 2005; 32(10 suppl):S4–S10.
7. Hammett TM, Harmon P, Rhodes W. The burden of infectious disease among inmates and releasees from correctional facilities. In: Health Care for Soon-to-be-Released Inmates: A Report to Congress. Vol 2. Chicago: National Commission on Correctional Health Care; 2002.
8. Glaser JB, Greifinger RB. Correctional health care: a public health opportunity. Ann Intern Med 1993; 118:139–145.
9. Shuter J. Communicable diseases in inmates: Public health opportunities. In: Health Care for Soon-to-be-Released Inmates: A Report to Congress. Vol 2. Chicago: National Commission on Correctional Health Care; 2002.
10. Kraut JR, Haddix AC, Carande-Kulis V, et al. Cost-effectiveness of routine screening for sexually transmitted diseases among inmates in United States prisons and jails. In: Health Care for Soon-to-be-Released Inmates: A Report to Congress. Vol 2. Chicago: National Commission on Correctional Health Care; 2002.
11. Varghese B, Peterman TA. Cost-effectiveness of HIV counseling and testing in US prisons. J Urban Health 2001; 78:304–312.
12. Chen JL, Callahan DB, Kerndt PR. Syphilis control among incarcerated men who have sex with men: public health response to an outbreak. Am J Public Health 2002; 92:1473–1474.
13. Chen JL, Kodagoda D, Lawrence AM, et al. Rapid public health interventions in response to an outbreak of syphilis in Los Angeles. Sex Transm Dis 2002; 29:277–284.
14. Chen JL, Bovee MC, Kerndt PR. Sexually transmitted diseases surveillance among incarcerated men who have sex with men – an opportunity for HIV prevention. AIDS Educ Prev 2003; 15(1 suppl A):117–126.
15. Holmberg SD. The estimated prevalence and incidence of HIV in 96 large US metropolitan areas. Am J Public Health 1996; 86:642–54.
16. Hethcote HW, Yorke JA. Gonorrhea Transmission Dynamics and Control. Lecture Notes in Biomathematics, 56. Berlin: Springer-Verlag; 1984.
17. Over M, Piot P. HIV infection and sexually transmitted diseases. In: Jamison DT, Mosley WH, Measham AR, Bobadilla JL, eds. Disease Control Priorities in Developing Countries. New York: Oxford University Press; 1993.
18. Brunham RC, Plummer FA. A general model of sexually transmitted disease epidemiology and its implications for control. Med Clin North Am 1990; 74:1339–52.
19. Holtgrave DR, Pinkerton SD. Updates of cost of illness and quality of life estimates for use in economic evaluations of HIV prevention programs. J Acquir Immun Defic Syndr 1997; 16:54–62.
20. Centers for Disease Control and Prevention. Screening tests to detect Chlamydia trachomatis and Neisseria gonorrhoeae infections–2002. MMWR Morb Mortal Wkly Rep 2002; 51(RR-15):1–38.
21. Atkins D. Screening for human immunodeficiency virus infection. In: Guide to Clinical Preventive Services: Report of the US Preventive Services Task Force. Collingdale: Diane Publishing Company; 1996.
22. Kahn JG, Washington AE. Screening for syphilis. In: Guide to Clinical Preventive Services: Report of the US Preventive Services Task Force. Collingdale: Diane Publishing Company; 1996.
23. Begley CE, McGill L, Smith PB. The incremental cost of screening, diagnosis, and treatment of gonorrhea and chlamydia in a family planning clinic. Sex Transm Dis 1989; 16:63–67.
24. Magid D, Douglas JM Jr, Schwartz JS. Doxycycline compared with azithromycin for treating women with genital Chlamydia trachomatis
infections: an incremental cost-effectiveness analysis. Ann Intern Med 1996; 124:389–399.
25. Centers for Disease Control and Prevention. Sexually transmitted diseases treatment guidelines 2002. MMWR Morb Mortal Wkly Rep 2002; 51(RR-6):1–80.
26. Haddad LM. Clinical Management of Poisoning and Drug Overdose. Philadelphia: W.B. Saunders Company; 1990.
27. McEvoy GK. American Hospital Formulary Service – Drug Information 95. Bethesda, MD: American Society of Hospital Pharmacists, Inc.; 1995.
28. Medical Economics Company. Physicians Desk Reference 2003. Montvale, NJ; 2002.
29. Rowley J, Berkley S. Sexually transmitted diseases. In: Murray CJL, Lopez AD, eds. Health Dimensions of Sex and Reproduction: The Global Burden of Sexually Transmitted Disease, HIV, Maternal Conditions, Perinatal Disorders, and Congenital Anomalies. Cambridge: Harvard University Press; 1998.
30. Washington AE, Johnson RE, Sanders LL Jr. Chlamydia trachomatis
infections in the United States. What are they costing us? JAMA 1987; 257:2070–2072.
31. Chesson HW, Rein D, Kassler WJ, et al. Direct medical costs of syphilis in the United States: The potential for a cost-saving national elimination program. Paper presented at: 1998 National STD Prevention Conference, Dallas, TX, 1998.
32. Pinkerton SD, Holtgrave DR, DiFranceisco W, et al. Cost-threshold analyses of the National AIDS Demonstration Research HIV prevention interventions. AIDS 2000; 14:1257–1268.
33. Pinkerton SD, Abramson PR. Effectiveness of condoms in preventing HIV transmission. Soc Sci Med 1997; 44:1303–1312.
34. Weller SC. A meta-analysis of condom effectiveness in reducing sexually transmitted HIV. Soc Sci Med 1993; 36:1635–1644.
35. Gold MR, Siegel JE, Russell LB, Weinstein MC, eds. Cost-effectiveness in Health and Medicine. New York: Oxford University Press; 1996.
37. Centers for Disease Control and Prevention. Increases in fluoroquinolone-resistant Neisseria gonorrhoeae
–Hawaii and California. MMWR Morb Mortal Wkly Rep 2002; 51:1041–1044.
38. Centers for Disease Control and Prevention. Notice to readers: Discontinuation of cefixime tablets – United States. MMWR Morb Mortal Wkly Rep 2002; 51:1052.
39. 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 Immun Defic Syndr 2000; 24:48–56.
40. Chesson HW, Pinkerton SD, Irwin KL, et al. New HIV cases attributable to syphilis in the USA: estimates from a simplified transmission model. AIDS 1999; 13:1387–1396.