Shrestha, Ram K. PHD*; Begley, Elin B. MPH*; Hutchinson, Angela B. MPH, PHD*; Sansom, Stephanie L. MPP, MPH, PHD*; Song, Binwei MS*; Voorhees, Kelly MSPH†; Busby, Amy BA‡; Carrel, Jack MPH‡; Burgess, Samuel MA, MSHCM‡
Partner Counseling and Referral Services (PCRS) have long been used by state and local health departments and public health clinics to notify partners of possible exposure to HIV or other sexually transmitted diseases.1 PCRS include notification of partners of HIV-infected persons (referred to as index patients), HIV counseling and testing of partners, and referral to care and treatment services.2,3 Particularly when the partners are notified by health care providers or health department staff (provider referral), rather than by index patients, PCRS has increased the number of HIV-exposed partners who have been tested and has potentially decreased HIV transmission in the community.1,2 In recent years, point-of-care rapid HIV testing has enabled providers to offer HIV counseling and testing in places that are convenient for the exposed partners.2,4
An estimated 20 million HIV tests are performed each year in the United States. However, 21% of the estimated 1.1 million HIV-infected individuals are still unaware of their infection.5 Individuals unaware of their HIV infection are at higher risk of engaging in unsafe sexual or needle-sharing activities and unknowingly transmitting HIV to their partners.6 PCRS aims to identify at-risk partners and provide them with HIV prevention and treatment services. However, more than a third of the partners who were notified of their potential exposure to HIV do not receive the test.1,7,8 PCRS programs have often used conventional HIV testing (enzyme immunoassay), which requires trained staff to collect specimens by using phlebotomy or oral swab and then to send those specimens to a certified laboratory for processing. This process takes from 1 day to 2 weeks, and partners are required to return for results. Rapid HIV tests, in contrast, can be administered relatively easily in the field or at places convenient for partners, and the results can be disclosed in as little as 20 minutes.9–11
Under the Advancing HIV Prevention (AHP) initiative, the Centers for Disease Control and Prevention (CDC) launched demonstration projects offering PCRS with rapid HIV testing in various settings.12 The AHP initiative provided 6 health departments with the opportunity to use rapid HIV tests.13 Of the 6 health departments, 2 participated in our cost analysis. We assessed the costs and effectiveness of PCRS with rapid HIV testing as implemented by the state health departments of Colorado and Louisiana.
Partner Notification and HIV Testing
Under the AHP demonstration project, the Colorado Department of Public Health and Environment and the Louisiana Office of Public Health initiated rapid HIV testing within existing PCRS programs. Health department staff provided HIV counseling and testing to the partners of HIV-infected individuals in their homes or workplaces, or in health department facilities or vehicles. The Colorado health department provided PCRS to the index patients and partners statewide; the Louisiana health department provided the services to individuals in 2 public health regions—Baton Rouge and New Orleans.13
PCRS with rapid HIV testing took place from April 2004 through August 2005 in Colorado and from April 2004 through January 2006 in Louisiana. The staff of the department first identified index patients using confidential health department records, then located and interviewed them to obtain information about their sex and needle-sharing partners. The staff then searched public health records to identify the partners who had been reported as HIV infected, and the staff contacted the remaining partners to inform them that they might have been exposed to HIV (provider referral), or the index patients referred partners to the health department for testing (self-referral). Partners who had already been reported as HIV-positive were contacted by the health department to initiate PCRS, but were excluded from the rapid testing demonstration project. The index patients were asked to identify all partners starting 12 months before their HIV diagnosis. All notified partners were offered rapid HIV testing (at a testing site or in the field), counseling, and referral to care. Rapid HIV testing was performed using OraQuick HIV-1 or OraQuick Advance HIV-1 or 2 antibody test (OraSure Technologies, Bethlehem, PA) on either whole blood or oral fluid specimens. The health department staff provided pretest counseling and posttest risk-reduction counseling, regardless of test results. Partners younger than 13 years or known to have a diagnosis of HIV infection were not eligible for testing through this project. Participation in the PCRS program was voluntary. Program details, including demographic characteristics of index patients and partners, have been published.13
Program Costs and Effectiveness
We analyzed the costs and the effectiveness of the PCRS programs in Colorado and Louisiana separately. We obtained annual total program costs of each intervention from a provider’s perspective (i.e., we included the costs incurred by the health departments but did not measure participants’ costs). Costs are expressed in 2007 US dollars. The key outcome measure was the average cost per partner notified of a new HIV diagnosis after a rapid test,14 obtained by dividing annual total program cost by the number of partners notified of a new HIV diagnosis. We also estimated the average costs per index patient identified and per partner tested.
Using microcosting methods, we identified the cost of each program element, such as personnel, facilities, equipment, supplies, and other program inputs, to estimate the total program cost.15–19 Fixed costs, those that remained constant within a relevant time frame regardless of the number of clients served, included costs attributable to program management (planning, administration, and supervision), training, travel between sites, and durable goods and equipment. Variable costs, those that vary with the number of persons served, included those attributable to the identification and interview of index patients, partner notification, counseling and testing, and nondurable goods and supplies, such as rapid HIV test kits and confirmatory tests.
In general, PCRS programs in both state health departments provided staff time and cost data retrospectively. However, in Colorado, we used time-diaries of 2 disease intervention specialists (professionals trained to intervene in the spread of communicable and chronic diseases of public health importance), who recorded the time required to identify and notify a convenience sample of 10 index patients and 10 partners, to estimate partner notification costs. We were unable to use time-diaries in Louisiana because cost collection took place at the end of the project, and project staff had already left or relocated in the aftermath of Hurricane Katrina. We calculated personnel costs based on the amount of time the program staff spent in program-related activities. We multiplied the staff time associated with each activity by the compensation (wage plus benefits) received by the staff person who performed the activity.
The cost of the OraQuick rapid HIV test kit was based on the bulk purchase price ($8.69/kit) available to CDC for the AHP demonstration projects.20 The cost of confirmatory Western blot testing ($41.17/test) was based on data from a national commercial reference testing laboratory.21
To estimate the overhead costs (i.e., utilities and facility space) attributable to PCRS, we multiplied the total amount that health departments spent on overhead by the fraction of the department staff time allocated to PCRS. Travel costs included round-trip travel time and vehicle mileage between sites. We used a 3% discount rate to amortize the costs of purchased equipment over the useful life of the equipment.15 We included the rental cost of facility space but excluded costs related to program evaluation. CDC determined that this project was a public health program activity, not research, so review by a CDC institutional review board was not required.
The PCRS programs in Colorado and Louisiana (Baton Rouge and New Orleans) identified, on average, 328 and 81 index patients per year, respectively (Table 1), and 320 and 81 of them were eligible (within jurisdiction and age ≥13 years). Of those eligible, 79% in Colorado and 94% in Louisiana were interviewed, and 77% and 93% of those interviewed provided partner information. The programs identified, on average, 253 and 138 partners per year in Colorado and Louisiana, respectively. Of those, 75% in Colorado and nearly 100% in Louisiana were assigned to health department staff for partner notification. Because some partners died, moved out of the department’s jurisdiction, or could not be located, only 62% of the assigned partners in Colorado and 86% of those in Louisiana were notified of a potential HIV exposure. Of those notified and not known to be HIV infected, 48% in Colorado and 84% in Louisiana received a rapid HIV test. Among the partners in Colorado who did not receive the test, the primary reasons were that they had already been identified as HIV-positive (49.7%) or had recently been tested and found to be HIV-negative (26.7%). Of the partners tested, 3 (seropositivity: 6.6% of partners tested) in Colorado and 8 (seropositivity: 9.9% of partners tested) in Louisiana, were diagnosed with new HIV infection. All partners with new HIV diagnosis after rapid testing received their test results.
The total annual program cost of offering PCRS with rapid testing was $62,802 in Colorado and $59,161 in Louisiana (Table 1). We estimated the average cost per partner tested at $1459 in Colorado and $714 in Louisiana. Variation in these cost ratios was mainly due to the differences in the number of partners tested in each program. The cost per partner notified of a new HIV diagnosis was $22,243 in Colorado and $7231 in Louisiana. The variability in these costs reflects the differences in previously undiagnosed HIV infection among partners tested in the 2 programs.
Fixed costs comprised 48% of the total program costs in Colorado and 65% of those in Louisiana (Table 2). The key components of the fixed costs were program management, facility space, utilities, and staff training. The program management cost (program planning, administration, and supervision) was the highest among fixed costs in both sites, $383.23/partner tested in Colorado and $326.23/partner tested in Louisiana.
Important variable costs in the analysis included those associated with index patient identification and partner notification. The average cost attributable to staff time spent on identifying an index patient, including travel, was $76.76 ($43.07, including $33.69 in travel) in Colorado and $67.86 ($37.58, including $30.28 in travel) in Louisiana, and the average cost of notifying a partner was $35.41 ($14.55, including $20.86 in travel) in Colorado and $75.77 ($49.29, including $26.48 in travel) in Louisiana. The higher cost of partner identification and notification in Louisiana was due to greater staff time devoted to these activities (Table 3). The costs due to staff time spent on counseling and testing partners whose test results were negative were $25.50 in Colorado and $14.32 in Louisiana; the costs for partners whose test results were preliminarily positive were $33.51 in Colorado and $21.48 in Louisiana. The higher cost of counseling and testing in Colorado was partly due to higher staff wages (the hourly wage rate for counselors was approximately 22% higher in Colorado than in Louisiana).
In this study, we evaluated the costs and the effectiveness of PCRS with rapid HIV testing in 2 states. The average cost of notifying a partner with previously undiagnosed HIV infection after a rapid test was $22,243 (seropositivity, 6.6%) in Colorado and $7231 (seropositivity, 9.9%) in Louisiana. Differences in results were primarily because of the variation in the number of HIV-exposed partners notified and tested per index patient, prevalence of undiagnosed HIV infection among partners tested, and the variation in facility and other-fixed costs.
The Louisiana PCRS staff spent more time, on average, identifying, locating, and interviewing index patients, and considerably more time locating and notifying partners. However, time spent on these activities in Louisiana, compared with Colorado, may have contributed to a higher rate of partners notified of a potential HIV exposure per index patient (1.46 vs. 0.36) and tested (1.02 vs. 0.13). For every 100 index patients, 10 new HIV cases were diagnosed in Louisiana compared with 1 in Colorado. Compared with Colorado, the program in Louisiana appeared to be much more successful in notifying and testing potentially HIV-exposed partners and finding new HIV-positive cases, thus substantially reducing the cost per partner notified of new HIV diagnosis.
The cost and cost-effectiveness of PCRS can vary by the types of costs reported and the outcomes considered.22–25 Studies using decision analytic models showed that PCRS can be cost-effective, and in some instances, cost-saving.22,26 In many studies, fixed costs were excluded, and the main outcome reported was the cost per new HIV diagnosis (not the cost per person notified of new HIV diagnosis). These 2 outcome measures are likely to be similar if rapid tests are used because test results can be disclosed in as little as 20 minutes. However, when conventional HIV testing (enzyme immunoassay) is used, it may take up to 2 weeks to obtain and disclose the results, and some persons may not return to receive them. When we excluded fixed costs in our analysis, we estimated the cost per person notified of new HIV diagnosis at $11,626 in Colorado and $2545 in Louisiana.
In a recent study of the costs and the effectiveness of PCRS at a sexually transmitted disease clinic in San Francisco, the authors estimated the variable cost per partner identified with a new HIV diagnosis (using rapid HIV antibody testing) at $7145 (costs adjusted to US $2007). The HIV seropositivity among partners was 13.0%.27 The authors of cost-effectiveness studies of PCRS using conventional HIV testing in Colorado, FL, NJ, and Utah (published during 1993–1998) reported that variable and some fixed costs per partner identified with previously undiagnosed HIV infection ranged from $3835 to $6911. In these studies, the HIV seropositivity among partners tested ranged from 14.0% to 21.8%.8,23,25 In other rapid HIV testing interventions in jails and community-based organizations, in which PCRS was not a component but in which the methods used to measure costs and outcomes were similar to those in our study, the total cost (viable and fixed) per person notified of new HIV diagnosis ranged from $2600 to $21,900.28,29
Our analysis has some limitations. Most costs were collected retrospectively, thus may be subject to recall bias. In particular, we used time-diaries to prospectively collect staff time required to identify and interview a convenience sample of 10 index patients and 10 partners in Colorado, although we collected the time data retrospectively using standardized forms in Louisiana. The differences in methods may have contributed to some of the differences in costs.
The PCRS programs involved a program manager, 2 disease intervention specialists, administrative staff, and other material resources to implement the program. However, the program staff time and resources were often distributed across different projects and program activities. Using microcosting methods, we collected costs attributable to staff time spent on each PCRS related program activity and the amount of material resources used in the intervention, and estimated the total program cost. This implies that the staff time or resources not used in the PCRS program would have been used in other projects. To the extent the staff time was not spent on other projects, but represented in idle time during the PCRS program, our costs underestimate true program costs.
It is important to note that in some cases, the partners might have been notified regardless of the PCRS intervention, leading to a potential overestimation of the program effectiveness, and in other cases the partners may have been notified outside PCRS jurisdiction, even if the notification was due to the information or services received from the program, resulting in a potential underestimation of the effectiveness. In addition, in this analysis we did not model the number of secondary HIV infections prevented through early diagnosis of HIV in partners and due to the partners’ subsequent reduction of risky behaviors.
Our rapid test kit cost was based on the bulk price available to CDC at the time, and the cost of confirmatory testing was based on data from a national commercial reference testing laboratory. These costs may vary by type of rapid test, over time, and across jurisdictions. Because our analysis was based on the costs and effectiveness data from PCRS programs in only 2 states, the results may not be generalizable to some other communities. We reported the cost of an intermediate health outcome, the partner notified of a new HIV diagnosis, and that can not be compared with cost-effectiveness studies that estimate the cost per life year or quality-adjusted life year saved.
In summary, PCRS programs in conjunction with rapid HIV testing successfully notified partners who were unaware of their possible exposure to HIV and provided them with rapid testing and counseling services. PCRS has the potential to decrease HIV transmission through the practice of safer behavior by partners who learn that they have been infected. However, PCRS has been underused, in part because of concerns about costs.3 Our analysis shows that the cost per new case of HIV infection detected through partner notification can vary substantially across sites and by undiagnosed HIV prevalence among partners. Our analysis helps program managers and health care providers in different settings to better understand the full costs of implementing PCRS, and to determine if PCRS is an efficient use of HIV prevention dollars in their community.
1. Hogben M, McNally T, McPheeters M, et al. The effectiveness of HIV partner counseling and referral services in increasing identification of HIV-positive individuals a systematic review. Am J Prev Med 2007; 33(suppl 1):S89–S100.
2. Centers for Disease Control and Prevention. Recommendations for partner services programs for HIV infection, syphilis, gonorrhea, and chlamydial infection. MMWR Recomm Rep 2008; 57(RR-9):1–83.
3. Dooley SW Jr, Douglas JM Jr, Janssen RS. Partner counseling and referral services for HIV infection: New endorsement of an old approach. Am J Prev Med 2007; 33(suppl 1):S81–S83.
4. Centers for Disease Control and Prevention. Advancing HIV prevention: New strategies for a changing epidemic—United States, 2003. Morb Mortal Wkly Rep 2003; 52:329–332.
5. Centers for Disease Control and Prevention (CDC). HIV prevalence estimates—United States, 2006. Morb Mortal Wkly Rep 2008; 57:1073–1076.
6. Marks G, Crepaz N, Janssen RS. Estimating sexual transmission of HIV from persons aware and unaware that they are infected with the virus in the USA. AIDS 2006; 20:1447–1450.
7. Landis SE, Schoenbach VJ, Weber DJ, et al. Results of a randomized trial of partner notification in cases of HIV infection in North Carolina. N Engl J Med 1992; 326:101–106.
8. Spencer NE, Hoffman RE, Raevsky CA, et al. Partner notification for human immunodeficiency virus infection in Colorado: Results across index case groups and costs. Int J STD AIDS 1993; 4:26–32.
9. Greenwald JL, Burstein GR, Pincus J, et al. A rapid review of rapid HIV antibody tests. Curr Infect Dis Rep 2006; 8:125–131.
10. Spielberg F, Branson BM, Goldbaum GM, et al. Choosing HIV counseling and testing strategies for outreach settings: A randomized trial. J Acquir Immun Defic Syndr 2005; 38:348–355.
11. Hutchinson AB, Branson BM, Kim A, et al. A meta-analysis of the effectiveness of alternative HIV counseling and testing methods to increase knowledge of HIV status. AIDS 2006; 20:1597–1604.
12. Centers for Disease Control and Prevention. Advancing HIV prevention: New strategies for a changing epidemic. Centers for Disease Control and Prevention. Available at: http://www.cdc.gov/hiv/topics/prev_prog/AHP
. Accessed September 15, 2008.
13. Begley EB, Oster AM, Song B, et al. Incorporating rapid HIV testing into partner counseling and referral services. Public Health Rep 2008; 123(suppl 3):126–135.
14. Farnham PG, Gorsky RD, Holtgrave DR, et al. Counseling and testing for HIV prevention: Costs, effects, and cost-effectiveness of more rapid screening tests. Public Health Rep 1996; 111:44–53.
15. Haddix AC, Teutsch SM, Corso PS. Prevention Effectiveness: A Guide to Decision Analysis and Economic Evaluation. New York, NY: Oxford University Press, 2003.
16. Gorsky RD. A method to measure the costs of counseling for HIV prevention. Public Health Rep 1996; 111(suppl 1):115–122.
17. Drummond MF, O’Brien B, Stoddart GL, et al. Methods for the Economic Evaluation of Health Care Programmes. Oxford, UK: Oxford University Press, 1996.
18. Levin HM, McEwan PJ. Cost-Effectiveness Analysis: Methods and Applications. New Delhi, India: SAGE Publications, 2001.
19. Gold MR, Siegel JE, Russell LB, Weinstein MC, eds. Cost-effectiveness in Health and Medicine. New York, NY: Oxford University Press, 1996.
20. Centers for Disease Control and Prevention. Rapid HIV test distribution—United States, 2003–2005. Morb Mortal Wkly Rep 2006; 55:673–676.
21. Farnham PG, Hutchinson AB, Sansom SL, et al. Comparing the costs of HIV screening strategies and technologies in health-care settings. Public Health Rep 2008; 123(suppl 3):51–62.
22. Varghese B, Peterman TA, Holtgrave DR. Cost-effectiveness of counseling and testing and partner notification: A decision analysis. AIDS 1999; 13:1745–1751.
23. Pavia AT, Benyo M, Niler L, et al. Partner notification for control of HIV: Results after 2 years of a statewide program in Utah. Am J Public Health 1993; 83:1418–1424.
24. Golden MR, Gift TL, Brewer DD, et al. Peer referral for HIV case-finding among men who have sex with men. AIDS 2006; 20:1961–1968.
25. Toomey KE, Peterman TA, Dicker LW, et al. Human immunodeficiency virus partner notification: Cost and effectiveness data from an attempted randomized controlled trial. Sex Transm Dis 1998; 25:310–316.
26. Ekwueme DU, Hutchinson AB, Dean H, et al. Estimating the cost and effectiveness of three referral strategies for HIV partner counseling and referral services. Poster presented at: 26th
Annual Meeting of the Society for Medical Decision Making; October 17–20, 2004; Atlanta, GA. Available at: http://smdm.confex.com/smdm/2004ga/techprogram/P1447.HTM
. Accessed May 26, 2009.
27. Ahrens K, Kent CK, Kohn RP, et al. HIV partner notification outcomes for HIV-infected patients by duration of infection, San Francisco, 2004 to 2006. J Acquir Immun Defic Syndr 2007; 46:479–484.
28. Shrestha RK, Clark HA, Sansom SL, et al. Cost of providing new HIV diagnoses using rapid HIV testing in community-based organizations. Public Health Rep 2008; 123(suppl 3):94–100.
29. Shrestha RK, Sansom SL, Richardson-Moore A, et al. Costs of voluntary rapid HIV testing and counseling in jails in 4 states—Advancing HIV Prevention demonstration project, 2003–2006. Sex Transm Dis 2009; 36(suppl 2):5–8.