The Centers for Disease Control and Prevention recommends that health departments provide partner services (PS) to persons diagnosed with early syphilis, and some provide PS to selected individuals diagnosed with gonorrhea and chlamydial infection.1 Partner service generally involves health department staff interviewing infected persons (original patients [OPs]) to obtain information about their exposed sex partners to ensure that those partners are tested and treated appropriately.1 Beyond that, PS resources, practices, and objectives vary substantially.2–5 Some programs use PS as an human immunodeficiency virus (HIV) prevention intervention and have defined HIV testing, linkage and relinkage of HIV-positive persons to treatment, and referral of HIV-uninfected individuals—particularly men who have sex with men (MSM)—to HIV preexposure prophylaxis (PrEP) as explicit PS outcomes.5–7
Despite rising rates of bacterial sexually transmitted diseases (STDs) in the United States,8 Centers for Disease Control and Prevention's annual STD budget declined 40% between 2003 and 2017, adjusted for inflation.9 Understanding the costs of STD PS programs could help ensure that limited resources are used efficiently.10 Prior cost studies focus on either syphilis or gonorrhea, chlamydia, and nongonococcal urethritis, or fail to capture heterogeneity between local jurisdictions or include HIV prevention objectives.11–14
Along with the United States, as a whole, bacterial STD incidences are increasing throughout Washington State.15,16 We described the epidemic characteristics and estimated the share of personnel time and financial costs associated with STD PS programs and activities at three high-burden local health jurisdictions that represent the geographic and resource diversity of Washington State. Results can inform both local providers and programs around the country.
Study Setting and Summary
In Washington State, STD PS interviews are primarily conducted over the telephone, integrate expedited partner therapy (EPT) for heterosexuals with gonorrhea or chlamydia,17 and promote HIV-specific interventions for MSM and transgender persons with an STD diagnosis or exposure. The HIV interventions include HIV testing,7,18 linkage to treatment,19 and linkage to PrEP for high-risk individuals.20 Partner services are provided to all cases of early syphilis and select cases of gonorrhea/chlamydia, including all pregnant women and persons with untreated infections, and in some cases, MSM, transgender persons, and adolescents depending on case risk factors and resource availability. Field visits were conducted on priority cases, only when efforts to reach the individual via several telephone and text attempts, emails, and letters had failed. Spokane also used Facebook messenger to make contact and conducted field visits when individuals with syphilis were incarcerated at the nearby local jail. King only provided field visits to untreated cases.
This study estimates the costs for STD PS delivery from the perspective of the health department at three local health jurisdictions: Public Health – Seattle & King County (King), Tacoma-Pierce County Health Department (Pierce), Spokane Regional Health District (Spokane). These high-burden locations were selected to represent the diversity within the State. King serves the dense urban city of Seattle and the surrounding area, Pierce serves the city of Tacoma and the more rural surrounding area in the county directly south of King, and Spokane serves the less dense city of Spokane and the rural surrounding area in the eastern most part of the state. For each location, we present the population size reported by the US census bureau and the number of reported STD cases by risk characteristic (MSM, men who have sex with women [MSW], men with unknown risk, and women) using county-level disease surveillance data from 2016. We conducted a microcosting analysis in 2016 to 2017 to estimate staff time allocation for specific components of PS delivery and estimated the total time personnel allocated to STD PS. Analyses included a time study and financial expenditure assessment. We used surveillance and programmatic data to estimate health service delivery indicators, outcomes, and associated costs. Results are stratified by jurisdiction to investigate heterogeneity in disease burden, population characteristics, costs, and program models.
These activities were conducted as part of public health program evaluation and therefore not considered human subjects research. Analyses were conducted using Microsoft Excel 2017 and Stata/SE 14.2.
Health Service Delivery Indicators
We used county-level surveillance and PS data from 2016 to determine all health services indicators unless otherwise stated, including the total numbers of STD cases diagnosed, assigned for PS, and interviewed by DIS, and the numbers of partners named by OPs, notified by DIS or OP, tested, diagnosed, and treated after the OP's PS interview. We also determined the number of OPs provided with EPT, and HIV-negative clients (OPs and partners) tested for HIV and referred to PrEP through PS, referred to as enhanced PS outcomes. Information for PrEP referrals was collected empirically for OPs in King County and via DIS self-report for STD PS clients combined (OPs and partners) in Pierce and Spokane, so are not directly comparable. The number of HIV-infected OPs identified as inadequately engaged in HIV care (defined as not being on antiretroviral therapy or, if treatment status is unknown, reporting having no provider or visit in the prior 6 months/scheduled future visit at the time of interview) and the number linked to care following PS (virally suppressed 6 months after interview) were estimated when available. Partner outcome indices were calculated by dividing the total number of partners with each outcome (number notified, tested, treated, etc.) by the number of OPs interviewed.21 Cases with multiple STDs (coinfections) were included as the higher priority STD in the following order: syphilis, gonorrhea, chlamydia.
We conducted a time study of all DIS and other PS staff. First, an independent observer assessed how much time each staff member spent on each activity over several hours during a typical workday. To complement this information, each staff member was interviewed before the observations (and after if observations revealed omissions) to gain a more complete understanding of workflow, time allocation, and if they had unrelated job responsibilities; these data were used to estimate the proportion of administrative staff and DIS labor allocated to STD PS based on their self-report and confirmed by the supervisor. Second, PS staff conducted individual case tracking and reported specific activities from case report, assignment, through to case closure. Tracked cases were purposively sampled to include cases representing a variety of STDs and risk characteristics.
We estimated total personnel hours spent per week on key activities by STD. We also summarized tracking form data to present the means, medians, and variability in time spent per case by STD and population. The total time allocated per case was estimated by dividing the average weekly number of hours allocated for work related to STD PS by the average number of cases assigned to STD PS in a week (calculated by dividing the total number of cases worked in 2016 by 44 work weeks, the average number of weeks worked by staff in a year).
Financial Expenditures Assessment
The 2016 financial records for personnel salaries and benefits and overhead were adjusted based on the percentage of time staff allocated to STD PS to estimate the total STD PS program costs by STD type at each jurisdiction. Benefit costs were provided by the PS supervisor for each staff member and overhead costs were calculated as a percent of salaries, defined as the amount charged against labor in each county (33% in King and 30% in Pierce and Spokane). Activities and time unrelated to STD PS were excluded from analyses. The total costs per STD PS outcomes (eg, OP interview, partner notified, tested, and treated after DIS interview, etc.) were calculated by dividing the estimated total program cost by the number of individuals known to have received that service or have that outcome in 2016. Outcomes achieved per US $100,000 invested in STD PS were calculated by dividing the number of individuals known to have received that service by total program cost and multiplying by US $100,000. All costs were adjusted for inflation to 2016 US dollars.
At all three jurisdictions, administrative/data entry staff process laboratory and case reports for all gonorrhea/chlamydia cases. Once the initial case report is entered into the statewide web-based STD surveillance/PS database, cases are assigned to DIS for PS depending on STD, risk criteria, and available resources (Fig. 1). For new syphilis cases, the State Department of Health (DOH) staff assign possible cases to DIS for review and PS.
Observations were conducted on select days in 2016 to 2017 in each health department. Preassignment administrative and data entry work for gonorrhea/chlamydia was conducted by three staff members in King and one each in Pierce and Spokane. There were 7, 5, and 4 DIS working on STD PS in King, Pierce, and Spokane, respectively (see supplement for additional details on staff time, https://links.lww.com/OLQ/A378).
In King, Pierce, and Spokane, respectively, DIS combined spent an average total of 73, 17, and 73 hours per week on syphilis PS overall; worked an average of 11, 3, and 3 cases per week; and allocated an average of 6.5, 6.4, and 28.8 hours per case (Fig. 2). Disease intervention specialist combined spent an average total of 142, 88, and 27 hours per week on gonorrhea/chlamydia PS work overall; worked an average of 96, 55, and 9 cases per week; and allocated an average of 1.5, 1.6, and 2.9 hours per case, respectively.
Estimated aggregate time in hours and percent of time spent per week on PS activities varied across STD and jurisdiction, but some trends were observed (Supplemental Figure 1, https://links.lww.com/OLQ/A379 and Supplemental Table 1, https://links.lww.com/OLQ/A380). Administrative staff's work was similar for all reported gonorrhea/chlamydia cases regardless of whether the case met criteria for PS follow-up and DIS investigation, spending the majority of their time on data entry at all jurisdictions (>53%). King administrative staff and DIS spent the largest percentage of time (17% and 9%, respectively) contacting health care providers to obtain and complete case reports versus Pierce and Spokane (<3%). For syphilis STD PS, DIS spent a similar share of time contacting health care providers in King (22%) and Pierce (19%) versus less time in Spokane (3%). Spokane staff directly accessed electronic medical records (EMR) and could look up case information directly.
For syphilis, DIS spent 6%, 31%, and 11% of time conducting case interviews and partner notification versus 21%, 39%, and 16% of gonorrhea/chlamydia work time in King, Pierce, and Spokane, respectively. A DIS in Spokane spent a greater proportion of time conducting field work for gonorrhea/chlamydia (18% vs. <6% in King and Pierce) and reported that interviews were often conducted at the nearby county detention center. A DIS working on gonorrhea/chlamydia in Spokane and Pierce spent a greater percent of time using electronic resources (eg, EMRs and Facebook) to locate and contact OPs and partners than in King. Facebook was generally used after other methods (eg, phone calls, texting) were unsuccessful. Only a DIS in Spokane frequently used local police blotters. In all jurisdictions, not all staff had access to commonly used electronic data sources (eg, Accurint or the state syphilis serology database), resulting in interruptions to staff with access to obtain information.
Time spent directly on casework captured by individual case tracking varied substantially across cases (Supplemental Table 2, https://links.lww.com/OLQ/A381). More DIS time was spent on syphilis than gonorrhea/chlamydia casework across jurisdictions. Among cases with substance use information, casework for methamphetamine users took longer than nonusers for Spokane syphilis cases (limited data regarding methamphetamine use was available from tracked cases in King and Pierce). Although complicated high-priority cases like neurological and neonatal syphilis are rare, DIS spent a much larger amount of time conducting related casework, resulting in extreme outliers for time spent per case.
STD PS objectives such as EPT, HIV testing, linkage and relinkage to HIV care, and PrEP referrals took minimal time relative to overall work (30 seconds to 2 min per topic per interview), except for PrEP referrals in Pierce. Clients often already had a documented HIV test, were not HIV-positive, or were already linked to care and on antiretroviral treatment if they were HIV-positive, minimizing additional DIS interventions. Some PrEP eligible clients had already discussed PrEP with their health care provider. King and Spokane had public health PrEP coordinators so DIS typically took only a few minutes to assess client eligibility and interest in PrEP and then referred the client to the coordinator. However, DIS in Pierce coordinated PrEP referrals and initiations themselves, including counseling and collecting blood samples for PrEP eligibility required for initiation, which DIS reported took approximately 69 to 155 minutes per referral for 3 to 6 clients per week.
Disease Burden, Characteristics, and PS Outcomes
King has the largest population and most reported STD diagnoses (Table 1). Pierce had approximately twice the number of reported gonorrhea/chlamydia cases than Spokane, but a similar number of syphilis cases. The majority (>80%) of syphilis cases in King and Pierce were identified as MSM versus 38% in Spokane. For gonorrhea, about half (52%) of cases in King were identified as MSM versus less than 20% in Pierce and Spokane.
Nearly all (>96%) of reported syphilis cases, 66% to 95% of reported gonorrhea cases, and a minority of chlamydia cases were assigned to PS across the jurisdictions (Table 2). Approximately 270, 268, and 61 syphilis and 1177, 1105, and 769 gonorrhea/chlamydia cases were assigned for PS per full-time DIS, in King, Pierce, and Spokane, respectively. Of reported syphilis cases, 70% to 91% were successfully interviewed by DIS, and 35% to 70% of reported gonorrhea cases were successfully interviewed (Table 3A). Partner outcome indices were lowest in King for syphilis cases and relatively similar across jurisdictions for gonorrhea/chlamydia. In 2016, median time from laboratory diagnosis to DIS interview in King, Pierce, and Spokane, respectively, for syphilis was 13 days (interquartile range [IQR], 6–32), 5 (IQR, 1–12), and 5 (IQR, 0–13), and for gonorrhea/chlamydia was 8 (IQR, 3–18), 4 (IQR, 0–10), and 8 (IQR, 4–20).
Financial Expenditures and Costs per Outcome
We estimated that the annual costs for STD PS were US $798,141, US $416,098, and US $400,759 in King, Pierce, and Spokane, respectively (Table 2). At the time of observations, syphilis-related work accounted for an estimated 28%, 15%, and 57% of total STD PS resources, respectively. Spokane was uniquely supported by an on-site DOH-funded syphilis consultant (costs were included in estimate as the consultant's contributions were necessary for service delivery to address the epidemic at this location). The incremental costs of PS activities can be derived by multiplying the program costs (Table 2) by the percent of personnel time spent on activities (Supplemental Table 1, https://links.lww.com/OLQ/A380).
The estimated total costs per OP interview were lowest in Pierce and highest in Spokane for all STDs, and higher for syphilis versus gonorrhea/chlamydia (Table 3A). The total cost per HIV test of all STD PS clients combined after the DIS interview ranged from US $988 to US $5467 (Table 3B), per client newly diagnosed with HIV was greater than US $130,000 for all jurisdictions, per OP identified as HIV-infected and inadequately engaged in care ranged from US $10,203 to US $111,837, and per HIV-uninfected client accepting a PrEP referral ranged from US $2551 to US $57,467. Outcomes achieved per US $100,000 invested in STD PS varied substantially by jurisdiction (Table 3C). For syphilis and gonorrhea, more outcomes were achieved in Pierce per US $100,000 invested than in King and Spokane.
Our study describes the nuances of STD PS work and the costs associated with service delivery. Our findings can be used to implement and improve similar programs to best address STD epidemics with limited resources. We found large variability in the time and cost required to provide PS to clients with different STDs across local health jurisdictions in Washington State, which was dependent on sexually transmitted infection epidemiologic characteristics and available support services. Program cost-per-interview ranged from US $527 to US $2210 for syphilis, US $219 to US $484 for gonorrhea, and US $164 to US $547 for chlamydia. Our study found similar trends in the relative costs between STDs compared with a similar, state-level, costing analysis of PS in New York in 201413,22 that estimated their cost per interview was US $1072, US $608, US $635, for syphilis, gonorrhea, and chlamydia, respectively.13 In our study, the additional resources required for syphilis PS were partially explained by the work exclusively being performed by highly trained DIS compared with administrative staff.
Syphilis PS consistently involved greater resources than gonorrhea/chlamydia, and heterosexual syphilis cases were consistently more resource-intensive than MSM. The cost per syphilis case assigned in Spokane was approximately four times that in King and Pierce. Most syphilis cases in Spokane were among heterosexuals, many of whom were suspected methamphetamine users, compared with King and Pierce where cases were primarily among MSM. Spokane DIS also spent a greater proportion of time in the field compared to King and Pierce. The higher costs per case, observed in Spokane, largely reflect epidemic characteristics and inherent difficulties engaging a highly socially marginalized and high-priority heterosexual population. Health departments often place higher priority on syphilis cases due to the greater morbidity associated with this infection. Syphilis is especially concerning among heterosexuals and pregnant women given the risk and severity of congenital syphilis, which requires immediate DIS response and a large allocation of staff resources and time. Syphilis rates among heterosexuals in the United States are now rising,8 which could increase the cost of PS.
Our findings highlight opportunities for improving the efficiency of PS. We observed that the majority of DIS time was spent on data entry and searching for clients before successfully interviewing cases and contacting partners at all locations. More efficient surveillance mechanisms could help providers and laboratories complete case reporting in a timely manner, resulting in faster PS responses, less time spent contacting providers, collecting data, and on data entry, and more time spent on casework. An electronic reporting system to submit and view case reports, and full access to EMR, would greatly reduce the burden on PS staff and health care providers. Additionally, increased access to restricted databases (eg, Accurint) and better data sharing between the state DOH and across neighboring jurisdictions could facilitate faster casework.
Adding HIV-related services was generally not time-consuming relative to overall work and resulted in substantial numbers of MSM accepting referrals for PrEP in King and Pierce. The PrEP referrals required substantially more DIS resources in Pierce, which lacked dedicated PrEP referral coordinators. The total STD PS program cost per HIV test after PS intervention found in our study (US $998–US $5467) was within the range of estimates of costs per partners tested in HIV PS programs in studies conducted in other locations across the United States.23–25 Seattle has achieved the Joint United Nations Programme on HIV/AIDS goal of 90-90-90,26 which likely contributed to the high cost per new HIV diagnosis after STD PS intervention (US $133,586–US $416,098) in our study compared with other programs in the United States.13,23,24,27 In contrast to the high total cost per HIV diagnosis, the total cost per client accepting a PrEP referral in our study ranged from US $2551 to US $57,467 for syphilis and US $3726 to US $6118 for gonorrhea/chlamydia cases.
The low incremental cost associated with adding HIV interventions to STD PS should prompt health departments to integrate them into existing programs. Although HIV case finding and the identification of out-of-care people with HIV were rare outcomes in our STD PS programs (and thus had a high cost per outcome), this was likely due to low frequencies of undiagnosed and untreated HIV cases within these jurisdictions,28 and other areas may have very different experiences. Human immunodeficiency virus prevention benefits can help justify expanding STD PS resources for the explicit purpose of HIV case finding. Whether program costs are a good value requires cost-effectiveness analyses incorporating the various outcomes measured.
This study has several strengths. We conducted direct observations and tracked individual cases across multiple health jurisdictions across Washington State with different epidemic characteristics, program models, and resources. Study limitations include that STD PS work vary over time and are not captured well by short observation periods. In addition, though efforts were made to minimize imposition of the time study itself, staff likely work differently while being observed.29 We attempted to address this by triangulating data from interviews, case-tracking forms, and observations. Additionally, the estimates of cases interviewed and partners notified derived from the time study were similar to the empirical case counts from surveillance that we used in our analysis. Not all partners treated after DIS interview may have been the result of DIS intervention, as some partners may have sought treatment regardless.30 These data may not be generalizable to programs in different settings. However, the detailed findings may be informative to any health department program.
Our results can inform the operational costs, areas to improve efficiency, and budget impact of STD PS in US health jurisdictions. Given that the lifetime cost of one HIV case is estimated to be US $400,000,31 efforts to integrate interventions to prevent HIV transmission within STD PS may be highly cost-effective. A cost-effectiveness analysis of STD PS that will balance incurred costs against savings from HIV and STD prevention is needed.
1. Centers for Disease Control and Prevention (CDC). Recommendations for partner services programs for HIV infection, syphilis, gonorrhea, and chlamydial infection. MMWR Recomm Rep 2008; 57:1–83; quiz CE1-4.
2. Desir FA, Ladd JH, Gaydos CA. Survey of partner notification practices for sexually transmissible infections in the United States. Sex Health 2016; 13:162–169.
3. Hogben M, Collins D, Hoots B, et al. Partner services in sexually transmitted disease prevention programs: A review. Sex Transm Dis 2016; 43(2 Suppl 1):S53–S62.
4. Golden MR, Hogben M, Handsfield HH, et al. Partner notification for HIV and STD in the United States: Low coverage for gonorrhea, chlamydial infection, and HIV. Sex Transm Dis 2003; 30:490–496.
5. Golden MR, Katz DA, Dombrowski JC. Modernizing field services for human immunodeficiency virus and sexually transmitted infections in the United States. Sex Transm Dis 2017; 44:599–607.
6. Katz DA, Dombrowski JC, Bell TR, et al. HIV incidence among men who have sex with men after diagnosis with sexually transmitted infections. Sex Transm Dis 2016; 43:249–254.
7. Katz DA, Dombrowski JC, Kerani RP, et al. Integrating HIV testing as an outcome of STD partner services for men who have sex with men. AIDS Patient Care STDS 2016; 30:208–214.
8. Centers for Disease Control and Prevention. Sexually Transmitted Disease Surveillance 2016 Atlanta: U.S. Department of Health and Human Services; 2017.
9. National Center for HIV/AIDS VH, STD, and TB Prevention (NCHHSTP) Newsroom. Telephone press briefing: New CDC analysis shows steep and sustained increases in STDs. David Harvey, MSW, executive director National Coalition of STD directors (NCSD). Telebriefing transcript: 2018 STD prevention conference. Accessed at https://www.cdc.gov/nchhstp/newsroom/2018/press-briefing-transcript-2018-std-prevention-conf.html
10. Rabarison KM, Bish CL, Massoudi MS, et al. Economic evaluation enhances public health decision making. Front Public Health 2015; 3:164.
11. Rahman MM, Khan M, Gruber D. A low-cost partner notification strategy for the control of sexually transmitted diseases: A case study from Louisiana. Am J Public Health 2015; 105:1675–1680.
12. Katz BP, Danos CS, Quinn TS, et al. Efficiency and cost-effectiveness of field follow-up for patients with Chlamydia trachomatis
infection in a sexually transmitted diseases clinic. Sex Transm Dis 1988; 15:11–16.
13. Johnson BL, Tesoriero J, Feng W, et al. Cost analysis and performance assessment of partner services for human immunodeficiency virus and sexually transmitted diseases, New York state, 2014. Health Serv Res 2017; 52(Suppl 2):2331–2342.
14. Reynolds SL, Kapadia AS, Leonard L, et al. Examining the direct costs and effectiveness of syphilis detection by selective screening and partner notification. J Public Health Med 2001; 23:339–345.
15. Assessment Unit Office of Infectious Disease Disease Control and Health Statistics WSDoH. STD Fast Facts: Washington State 2017. 2018.
16. Assessment Unit Office of Infectious Disease Disease Control and Health Statistics WSDoH. STD Fast Facts: Washington State 2016. 2018.
17. Golden MR, Kerani RP, Stenger M, et al. Uptake and population-level impact of expedited partner therapy (EPT) on Chlamydia trachomatis
and Neisseria gonorrhoeae
: The Washington state community-level randomized trial of EPT. PLoS Med 2015; 12:e1001777.
18. Katz DABT, Dombrowski JC, Kerani RP, et al. Sexually Transmitted Disease Partner Services Increases HIV Testing among Partners of Men Who Have Sex with Men. Atlanta, GA: STD Prevention Conference, 2016.
19. Katz DAGM, Bell TR, Kerani RP, et al. Using Sexually Transmitted Disease Partner Services to Promote Engagement in HIV Care among Persons Living with HIV. Atlanta, GA: STD Prevention Conference, 2016.
20. Katz DA, Dombrowski JC, Barry M, et al. STD partner services to monitor and promote HIV pre-exposure prophylaxis use among men who have sex with men. J Acquir Immune Defic Syndr 2019; 80:533–541.
21. Heumann CL, Katz DA, Dombrowski JC, et al. Comparison of in-person versus telephone interviews for early syphilis and human immunodeficiency virus partner Services in King County, Washington (2010-2014). Sex Transm Dis 2017; 44:249–254.
22. Martin EG, Feng W, Qian F, et al. Delivering partner services to reduce transmission and promote linkage to care: Process outcomes varied for chlamydial infection, gonorrhea, HIV, and syphilis cases. J Public Health Manag Pract 2017; 23:242–246.
23. Li XC, Kusi L, Marak T, et al. The cost and cost-utility of three public health HIV case-finding strategies: Evidence from Rhode Island, 2012-2014. AIDS Behav 2018; 22:3726–3733.
24. Shrestha RK, Begley EB, Hutchinson AB, et al. Costs and effectiveness of partner counseling and referral services with rapid testing for HIV in Colorado and Louisiana, United States. Sex Transm Dis 2009; 36:637–641.
25. Zulliger R, Maulsby C, Solomon L, et al. Cost-utility of HIV testing programs among men who have sex with men in the United States. AIDS Behav 2017; 21:619–625.
26. HIV/AIDS epidemiology unit public health—Seattle & King County and the Infectious Disease Assessment Unit—Washington State Department of Health. HIV/AIDS Epidemiology Report 2017:86. 2017.
27. Huang YL, Lasry A, Hutchinson AB, et al. A systematic review on cost effectiveness of HIV prevention interventions in the United States. Appl Health Econ Health Policy 2015; 13:149–156.
28. HIV/AIDS epidemiology unit—public health—Seattle & King County, infectious disease assessment unit—Washington State Department of Health. HIV/AIDS Epidemiology Report 2016; 85.
29. Chen LF, Vander Weg MW, Hofmann DA, et al. The Hawthorne effect in infection prevention and epidemiology. Infect Control Hosp Epidemiol 2015; 36:1444–1450.
30. Dombrowski JC, Hughes JP, Buskin SE, et al. A cluster randomized evaluation of a health department data to care intervention designed to increase engagement in HIV care and antiretroviral use. Sex Transm Dis 2018; 45:361–367.
31. Centers for Disease Control and Prevention 2017. Accessed at https://www.cdc.gov/hiv/programresources/guidance/costeffectiveness/index.html
. Accessed Feb 2, 2018.