# HIV Cost-effectiveness

The CDC Division of HIV Prevention is pleased to provide a basic guide to the cost-effectiveness analysis of prevention interventions for HIV infection and AIDS. The purpose of this guide is to help prevention program staff and planners become more familiar with potential uses of economic evaluation.

This guide consists of two sections. The first section introduces the basic concept of cost-effectiveness analysis. The second section provides the sources of basic model inputs commonly used in the literature. Significant publications in the field and other related sources are also provided at the end.

What is cost-effectiveness analysis?
Cost-effectiveness analysis (CEA) is a type of economic analysis where both the cost and the outcome (impact, result, effect, benefit, health gain …) of an intervention are evaluated and then expressed in the form of a cost-effectiveness ratio. The numerator of the cost-effectiveness (CE) ratio represents the cost of the intervention associated with one unit of “outcome”. The denominator is the unit of outcome. It can be expressed using many types of measures including: years of life gained, quality-adjusted life years gained (QALYs), new diagnoses, infections averted, and deaths averted. CEA is usually conducted on interventions that are known to be effective.

The CE ratio is a fraction used to compare the relative costs and outcomes of two or more interventions. In Example 1, the outcome measure chosen is “new HIV diagnoses” and the CE ratio of the programs evaluated is expressed in terms of “cost per new HIV diagnosis”. The CE ratio of Program A is \$41,667 per new HIV diagnosis. This ratio does not reveal the cost of implementing the program nor the number of new HIV diagnoses detected by the program. However, when comparing the CE ratio of Program A to that of Program B, we can say that Program B is more cost-effective than Program A when CE is measured in terms of “cost per new HIV diagnosis,” because at \$7,400 per new HIV diagnosis, Program B is less costly for the same outcome.

How to interpret a CE ratio?
Example 1 [a] Annual program cost [b] Annual number of new HIV diagnoses detected by program CE ratio:  Cost per new HIV diagnosis ([a]/[b])
Program A \$500,000 12 \$41,667 / new HIV diagnosis
Program B \$37,000,000 5,000 \$7,400 / new HIV diagnosis

Cost per new HIV diagnosis
HIV interventions, such as screening and partner services, are intended to identify HIV-positive persons who are unaware of their infection. When evaluating several such programs in CE analysis, the outcome “new HIV diagnoses” is often used to enable a comparison across these programs; so the CE ratio is expressed in terms of cost per new HIV diagnosis.

Cost per infection averted (IA)
HIV prevention interventions such as syringe exchange programs, counseling for at-risk youth or post-exposure prophylaxis are intended to prevent (avert) infection in HIV-negative persons. Such programs can be evaluated to determine the number of infections prevented that would have otherwise occurred had the intervention not been provided. When evaluating several such programs in CE analysis, the outcome “HIV infections averted” is often used to enable a comparison across these programs; so the CE ratio can be expressed in terms of cost per infection averted.

The lifetime treatment cost of an HIV infection can be used as a conservative threshold value for the cost of averting one infection. Currently, the lifetime treatment cost of an HIV infection is estimated at \$379,668 (in 2010 dollars), therefore a prevention intervention is deemed cost-saving if its CE ratio is less than \$379,668 per infection averted.

As an outcome, the number of HIV infections averted due to a program can be evaluated using different mathematical techniques that vary in complexity and the amount of data or number of assumptions required. Attention should be paid to the timeframe of the intervention effect considered in the evaluation. For example, if the timeframe is one year, then the cost may have to be incurred annually in order to continue to avert the HIV infections.

Cost per life year (LY) gained
HIV treatment interventions, including retention in care and treatment adherence, are in part intended to extend the lives of HIV-positive persons. Such programs can be evaluated to determine the number of additional life years gained (or saved) that otherwise would have been lost to premature death. When evaluating these types of  programs in CE analysis, the outcome “life years” often is used to compare them; so the CE ratio can be expressed in terms of cost per life year gained.

Cost per quality-adjusted life year (QALY) gained
As an outcome, life years do not reflect any of the positive or negative effects on the quality of life of the patients receiving an intervention. For example, drug treatment A may provide an additional 2 years of life dominated by hospitalization while drug treatment B may provide an additional 1 year of life without any significant ill effects.

A quality-adjusted life year (QALY) is an outcome measure that considers both the quality and the quantity of life lived. The QALY is based on the number of years of life added by the intervention. Each year in perfect health is assigned the value of 1.0.  Each year of less-than-perfect health is assigned a value less than 1.0 down to a value of 0.0 for death. If the extra years would not be lived in full health, for example if the patient would lose a limb, be blind or suffer from worse mental health, then the extra life-years may be given a value of less than 1 to account for this.

HIV interventions intended to improve and/or extend the lives of HIV positive persons can be evaluated to determine the number of additional QALYs gained (or saved) that would have otherwise been lost. When evaluating several such programs in CE analysis, the CE ratio can be expressed in terms of cost per QALY gained.

Most outcome measures, including infections averted, life years gained and new HIV diagnoses, can be translated into QALYs, thereby providing a consistent measure of comparison across many different types of intervention programs.

Cost-effectiveness thresholds

A cost-effectiveness ratio of \$50,000 to \$100,000 per QALY gained has been long cited in the literature as a conservative threshold for a cost-effective intervention. Traditionally, if an intervention was estimated to cost less than \$50,000 to \$100,000 per QALY gained, it would be considered cost-effective. However, recent studies have argued that this benchmark is likely too low since the threshold has not been reassessed over time.1 To reflect the advances of modern health care, Braithwaite et al reevaluated the threshold and estimated the plausible range for a cost-effectiveness decision rule to be between \$109,000 and \$297,000 per QALY saved (in 2003 dollars; \$143,000-\$388,000 in 2010 dollars).2

What does “cost-saving” mean?

When two or more programs are being compared (intervention vs. comparator), the intervention is labeled as “cost-saving” when both the net outcome of the intervention is greater than or equal to that of the comparator and the cost of the intervention is less than the cost of the comparator. A program can only be deemed cost-saving when it is compared to an alternative. The alternative is typically the status quo or the current standard of care.

In Example 2, Program A is both cheaper and more beneficial than the current standard of care and is therefore a cost-saving alternative. CE ratios cannot be negative.

How to interpret a CE ratio?
Example 2 [a] Annual program cost [b] Annual number of QALYs gained CE ratio:  Cost per QALY gained ([a]/[b])
Program A (intervention) \$750,000 50 \$15,000 / QALY gained
Standard of care (comparator) \$1,000,000 40 \$25,000 / QALY gained
Difference \$(250,000) 10 Cost-saving

If the costs of Program A and the Standard of care are borne by the same institution, then the savings will be reaped by that institution. Often, however, the costs of HIV interventions are borne by many distinct entities, including government, health care systems and individuals, and the savings are not realized by any single entity. In addition, the savings may occur over many years.

How to interpret a CE ratio?

At \$100,000 per QALY (or at higher thresholds), a program may be considered cost-effective. However, this ratio contains a numerator and a denominator and thus no interpretation can be made as to the annual cost of this program.

How to interpret a CE ratio?
Example 3 [a] Annual program cost [b] Number of persons served by program [c] Sum of QALYs gained by program Cost per person served ([a]/[b]) Cost per QALY gained ([a]/[c])
Program A \$400,000 4,000 10 \$100 \$40,000
Program B \$50,000,000 5,000 1,250 \$10,000 \$40,000

In Example 3, both programs A and B have the same measure of cost-effectiveness in terms of cost per QALY gained, however, Program B is more costly to implement than A. Investment in Program B may nonetheless be justified depending on budgetary constraints and the ability to implement for the program in the population and setting considered.

If A and B are complementary rather than alternative programs, then they can both be implemented. Implementing Program A and/or B in a particular population and setting requires an evaluation of the number of persons that potentially could be served by the intervention and the resulting overall costs.

In this section, we list some of the most recent and significant publications that include key input parameters researchers might use in model-based cost-effectiveness analyses. It is not intended to provide a comprehensive overview of these topic areas—only to give readers an idea of some key works in the field.

Cost of HIV treatment
A large fraction of the economic burden of HIV/AIDS is the medical costs of treating persons with HIV. Medical cost estimates are often based on health care utilization by persons with HIV disease. The costs associated with health care utilization in each disease stage are summed across all disease stages from infection to death. The average annual cost of HIV care in the ART era was estimated to be \$19,912 (in 2006 dollars; \$23,000 in 2010 dollars).3 The most recent published estimate of lifetime HIV treatment costs was \$367,134 (in 2009 dollars; \$379,668 in 2010 dollars).4

Testing in health care settings
Several US-based studies have evaluated the cost-effectiveness of routine opt-out HIV screening in clinical settings. These settings included emergency departments, primary care settings, urgent care centers, and STD clinics. The results were generally consistent. The cost per new diagnosis ranged from \$1,900 to \$10,000 (in 2010 dollars), and varied by setting and testing implementation strategy.5-9

Testing in non-health care settings
Non-health care settings, such as jails/prisons, community-based organizations (CBOs), and outreach venues, are also common places to implement HIV testing programs. Individuals eligible for testing in those settings could be identified through partner services or social networks. Cost-effectiveness studies of these strategies have found the results generally consistent within similar settings. For example, the cost per new HIV diagnosis associated with CBO-sponsored activities ranged from \$10,334 to \$20,413 (2010 dollars).10-11 Variance in the cost per new HIV diagnosis was more pronounced when evaluating HIV testing programs in jails  (from \$2,946 per new diagnosis in Florida jails to \$30,392 in Wisconsin jails),12 reflecting the differences in undiagnosed HIV prevalence among inmates as well as differences in implementation costs.

HIV survival
The use of highly active antiretroviral therapy (HAART) since 1996 has significantly improved survival for persons infected with HIV. Schackman et al. estimated life expectancy from the time of infection to be 32.1 years from a large dataset of persons in routine outpatient care in the current treatment era.4 Using US national HIV surveillance data, another study estimated that average life expectancy after an HIV diagnosis increased from 10.5 to 22.5 years from 1996 to 2005.13

HIV survival data have been reported slightly differently in the literature because of various definitions of timeframe, e.g., time from HIV seroconversion to AIDS, time from seroconversion to death, and time from HIV diagnoses to death. Survival also varies by gender, age at infection, mode of infection, and the timing of initiation of antiretroviral therapy.14-17

Recent HIV incidence estimates
CDC published new incidence estimates in 2011 using a refined methodology that allowed for an updated 2006 incidence estimate (previously 56,300) as well as new estimates for 2007, 2008, and 2009. These new estimates showed that the annual number of new HIV infections was stable overall from 2006 through 2009:18

• In 2006 there were an estimated 48,600 new HIV infections in the United States (95% confidence interval: 42,400-54,700)
• In 2007 there were an estimated 56,000 new HIV infections (95% confidence interval: 49,100-62,900)
• In 2008 there were an estimated 47,800 new HIV infections (95% confidence interval: 41,800-53,800)
• In 2009 there were an estimated 48,100 new HIV infections (95% confidence interval: 42,200-54,000)

More HIV surveillance reports can be found in HIV Surveillance Reports.

HIV transmission rate estimates
HIV transmission risk varies by different modes of transmission. The most common transmission modes include unprotected receptive and insertive anal intercourse, unprotected receptive and insertive vaginal intercourse, and contaminated needle sharing. The estimates of these and other per-act or per-partner transmission probabilities can be found in the listed references of systematic reviews and meta-analyses.19-22

Utility Estimate for HIV/AIDS
Many studies have reported quality-of-life estimates for HIV infection and AIDS. Published estimates vary by study design and assessment method.23, 24 Tengs et al conducted a meta-analysis of utility estimates for HIV/AIDS to elicit utilities from patients on a scale ranging from 0.0 for death to 1.0 for perfect health. The study is commonly cited for reporting a pooled estimate of utility of 0.70 for AIDS patients, 0.82 for symptomatic HIV patients, and 0.94 for asymptomatic HIV patients [3].25

Annual cost of HIV by state

We estimated the annual cost of HIV by state based on the number of new HIV diagnoses in each state, multiplied by the lifetime treatment cost discounted to the time of infection for each new case (Table 1). Our cost estimates assume that a diagnosis occurs within the same year as infection, and thus an individual incurs treatment costs over many years. The states with highest number of new diagnoses in 2009, and thus the greatest financial burden, were Florida, California, New York, and Texas. In all, the total lifetime treatment cost for HIV based on new diagnoses in 2009 was estimated to be \$16.6 billion.

Table 1: State-Specific Costs from New Diagnoses of HIV Infection in 2009—
United States and 5 U.S. dependent areas

Table 1: State-Specific Costs from New Diagnoses of HIV Infection in 2009—
United States and 5 U.S. dependent areas
State Nb. of New Diagnosesa Total Lifetime Treatment Costb
(in million)
Alabama 690 \$253
Alaska 21 \$8
Arizona 653 \$240
Arkansas 214 \$79
California 4,886 \$1,794
Colorado 391 \$144
Connecticut 366 \$134
Delaware 168 \$62
District of Columbia 713 \$262
Florida 5,775 \$2,120
Georgia 2,073 \$761
Hawaii 70 \$26
Idaho 42 \$15
Illinois 1,708 \$627
Indiana 483 \$177
Iowa 125 \$46
Kansas 150 \$55
Kentucky 361 \$133
Louisiana 1,247 \$458
Maine 57 \$21
Maryland 1,400 \$514
Massachusetts 484 \$178
Michigan 827 \$304
Minnesota 393 \$144
Mississippi 559 \$205
Missouri 547 \$201
Montana 30 \$11
Nebraska 105 \$39
Nevada 386 \$142
New Hampshire 43 \$16
New Jersey 1,252 \$460
New Mexico 170 \$62
New York 4,649 \$1,707
North Carolina 1,719 \$631
North Dakota 14 \$5
Ohio 1,144 \$420
Oklahoma 297 \$109
Oregon 235 \$86
Pennsylvania 1,736 \$637
Rhode Island 123 \$45
South Carolina 789 \$290
South Dakota 23 \$8
Tennessee 999 \$367
Texas 4,291 \$1,575
Utah 125 \$46
Vermont 11 \$4
Virginia 997 \$366
Washington 557 \$204
West Virginia 80 \$29
Wisconsin 305 \$112
Wyoming 19 \$7
Subtotal 44,502 \$16,338
U.S. dependent areas
American Samoa 0 \$0
Guam 3 \$1
Northern Mariana Islands 1 \$0
Puerto Rico 671 \$246
U.S. Virgin Islands 25 \$9
Subtotal 700 \$257
Total 45,202 \$16,595

Note:
a Source: CDC HIV Surveillance Report 2009, Vol 21. Table 19
Note that the numbers of new diagnoses listed in this table do not adjust for reporting delay, and thus are likely underestimated.
b Total cost = Nb. of new diagnoses* Lifetime treatment cost per person
Life treatment cost per person=\$367,134 (in 2009 dollars)
Source: Schackman BR, Gebo KA, Walensky RP, et al. The lifetime cost of current human immuno-deficiency virus care in the United States. Medical Care 2006; 44: 990-997.

Savings from prevention efforts

Farnham et al. (2010) measured the value of HIV prevention efforts in the United States by comparing the difference between the number of infections that have occurred with the number that might have occurred in the absence of prevention programs. Combined with estimates of lifetime treatment costs4 (2009 dollars), the study estimated the medical savings from infections averted by US prevention programs from 1991-2006 to be \$129.9 billion with 361,878 HIV infections averted.26

## References and Resources

• Ubel PA, Hirth RA, Chernew ME, Fendrick AM. What is the price of life and why doesn’t it increase at the rate of inflation? Arch Intern Med 2003; 163(14): 1637-1641.
• Braithwaite RS, Meltzer DO, King JT Jr, Leslie D, Roberts MS. What does the value of modern medicine say about the \$50,000 per quality-adjusted life-year decision rule? Med Care 2008; 46(4): 349-356.
• Gebo KA, Fleishman JA, Conviser R, Hellinger J, Hellinger FJ, Josephs JS, Keiser P, Gaist P, Moore RD; HIV Research Network. Contemporary costs of HIV healthcare in the HAART era. AIDS 2010; 24(17): 2705-2715.
• Schackman BR, Gebo KA, Walensky RP, Losina E, Muccio T, Sax PE, Weinstein MC, Seage GR 3rd, Moore RD, Freedberg KA. The lifetime cost of current human immunodeficiency virus care in the United States. Med Care 2006; 44(11):990-997.
• Phillips KA, Fernyak S. The cost-effectiveness of expanded HIV counseling and testing in primary care settings: a first look. AIDS 2000; 14(14): 2159-2169.
• Walensky RP, Losina E, Malatesta L, Barton GE, O’Connor CA, Skolnik PR, Hall JM, McGuire JF, Freedberg KA. Effective HIV case identification through routine HIV screening at urgent care centers in Massachusetts. Am J Public Health 2005; 95(1): 71-73.
• Silva A, Glick NR, Lyss SB, Hutchinson AB, Gift TL, Pealer LN, Broussard D, Whitman S. Implementing an HIV and sexually transmitted disease screening program in an emergency department. Ann Emerg Med 2007; 49(5): 564-572.
• Mehta SD, Hall J, Greenwald JL, Cranston K, Skolnik PR. Patient risks, outcomes, and costs of voluntary HIV testing at five testing sites within a medical center. Public Health Rep 2008; 123(5): 608-617.
• Farnham PG, Hutchinson AB, Sansom SL, Branson BM. Comparing the costs of HIV screening strategies and technologies in health-care settings. Public Health Rep 2008; 123 Suppl 3: 51-62.
• Shrestha RK, Clark HA, Sansom SL, Song B, Buckendahl H, Calhoun CB, Hutchinson AB, Heffelfinger JD. Cost-effectiveness of finding new HIV diagnoses using rapid HIV testing in community-based organizations. Public Health Rep 2008; 123 Suppl 3: 94-100.
• Golden MR, Gift TL, Brewer DD, Fleming M, Hogben M, St Lawrence JS, Thiede H, Hnadsfield HH. Peer referral for HIV case-finding among men who have sex with men. AIDS; 20(15): 1961-1986.
• Shrestha RK, Sansom SL, Richardson-Moore A, French PT, Scalco B, Lalota M, Llanas M, Stodola J, Macgowan R, Margolis A. 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(2 Suppl): S5-S8.
• Harrison KM, Song RG, Zhang XJ. Life expectancy after HIV diagnosis based on national HIV surveillance data from 25 states, United States. JAIDS 2010; 53(1): 124-130.
• Losina E, Schackman BR, Sadownik SN, Gebo KA, Walensky RP, Chiosi JJ, Weinstein MC, Hicks PL, Aaronson WH, Moore RD, Paltiel AD, Freedberg KA. Racial and sex disparities in life expectancy losses among HIV-infected persons in the United States:  impact of risk behavior, late initiation, and early discontinuation of antiretroviral therapy. Clin Infect Dis 2009; 49(10): 1570-1578.
• Phillips AN, Gazzard B, Gilson R, Easterbrook P, Johnson M, Walsh J, Leen C, Fisher M, Orkin C, Anderson J, Pillay D, Delpech V, Sabin C, Schwenk A, Dunn D, Gompels M, Hill T, Porter K, Babiker A; UK Collaborative HIV Cohort Study. Rate of AIDS diseases or death in HIV-infected antiretroviral therapy-naïve individuals with high CD4 cell count. AIDS 2007; 21(13): 1717-1721.
• The Antiretroviral Therapy Cohort Collaboration. Life expectancy of individuals on combination antiretroviral therapy in high-income countries: a collaborative analysis of 14 cohort studies. Lancet 2008; 372: 293–299.
• May M, Compels M, Sabin CA. Impact on life expectancy of late diagnosis and treatment of HIV-1 infected individuals: UK CHIC. Journal of the International AIDS Society 2010; 13 (Suppl 4): O27.
• Prejean J, Song R, Hernandez A, Ziebell R, Green T, Walker F, Lin LS, Mermin J, Lansky A, Hall HI; HIV Incidence Surveillance Group. Estimated HIV Incidence in the United States, 2006-2009. PLoS ONE 2011; 6(8): e17502.
• Boily MC, Baggaley RF, Wang L, Masse B, White RG, Hayes RJ, Alary M. Heterosexual risk of HIV-1 infection per sexual act: systematic review and meta-analysis of observational studies. Lancet Infect Dis 2009; 9(2): 118-129.
• Baggaley RF, White RG, Boily MC, HIV transmission risk through anal intercourse: systematic review, meta-analysis and implications for HIV prevention. Int J Epidemiol 2010; 39(4): 1048-1063.
• Vittinghoff E, Douglas J, Judson F, McKirnan D, MacQueen K, Buchbinder SP. Per-contact risk of human immunodeficiency virus transmission between male sexual partners. Am J Epidemiol 1999; 150(3): 306-311.
• Baggaley RF, Boily MC, White RG, Alary M.  Risk of HIV-1 transmission for parenteral exposure and blood transfusion: a systematic review and meta-analysis. AIDS 2006; 20(6): 805-812.
• Bayoumi AM, Redelmeier DA. Economic methods for measuring the quality of life associated with HIV infection. Qual Life Res 1999; 8(6): 471-480.
• Schackman BR, Goldie SJ, Freedberg KA, Losina E, Brazier J, Weinstein MC. Comparison of health state utilities using community and patient preference weights derived from a survey of patients with HIV/AIDS. Med Decis Making 2002; 22(1): 27-38.
• Tengs TO, Lin TH. A meta-analysis of utility estimates for HIV/AIDS. Med Decis Making 2002; 22(6): 475-481.
• Farnham PG, Holtgrave DR, Sansom SL, Hall HI. Medical costs averted by HIV prevention efforts in the United States, 1991-2006. JAIDS 2010; 54(5): 565-567.

Cost-Effectiveness Databases

This US-based database offers detailed information on more than 2,500 English-language cost-effectiveness analyses published in the peer-reviewed medical and economic literature.  The original cost per QALY estimates from a wide array of diseases and intervention types are retrieved and updated to the most recent year for comparison purposes.

NICE serves the United Kingdom’s National Health Services and is well-known for developing and publishing guidelines on public health, health technologies and clinical practice based on evaluations of efficacy and cost-effectiveness evidence. It also sets quality standards and manages a national database for high-quality, cost-effective patient care, covering the treatment and prevention of different diseases and conditions. NICE provides access to quality information and best practices so that care decisions can be made based on the best possible evidence.

This UK-based database focuses on the economic evaluation of health care interventions, including cost-benefit analyses, cost-utility analyses, and cost-effectiveness analyses. Extensive literature searches are undertaken each week, and brief details from eligible studies are published on the database. Studies that are relevant to the UK health care system are considered priorities for writing abstracts, which include a non-technical summary of the topic, conclusions, and a brief description of the effectiveness information. A critical commentary summarizes the overall reliability and generalizability of the study.

Cost Inflation Tool

Consumer Price Index (CPI)
The Consumer Price Index (CPI) is a measure of the average change over time in the prices paid by urban consumers for a market basket of consumer goods and services. Every month the Bureau of Labor Statistics (BLS) surveys prices and generates the CPI. The CPI includes all consumer expenditure items in more than 200 categories, arranged into eight major CPI components, such as housing, transportation, medical care, etc. Researchers often use the medical care component of the CPI to adjust health care costs reported in previous years to their value in current dollars.

Historic CPI series and component data can be found at the Bureau of Labor Statistics website: http://www.bls.gov/cpi/.

How to adjust medical costs using CPI
The formula for calculating current costs using the CPI is relatively simple. Assume the medical care component of CPI for year 1990 is 125. A CPI for year 2000 of 175 indicates a 1.4 cost adjustment or a 40% increase in medical-related expenditures since 1990.

Approaches to Address Barriers
Year Medical care component of CPI Inflation rate from
year 1 to year 5
1990 125 175/125=1.4
2000 175

Assume the medical costs for condition X are estimated to be \$10,000 per patient in year 1990. By multiplying \$10,000 by 1.4, we get the value of the medical costs for condition X in 2000. In other words, treatment valued at \$10,000 in 1990 would cost \$14,000 in 2000.

## Additional Prevention Modeling and Economics Team (PMET) Publications

1. Prabhu VS, Farnham PG, Hutchinson AB, Soorapanth S, Heffelfinger JD, Golden MR, Brooks JT, Rimland D, Sansom SL. Cost-effectiveness of HIV screening in STD clinics, emergency departments, and inpatient units: a model-based analysis. PLoS One 2011; 6(5): e19936.
2. Hutchinson AB, P. Patel, S. L. Sansom, P. G. Farnham, T. J. Sullivan, B. Bennett, P. R. Kerndt, R. K. Bolan, J. D. Heffelfinger, V. S. Prabhu, and B. M. Branson. Cost-effectiveness of pooled nucleic acid amplification testing for acute HIV infection after third-generation HIV antibody screening and rapid testing in the United States: a comparison of three public health settings. PLoS Med 2010; 7: e1000342.
3. Prabhu VS, Hutchinson AB, Farnham PG, Sansom SL.  Sexually acquired infections in the United States due to acute-phase HIV transmission: an update. AIDS 2009;23(13): 1792-1794.
4. Sansom SL, Hutchinson AB, An Q, Hall I, Shrestha RK, Prabhu VS, Lasry A, Taylor A. Cost-effectiveness of newborn circumcision in preventing HIV among U.S. males. PLoS One 2010; 5(1): e8723.
5. Hutchinson AB, Farnham PG, Duffy N, Wolitski RJ, Sansom SL, Dooley SW, Cleveland JC, Mermin JH. Return on public health investment: CDC’s expanded HIV testing initiative. JAIDS 2011 (Epub ahead of print).