PERSPECTIVE

How to Think about Future Health Care Spending

and it is particularly irresponsible when the interventions are provided at public expense. On the other hand, we must avoid an innovation policy that cuts off new interventions prematurely. Some interventions that are not cost-effective at first may prove to be so over time and with greater experience in implement-

ing them. It is in gathering this experience that the private part of the should question becomes important. Given a substantial market for untried, cutting-edge interventions that are not subsidized by the public purse, some innovations may prove to be costeffective in the long run. Such innovations should then be in-

cluded in the publicly financed benefit package.

Disclosure forms provided by the author are available with the full text of this article at NEJM.org. From Stanford University, Stanford, CA. This article (10.1056/NEJMp1000200) was published on March 10, 2010, at NEJM.org.
Copyright 2010 Massachusetts Medical Society.

AIDS in America Forgotten but Not Gone

Wafaa M. El-Sadr, M.D., M.P.H., Kenneth H. Mayer, M.D., and Sally L. Hodder, M.D. ver the past decade, limited attention has been paid to the human immunodeficiency virus (HIV) epidemic in the United States. The global epidemic particularly the epidemic in subSaharan Africa, where approximately two thirds of the worlds population living with AIDS resides has rightfully received most of the focus. Meanwhile, however, the prevalence of HIV infection within some U.S. populations now rivals that in some sub-Saharan African countries (see graph). For example, more than 1 in 30 adults in Washington, D.C., are HIV-infected a prevalence higher than that reported in Ethiopia, Nigeria, or Rwanda.1 Certain U.S. subpopulations are particularly hard hit. In New York City, 1 in 40 blacks, 1 in 10 men who have sex with men, and 1 in 8 injection-drug users are HIV-infected, as are 1 in 16 black men in Washington, D.C.2 In several U.S. urban areas, the HIV prevalence among men who have sex with men is as high as 30%3 as compared with a general-population prevalence of 7.8% in Kenya and 16.9% in South Africa. During the first two decades

of the epidemic, remarkable advances in preventing mother-tochild transmission, screening of blood and blood products, and behavior change among men who have sex with men resulted in significant decreases in new HIV infections in the United States from approximately 130,000 in 1984 to about 60,000 in 1991. For the past decade, however, progress has been stalled. It had been anticipated that effective antiretroviral therapy, with its suppressive effect on viral replication, would reduce the overall rate of new infections, but this expectation has not been realized. More than half a million Americans became infected with HIV in the past decade, including about 56,000 in the past year.4 It is estimated that there are now more than 1 million HIV-infected Americans, more than 20% of whom are unaware of their infection. Unlike the generalized HIV epidemics in sub-Saharan Africa, the U.S. epidemic primarily affects certain discrete geographic areas especially urban areas of the Northeast and West Coast and cities and small towns in the South (see U.S. map). Within

these areas, specific neighborhoods are often disproportionately affected (see New York City map), in part because of residents engagement in unprotected sex within relatively insular social sexual networks. Many of the populations most affected tend to have limited social mobility; thus, partner selection tends to concentrate transmission patterns and amplify spread within defined geographic areas. Traditionally, researchers and policymakers concerned with HIV acquisition have concentrated on specific high-risk transmission behaviors, including injection-drug use, sex with multiple partners, and failure to use protective measures such as condoms or safe injection practices. It is now evident that among men who have sex with men, the use of drugs such as crystal methamphetamine especially at sex parties and in venues such as bathhouses has contributed to risky behavior and HIV acquisition. Other disinhibiting substances, including alcohol and cocaine, are also associated with increased risk taking in these populations. However, the extent of the risk of acquiring HIV in the United

n engl j med 362;11

nejm.org

march 18, 2010

967

Downloaded from www.nejm.org on May 12, 2010 . Copyright 2010 Massachusetts Medical Society. All rights reserved.

PERS PE C T IV E

AIDS in America Forgotten but Not Gone

18 16 14 12

10 8 6 4 2 0

Population

HIV Prevalence in Adults from Selected Countries in Sub-Saharan Africa and Subpopulations in the United States. Data are from the Centers for Disease Control and Prevention, the District of Columbia Department of Health, the New Jersey Department of Health, the New York City Department of Health and RETAKE: 1st AUTHOR: El-Sadr Mental Hygiene, and the Joint United Nations Program on HIV/AIDS (UNAIDS). MSM denotes 2nd FIGURE: men who have sex with men. 1 of 2 3rd
ARTIST: ts
Revised

States today is largely defined 4-C H/T the potential benefit of such by its TYPE: Line 2x col Combo a persons sexual network rather therapy as a preventive strategy. AUTHOR, PLEASE NOTE: than his or her individual behav- type hasThe reset. Figure has been redrawn and been situation is similar for Please check carefully. iors. Understanding the context black and Hispanic women, whose and settings in which risk is in- increased risk of HIV acquisition JOB: 36211 ISSUE: 03-18-09 creased may lead to more robust is attributable in greater part to and effective preventive interven- their vulnerable social and ecotions. For example, black men nomic situations and their sexual who have sex with men are at networks than to their own risky increased risk for HIV infection behaviors. Socioeconomic disadin part because of its high prev- vantage and instability of partalence in their sexual networks nerships due to high rates of inand their likelihood of choosing carceration among men in their racially similar partners; they have communities may lead women to also been shown to be less likely engage in concurrent relationthan their white counterparts to ships or serial monogamy. In adbe aware of their HIV status and dition, they may be unaware of thus are more likely to unknow- their partners HIV status or may ingly transmit HIV.5 Moreover, be involved in abusive or economeven those who are aware of their ically dependent relationships and HIV infection may be less en- thus be unable to negotiate safer gaged in HIV care and less likely sex with their partners. The specific characteristics of to avail themselves of antiretroviral therapy behavior that lim- the U.S. HIV epidemic low
968

SIZE

prevalence in the general population, high prevalence among the disenfranchised and socially marginalized, with a concentration in geographic hotspots in combination with the various structural impediments to prevention create unique challenges for the design and implementation of effective interventions. Thus, a nuanced and targeted approach that avoids stigmatization of these populations is necessary. Structural interventions might include tackling the disproportionate incarceration of black and Hispanic men, urging health insurers to reimburse providers for preventive care, and using microcredit to help women out of poverty so that they avoid the perceived need to engage in commercial sex or other coercive sex. Research tailored to specific populations is required if we are to gain the understanding needed to move forward. For example, how do we identify those people in the United States who are at greatest risk for HIV acquisition, especially among women? Although more than a quarter of new HIV infections in the United States occur in women (predominantly black or Hispanic women), identifying such at-risk women to engage them in prevention studies has proven particularly challenging. Research is also needed to identify interventions that will persuade men who have sex with men to undergo HIV testing, facilitate their disclosure of their HIV status to sexual partners, and promote negotiations for safer sexual practices; such interventions need to be implemented in the settings where such men may meet (e.g., in bars or on the Internet). Additional qualitative research is needed to understand how the targeted community uses various sources of information

Prevalence (%)
Un W

as hi n D gto .C . ( n, D bl ac . C. km en N ew Ne ) ar wa k, rk ,N N .J. (b .J. la N ck ew Ne s) Yo w Y or rk N kC ew Ci Yo ty ( ity bl rk Ci ack D ty s em (M ) oc SM ra tic ) Re pu S en bl e ic of gal Co ng Et o hi op Rw ia an d Bu a ru nd N i ig er ia Ke So ny ut a h Af ric a to n,

as

hi

ng

ite d

St at es

n engl j med 362;11

nejm.org

march 18, 2010

PERSPECTIVE

AIDS in America Forgotten but Not Gone

Neighborhoods in New York City

No. of Cases
2731000 10012000 20014000 40018000 25,001 199,402 8001 25,000

Proportion of persons living with HIVAIDS in 2008 0.20.5% 0.61.0% 1.11.8% 1.94.8%
Harlem Manhattan

South Bronx

Queens Brooklyn

Staten Island

Washington Montana Oregon Idaho Wyoming Nevada Utah California Colorado Kansas Missouri Arizona New Mexico Oklahoma Kentucky Tennessee Arkansas Mississippi Texas Louisiana Nebraska Iowa Illinois Indiana Ohio West Virginia Virginia North Carolina South Carolina Washington, D.C. Alabama Georgia Florida South Dakota North Dakota Minnesota Wisconsin Michigan Pennsylvania Vermont New York Maine New Hampshire Massachusetts Rhode Island Connecticut New Jersey Delaware Maryland

Alaska

Hawaii

Puerto Rico

Numbers of Reported AIDS Cases According to Metropolitan Statistical Area of Residence, Cumulative through 2007. Only metropolitan statistical areas with a population of more than 500,000 were included. Data are from the HIV/AIDS Surveillance Report, 2007, Centers for Disease Control and Prevention. The inset shows the proportion of persons living with HIVAIDS in New York City. Data are from the New York City Department of Health and Mental Hygiene, Epidemiology and Field Services Program, October 2009. RETAKE: 1st AUTHOR: El-Sadr
FIGURE: 2 of 2
2nd 3rd Revised

in making decisions about sexual risk taking. Most glaringly, HIV disproportionately affects poor black Americans who have substandard education, unstable housing, and limited social mobility. This confluence of factors may result in high rates of incarceration, which threaten a communitys social fabric. Such vulnerable populations must be engaged in research, program development, and interven-

are culturally relevant SIZE transfer (providing financial inand address Combosocioeconomic centives for healthy behavior) or the 4-C TYPE: Line H/T milieu in which HIV transmis- to support work in high-risk AUTHOR, PLEASE NOTE: sion occurs. redrawn and type has been reset. venues, such as bathhouses, has Figure has been PreventivePlease check carefully.must 3 col interventions hampered progress. Cash transbe rooted in science, not driven fer has proved effective in achievJOB: 36211 ISSUE: 03-18-09 by ideological concerns. Homo- ing desirable health outcomes, phobia may have impeded the including weight control, smokdevelopment of sexually appro- ing cessation, and decreased use priate prevention studies among of crystal methamphetamine, but men who have sex with men. Re- until recently it was not being luctance to fund studies of needle studied for use in HIV prevention exchange or conditional cash in the United States.
969

tions that ARTIST: ts

n engl j med 362;11

nejm.org

march 18, 2010

Downloaded from www.nejm.org on May 12, 2010 . Copyright 2010 Massachusetts Medical Society. All rights reserved.

PERSPECTIVE

AIDS in America Forgotten but Not Gone

What will it take to control the U.S. HIV epidemic? First, there is an urgent need to acknowledge that HIV remains a major health threat in the United States. Second, concerted effort and substantial resource investment especially in innovative and courageous approaches are necessary. Focused studies of the sociocultural dynamics that facilitate transmission are needed, as well as large studies assessing the effectiveness of multidimensional interventions, including behavioral, biomedical, and structural components. Disenfranchised communities must be engaged as partners in such efforts, along with new researchers drawn from the affected populations, if the

nuances of local epidemics are to be addressed. The time has come to confront this largely forgotten and hidden epidemic.
Disclosure forms provided by the authors are available with the full text of this article at NEJM.org. From the International Center for AIDS Care and Treatment Programs, Mailman School of Public Health, and the Department of Medicine, Harlem Hospital and College of Physicians and Surgeons, Columbia University, New York (W.M.E.-S.); the Fenway Institute at Fenway Health, Boston, and the Departments of Medicine and Community Health, Brown University, Providence, RI (K.H.M.); and the Department of Medicine, University of Medicine and Dentistry of New Jersey, Newark (S.L.H.). This article (10.1056/NEJMp1000069) was published on February 10, 2010, at NEJM.org. 1. District of Columbia HIV/AIDS epidemiology update 2008. Washington, DC: Gov-

ernment of the District of Columbia, Department of Health, HIV/AIDS Administration; 2008. (Accessed February 8, 2010, at http:// dchealth.dc.gov/DOH/frames.asp?doc=/ doh/lib/doh/pdf/dc_hiv-aids_2008_ updatereport.pdf.) 2. Nguyen TQ, Gwynn RC, Kellerman SE, et al. Population prevalence of reported and unreported HIV and related behaviors among household adult population in New York City, 2004. AIDS 2008;22:281-7. 3. HIV prevalence among selected populations: high-risk populations. Atlanta: Centers for Disease Control and Prevention, 2007. (Accessed February 8, 2010, at http://www .cdc.gov/hiv/topics/testing/resources/ reports/hiv_prevalence/high-risk.htm.) 4. Hall HI, Song R, Rhodes P, et al. Estimation of HIV incidence in the United States. JAMA 2008;300:520-9. 5. Millett GA, Flores SA, Peterson JL, Bakeman R. Explaining disparities in HIV infection among black and white men who have sex with men: a meta-analysis of HIV risk behaviors. AIDS 2007;21:2083-91.
Copyright 2010 Massachusetts Medical Society.

Five Next Steps for a New National Program for Comparative-Effectiveness Research
Jordan M. VanLare, A.B., Patrick H. Conway, M.D., and Harold C. Sox, M.D. he American Recovery and Reinvestment Act appropriated $1.1 billion to fund comparative-effectiveness research (CER) unprecedented generosity for a program for evaluating health care practices. The legislation established the Federal Coordinating Council for Comparative Effectiveness Research and charged it with advising the secretary of health and human services on the allocation of CER funds. It also mandated an Institute of Medicine (IOM) study to recommend initial national priorities for CER. Both the Federal Coordinating Council and the IOM reported to Congress on June 30, 2009. Both organizations solicited input from stakeholders. The IOM committee issued an open solicitation asking the public to nomi970

nate research topics. It received 1546 nominations, which it narrowed to 100 highest-priority research questions. The Federal Coordinating Council hosted three public listening sessions to identify priorities and posted drafts of its work on its Web site for public comment. By establishing a national CER agenda with input and support from diverse stakeholders, the two reports moved the United States closer to creating a sustained national CER program.1,2 Both reports recognized the need for a robust CER enterprise. The IOM made 10 recommendations for its development (see box). The Federal Coordinating Councils report included a definition of CER, a strategic framework, priority-setting criteria, and rec-

ommendations for investing the $400 million that Congress allocated to the Department of Health and Human Services for CER. Both reports recommended creating CER data networks and conducting research on practitioners adoption of changes based on CER findings.3 Federal agencies and Congress appear willing to implement these recommendations. The National Institutes of Health, the Agency for Healthcare Research and Quality (AHRQ), and the secretary of health and human services have begun to allocate their Recovery Act funds and coordinate their efforts. The AHRQ has requested proposals for studying the IOMs high-priority research questions that fit within its own priorities,4 and the secretary of health

n engl j med 362;11

nejm.org

march 18, 2010

Downloaded from www.nejm.org on May 12, 2010 . Copyright 2010 Massachusetts Medical Society. All rights reserved.