Image via Wikipedia

Routine HIV Screening — What Counts in Evidence-Based Policy?

Ronald Bayer, Ph.D., and Gerald M. Oppenheimer, Ph.D., M.P.H.

N Engl J Med 2011; 365:1265-1268October 6, 2011

Article
References: In January 2011, the President's Advisory Council on HIV/AIDS called on the U.S. Preventive Services Task Force (USPSTF) to reconsider its 2005 determination that the scientific evidence did not justify routine testing for human immunodeficiency virus (HIV) in adolescents and adults in the general U.S. population. In July, the Institute of Medicine issued a report calling for annual HIV screening in all sexually active women. These are the most recent episodes in a struggle that began in 2006, when the Centers for Disease Control and Prevention (CDC) recommended routine HIV testing. Normally, USPSTF decisions receive little attention. Sometimes, however — for example, when the Task Force recommended against routine mammography for women in their 40s — there is political fallout, and observers question what kind of evidence should count and who should make those determinations. The battle over HIV screening, though less public and heated than the mammography debate, demonstrates how the same evidence can lead the country's top public health agency and the body charged by the government with providing scientifically grounded recommendations on clinical preventive interventions to reach dramatically different conclusions.
The USPSTF reflects the legacy of British clinician Archie Cochrane, who criticized medicine for failing to systematically weigh clinical and epidemiologic evidence in making health care decisions. Established in 1984 and sponsored by the Agency for Health Care Research and Quality since 1998, the Task Force evaluates clinical preventive services on the basis of systematic reviews of empirical evidence, determining the magnitude of benefits and harms and issuing grades ranging from A, strongly recommended, to D, not recommended. Capturing the Task Force's ethos, Mark Helfand of the Oregon Evidence-based Practice Center said, “It is never rational to say we must do something [now] because it will take too long to get the evidence. Such actions time and again have been wrong.”
In 2005, the Task Force concluded that although “targeted [HIV] screening misses a substantial proportion of HIV positive patients . . . universal screening would result in large numbers of patients screened for each clinical outcome prevented.”1 Regarding the potential effects of identifying many more people who were unaware of being HIV-positive, the Task Force declared, “Despite evidence that knowledge of HIV positive status reduces some high risk behaviors, there is insufficient evidence with which to accurately estimate the effects on transmission.” Given the potential benefits and harms associated with HIV diagnosis, “including fears of rejection, abandonment, verbal abuse and physical assault,” the Task Force gave general HIV screening a C rating, indicating no recommendation for or against. By stating that individual physicians should determine whether to test, the Task Force rejected the notion of making HIV testing routine practice.
In a published exchange encapsulating the ensuing debate about what evidence regarding HIV testing should be required for changing public policy, a group of clinicians with expertise in AIDS wrote that the benefits of expanding routine HIV testing were “unmistakable.”2 Knowledge of infection provided an opportunity to prevent transmission. Claiming that the potential harms of HIV screening were “minimal,” they argued that the risks did not warrant the withholding of testing. Pointing to current practices' failure to meet the public health and clinical challenges posed by HIV, they concluded, “We cannot allow ineffective testing policies to be perpetuated when experience demonstrates that change is required.”
The chair of the Task Force responded by reemphasizing the burdens of testing for individual patients and clinicians and declaring that as many as 11,000 non–high-risk patients would have to be screened over 3 years to prevent one clinical progression to death. Given the limited duration of primary care visits and what he characterized as the very small likelihood that “any one clinician would provide health benefits to an otherwise undetected seropositive patient,” he concluded that “the physician should consider whether this time might be better spent providing other preventive services that carry the potential to improve outcomes for more patients.”
In September 2006, the CDC issued new recommendations for routine screening that differed sharply from those of the Task Force.3 Haunted by the persistent failure to diagnose HIV in the 250,000-plus people in this country who are infected but unidentified (see tableEstimated Number of People in the United States 13 Years of Age or Older Living with HIV Infection and Number and Percentage Whose HIV Infection was Undiagnosed, 2008.), and on the basis of its own data analysis, the CDC concluded that the public health benefit of reduced transmission justified a departure from its own prior guidance. In fact, it asserted that infected people who were ignorant of their status were a public health hazard, contributing disproportionately to HIV's spread. Citing a recent meta-analysis, the CDC found evidence that people who were aware of their HIV infection substantially curtailed their high-risk behaviors.
The crafters of the new recommendations rejected the Task Force analysis, arguing that an analysis of risk and burden focused on each individual screened ignores significant likely benefits at the population level. The question was what course to pursue when evidence did not meet the most exacting scientific standards. Not taking action, permitting an ineffective policy to persist, was in fact an action. As Bernard Branson of the CDC Division of HIV/AIDS Prevention noted, “The Task Force has the luxury of making no recommendation for or against. The CDC does not really have that option.”
Although the USPSTF usually doesn't revisit its determinations for 5 years, considerable pressure and claims about new evidence led to a “focused update” in April 2007.4 Yet it didn't modify the earlier conclusion.
After examining the nature of the evidence used by the CDC and what it considered methodologic flaws in relevant studies, the USPSTF concluded that the CDC had overstated routine HIV testing's potential effects on transmission. The Task Force also acknowledged the difficulty of obtaining the evidence it deemed crucial: “Studies directly linking screening to decreased rates of transmission would require very large populations with long duration of follow up and are difficult to perform. Not surprisingly no such studies have been published since [the] 2005 evidence synthesis was completed.”
And there the issue remained, despite a 2009 analysis by the American College of Physicians and the HIV Medicine Association supporting the CDC recommendation for routine testing.5 Ironically, what rekindled consideration in 2010 was not new evidence, but the passage of the Affordable Care Act (ACA), which requires health plans to cover preventive services, without charge to patients, if they've received an A or B rating from the USPSTF. Routine HIV screening would thus remain uncovered. In response to that ACA requirement, the Institute of Medicine Committee on Preventive Series for Women recommended that annual HIV screening be included among covered services. Furthermore, 2009 recommendations by the U.S. Public Health Service that antiretroviral therapy be initiated earlier underscored the clinical importance of identifying persons with asymptomatic HIV infection, and in 2011, the HIV Prevention Trials Network's HPTN052 trial demonstrated the powerful effect of therapy on transmission rates.
The Task Force is revisiting its 2005 determination and may reach a decision by the end of 2011. But this saga underscores a broader matter of critical importance to anyone committed to evidence-based practice in medicine and public health. What is from one perspective methodologic rigor may be viewed by others with public health responsibilities as methodologic zealotry. An emphasis on the importance of contextual flexibility is to some an invitation to perilous inconsistency.
Public health practitioners confront the challenge of choosing policies to pursue in the face of uncertainty. Such decisions must take account of the severity of the potential harms from action or inaction. Certainly, such determinations require a strong evidence base, but they will be different for those whose first priority is scientific exactitude and those who focus on making policy in the context of public health, where a precautionary perspective mandates action to prevent foreseeable harms, even when the evidence does not meet the most rigorous demands of science.
Disclosure forms provided by the authors are available with the full text of this article at NEJM.org.
This article (10.1056/NEJMp1108657) was published on September 21, 2011, at NEJM.org.
SOURCE INFORMATION
From the Center for the History and Ethics of Public Health, Department of Sociomedical Sciences, Mailman School of Public Health, Columbia University, New York.