The U.S. Preventive Services Task Force (USPSTF), a nonfederal, independent panel of scientists with notable experience in primary care and evidence-based medicine, is convened and supported by the Agency for Health-care Research and Quality and charged by Congress to develop evidence-based recommendations for the health care community. These recommendations generally are aimed at primary care physicians, who deliver a range of preventive services to asymptomatic persons in a typical ambulatory practice setting.
In 2004, the USPSTF issued two recommendations1 on screening for hepatitis C virus (HCV) infection: a recommendation against routine screening for HCV infection in asymptomatic adults who are not at increased risk (grade D recommendation), and a finding of insufficient evidence to recommend for or against routine screening for HCV infection in adults at high risk (grade I recommendation). The latter recommendation has elicited some controversy, and it may be useful to clarify the reasoning behind the recommendation.
The USPSTF examines the evidence for the impact of screening and associated interventions on both favorable and adverse health outcomes, then weighs the benefits and harms to arrive at a net health benefit at the population level to assist the physician who makes decisions at the individual level. Recommendations with A, B, C, and D grades reflect net health benefits that are (respectively) substantial, moderate, small, and none. Grade I reflects substantial gaps in available evidence such that the USPSTF is unable to assess the net health benefit; therefore, it is a call for further research to clarify the net health benefit.2
When the natural history of a disease is known and the relationship of intermediate outcome measures (e.g., biochemical marker levels, disease stage, viral titers) and health outcome measures (e.g., morbidity and mortality rates) is well established, improvement in intermediate outcomes is a valid surrogate measure for improvement in health outcomes. However, the natural history of HCV infection is not well described in the literature. The great majority of persons infected with HCV apparently do not progress to cirrhosis, with only 10 to 20 percent of infected persons developing cirrhosis after 20 to 30 years of infection.3,4 Currently, no method can reliably predict which patients will progress to cirrhosis. In the absence of the ability to target therapy to patients who need it, treating all HCV-infected persons would be worthwhile only if the treatment has proven benefit in reducing or preventing disease progression to cirrhosis or cancer, and if the treatment has minimal or no adverse effects. In addition, the maximal potential public health benefit would be reduced if all HCV-infected patients do not receive or respond to treatment. Factors influencing treatment eligibility and response rates include severity of liver damage, active alcohol abuse, illicit drug use, and other serious physical or psychologic comorbidities.
A review3 of the available evidence on HCV infection shows that only 30 to 40 percent of infected persons who are referred for treatment are eligible to receive it, and that only 54 to 60 percent of treated patients have sustained reduced viremia. The duration of this response beyond the length of the current studies (i.e., a few years) is not known. Most importantly, the efficacy of treatment in reducing or preventing disease progression to cirrhosis has not been established. Adverse effects are experienced by 50 to 60 percent of treated patients; these effects are severe enough for up to 22 percent of patients who receive combination therapy with pegylated interferon and ribavirin to discontinue treatment.3 Additional harms are associated with the diagnostic work-up (e.g., complications from liver biopsy) and screening (e.g., psychologic harms such as anxiety, negative effects on partner relationships). The magnitude of harms of screening is unclear.
It is not known whether counseling HCV-infected persons to change their behavior decreases rates of disease transmission or if it leads to improved health outcomes. There is no evidence showing that vaccination against hepatitis A virus or counseling against alcohol use in HCV-infected persons leads to reduced rates of cirrhosis. Although there may be other reasons to advocate HCV testing (e.g., disease surveillance, research, disease management in individual patients), it does not alter the fact that on the basis of proven health benefit to individual patients, the evidence to support screening is insufficient.
Primary care physicians have limited time and resources to deliver preventive services. The USPSTF believes that services with adequate evidence for substantial to moderate net health benefit (grades A and B) should receive the highest priority for delivery. After delivering these services, physicians can decide how to prioritize other interventions. This strategy would be expected to yield the greatest health benefit to individual patients and to the entire population while making the most prudent use of time and resources.