to the editor: I read with interest and some disappointment the recent editorial by Drs. Brown and Kennelly regarding the American Academy of Family Physicians' (AAFP's) recommendations on screening for human immunodeficiency virus (HIV). As the authors state, the U.S. Preventive Services Task Force (USPSTF) recommended in April 2013 that universal screening be performed in all persons 15 to 65 years of age in the United States. This recommendation received an A grade (high certainty that the net benefit is substantial) based on new studies published over the past few years.¹ Previously, the USPSTF gave routine HIV screening a grade C recommendation, and the AAFP maintained a neutral stance.

The American College of Physicians, the American Academy of Pediatrics, and the American Congress of Obstetricians and Gynecologists have endorsed the USPSTF recommendations. Data from the Centers for Disease Control and Prevention (CDC) indicate that almost 50% of teenagers in the United States are sexually active, and they account for a large number of sexually transmitted infections.² I believe the AAFP's concern that false-positive test results can lead to significant emotional distress is unfounded. We deal with false-positive test results in our clinical practices on a regular basis and must provide this information to patients before and after we perform screening tests. Most patients are relieved to hear a result was a false positive; thus, this occurrence should not affect the physician-patient relationship or influence the patient's willingness to undergo future screening tests, as the authors claim.

Regarding testing protocol, the CDC now endorses a new testing algorithm that uses a fourth-generation assay and no longer includes Western blot. This has shortened the turnaround time of results and improved the accuracy of testing.³

Although I would never argue against pretest HIV counseling, the 2006 CDC recommendations on HIV screening purposely removed this requirement, noting that it is a barrier to HIV screening based on time constraints.⁴ Unfortunately, pretest counseling does not appear to alter the risk behaviors of those who test negative for HIV.

I agree that we should continue to address sexuality with our adolescent patients and assess for risks. I disagree that we should screen patients 15 to 17 years of age based on risk factor assessment alone.

in reply: We appreciate Dr. Kirchner's interest in our editorial. Although sexually transmitted infections in adolescents may be common,¹ the prevalence of HIV remains low.²

We found no studies that evaluated the emotional impact of false-positive HIV screening results. However, the negative impact of false-positive results has been shown in other screening tests, such as mammography.^3,4 Women who have false-positive mammograms experience emotional distress, which may affect their willingness to undergo future screening. The emotional and health care burden of false-positive HIV test results is an area for future study, but research on other screening tests suggests the possibility of unintended consequences.

As Dr. Kirchner notes, the CDC recommends using a testing algorithm consisting of a fourth-generation immunoassay followed by an antibody differentiation immunoassay for persons with positive results. This algorithm has a specificity of 99.91%.⁵ Using this specificity and the incidence data from our editorial, the positive predictive value for 15-year-olds is 1.2%; in other words, there are 81 false-positive results for every true positive. Despite improved test characteristics, the performance of this test remains poor because of the low prevalence of the disease. False-positive results occur at a rate of only one in 1,000, but they add up when it is necessary to screen almost 100,000 15-year-olds or 12,500 17-year-olds to find one with true HIV infection.

It was not our intention to imply that conventional HIV pretest written consent is still required or recommended. Ethical medical care requires explaining to patients what tests are being done and why. This process takes less time than historical HIV pretest consent, but still takes time and is poorly understood.⁶ In an era of limited time and resources, we believe that a family physician's time can be better spent on tasks that are more likely to benefit adolescent patients.

Depending on regional prevalence of HIV infection, family physicians have the option to test all 15- to 17-year-old patients in their own practices, but universal screening of this age group should not be prioritized.