After reading the article by Montalto on implementing the Guidelines for Adolescent Preventive Services (GAPS) in this issue of American Family Physician,¹ the family physician is likely to ask: Should we begin implementing these guidelines in our practice? The answer: Not yet. The guidelines should be reviewed; the issues covered should promote thoughtful discussion about the relevance of the guidelines to the adolescents in our practice, to their families and to our local communities. But before embarking on the comprehensive and time-consuming task of implementing these recommendations, physicians should have further evidence that these guidelines make a difference to adolescent patients.

Why the caution about adopting GAPS? As family physicians, we should adopt guidelines that are explicitly evidence-based. For GAPS, the quality of evidence is not provided. Demonstration of improved health outcomes is necessary before we proceed with wide implementation of any preventive health care guidelines. This is especially important because guidelines can become measures of quality of care. Guidelines based primarily on expert opinion may not be appropriate in a typical primary care practice. Family physicians need to make decisions about how much of our time and other resources we want to invest in following GAPS or any other practice guideline based on the needs of our patients.

Where did GAPS come from and what quality of evidence supports the recommendations? The American Medical Association's Department of Adolescent Health convened a scientific advisory board of 11 individuals with a special interest in adolescents to create GAPS.² Department staff conducted a literature review and gathered evidence for all kinds of preventive interventions. The advisory board was later expanded to include representatives of various professional groups targeted by these guidelines, including a representative from the American Academy of Family Physicians (AAFP). Established guidelines affecting adolescents, such as for immunizations or lipid screening, were simply incorporated into GAPS, regardless of whether the guideline was evidence-based. It is not clear how the evidence was used in developing GAPS recommendations. No reference to the quality of evidence for the recommendations is provided.³ Most of the evidence presented that suggests interventions in adolescents are effective comes from community studies and not from typical practices.

Improving preventive services in adolescents is an important goal. If GAPS recommendations seem laudable, why does it matter whether GAPS are evidence-based? First, our time and resources are limited; it is increasingly important that we measure the usefulness of any intervention in producing outcomes that matter to patients.⁴ Every new guideline must be scrutinized through this lens. Adopting time-consuming strategies, such as the annual preventive health visit recommended in GAPS for every adolescent, will displace other activities. Is such a displacement warranted without evidence that it makes a difference to patients? Second, once codified and accepted, a guideline may become a community standard of care. If measures of quality of care include compliance with GAPS, even more practice-wide effort will go into ensuring that the guidelines are followed, despite inadequate evidence of improved health outcomes for adolescents.

Most of what we do in our therapeutic practice is not well supported by evidence and yet we proceed with our best judgment based on what we know. However, prevention guidelines such as GAPS, designed for healthy patients, impose special obligations on us. They need to be supported by high-quality evidence that they improve on the asymptomatic state. Prevention strategies must provide clear benefit to patients before they are widely adopted. Whether from double-blind randomized controlled trials or from expert opinion, the quality of evidence that underlies a recommendation should be provided with any guideline. The U.S. Preventive Services Task Force (USP-STF) guidelines for individuals ages 11 to 24 years have this annotation for each of their health-risk counseling recommendations: “The ability of clinician counseling to influence this behavior is unproven.”⁵ This information allows the physician to recognize gaps in knowledge and weigh the usefulness of new information as it becomes available. The AAFP Policy Recommendations for Periodic Health Examinations, which were based on the USPSTF guidelines, can be found on the AAFP Web site (https://www.aafp.org/exam/).

As family physicians, we need to be particularly careful with recommendations based solely on expert opinion. Our subspecialty colleagues often see a different spectrum of illness than we do, and their expertise may have limited value in our settings. Will the adolescents who come for their recommended annual health check-up in our practices be the adolescents at risk for all of the unhealthy behaviors discussed in GAPS? In my practice, the patients who present regularly for preventive services represent a different population from those who come only for acute problems. For instance, most adolescents requesting contraception in my practice come in at the request of a parent, usually the mother. In more conflicted or troubled situations, the adolescent is likely to seek care from a more anonymous source like an adolescent clinic or health department facility. We all hope that the interventions shown to reduce high-risk behaviors in community settings will be as useful in a typical medical practice, but we do not yet have evidence to support this.

The GAPS recommendations are a comprehensive reminder of all of the issues that face our adolescent patients and will be useful in framing the care of patients in this age group. However, comprehensive adoption of the guidelines in our practices should await clearer evidence of benefits to our adolescent patients.