outcome measures may be in a normal range even if they are not fit. Methods of setting cut-points based on evidence in adults assume that a fit child will likely become a fit adult. While there is some evidence to support this assumption for certain fitness components (see, e.g., Beunen et al., 1992; Campbell et al., 2001; Malina, 1996, 2001), more research, especially longitudinal studies, is needed to confirm this assumption.
When a relationship can be confirmed only in adults, there are two methods for estimating the cut-points for health-related fitness tests in youth—the relative position and the panel-driven methods. With the relative position method, the percentile of adults considered to be at risk based on their performance on a fitness test is taken as the fitness standard in youth. For example, the lowest 20th percentile for performance on a cardiorespiratory endurance test item could be selected based on the demonstration by Blair and colleagues (1989) that morbidity and mortality are disproportionately elevated in the lowest quintile for performance on a maximal treadmill test in adults. In the panel-driven method, a panel of experts uses the cut-points from adults and all available information (e.g., growth curves and performance characteristics for different ages and genders) to derive the cut-points for youth. For example, the criterion maximal oxygen uptake (VO2max) value in youth could be determined in various ways, ranging from expert opinion to extrapolation from associations between VO2max and health outcomes in adults. The panel-driven method was used to set the Fitnessgram® standards for cardiorespiratory endurance test items (Cureton, 1994; Cureton and Warren, 1990).
Establishing Cut-Points When the Relationship Between a Fitness Test and Health Outcomes Is Not Confirmed in Youth or Adults
While the importance of some fitness components to health has been suggested, the relationship between specific fitness test items and health outcomes may not be confirmed. For example, while the validity and reliability of commonly used tests have generally been well established, evidence for the importance of muscular strength for health in adults is still growing and may be equivocal for some tests, and for youth remains largely unconfirmed. Until these relationships are confirmed, an alternative approach for setting cut-points is to use the comparatively relative position method, in which a percentile established for another measure is borrowed. If the percentile from another test is borrowed, the two tests should be as comparable as possible in their nature (e.g., both require movement of the body) and in the dimension they measure (e.g., upper-body strength). For example, if the cut-points for tests of the cardiorespiratory endurance component derived through a criterion-referenced evaluation procedure were set at about the 20th percentile, the cut-points for tests of the musculo-