The Healthy Fitness Zone Continuum Score as a Measure of Change in Body Mass Index of School-Aged Children and Adolescents,Georgia,2012-2014

Abstract

Objectives:

The FitnessGram Healthy Fitness Zone continuum (HFZc) score reflects the relative difference of a person’s body mass index (BMI) from the established FitnessGram standard. As such, it may provide added utility for public health programming and research on obesity among school-aged children and adolescents. We used the standard BMI Z (BMIz) score and the alternative HFZc score to describe changes in BMI of school-aged children and adolescents in Georgia over time.

Methods:

We compiled 2012-2014 BMI data from the Georgia FitnessGram database. The sample included 162 992 boys and 141 711 girls enrolled in 239 schools from a large urban district in Georgia. We analyzed trends in BMIz and HFZc scores separately for normal-weight, overweight, and obese categories for school-aged children and adolescents using hierarchical linear models.

Results:

From 2012 to 2014, the BMIz score shifted favorably in up to 40.7% (2052/5047) of normal-weight, 51.0% (758/1485) of overweight, and 52.8% (5430/10 279) of obese students. We also found favorable shifts in HFZc score in up to 69.8% (105 831/151 739) of normal-weight, 78.3% (3605/4603) of overweight, and 80.8% (8305/10 279) of obese students.

Conclusions:

Compared with the BMIz score, the HFZc score may be a better indicator of favorable changes in BMI over time among school-aged children and adolescents with different baseline BMI levels, making it potentially valuable for use in individualized assessments, school programs, obesity research, and public health curriculum and policy development.

Keywords

BMI FitnessGram Healthy Fitness Zone obesity tracking

The Cooper Institute developed the FitnessGram program to provide schools with tools and resources to facilitate the tracking and reporting of health-related fitness.¹ FitnessGram has been used for 30 years and has become the most widely used physical fitness assessment, education, and reporting tool for school-aged children and adolescents in the world and the default national fitness battery for school-aged children and adolescents in the United States. The FitnessGram battery includes field tests of health-related fitness including aerobic capacity, muscular strength, muscular endurance, flexibility, and body composition. Each fitness indicator is evaluated using age- and sex-specific criterion-referenced standards; standards that were developed against a set health criteria. For example, the standards for body mass index (BMI) and aerobic capacity were developed using nationally representative data and were established for detecting potential risk for metabolic syndrome.^2

–5 For each measure of fitness, students who meet the standards are classified as being in the Healthy Fitness Zone (HFZ), whereas students who fall below the standards are classified as being in either the Needs Improvement Zone (NIZ) or the Needs Improvement–Health Risk (NIHR) Zone. FitnessGram offers an online software tool that enables school personnel to provide individualized feedback to students and to monitor school- and district-level patterns and trends over time.

The recommended FitnessGram measure for body composition is based on BMI. Several studies have shown that excess BMI has important negative implications for young people’s health.^6

–10 The FitnessGram standards for BMI were originally established empirically using data from the National Health and Nutrition Examination Survey^2,3 and fell within 1 to 2 percentiles of the more widely used BMI cutoffs defined by the Centers for Disease Control and Prevention (CDC).¹¹ Because CDC and FitnessGram BMI standards were so similar and showed similar predictive utility for health risk,¹² in 2013, the FitnessGram standards were subsequently aligned with the CDC thresholds. As a result, the FitnessGram BMI categories (HFZ, NIZ, and NIHR Zone) now correspond exactly with the CDC thresholds (ie, underweight for those <5th percentile, normal weight for those between the 5th and <85th percentile, overweight for those between the 85th and <95th percentile, and obese for those ≥95th percentile). This standardization provided consistency between FitnessGram BMI values and BMI values reported by physicians and allowed childhood BMI data obtained through schools to be used for state and national surveillance of health-related fitness among school-aged children and adolescents.^13,14

Although these BMI classifications are easy to interpret and offer a simple way to assess individuals or schools at a single point in time, they are not as useful for longitudinal analyses and other research applications. Another limitation of these classifications is that they divide a continuous scale into several distinct categories, which generally results in a substantial loss of statistical power and an attenuation of the magnitude of associations with other variables.¹⁵ Because of these limitations, some obesity researchers have used either CDC BMI percentiles (without the CDC categories) or CDC BMI Z (BMIz) scores.¹⁶ However, BMIz scores, which are based on the number of standard deviations (SDs) that an individual’s BMI is above or below the average age- and sex-specific BMI, are particularly problematic for longitudinal use because of the variable absolute intervals between percentiles. Alternative approaches, such as the Percent OverBMI¹⁷ and BMI Sympercent¹⁸ methods, have been shown to be more sensitive than are BMIz scores to BMI changes over time, but they provide little clinical or diagnostic information, are not widely used in clinical practice, and are difficult for individuals and medical providers to understand.

To address the need for a continuous BMI index that is intuitive, clinically useful, and suitable for use in longitudinal surveillance, we propose the Healthy Fitness Zone Continuum (HFZc) score. The HFZc score reflects the relative difference (as a percentage) of a student’s BMI from the age- and sex-specific HFZ BMI threshold for good health. The HFZc score is based conceptually on the same principles as other BMI indices,^17,18 but its use has not been reported in the literature.

In this study, we provide a descriptive evaluation of changes in BMI over time, using both BMIz and HFZc scores. We applied these scores to longitudinal data from a large sample of school-aged children and adolescents taken from the Georgia FitnessGram database. The use of the FitnessGram program in Georgia grew from a mandate passed in 2009, the Georgia Student Health and Physical Education (SHAPE) Act, which required all public school students in grades 1 through 12 enrolled in physical education to participate, starting with the 2011-2012 school year.¹⁹ Our hypothesis was that the HFZc score would be a sensitive and robust indicator of favorable progress in BMI over time for school-aged children and adolescents, making it potentially valuable for various public health uses.

Methods

Through an active collaboration with the Cooper Institute, we obtained de-identified data from the FitnessGram database for school-aged children and adolescents in Georgia. We used data from students in grades 1 through 12 (ie, aged 6-18) from a large urban school district in Georgia for the study. The Iowa State University Institutional Review Board approved this study.

Measurement of BMI

Trained physical education teachers measured the heights and weights of students to calculate BMI, using the standard formula, BMI = weight (kg)/height (m²), and standard procedures as recommended in the FitnessGram Test Administration Manual.²⁰ The teachers collected and entered the data into the FitnessGram online platform using individualized identification numbers for each student, which allowed tracking of BMI over time.

Data Processing

We excluded from analysis records that were missing data on age, sex, height, or weight, and we excluded students aged ≤5 or >18. We further screened data for quality (eg, identifying abnormal weight or height values) by using standardized procedures described elsewhere.²¹ We combined students from grades 1 through 5 into the elementary school group, grades 6 through 8 into the middle school group, and grades 9 through 12 into the high school group. We used longitudinal, individually tracked data on BMI from students during the 2012, 2013, and 2014 school years. We calculated BMI scores using both the CDC BMIz scoring method and our proposed HFZc scoring method. The details of each method are described later.

BMIz Scoring Method

We converted BMI to a BMIz score for each student using SAS macro.²¹ This score reflected the number of SDs that a student’s BMI was above or below the average BMI of students of the same age and sex.^11,22

HFZc Scoring Method

The HFZc score is computed using a method similar to that used for the Percent OverBMI score, an alternative BMI index that reflects a student’s BMI percentage above the 50th percentile BMI for students of the same age and sex.¹⁷ Instead, the HFZc score incorporates the FitnessGram HFZ BMI threshold in the calculation; the score reflects the relative difference of a student’s BMI from the age- and sex-specific HFZ thresholds for BMI. It is calculated as follows:

\begin{array}{l} HFZc score = (BMI−HFZ BMI threshold) \\ \times 100 HFZ BMI threshold \end{array}

The HFZc score is reported as the percentage above or below the recommended HFZ BMI threshold for good health. For example, an HFZc score of +10% indicates that a student’s BMI score is 10% above the HFZ BMI standard, whereas a score of –10% indicates that the student’s BMI is within the HFZ BMI threshold by a 10% margin.

Statistical Methods

We stratified the baseline (2012) ages, heights, weights, BMIz scores, and HFZc scores of students by sex (boys or girls) and school group (elementary, middle, or high school), and we presented the results as means and SDs. We divided the baseline BMI levels of students into the 3 groups used by CDC (normal weight, overweight, obese), stratified them by sex and school group, and reported the results as proportions of students in each group. We also divided the students into the 3 FitnessGram fitness zones (ie, HFZ, NIZ, and NIHR Zone) for 2012, 2013, and 2014, and we reported the results as proportions of students in each zone for each year.

We first examined normative distributions among students using hierarchical linear models to obtain adjusted mean changes of both BMIz and HFZc scores. We applied the hierarchical models separately to each school group (elementary, middle, and high school), and we used the models to assess linear changes in BMIz and HFZc scores from 2012 to 2013 and from 2012 to 2014. We included baseline BMI group (normal weight, overweight, obese) and sex as covariates in the models. However, because the trends over time among boys and girls were similar (unpublished results), and for the sake of simplicity, we stratified only the changes in BMIz and HFZc scores over time by baseline BMI group (while controlling for sex). We also accounted for the possible correlation among results due to clustered school data (ie, the cluster effect) in all of the models used.

We then performed criterion-referenced evaluations of BMIz and HFZc scores over time to compare the ability of the 2 scores to capture favorable shifts in BMI over time. For the BMIz score, we defined favorable shifts as reductions in the score of ≥0.1. For the HFZc score, we defined favorable shifts as reductions over time in the scores of ≥0.1%. As we did for the normative evaluations, we described the results separately for each school group, and we stratified the changes in BMIz and HFZc scores over time by the baseline BMI groups.

We considered P < .05 to be significant. We analyzed all data using SAS version 9.2.²³

Results

A total of 1 194 212 (for 2012), 1 582 295 (for 2013), and 1 517 217 (for 2014) Georgia student FitnessGram records were available for review. Of these, 162 992 boys and 141 711 girls had valid data for all 3 years and were enrolled in the 239 schools that were part of the large urban district in Georgia that we studied.

For both sexes and all school groups, most of the baseline mean BMIz scores were 0.8 (indicating that the mean BMI in the population was 0.8 SDs above the mean BMI of all students of the same age and sex), and we found no substantial differences by school groups or sex (Table 1). On the other hand, most of the baseline mean HFZc scores were negative (indicating that the mean BMI in the population was already within the HFZ). The mean HFZc score ranged from 0.3% for elementary school boys (the only group not within the HFZ) to –5.4% for high school girls.

Table 1.

Baseline age, height, weight, BMIz score,^a HFZc score,^b and BMI group^c of school-aged children and adolescents (n = 304 703), by school group and sex, Georgia, 2012

	Elementary School (Grades 1-5)		Middle School (Grades 6-8)		High School (Grades 9-12)
Variables	Boys (n = 135 923)	Girls (n = 127 053)	Boys (n = 20 043)	Girls (n = 12 831)	Boys (n = 7026)	Girls (n = 1827)
Age, y, mean ± SD	8.2 ± 1.4	8.1 ± 1.3	12.5 ± 1.0	12.3 ± 0.9	15.1 ± 0.8	15.0 ± 0.8
Height, cm, mean ± SD	132.9 ± 9.8	132.3 ± 10.2	159.3 ± 10.6	156.5 ± 7.8	174.1 ± 7.9	162.8 ± 7.4
Weight, kg, mean ± SD	33.4 ± 10.5	33.4 ± 10.7	55.9 ± 16.0	54.4 ± 14.3	72.0 ± 15.6	61.1 ± 12.8

BMIz score,^a SD, mean ± SD	0.8 ± 1.2	0.8 ± 1.1	0.8 ± 1.1	0.8 ± 1.0	0.8 ± 0.9	0.6 ± 0.8
HFZc score,^b %, mean ± SD	0.3 ± 20.2	–0.6 ± 20.6	–0.4 ± 22.4	–1.3 ± 22.2	–1.0 ± 18.9	–5.4 ± 17.8

BMI group^c
Normal weight, No. (%)	78 303 (57.6)	73 436 (57.8)	10 944 (54.6)	7048 (54.9)	3853 (54.8)	1194 (65.4)
Overweight, No. (%)	16 007 (11.8)	10 336 (8.1)	3020 (15.1)	1583 (12.3)	1251 (17.8)	234 (12.8)
Obese, No. (%)	41 613 (30.6)	43 281 (34.1)	6079 (30.3)	4200 (32.7)	1922 (27.4)	399 (21.8)

Abbreviations: BMI, body mass index; BMIz, Body Mass Index Z; HFZc, Healthy Fitness Zone Continuum; SD, standard deviation.

^aThe BMIz score is expressed as the number of SDs that a student’s BMI is above or below the average BMI of students of the same age and sex.

^bThe HFZc score is expressed as a percentage and reflects the relative difference of a student’s BMI from the age- and sex-specific FitnessGram Healthy Fitness Zone thresholds for BMI.¹⁴

^cCenters for Disease Control and Prevention BMI groups: underweight (<5th percentile), normal weight (5th to <85th percentile), overweight (85th to <95th percentile), obese (≥95th percentile).

Of the 304 703 students in our study, the proportion who were in the HFZ increased by about 1 percentage point from 2012 to 2013 and remained unchanged from 2013 to 2014. When stratified by school group, the proportions of middle and high school students in the HFZ increased steadily from 2012 to 2014 (from 54.7% to 57.3% among middle school students and from 57.0% to 59.2% among high school students), but the proportion of elementary school students in the HFZ declined slightly from 57.7% in 2012 to 57.2% in 2013 and substantially to 54.7% in 2014 (Figure 1).

Figure 1.

Proportion of school-aged children and adolescents (n = 304 703) in FitnessGram body mass index (BMI) zones, by school group and year, Georgia, 2012-2014. FitnessGram Healthy Fitness Zones for BMI include Needs Improvement–Health Risk Zone (NIHR), Needs Improvement Zone (NIZ), and Healthy Fitness Zone (HFZ).¹⁴

Normative Comparisons

The mean BMIz scores ± standard error in 2012, 2013, and 2014 were 1.1 ± 0.0, 1.0 ± 0.0, and 1.1 ± 0.0 for elementary school students; 1.1 ± 0.0, 0.9 ± 0.0, and 0.9 ± 0.0 for middle school students; and 1.2 ± 0.0, 1.1 ± 0.0, and 1.1 ± 0.0 for high school students, respectively. Hierarchical linear regression analysis showed that mean BMIz scores for elementary school students changed by –0.1 ± 0.0 from 2012 to 2013 (P < .001) but only by –0.02 ± 0.0 (P < .001) from 2012 to 2014. Mean BMIz scores for middle school and high school students changed identically, by –0.1 ± 0.0 (P < .001) from 2012 to 2013 and by –0.1 ± 0.0 (P < .001) from 2012 to 2014.

Favorable shifts in mean BMIz score during the 3-year period were noted among both the overweight and obese groups and across all school groups but not among the normal-weight group (Figure 2). The most substantial favorable shift in mean BMIz score occurred in obese middle school students from 2012 to 2013 (–0.2 ± 0.0, P < .001) and from 2012 to 2014 (–0.2 ± 0.0, P < .001). In the normal-weight group, a subtle favorable shift in mean BMIz score occurred only among high school students and only from 2012 to 2013 (0.05 ± 0.0, P = .008). Otherwise, most mean BMIz score changes among students in the normal-weight group were unfavorable, and the change for high school students from 2012 to 2014 was not significant (0.04 ± 0.0, P = .158).

Figure 2.

Trends of mean Body Mass Index Z (BMIz) and Healthy Fitness Zone Continuum (HFZc) scores of school-aged children and adolescents (n = 304 703), by baseline body mass index (BMI) group and school group, Georgia, 2012-2014. The BMIz score is expressed as the number of standard deviations (SDs) that a student’s BMI is above or below the average BMI of students of the same age and sex, and it represents a favorable trend when the score declines. The HFZc score, expressed as a percentage, reflects the relative difference of a student’s BMI from the age- and sex-specific FitnessGram Healthy Fitness Zone thresholds for BMI,¹⁴ and it represents a favorable trend when the score declines. Centers for Disease Control and Prevention BMI groups: underweight (<5th percentile), normal weight (5th to <85th percentile), overweight (85th to <95th percentile), obese (≥95th percentile).

The mean HFZc scores in 2012, 2013, and 2014 were 5.7% ± 0.1%, 5.4% ± 0.1%, and 6.5% ± 0.1% for elementary school students; 5.0% ± 0.2%, 3.6% ± 0.2%, and 3.4% ± 0.2% for middle school students; and 6.2% ± 0.3%, 5.5% ± 0.3%, and 6.2% ± 0.3% for high school students, respectively. Hierarchical linear regression showed different HFZc change trends for each school group. Mean HFZc scores for elementary school students changed favorably by –0.3% ± 0.0% (P < .001) from 2012 to 2013 but unfavorably by 0.8% ± 0.1% (P < .001) from 2012 to 2014. Mean HFZc scores for students in middle school changed favorably from both 2012 to 2013 (–1.4% ± 0.1%, P < .001) and 2012 to 2014 (–1.6% ± 0.1%, P < .001). Mean HFZc scores for high school students changed favorably by –0.8% ± 0.2% from 2012 to 2013 (P < .001) but were unchanged from 2012 to 2014 (P = .998) (Figure 2).

We noted unfavorable shifts in mean HFZc score during the 3-year period among students in the normal-weight group, whereas mean HFZc scores shifted either favorably or not at all in the overweight and obese groups, depending on the school group. Similar to the patterns for BMIz scores, favorable changes in mean HFZc score were of greatest magnitude among the middle and high school obese student groups, from 2012 to 2013 (middle school: –4.3% ± 0.2%, P < .05; high school: –3.3% ± 0.3%, P < .001) and from 2012 to 2014 (middle school: –5.7% ± 0.2%, P < .001; high school: –2.9% ± 0.3%, P < .001) (Figure 2).

Criterion-Referenced Comparisons

The proportion of students with favorable shifts in BMIz score varied by BMI baseline group (ie, normal weight vs overweight vs obese). We found that 35.0% (6291/17 992) and 40.7% (2052/5047) of normal-weight students had favorable shifts in their BMIz score during the 3-year period. However, the proportions of students with favorable shifts in BMIz score were highest in the overweight (range, 43.4% [11 422/26 343] to 51.0% [758/1485]) and obese (range, 46.5% [39 469/84 894] to 52.8% [5430/10 279]) groups (Table 2).

Table 2.

School-aged children and adolescents (n = 304 703) with favorable shifts in BMIz^a and HFZc^b scores, by baseline BMI group^c and school group, Georgia, 2013-2014

	Elementary School (n = 262 976), No. (%)		Middle School (n = 32 874), No. (%)		High School (n = 8853), No. (%)
Variable	2013	2014	2013	2014	2013	2014
BMIz score^b
Normal weight^d	60 556 (39.9)	57 700 (38.0)	6388 (35.5)	6291 (35.0)	1888 (37.4)	2052 (40.7)
Overweight^e	11 827 (44.9)	11 422 (43.4)	2217 (48.2)	2313 (50.3)	708 (47.7)	758 (51.0)
Obese^f	37 307 (44.0)	39 469 (46.5)	4713 (45.9)	5430 (52.8)	988 (42.6)	1106 (47.7)
HFZc score³
Normal weight^d	105 831 (69.8)	101 630 (67.0)	12 416 (69.0)	11 155 (62.0)	3214 (63.7)	2803 (55.5)
Overweight^e	17 657 (67.0)	15 897 (60.4)	3605 (78.3)	3422 (74.3)	1101 (74.1)	1012 (68.2)
Obese^f	54 358 (64.0)	47 661 (56.1)	8305 (80.8)	8109 (78.9)	1773 (76.4)	1679 (72.3)

Abbreviations: BMI, body mass index; BMIz, Body Mass Index Z; HFZc, Healthy Fitness Zone Continuum.

^aThe BMIz score is the number of standard deviations that a student’s BMI is above or below the average BMI of students of the same age and sex.

^bThe HFZc score is the relative percentage difference of a student’s BMI from the age- and sex-specific FitnessGram Healthy Fitness Zone thresholds for BMI.¹⁴

^cCenters for Disease Control and Prevention BMI: underweight (<5th percentile), normal weight (5th to <85th percentile), overweight (85th to <95th percentile), obese (≥95th percentile).

^dNormal-weight populations: 151 739 in elementary school, 17 992 in middle school, and 5047 in high school.

^eOverweight populations: 26 343 in elementary school, 4603 in middle school, and 1485 in high school.

^fObese populations: 84 894 in elementary school, 10 279 in middle school, and 2321 in high school.

Favorable shifts in HFZc scores were more dramatic than were those in BMIz scores. Most normal-weight students (range, 55.5% [2803/5047] to 69.8% [105 831/151 739]) had favorable HFZc score shifts, and the proportion of students with favorable shifts was even higher for the overweight (range, 60.3% [15 897/26 343] to 74.3% [3422/4603]) and obese (range, 56.1% [47 661/84 894] to 78.9% [8109/10 279]) groups (Table 2).

Discussion

The HFZc score is a new alternative measure of BMI that reflects the relative difference of a person’s BMI from the age- and sex-specific FitnessGram HFZ BMI thresholds for good health. This study compared the utility of using this HFZc score and the more traditional indicator (BMIz score) to evaluate changes in BMI of school-aged children and adolescents over time. Using data available through the statewide FitnessGram tracking system in Georgia, we analyzed trends of change for both indicators between 2012 and 2014.

We found that although the general patterns of change of the BMIz and HFZc scores in the study population were similar, the scale of these changes was more pronounced and the changes were more easily detected when using the HFZc score than when using the BMIz score. These differences were most evident in our criterion-referenced comparison results. In our study, favorable shifts in BMI in most of the school groups and baseline BMI groups were identified almost twice as often when using the HFZc score as they were when using the BMIz score. In a supplemental analysis that used a more precise definition of change for the BMIz score (ie, a reduction of ≥0.01 instead of ≥0.1), while maintaining our definition of change for the HFZc score, we still found substantial differences between the 2 indicators (data not reported). This finding suggests that, when compared with other commonly used indicators such as the BMIz score, the HFZc score may provide a more sensitive and robust indicator of favorable progress with BMI over time, making it potentially useful not only for individual students but also for schools, researchers, and public health personnel.

We found that BMIz scores lacked sensitivity, particularly when trying to detect small shifts in the BMI of those students with high baseline BMI levels (ie, overweight or obese). For example, when shifts were favorable, the mean BMIz scores for most groups of students improved by about 0.1 SDs. This shift was the equivalent of a 4-percentile improvement for students with a BMIz score of 1.0 (1 SD above the average BMI) but only equivalent to a 1-percentile improvement for students with a BMIz score of 2.0 (2 SDs above the average BMI). As such, by using the BMIz score, favorable BMI changes were more muted in heavier students. This limitation of the BMIz score (and of BMI percentiles) has been noted by others.^6,16,17

Along these same lines, when we stratified favorable changes in BMI from 2012 to 2014 by BMI group, we found that HFZc scores had improved in 56.1% to 80.8% of overweight and obese students, whereas BMIz scores had improved in only 43.4% to 52.8% of the same students. These differences in sensitivity to BMI changes between BMIz and HFZc scores suggest that although the degree of change in BMIz scores appears to be muted for students in the overweight and obese groups, differences were more pronounced when we used HFZc scores. Thus, the HFZc score may be able to fill an important gap in public health efforts aimed at tracking obesity profiles among school-aged children and adolescents.²⁴

Published studies tracking BMI changes have tended to use measurements of obesity or overweight prevalence or other indices, such as BMIz scores.^25

–28 The use of BMI category prevalence (ie, percentage normal weight, overweight, or obese) is particularly limited by the ability to detect only transitions between (but not within) body composition categories. For example, it is possible that a student would remain categorized as overweight despite achieving a substantial reduction in BMI. Furthermore, as noted previously, the lack of sensitivity of indices such as the BMIz score to changes in BMI may also substantially limit their ability to identify either favorable or unfavorable changes. Diminished sensitivity in detecting BMI changes could certainly adversely affect the monitoring of responses to public health efforts aimed at reducing childhood obesity.

Our proposed HFZc score provides a potentially better alternative. The HFZc score is continuous and reflects the relative difference from an established criterion-referenced standard for health. It reflects the number of percentage points that a young person’s BMI is above or below the recommended age- and sex-specific BMI for health (which corresponds to both the CDC 85th percentile BMI and the FitnessGram HFZ). An advantage of criterion-referenced standards over percentile norms is that they are based on levels of fitness needed to reduce health risks. In addition, the HFZc score is purposely designed to be sensitive to small changes in BMI over time, while also accounting for normal growth and maturation. Furthermore, unlike some other indices, the HFZc score appears to be sensitive to changes in BMI among school-aged children and adolescents of varying BMI levels. An additional advantage of the HFZc method is that similar scores can be generated for any of the other FitnessGram fitness measures. Finally, and perhaps most important, the strong link of the HFZc score to health outcomes makes it particularly relevant for personal and public health monitoring.¹

Limitations

This study had several limitations. First, we assessed only BMIz and HFZc scores. Alternative indices, such as Percent OverBMI¹⁶ and BMI Sympercent,¹⁷ have also been shown to be sensitive to changes in BMI and to address the attenuation in BMI change that occurs with high baseline BMI. However, they reflect the relative difference of BMI from an arbitrary level, such as the 50th percentile. Future studies might evaluate and compare all of these alternative BMI indices. Second, our study was not designed to assess the use of HFZc scores to measure responses to individual or public health interventions. Future studies designed for this purpose would be informative. Finally, because our study focused on only one population (school-aged children and adolescents) in a large urban school district in Georgia, they may not be generalizable to all students in other areas of Georgia or the United States.

Conclusions

The HFZc score reflects the difference of an individual’s BMI from an criterion-referenced standard, the age- and sex-specific FitnessGram HFZ threshold for BMI. The HFZc score is sensitive to changes in BMI among school-aged children and adolescents of various baseline BMI levels. When compared with the BMIz score, the HFZc score may be a more sensitive and robust indicator of favorable progress with BMI over time, making it potentially valuable for use in individualized assessments, school programs, obesity research, and public health policy and curriculum development.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) declared no funding with respect to the research, authorship, and/or publication of this article.

References

Plowman

Sterling

Corbin

Meredith

Welk

Morrow

Jr . The history of FitnessGram. J Phys Act Health. 2006;3(suppl 2):S5–S20.

Laurson

Eisenmann

Welk

. Development of youth percent body fat standards using receiver operating characteristic curves. Am J Prev Med. 2011;41(4 suppl 2):S93–S99.

Laurson

Eisenmann

Welk

. Body mass index standards based on agreement with health-related body fat. Am J Prev Med. 2011;41(4 suppl 2):S100–S105.

Welk

Laurson

Eisenmann

Cureton

. Development of youth aerobic-capacity standards using receiver operating characteristic curves. Am J Prev Med. 2011;41(4 suppl 2):S111–S116.

Laurson

Saint-Maurice

Karsai

Csányi

. Cross-validation of FitnessGram health-related fitness standards in Hungarian youth. Res Q Exerc Sport. 2015;86(suppl 1):S13–S20.

Lobstein

Baur

Uauy

; IASO International Obesity TaskForce. Obesity in children and young people: a crisis in public health. Obes Rev. 2004;5(suppl 1):4–104.

National Institutes of Health. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults—the evidence report [published erratum appears in Obes Res. 1998;6(6):464]. Obes Res. 1998;6(suppl 2):51S–209S.

Freedman

Mei

Srinivasan

Berenson

Dietz

. Cardiovascular risk factors and excess adiposity among overweight children and adolescents: the Bogalusa Heart Study. J Pediatr. 2007;150(1):12–17.

Singh

Mulder

Twisk

van Mechelen

Chinapaw

. Tracking of childhood overweight into adulthood: a systematic review of the literature. Obes Rev. 2008;9(5):474–488.

10.

Herman

Craig

Gauvin

Katzmarzyk

. Tracking of obesity and physical activity from childhood to adulthood: the Physical Activity Longitudinal Study. Int J Pediatr Obes. 2009;4(4):281–288.

11.

Kuczmarski

Ogden

, Guo SS, et al. 2000 CDC growth charts for the United States: methods and development. Vital Health Stat 11. 2002;246:1–190.

12.

Laurson

Welk

Eisenmann

. Diagnostic performance of BMI percentiles to identify adolescents with metabolic syndrome. Pediatrics. 2014;133(2):e330–e338.

13.

Bai

Saint-Maurice

Welk

. Fitness trends and disparities among school-aged children in Georgia, 2011-2014. Public Health Rep. 2017;132(suppl 2):39S-47S.

14.

Bai

Saint-Maurice

Welk

Allums-Featherston

Candelaria

Anderson

. Prevalence of youth fitness in the United States: baseline results from the NFL PLAY 60 FitnessGram Partnership Project. J Pediatr. 2015;167(3):662–668.

15.

Ragland

. Dichotomizing continuous outcome variables: dependence of the magnitude of association and statistical power on the cutpoint. Epidemiology. 1992;3(5):434–440.

16.

Barlow

. Expert committee recommendations regarding the prevention, assessment, and treatment of child and adolescent overweight and obesity: summary report. Pediatrics. 2007;120(suppl 4):S164–S192.

17.

Paluch

Epstein

Roemmich

. Comparison of methods to evaluate changes in relative body mass index in pediatric weight control. Am J Hum Biol. 2007;19(4):487–494.

18.

Cole

Faith

Pietrobelli

Heo

. What is the best measure of adiposity change in growing children: BMI, BMI %, BMI z-score or BMI centile? [published erratum appears in Eur J Clin Nutr. 2005;59(6):807]. Eur J Clin Nutr. 2005;59(3):419–425.

19.

Official Code of Georgia 20-2-777 (2009).

20.

Cooper Institute. FitnessGram & ActivityGram: Test Administration Manual. 4th ed. Champaign, IL: Human Kinetics; 2010.

21.

Saint-Maurice

Welk

Bai

Allums-Featherston

. Comparison of data screening methods for evaluating school-level fitness patterns in youth: findings from the NFL PLAY 60 FitnessGram Partnership Project. Open J Prev Med. 2014;4(11):876–886.

22.

Kuczmarski

Ogden

Grummer-Strawn

. CDC growth charts: United States. Adv Data. 2000;314:1–27.

23.

SAS Institute Inc. SAS Version 9.2. Cary, NC: SAS Institute Inc; 2008.

24.

Dietz

Story

Leviton

. Issues and implications of screening, surveillance, and reporting of children’s BMI. Pediatrics. 2009;124(suppl 1):S98–S101.

25.

Boddy

Hackett

Stratton

. Changes in BMI and prevalence of obesity and overweight in children in Liverpool, 1998-2006. Perspect Public Health. 2009;129(3):127–131.

26.

Vignerová

Humeníkova

Brabec

Riedlová

Blâha

. Long-term changes in body weight, BMI, and adiposity rebound among children and adolescents in the Czech Republic. Econ Hum Biol. 2007;5(3):409–425.

27.

Hedley

Ogden

Johnson

Carroll

Curtin

Flegal

. Prevalence of overweight and obesity among US children, adolescents, and adults, 1999-2002. JAMA. 2004;291(23):2847–2850.

28.

Chrzanowska

Koziel

Ulijaszek

. Changes in BMI and the prevalence of overweight and obesity in children and adolescents in Cracow, Poland, 1971-2000. Econ Hum Biol. 2007;5(3):370–378.