1. Reference Population: Standard Normal Curve
50th Percentile
5th
95th
Three factors that are important for assessing a child’s growth are an appropriate growth reference, accurate measurements, and accurate calculation of a child’s age. A growth reference allows comparison of a child’s growth with that of a well-nourished population. Growth measured by weight and height is distributed in the population in a continuous manner and can be represented as a normal or bell shaped curve. For example, children are not just “short” or “tall”; they represent a continuous gradation from shortest to tallest within each age and gender group. Similarly, children are not just lean or fat; they vary continuously in leaness and fatness. The percentiles shown on this bell curve correspond with the percentile curve lines on the BMI-for-age chart as well as the other growth charts. Percentiles are commonly used in clinical settings to indicate where a child fits in the context of the reference population. For example, if a child has a BMI-for-age at the 95 percentile, then 5% of the population of the same age and gender have a higher BMI-for-age or weigh more relative to their height.

2. Indicators of Nutritional Status
<5th percentile
>95th percentile
Head circumference-for-age
Stunting/shortness
length or height-for-age
<5th percentile
Underweight
weight-for-length
BMI-for-age
<5th percentile
The most common measures for monitoring a child’s growth are head circumference, length or height, and body weight.
The most common indices to compare weight and height measurements with reference curves are length or height-for-age, weight-for-age, weight-for-length for infants at birth to 2 years of age, and BMI-for-age for children 2 to 20 years of age. BMI is calculated by dividing weight in kilograms by height in meters squared and it varies with a child’s age.
Although evaluating a child’s growth pattern over time is more important than a single measurement, single measurements can be used to screen children who may be at nutritional risk and need additional assessment.
Head circumference is closely related to brain size and is often used to screen for potential developmental or neurological disabilities among infants at birth to 24 months old. Children with a head circumference less than the 5th percentile or above the 95th percentile have health or developmental risks that need further medical assessment.
Infants and children whose length- or height-for-age is less than the 5th percentile may be short because their parents are short or they may be stunted because of long-term malnutrition, delayed maturation, chronic illness, or genetic disorder.
Underweight defined as weight-for-length or BMI-for-age less than the 5th percentile may be indicative of recent malnutrition, dehydration, or a genetic disorder.
The cutoff of less than the 5th percentile and above the 95th percentile is used only to screen for potential health or nutrition problems and identify children who should receive further medical assessment.

3. Indicators of Nutritional Status
Overweight
Weight-for-length
BMI-for-age
>95th percentile
Risk of overweight
BMI-for-age
85th to 95th percentile
“Overweight” rather than obesity is the term preferred for describing infants or children greater than or equal to the 95th percentile of weight-for-length or BMI-for-age. The 85th percentile is included on the BMI-for-age and the weight-for-stature charts. Expert committees have indicated that children and adolescents aged 2 to 20 years between the 85th and 95th percentiles are at risk of being overweight.
Evaluating a child’s pattern of growth over time is more important than a single measure of size. The pattern of growth is based on periodic measurements which are tracked on a percentile line as a child grows.

4. Reference Data Sets: Birth to 36 Months
For the first time, nationally representative data have been used to construct growth charts for infants from birth to 36 months of age. The new growth charts were developed primarily from anthropometric measurements collected during a series of National Health and Nutrition Examination Surveys, referred to as NHANES I to III, conducted by The National Center for Health Statistics from 1971 to The three surveys that are included in the reference population collected data for children at different ages. Because none of the surveys included data between birth to 2 months of age, supplementary data were incorporated. These data were combined to create the infant growth chart data set.

5. Reference Data Sets: 2 to 20 Years
The reference population used for older children in the 1977 growth charts included cycles 2 and 3 of the National Health Examination Surveys or NHES surveys conducted from 1963 to 1970 and the first NHANES conducted from 1971 to The new CDC growth charts have an improved reference population because of the addition of the NHANES II and III survey data, which were combined with data from previous surveys. Note that for weight-for-age and BMI-for-age, the NHANES III data were not included and this will be described in a moment. The large sample size in these surveys and the pooling of older data added precision for calculation of the outlying percentile estimates, especially the 3rd and 97th percentiles, to better assess children who are growing at the extremes.

6. Exclusions from the Reference Data
VLBW infants (<1500 g) were excluded because they have different growth patterns
NHANES III weight data for 6+ year olds were excluded to avoid an upward shift in weight and BMI-for-age curves
Very low birth weight (VLBW) infants were excluded from the reference data. This decision was based on a literature that shows that VLBW infants grow differently than non-VLBW infants.
There are several growth charts for VLBW infants. The most common limitation is that they are based on old data. The most current and best of these references is the one developed from the Infant Health and Development Program (IHDP) based on 1985 data.
A decision was made to exclude the NHANES III weight data for children 6 years and older to avoid the influence of an increase in body weight that occurred between the previous national surveys and the NHANES III survey. These data would have been reflected as an upward shift in the weight and BMI-for-age curves. Without this exclusion, the 85th and 95th percentile curves would have been higher and fewer children and adolescents would have been classified as overweight or at risk of overweight. A desirable feature of a reference is that it is stable over time.

7. Age-Adjusted Prevalence of Overweight* From NHANES I to III1
Percent
The rationale for excluding weight data for children 6 years and older is illustrated here. There were considerable increases in the prevalence of overweight using BMI-for-age from NHANES III compared with NHANES I and II. For boys and girls ages 6 to 11 and 12 to 17, when compared with NHANES I (purple bar) and II (white bar), there was a significant increase in the prevalence of overweight in the NHANES III data represented by the green bar. As indicated, it is desirable for a reference population to be stable over time.
If the data from NHANES III were included when the curves were being developed, fewer children would be identified as overweight.
Sex and Age Group
*>95th percentile BMI-for-age
1 Troiano et al, Arch Pediatr Adolesc Med 1995: 149:

8. CDC Growth Charts Are for All Racial and Ethnic Groups Combined
The effect of race and ethnicity on growth is controversial
Inadequate sample data for racial- and ethnic- specific charts
Environmental influences appear to contribute to variations in growth but more research is needed
CDC promotes one set of growth charts for all racial and ethnic groups. This decision was based on the lack of clear evidence that the differences in growth among multiracial groups in the United States are genetically determined.
Additionally, racial- and ethnic-specific charts were not recommended because the current reference population lacked sufficient numbers of specific racial and ethnic groups. Although some studies using BMI-for-age to evaluate at risk of overweight and overweight have found differences by ethnic and racial groups, factors that affect differences in growth among racial and ethnic groups, if they truly exist, remain unclear and more research is needed to clarify the issue.
Several studies support the premise that differences in growth among various racial and ethnic groups is largely a result of environmental inadequacies rather than genetics.

9. Age Adjusted Prevalence of Low Height-for-Age by Ethnic Groups, Children Aged 0 to 5 Years1
Percentage
For example, this graph shows the prevalence of low height-for-age or stunting of recently immigrated refugee children from Southeast Asia in the early 1980s represented by the yellow line. By the 1990s, children born later to these Asian families had heights for age almost identical to white children in the United States, represented by the red line. These observations illustrate the effect of environmental factors on growth. Changing socioeconomic status often is associated with improved growth.
Year of Visit
1 Yip and Mei, 1995

10. Breast-Fed vs. Formula-Fed Infants
Mode of infant feeding can influence growth
New charts represent the combined growth patterns of breast-fed and formula-fed infants
Separate charts are not recommended
Working group of the World Health Organization (WHO) is collecting data on infants following the WHO feeding recommendations to develop charts for infants and children through age 5.
Research has shown that the growth patterns of breast-fed infants differ from those of formula-fed infants. Generally, breast-fed infants grow more rapidly in the first 2 months of life. Growth is not as rapid at 3 to 4 months and breast-fed infants continue to grow less rapidly up to 12 months compared with the 1977 reference data, which were based on mainly formula-fed infants. Slower growth is seen in weight-for-age (Standard Deviation [SD]-0.6) than for length (SD-0.3).
The new reference represents the combined growth patterns of both breast- and formula-fed infants. About half the infants born were reported to have been breast-fed and about one-third were breast-fed 3 months or longer. However, because the duration of breast-feeding is generally short in these data sets, even the new reference shows somewhat different patterns of growth than typically are observed in healthy breast-fed infants. The growth in weight-for-age is still slower after 3 months of age when the new reference is used.
Because of insufficient sample size and the complexities of application for partially breast-fed children, separate charts for breast-fed and formula-fed infants were not recommended.
However, the World Health Organization (WHO) is currently collecting data in six countries worldwide to develop a set of international growth charts for infants and children through age 5 years. The charts will be based on the growth of infants fed according to WHO recommendations (breast-fed at least 12 months and complementary food introduced sometime between 4 and 6 months) This is a prescriptive growth reference. Data collection should be completed in Use of this reference among the US population will be evaluated.

11. Disjunction: Smoothed in New Charts
120
120
110
110
Old
New
100
100 90 90
Length/height in cm 80
Length/height in cm 80 70 70 60
In the 1977 charts, the infant and child curves for length-for-age and stature-for-age did not exactly join at the usual junction of 24 to 36 months. This disjunction occurred in part because recumbent length was from the Fels data set from upper- middle-class infants in Ohio. Stature was from the NCHS data sets. When the switch from the length to the height (stature) charts is made in a clinical setting, usually between 24 and 36 months, there appears to be a downward shift in the child's placement on the charts. This could be misinterpreted as inadequate growth, but it is actually an inherent artifact of the charts.
This unusual disjunction does not occur on the new charts because the same reference population of children 2 to 3 years of age was measured for both length and height. The disjunction is minimized but not eliminated because a child’s stature is shorter than the length measurement. The difference between the length and height of the same child is estimated to be 0.8 cm or approximately 1/3 inch; therefore the length and stature curves were constructed 0.8 cm apart as shown here for the new curve. 60 50 50 40 40 6 12 18 24 30 36 42 48 54 60 6 12 18 24 30 36 42 48 54 60
Age in month
Age in month

12. Reference Population for CDC Growth Charts
Racially and ethnically diverse
Birth to 36 months:
nationally representative
Breast-fed and formula-fed infants
In summary, the reference populations for CDC growth charts are:
Racially and ethnically diverse;
Nationally representative for infants at birth to 36 months;
Representative of breast-fed and formula-fed infants.

13. New Reference Curves Compared with Old Curves
New Reference Curves Compared with Old Curves* Prevalence of Nutrition Indicators
Birth to 36 Months
Nutrition Indicator
Change in Prevalence
Stunting/shortness
length-for-age <5th
2% to 4% lower from 6-24 mos
3% to 5% higher
Underweight
weight-for-length <5th
Clinicians have asked about the impact the new reference population will have on the prevalence of nutrition indicators including stunting or shortness, underweight, and overweight. To answer this question, data from NHANES III, the Pediatric Nutrition Surveillance System (PedNSS), and the World Health Organization (WHO) pooled breastfeeding data sets were used to compare the 1977 (old) reference with the revised reference. There are only slight differences in the prevalence rates of shortness, underweight, and overweight with the new reference. Fewer children will be classified as short or stunted, but a few more will be classified as underweight.
Specifically, among infants:
 the prevalence of stunting or shortness, defined as length-for-age less than the 5th percentile, is 2% to 4% lower among infants 6 to 24 months of age;
 Underweight, defined as weight-for-length less than the 5th percentile, is 3% to 5% higher among children at birth to 36 months; and
 Overweight, defined as greater than the 95th percentile, did not change among children at birth to 36 months.
Overweight
weight-for-length >95th
Same
* NHANES III, PedNSS, WHO

14. New Reference Curves Compared with Old Curves
New Reference Curves Compared with Old Curves* Prevalence of Nutrition Indicators
Children 2 to 5 years of age
Change in Prevalence
Nutrition Indicator
Stunting/shortness
height-for-age <5
2% to 3% lower
3% to 5% higher
Underweight** <5
For children 2 to 5 years of age
 the prevalence of stunting or shortness, defined as height-for-age less than the 5th percentile, is 2% to 3% lower;
 Underweight defined as weight-for-length less than the 5th percentile, is 3% to 5% higher; and
 Overweight, defined as greater than or equal to the 95th percentile, is 3% to 5% higher.
Overweight** 95th
3% to 5% higher
* NHANES III, PedNSS, WHO
**BMI-for-age, weight-for-stature

15. New in the CDC Growth Charts
 Charts extend to 20 years
 3rd and 97th percentiles available
 Lower limits of length (45 vs. 49 cm) and height (77 vs. 90 cm) extended
 BMI-for-age charts (2-20 years) added
 85th percentile (at risk of overweight) added
This slide summarizes the changes in the new charts that are clinically significant and include
extension of the charts to 20 years of age;
addition of 3rd and 97th percentiles to the charts;
lowering the scale for length from 49 to 45 cm and for height from 90 to 77 cm. Lowering the scale for height to 77 cm or approximately 30 inches instead of 35.5 inches allows almost all 2-year-old children to be plotted on the weight -for-stature chart;
addition of the BMI-for-age chart has major clinical implications. For the first time there is a reference population for adolescents and children can be screened for being at risk of overweight; and
addition of the 85th percentile to the BMI-for-age chart to identify children and adolescents at risk of overweight.

16. weight (kg)/height (m)2
What Is BMI?
Body mass index (BMI) =
weight (kg)/height (m)2
BMI is an effective screening tool; it is not a diagnostic tool
For children, BMI is age and gender specific, so BMI-for-age is the measure used
Body Mass Index (BMI) is an anthropometric index of weight and height that is defined as body weight in kilograms divided by height in meters squared. BMI is the commonly accepted index for classifying adiposity in adults and it is recommended for use with children and adolescents.
Like weight-for-height, BMI is a screening tool used to identify individuals who are underweight or overweight. BMI is not a diagnostic tool. For example, a child who is relatively heavy may have a high BMI for his or her age or high weight-for-stature. To determine whether the child has excess fat or is overweight, further assessment is needed that might include triceps skinfold measurements, assessments of diet, health, and physical activity.
BMI is gender specific for children. Furthermore, whereas a fixed BMI cutoff is used for adults, for children, because BMI changes substantially as they get older, BMI-for-age is the measure used for ages 2 to 20 years.

17. Advantages of BMI-for-Age
Provides a reference for adolescents that was not previously available
Consistent with adult standards so can be used continuously from 2 years of age to adulthood
Tracks childhood overweight into adulthood
There are several advantages to using BMI-for-age as a screening tool for overweight and underweight. BMI-for-age provides a reference for adolescents that was not previously available. When the 1977 NCHS growth charts were developed, weight-for-height percentiles were provided only for prepubescent girls up to 10 years and for boys up to 11.5 years. Age and stage of sexual maturation are highly related to body fatness. BMI-for-age is the only indicator that allows us to plot a measure of weight and height with age on the same chart. BMI-for-age was not available in the 1977 charts.
Another advantage is that BMI-for-age is the measure that is consistent with the adult index so it can be used continuously from 2 years of age to adulthood.
These characteristics allow us to use the BMI to track body size throughout the life cycle. BMI in childhood is a determinant of adulthood BMI.

18. Tracking BMI-for-Age from Birth to 18 Years with Percent of Overweight Children who Are Obese at Age 251
The tracking of BMI that occurs from childhood to adulthood is clearly shown in data from a study by Robert Whitaker (Children’s Hospital Medical Center in Cincinnati) and his colleagues. They examined the probability of obesity in young adults in relation to the presence or absence of overweight at various times during childhood. For example, in children 10 to 15 years old, 10% of those with BMI-for-age < 85th percentile were obese at age 25 whereas 75% of those with a BMI-for-age > 85th percentile were obese as adults and 80% of those with a BMI-for-age > 95th percentile were obese at age 25. (The sample size for the study was 854.) From this study, it is clear that an overweight child is more likely than a child of normal weight to be obese as an adult.
Other studies have shown this same trend of tracking occurring from childhood to adulthood.
Whitaker et al. NEJM: 1997;337:

19. Advantages of BMI-for-Age
BMI-for-age relates to health risks
Correlates with clinical risk factors for cardiovascular disease including hyperlipidemia, elevated insulin, and high blood pressure
BMI-for-age during pubescence is related to lipid levels and high blood pressure in middle age
Another advantage of using BMI-for-age to screen for overweight or at risk of overweight in children is that it correlates with clinical risk factors for cardiovascular disease including hyperlipidemia, elevated insulin and high blood pressure. Freedman and colleagues used data from the Bogalusa Heart Study and found that approximately 60% of 5 to 10 year-old children who were overweight had at least one biochemical or clinical risk factor for cardiovascular disease such as those just mentioned, and 20% had two or more risk factors.*
We know that risk factors in children become chronic diseases in adults. BMI-for-age during pubescence is related to lipid and lipoprotein levels and blood pressure in middle age.
*Freedman et al., The relation of overweight to cardiovascular risk factors among children and adolescents: the Bogalusa Heart Study. Pediatrics 1999;103:

20. BMI-for-Age Compares Well with
Weight-for-stature measurements
Measures of body fat
BMI-for-age compares well with both weight-for-stature measurements and measures of body fat.
A study completed by researchers at CDC compared the performance of BMI-for-age and weight-for-stature with fatness measured by dual energy x-ray absorptometry, a direct measure of adiposity. NHANES III data were used to test how well BMI-for-age predicts underweight (below 15th percentile) and overweight (>85th percentile) relative to the traditional weight-for-stature in children 2 to 19 years old. Both BMI-for-age and weight-for-stature performed equally well in screening for underweight and overweight among children 3 to 5 years of age. For school-aged children (6 to 11 and 12 to 19 age groups), BMI-for-age was slightly better than weight-for-stature in predicting underweight and overweight. Ratios of weight relative to stature such as BMI-for-age and weight-for-height may be used as indirect measures of overweight that correlate with more direct measures.
BMI-for-age is significantly correlated with subcutaneous and total body fatness in adolescents. It is not a measure of body fatness but rather a proxy for body fat.

21. Why Use BMI-for-Age?
Recommended by expert committees to evaluate overweight
Guidelines for Overweight in Adolescent Preventive
Services (Am J Clin Nutr 1994;59: )
Obesity Evaluation and Treatment: Expert Committee
Recommendations (Pediatrics 1998 Sept;(102)3:e 29)
Assessment of Childhood and Adolescent Obesity:
International Obesity Task Force (Am J Clin Nutr 1999, 70,suppl)
Because of the numerous advantages of using BMI-for-age for assessing overweight in children and adolescents several expert committees and advisory groups have recommended BMI-for-age as the accepted measure. The published references are given here.
In 1994, an expert committee was convened by the Maternal and Child Health Bureau (MCHB), American Academy of Pediatrics and the American Medical Association with support from the CDC, to advise Bright Futures: National Guidelines for Health Supervision of Infants, Children and Adolescents and Guidelines for Adolescent Preventive Services (GAPS). The committee recommended using BMI-for-age to routinely screen for overweight in adolescents.
Subsequently, an expert committee on pediatric obesity was convened by the MCHB to consider recommendations for children. They recommended that BMI-for-age be used to screen children aged 2 years and older for overweight and risk of overweight.
In 1999, a workshop convened by the International Obesity Task Force concluded that BMI is a reasonable measure assessing overweight in children and adolescents and they recommended using percentiles that correspond to a BMI of 25 and 30, respectively, in young adults to identify at risk of overweight and overweight in children and adolescents.

22. Shape of Growth Curves: Weight-for-Stature versus BMI-for-Age10 15 20 25 30 35 24 72
120
168
216
Age (months)
BMI 5 10 15 20 25 30 35 80 90
100
110
120
130
Stature (cm)
Weight (kg)
95th
95th
50th
5th
50th
5th
The shapes of the weight-for-stature and the BMI-for-age growth charts are shown here. As you can see, the shapes differ. The weight-for-stature chart shows how weight increases in relation to stature as a child gets older but is limited to the prepubescent period. The BMI-for-age chart shows age-related changes in growth; we consider weight, stature and age for a child, whereas with the weight-for-stature chart, only weight and height are used.
The BMI-for-age charts begin at 24 months of age because the value of using BMI at ages younger than 2 years is unclear. BMI-for-age values at younger ages have not been associated with adolescent or adult obesity.

23. For Children, BMI Differs by Age
Example: 95th Percentile Tracking
Age BMI
2 yrs
4 yrs
9 yrs
13 yrs
Boys: 2 to 20 years
Because adiposity varies with age and gender, BMI must be age and gender specific. As illustrated here, growth has been established along the 95th percentile with BMI-for-age reaching a minimum at 4 years of age and then increasing.
From age 6 to 19 years, values of BMI-for-age for females exceed those for males.
BMI
BMI

24. Shape of BMI-for-Age Growth Curve: “Adiposity” Rebound (AR)
Example: Early AR
Age (mos) BMI
Boys: 2 to 20 years
BMI
A section of the BMI-for-age chart for boys has been enlarged to show the shape of the curve in more detail. BMI changes substantially with age. After about 1 year of age, BMI-for-age begins to decline and it continues falling during the preschool years until it reaches a minimum around 4 to 6 years of age. After 4 to 6 years of age, it begins a gradual increase through adolescence and most of adulthood. BMI is not a direct measure of body fat but it parallels changes obtained by skinfolds or other direct measures of body fat. The rebound or increase in BMI that occurs after it reaches its lowest point is referred to as “adiposity” rebound. This is a normal pattern of growth that occurs in all children. Recent research has shown that the age when the “adiposity” rebound occurs may be a critical period in childhood for the development of obesity as an adult. “Adiposity” rebound, occurring before age 4, is associated with higher BMI in adolescence and adulthood. However, these studies have yet to determine whether the higher BMI in childhood is truly adipose tissue (versus lean body mass or bone). In addition, children must be measured at frequent intervals to determine at what age BMI reaches its lowest point before it starts to increase.
Here we have an example of adiposity rebound occurring at around age 3. BMI reached the lowest point at 32 months (2 years 8 months) and then began to increase.
BMI

25. BMI-for-Age Cutoffs > 95th percentile Overweight
85th to < 95th Risk of overweight
percentile
< 5th percentile Underweight
This slide reviews the expert committees’ recommendations to screen and classify BMI-for-age above the 95th percentile as overweight and between the 85th and 95th percentile as at risk of overweight.
The established cut point of the 95th percentile for BMI-for-age are similar to a BMI of 30 in a young U.S. adult, a BMI of 25 in a young U.S. adult corresponds to slightly less than the 85th percentile. So extending the adolescent criteria to younger children and adolescents seemed warranted.
The cut point for underweight of less than the 5th percentile is based on recommendations by the World Health Organization Expert Committee on Physical Status.*
*The World Health Organization Expert Committee on Physical Status. The Use and Interpretation of Anthropometry. Physical Status: Report of a WHO Expert Committee: WHO Technical Report Series 854, WHO, Geneva, 1996.

26. Performance of BMI-for-Age as a Screening Tool
Using the 85th and 95th percentiles as cut points, few children are incorrectly identified as over-fat but some over-fat children will be missed.
It is desirable to correctly identify those children not at risk of overweight or overweight.
“The validity of selected cutoff points to identify adolescents with the highest percentage of body fat has been investigated. In general, common cutoff points for BMI and relative weight have low sensitivities but high specificities. For example, BMIs > 85th percentile have sensitivities of 29% and 23% for identifying adolescent males and females, respectively, who are above the 90th percentile for percentage body fat; corresponding specificities are 99% and 100%. A relative weight > 120% has low sensitivities and high specificities. There are no corresponding data for validity of other BMI cutoff values, or for specific levels of body fat in adolescents that are most predictive of adverse outcomes. In screening for adolescent overweight, specificity may be more important than sensitivity. Maximizing specificity minimizes the proportion of adolescents who will be incorrectly considered overweight by the screen.”*
* Himes and Dietz, Guidelines for overweight in adolescent preventive services: Recommendations from an expert committee. Am J Clin Nutr 1994;59:

27. Can you see risk? This boy is 3 years, 3 weeks old. Is his BMI-for-age
- below the 5th percentile?
- 5th to <85th percentile: normal?
- >85th to <95th percentile: at risk for overweight?
- >95th percentile: overweight?
In the next three slides, we want you to do a self-test to see how well you can screen for risk of overweight in children by looking. We want you to try to identify children with a BMI-for-age equal to or greater than the 85th percentile and less than the 95th percentile.
It has been said that “few medical conditions can be diagnosed as confidently by untrained individuals as gross obesity.” Yet it is very difficult to distinguish children who are at risk of overweight from normal children. In childhood, the distinction is made more difficult by age-related physiological variations.
So, see how you do with the three photos we will show you. This first one is a boy who is 3 years old. Does he appear underweight, normal, at risk of overweight, or overweight?
Photo from UC Berkeley Longitudinal Study, 1973

28. Plotted BMI-for-Age Measurements: Age=3 y 3 wks
Boys: 2 to 20 years
BMI
Measurements:
Age=3 y 3 wks
Height=100.8 cm (39.7 in)
Weight=18.6 kg (41 lb)
BMI=18.3
BMI-for-age= >95th percentile overweight
This boy’s height is 39.7 inches and his weight is 41 pounds. Using his height and weight, we calculated his BMI-for-age to be 18.3.
Plotted on the BMI-for-age chart for boys, his BMI-for-age falls above the 95th percentile. Likewise, when plotted on the weight-for stature grid, it falls above the 95th percentile.

29. Can you see risk? This girl is 4 years, 4 weeks old.
Is her BMI-for-age
- below the 5th percentile?
- 5th to <85th percentile: normal?
- >85th to <95th percentile: at risk for overweight?
- >95th percentile: overweight?
Here is 4-year-old girl.
Photo from UC Berkeley Longitudinal Study, 1974

30. Plotted BMI-for-Age Measurements: Age= 4 y 4 wks
Girls: 2 to 20 years
Age= 4 y 4 wks
Height=106.4 cm (41.9 in)
Weight=15.7 kg (34.5 lb)
BMI=13.9
BMI-for-age= th percentile Normal
This girl’s height is 41.9 inches and her weight is 34.5 pounds. Using her height and weight we calculated BMI-for-age to be 13.9.
Plotted on the BMI-for-age chart for girls, her BMI-for-age falls on the 10th percentile. Likewise, when plotted on the weight-for stature grid, it falls around the 10th percentile.
BMI
BMI

31. Can you see risk? This girl is 4 years old. Is her BMI-for-age
- below the 5th percentile?
- 5th to <85th percentile: normal?
- >85th to <95th percentile: at risk for overweight?
- >95th percentile: overweight?
This is another 4-year-old girl.
Does she appear at risk of overweight?
Photo from UC Berkeley Longitudinal Study, 1973

32. Plotted BMI-for-Age Measurements: Age=4 y Height=99.2 cm (39.2 in)
Weight=17.55 kg (38.6 lb)
BMI=17.8
BMI-for-age= 94thpercentile At risk for overweight
BMI
BMI
Girls: 2 to 20 years
This girl’s height is 39.2 inches and her weight is 38.6 pounds. Using her height and weight we calculated BMI-for-age to be 17.8.
Plotted on the BMI-for-age chart for girls, her BMI-for-age falls on the 94th percentile. Likewise, when plotted on the weight-for-stature grid, it falls above the 94th percentile.
She is classified as at risk of overweight.
The point of this exercise is to demonstrate the difficulty of making an accurate visual assessment of at risk of overweight. BMI-for-age must be determined and plotted on the appropriate growth chart.
BMI
BMI

33. Accurate Measurements are Critical
Boys: 2 to 20 years
BMI
5 1/2 year old boy
Weight: 41.5 lb
Height: 43 in
BMI= 15.8
BMI-for-age=50th %tile
Inaccurate height measurement: 42.25
BMI=16.3
BMI-for-age=75th %tile
As you know, accurate measurements are critical. To illustrate the point of accurate data, we used the case of a 5.5-year-old boy, weighing 41.5 lb with a height of 43 inches. His calculated BMI-for-age is When plotted on the appropriate chart, this falls just below the 50th percentile. If his height were measured or recorded inaccurately at (3/4 inch below his actual height of 43 inches), his BMI-for-age would be 16.3 and would fall just below the 75th percentile. A measurement error of 3/4 inch in height resulted in a change of 25 percentiles. In this example, the measurement error did not cause a change in classification because growth remained within the normal range but you see what could happen.

34. Summary of Using BMI-for-Age
BMI-for-age is the recommended method for screening overweight and underweight
For children, BMI is age and gender specific; for adults there are fixed cut points
Accurate and periodic measurements are important elements of any anthropometric screening
To conclude this section on BMI-for-age, I want to reiterate a few points.
BMI-for-age is the method recommended for screening overweight and underweight from 2 years of age to adulthood. BMI-for-age is a screening tool that may require further assessment to diagnose a specific health condition.
For children, BMI is age and gender specific and nutritional status is identified based on percentiles. For adults nutritional status is defined by fixed cut points.
Periodic, accurate measurements and growth records are important elements of growth screening. An accurate interpretation of growth depends on the accuracy of weighing and measuring.

35. Available CDC Growth Charts
Birth to 36 months
Weight-for-length
Length-for-age
Weight-for-age
Head circumference-for-age
Clinical charts for infants and children 0 to 36 months of age and preschoolers and adolescents 2 to 20 years old are available electronically. The look of the clinical charts is similar to those in use previously. They include space for identifying information and a data entry table for recording anthropometric measurements as well as other data needed to assess growth. The chart includes the 5th, 10th, 25th, 50th, 75th, 90th, and 95th percentile curves, and the scales are provided in both the English and metric systems.
For children 0 to 36 months of age whose recumbent length has been measured, gender specific charts reflecting the anthropometric indices shown on this slide are available.
The weight-for-length chart reflects body weight relative to height independent of age and is used to indicate body size (i.e. over- or underweight). The length-for-age chart enables us to assess linear growth. Weight-for-age reflects body weight or mass relative to age. Head circumference-for-age is important for infants and generally is included as a screen for neurological or developmental problems.

36. Available CDC Growth Charts
2 to 20 years
BMI-for-age
Stature-for-age
Weight-for-age
Weight-for-height
For children 2 to 20 years of age whose standing height has been measured, gender-specific charts reflecting BMI-for-age, height-for-age, and weight-for-age are available. The primary difference between these charts and those developed for younger children is the documentation of BMI-for-age in the data entry table and the addition of the 85th percentile curve on the BMI-for- age chart.
The revised weight-for-stature charts were developed specifically to meet the needs of the WIC program because weight-for-stature is currently used to certify children and assess the need for intervention.
Because the WIC program functions as an adjunct to health care, use of the BMI-for-age chart will foster better coordination of care between the program and local health care providers who have begun to use the BMI-for-age chart. Without comparable WIC data, long-term tracking that enables health care providers to identify the occurrence of adiposity rebound would be difficult.

37. Steps to Plot BMI-for-Age
Select appropriate growth chart
Record data
Obtain accurate weight and height measurements
Calculate BMI
Plot measurements
Interpret plotted measurements
The six steps outlined here to plot BMI-for-age are similar to general growth assessment.
First, based on the child’s gender and age, the clinician or health care provider selects the growth chart that will be used.
On the appropriate chart, accurate measurements (i.e., weight, height and head circumference) are recorded immediately after being taken and the child’s age at the time of the visit is determined.
After weight and height are recorded, BMI is calculated.
Recorded measurements are plotted on the growth chart for the current visit.
Last, plotted measurements are interpreted based on the percentile ranking.

38. Case Study: “Sam” Name: Sam Weight: 37 lb 4 oz (16.9 kg)
Height: inches (105 cm)
DOB (date of birth): 9/15/1994
DOV (date of visit): 4/4/1998
Let’s take the next 5 minutes or so to assess Sam’s physical growth. We will use the steps just reviewed to determine and plot Sam’s BMI-for-age.
Sam has already been weighed and measured. He weighs 37 pounds and 4 ounces and is 41.5 inches tall. Because Sam is at least 2 years old and we have a standing height, we select the ‘Boys 2 to 20 BMI-for-age’ chart.
Sam’s date of birth is shown here as September 15, 1994.
The date of Sam’s visit is April 4, 1998.

39. Data Entry Table for BMI-for-Age
Mother’s Height________
Father’s Height________
Age
4/4/98
Gestational
Age________ Weeks
Weight
37 lb 4 oz
Comment
Date
Height
BMI
41.5 in
Sam’s identifying information has already been noted on the chart, and his weight and height were recorded on the data entry table immediately after being taken as 37 pounds and 4 ounces and 41.5 inches, respectively. The date of visit or measurements is shown here as April 4, 1998.

40. Calculating BMI with the Metric System
Formula: weight (kg)/[height (m)]2
Calculation: [weight (kg)/ height (cm)/
height (cm)] x 10,000
Example: A child’s weight=16.9 kg and height=105.4 cm
BMI = [16.9 kg / cm / cm] x 10,000 = 15.2
We will show you two methods for determining BMI, one using the metric system and one using the English numeric system. With the metric system, BMI is calculated using weight in kilograms divided by height in meters squared. Because height is rarely recorded in meters, it can be converted from meters to centimeters by multiplying meters by Then divide weight by centimeters squared, which is the same as dividing weight by the height twice as shown in the formula. Then multiply by 10,000. Suppose we know that Sam’s weight is 16.9 kg and his height is cm. When the calculations are completed we find his BMI to be 15.2.
BMI can also be determined by looking it up on a standard table of BMI values or using a Web calculator.
BMI can be calculated with a standard hand calculator and using a mathematical formula.

41. Calculating BMI with the English System
Formula: weight (lb)/[height (in)]2 x 703
Calculation: [weight (lb)/height (in)/height (in)] x 703
Example: A child’s weight = 37 pounds, 4 ounces and height = 41 1/2 inches
(convert fractions to decimal value)
BMI = [37.25 lb / 41.5 in / 41.5 in] x 703 = 15.2
When using English measurements, the formula is expressed as weight in pounds divided by height in inches squared multiplied by a conversion factor of 703.
Now we will calculate Sam’s weight with the English numeric system. His weight is 37 pounds and 4 ounces and his height is 41 1/2 inches. Before continuing, we must change the fractions and ounces to decimal values. Thus Sam’s weight becomes and his height becomes You then divide Sam’s weight, 37.25, by 41.5 squared or divide Sam’s weight by 41.5 two times. Multiply the value you get by 703. His BMI is 15.2, as we found with the metric system.

42. Data Entry Table for BMI-for-Age
Mother’s Height________
Father’s Height________
Age
3 1/2
4/4/98
Gestational
Age________ Weeks
Weight
37 lb 4 oz
Comment
15.2
Date
Height
BMI
41.5 in
Sam’s age and BMI at the time of the clinic visit can now be entered on the data entry table. He is 3 1/2 years old and has a BMI of 15.2.

43. Sam’s BMI Plotted on Boy’s BMI-for-Age Chart
Boys: 2 to 20 years
BMI
Interpretation:
Sam’s BMI-for-age is slightly below the 25th %tile so it falls within the normal range.
Of 100 boys who are the same age, fewer than 25 have a BMI-for-age lower than Sam’s.
All the necessary information is recorded and Sam’s BMI can be plotted. On the BMI-for-age chart, find Sam’s age on the horizontal axis and visually draw a vertical line up from that point, then find his BMI on the vertical axis and visually draw a horizontal line across from that point. The point where the two intersect represents Sam’s BMI-for-age.
When plotted on the growth chart, Sam’s BMI-for-age falls just below the 25th percentile curve. Percentile indicates the rank of a measure in a group of This means that of 100 children the same sex and age as Sam, fewer than 25 children will have a BMI lower than his.

44. BMI-for-Age Cutoffs > 95th percentile Overweight
85th to < 95th Risk of overweight
percentile
< 5th percentile Underweight
Sam is neither overweight, underweight nor at risk of overweight. When a child’s plotted measurement falls between the 5th and 95th percentiles it is considered to be in the normal range. Sam’s BMI-for-age is in the normal range. When a child’s percentile rank falls outside the normal range (i.e., outside the 5th or 95th percentiles), further evaluation is needed.

45. Interpreting the BMI-for-Age Chart
BMI-for-age indicates a child’s weight in relation to his/her height for a specific age and gender
Need a series of BMI plots to determine the growth trend
If indices deviate from normal growth patterns, further assessment may be needed
To summarize:
For children, BMI is age and gender specific
When assessing physical growth, it is necessary to have a series of accurate measurements to establish a growth pattern. Having a series of measurements takes into consideration short- and longer-term conditions.
Growth patterns that fall outside the established parameters, the 5th and 95th percentile for any given anthropometric indices, suggest the need to recheck measurements, plots, and calculations and make any necessary corrections or adjustments. If these are correct, further evaluation is required to determine the cause.

46. Future Directions for CDC Training
Continuation of training at national meetings
Development of a Web-based training module
Training focus: accurate weighing and measuring, calculating BMI, plotting and interpreting the growth charts
CDC will continue to provide training on the new growth charts at national meetings with the intent that participants will return to their states and train professionals, paraprofessionals, and clinicians to use the new charts.
The focus of the training is on accurate weighing, measuring, calculating BMI, plotting measurements, and interpreting the plotted measurements. The importance of obtaining accurate measurements cannot be overemphasized.
The Division of Nutrition and Physical Activity at CDC is collaborating with the Maternal and Child Health Bureau, National Center for Health Statistics, federal and state partners, and professional associations to develop education materials that include web-based training modules on the development of the CDC growth charts, using BMI for nutritional assessment and clinical applications of the growth charts. It is anticipated that the Web-based training modules will begin to be available on the Internet by summer 2001.

47. Distribution of CDC Growth Charts
Electronic Copies
Clinical growth charts can be downloaded or printed from this Web site ( You have the option of signing up for listserv at this site to be notified as training materials related to the growth charts become available.