Weighty Wisdom by: Elizabeth Jackson, M.S., R.D.


PART TWO: Weights are Up Across the Board, but a Child Isn't a Population

In the last Weighty Wisdom column (2, Part 1), we looked at the origins of weight classifications. In this column, Part 2, we’ll talk about what’s happened on a population-wide basis but give a few warnings about interpretation of weight classes. Part 3, coming at the end of August 2003, will tackle untangling normal from abnormal weight by looking at patterns of growth.

Why are we so worried about kids’ weights? How have weights changed in American children over the last 30 years?
In the beginning, the percentage of kids at a certain percentile basically matched the percentile itself. This means about 5% of American kids or so were at the 95th percentile and above for weight at any age. As we discussed in the last column, weight classifications have come from large national government surveys, called the National Health and Nutrition Examination Surveys (NHANES), in which thousands of children have been measured and weighed. The NHANES surveys tell us that 6% or fewer of children and teens aged 6-19 years old were defined as overweight up until the mid 1970s. Prevalence of overweight (defined at or above the 95th %) in younger kids, aged 6-11, nearly doubled from 4% of kids to 7% between NHANES I and NHANES II (the 1970s). The big increase in overweight for children of all ages came in 1980s. During the period beween the NHANES II and the NHANES III, the prevalence of overweight jumped from 7% and 5% for 6-11 and 12-19 year olds, respectively, to 11% for both groups. Then prevalence jumped again to 15% during the 1990s. This means that now, in a room of 100 American kids of the same age, about 15 (instead of 5) will have a weight that is at the 95% or greater. This is true across ethnic groups.

If kids across the board have gotten heavier, then wouldn’t the new weight curves shift so the percentiles would really reflect the percent of kids at those weights (meaning, only 5% of kids would be at or above the 95th %)?
In fact, because of the upward trend, weight data from NHANES III for kids over age 6 was not included in the 2000 weight curve revisions in order to keep weight for height standards comparable to earlier times. So kids are definitely getting heavier. But before we talk about why this is happening and what we can do about it, we need to know some basic things about weight and BMI in kids.

1) Overweight and underweight or simply high weight and low weight? It is a bit problematic to take a range of values for a population and say that the numbers at each end (the top and bottom 5%) are automatically abnormal. The top and bottom weights percentiles in a population certainly may have a higher likelihood of reflecting growth or medical problems, but in a group of individuals, some members always have to have the top and bottom numbers or degrees for characteristics—for example oldest and youngest, darkest and lightest skin or eye color, straightest and curliest hair, etc. There are many, many children for whom being at the top or bottom of the BMI scale has no negative health consequences. But we do need to know how to recognize actual growth problems (coming in Part 2 of this article!).

2) Remember the BMI is a calculation based on weight and height and is not a direct measurement of body fat. Medically, there is more worry about being overfat versus being overweight. BMI is generally a good estimate of fatness in people, but individuals who are very large boned or muscular (including very fit athletes) can have BMIs that fall in the overweight or even obese range. In other words, BMI is a measure of weight versus height, but higher weight does not always mean higher fat. To actually measure body fat directly, the only method you might see used in a clinic would be skinfold measurements. Fat under the skin is measured (by a trained clinician!) with a special instrument called a caliper at several sites on the body. This has been done in the NHANES surveys. There are other very high tech and even more accurate ways to measure body fat, but these methods are used primarily in research studies and include scanning/imaging devices and body density measuring.

3) Single BMI measurements can be misleading because kids often get fatter right before a growth spurt (again, we’ll discuss patterns of growth in Part 2).

4) Perhaps most importantly of all, no matter what is going on in our culture that may be making kids heavier, we can’t forget that weight is still very strongly determined by genetics, meaning the natural size of the parents. A very famous study from the 1970s, conducted by Garn & Clark, showed that weights of children strongly resembled weight combinations of parents. The leanest kids came from two lean parents and fattest kids came from two obese parents, with other combinations in-between. We might think this is simply because fatter people eat less well or are less active and so have bigger kids because of “bad habits” they pass down. Yet a study from the 1980s by Albert Stunkard looked at the weights of Danish adopted children and found their weights correlated largely to their biological parents’ weights rather than the adoptive parents’ weights. In 1990, Stunkard published another study based on Swedish adoptees with the same results. In other words, we’re not all meant to be thin. We certainly understand that short parents tend to produce short kids and tall parents, tall kids—but as a culture we haven’t been comfortable thinking that weight is also largely determined by genetics. At a time when we’re very legitimately worried about the increased numbers of very heavy kids, we can’t forget that there are always going to be some kids who are naturally fatter than their naturally thinner friends.

Also, kids from different ethnic groups naturally tend to be built a bit differently. As mentioned, the CDC revised growth charts of 2000 were based on measurements of groups of children that included ethnic minority groups in proportion to the total U.S. population. That would mean, for example, that since about 17% of kids in the U.S. are Hispanic/Latino, about 17 percent of measurements used when combining data to make the growth curves would have come from Hispanic/Latino children. About 61% of American kids are identified as white, so we can say that the majority of the measurements used for the growth charts were those of white children. But for exactly this reason, some researchers feel the BMI growth charts may not be the most accurate standards for non-white children. To give some interesting examples, Hispanic children naturally tend to be shorter and heavier than Caucasian children, and this even shows up in babies. Children of Chinese ethnicity in Singapore are built smaller than Chinese children in Beijing (same % of body fat but lower BMI). Children from the Chaldean ethnic group (of Middle Eastern heritage) in the Detroit area have denser bones and more lean body mass (e.g., muscle) in general, as well as more body fat, and so end up on average with higher BMIs than Caucasian kids. Up to a certain point, children with higher BMIs may reflect better nourished kids: for example, immigrant Mayan Indian children growing up in Florida and California are taller and heavier and have more muscle and fat than Mayan village children back in Guatemala, though the immigrant children are still generally shorter than the white, African-American and Mexican-American children in their cities. We have to be very careful when comparing children to standards.

So: if we presume that children do naturally come in different sizes based on genetics, and if this has been true forever, how do we explain why so many American kids are heavier now? And heavier to such a great degree? This would suggest that while some children are simply meant to be larger, others are indeed experiencing de-regulated weight patterns—they are not growing as nature intended. What’s going wrong and what are we to do about it to keep kids healthy? And, in the process, how do we address these issues without overreacting to normal patterns of growth and fatness in children? In Part 2 of Looking Beyond the Numbers we’ll discuss patterns of growth—many measurements taken over time. We’ll see how patterns of growth are the key to to determining when weight truly reflects a problem and how the best standard to compare a child to is him- or herself.


Sign-up to receive e-mail updates about Weighty Wisdom articles

If you have a question on early childhood feeding, eating or weight for the dietitian, please send it in Weighty Wisdom Question Form



You will need the Adobe Acrobat Reader to open PDF files.