Growth chart Q&A with Dr. Dad: moving beyond the growth chart hype

I’ve been thinking a lot about the mommy competition with growth charts, developmental charts, and general is-my-baby-keeping-up-with-your-baby worry.

And then I saw this article on the mommy madness around comparing our kids’ chart stats to others’. And then I saw a segment on the Today show about a finding that there is an increase in obesity in 9-month-old babies, but
they didn't talk about the context of this information – for instance, if you are just looking at weight-for-age measurements, my Pie is morbidly obese, and has been since birth. If you look at weight-for-height, she is between the 25th and 50th percentile. Were they looking at weight-for-height? Or just the charted weight dots? I know the obesity epidemic isn’t letting up, but I want some context for that info.

This post certainly isn’t all the information, and it certainly doesn’t take the place of any medical information you’ve received from your doctor. This is my understanding of the information I’ve collected, and is surely colored by my opinions about doctors in general and medical treatment. For the record, I utilize modern medicine and believe that doctors are working to create healthy babies and healthy people. I also believe that a lot of medical procedures are unnecessary, and that when parents feel that there is something wrong that they should be insistent about getting more information or seeing a specialist. However when parents are not fully informed but are insistent on receiving a specific treatment, it can end up doing no good, or even harm. This does not mean I think people should always trust every doctor’s diagnosis or opinion – if you feel something is wrong, do your research, get a second opinion, see a specialist.

I wanted to write this post about growth charts because I have been noticing mamas comparing their kids’ percentile stats like they’re batting averages, and becoming concerned, without really taking the stats in context.

Understanding Percentiles: Percentiles are similar to percentages, but not an exact match. The idea is that if you lined up 100 kids, the largest kid is the 99th percentile, meaning that 99 kids in the room are smaller. Most charts measure between 5th & 95th percentile, where almost all children will fall.

Understanding the Chart: (For the CDC chart, which is probably what you’ve all seen for your kids) There are different charts for boys and girls, and different charts for infants and children. On each chart are two curves – a curve for weight-for-age and a curve for height-for-age. What is most useful is the curve on the back of the chart, putting both of those points in context and showing weight-for-height.

At her two-month appointment, my Pie was 98th percentile for gender for height – meaning that if you lined 100 kids up, 98 of them would be smaller than her – and 96th for weight. Because of her dad’s size (more on that below) this is not a concern, although people have told me that I am “lucky” that my baby is growing well, or that I need to keep an eye on her weight, because she is already “obese”.

I would be concerned if she suddenly dropped below the 50th percentile (meaning half of the children would suddenly be larger than her – an indication that most children are expected to have grown significantly during a given time period, and she did not) or if she had started near the 25th percentile and suddenly soared to the 75th percentile, indicating that she is growing at an alarmingly fast rate as compared to the “average”. The comparison to the “average” child’s expected growth is useful, however the most useful indicator is steady growth (even if that growth is slow) following the general shape of the growth curve, regardless of the percentile markers on the child’s own curve.

All this information in the media about growth charts, revised charts, new findings, infant obesity, and getting over the competition surrounding charting is swirling around in my head. It is conflicting messaging, all containing some useful information but all incomplete information, or information taken out of context.

It was a start, I thought, but I wanted more. I sent some questions to a pediatrician I know who is a growth specialist, an endocrinologist with a reputation for a “thoroughly encyclopedic understanding of endocrinology” and the most dedicated clinician I know.

And I had access, because he’s my dad.

I sent him an email asking the following questions (surely incomplete questions, and this in no way offers complete understanding of the topic, but again, it’s a start), and got the following answers:

Mama: How were the growth charts we currently use (the CDC charts)developed?

Dr. Dad: They are cross sectional charts meaning that they took a bazillion children at different ages and measured them at ONE point in time and then created the curves rather than following a gazillion babies (which is 1/10th of a bazillion) from birth through 20 years old measuring them repeatedly and then creating a curve.

Since I have a 5-month-old girl, I am presenting the standard girls birth-36month chart, which is length-for-age and weight-for-age.

So the growth charts seem to me to actually be kind of a historical document (showing children’s size and assumed growth rates in the late 70’s) and comparative tool, as opposed to diagnostic tool. They show a cross-section of the heights and weights of children at different ages in the late 70’s, when the chart was developed. The charts do not show how any particular child grew, but are really just an “average” of the available children who were measured at the time. It would have been prohibitively expensive and would have take over 20 years to follow the growth of each child, and follow enough children to make the data statistically relevant, and then by the time the data was complete, it may well be outdated. But we use this chart, and we now compare our babies to this “average” as a way to track their growth.

From www.CDC.gov/growthcharts/background : The 1977 growth charts were developed by the National Center for Health Statistics (NCHS) as a clinical tool for health professionals to determine if the growth of a child is adequate. The 1977 charts were also adopted by the World Health Organization for international use.

Mama: Are they representative of breastfed babies, formula fed babies, or a combination? What does this mean in terms of how babies fall on the chart?

Dr. Dad:The curves were primarily based upon formula fed AND NOT breast fed babies, who tend to be “fatter” up until 6 months and then level off.

I wanted to look into this further – and it wasn’t until the CDC added the BMI chart in 2000 that breast-feeding was taken into consideration. From CDC.gov:

…the revised growth charts for infants contain a better mix of both breast- and formula-fed infants in the U.S. population. (On average, since 1970 approximately one-half of children born in the United States are reported to have been breast fed at some point, and about one-third have been breast fed for 3 months or more.) The addition of the BMI charts is probably the single most significant new feature of the revised growth charts.

Also from the CDC on BMI charts and weight-for-height measurements: These BMI-for-age charts were created for use in place of the 1977 weight-for-stature charts. BMI (wt/ht2) is calculated from weight and height measurements and is used to judge whether an individual's weight is appropriate for their height.

However even a BMI chart is not absolutely accurate for breastfed babies, who tend to to be heavier to 6 months of age. Some suspect that the BMI is not a perfect system.  Working toward improving the way we measure bodies, Professor Richard Telford, a leading Australian scientist at the Commonwealth Institute, is creating a new formula to measure body size called Body Mass Function (BMF) that not only takes body fat, age and development into account, but also considers height and weight. This is most useful for adults as they age and lose muscle mass, but BMI is also inaccurate for other segments of the population – those who tend to be quite muscular measure obese on BMI scales, because it’s measuring weight against height and does not take body fat percentages or muscle mass into account.

Dr. Dad: The new recommendation is to use the World Health Organization (WHO) growth curves for 0-2 yo as these are based primarily on breast fed babies (but includes those in third world countries).

This is the chart from the WHO website – not all that informative for giving us an idea of how the chart compares to the old CDC chart. It includes data from around the world as opposed to just the United States, which may well show that American infants are larger on average than the world average, but since the growth includes more breastfed babies the curve may be more accurate for those who are breastfeeding.  (image from http://www.who.int/childgrowth/en/)

Mama: How would you take the parents’ weight/height/health into consideration in terms of charting? Or is it taken into consideration?

Dr. Dad: One must always consider the familial genetic potential
- the gender-adjusted mid-parental height target. In the following estimations I am using 5' 3" for you and 6' 4" for the mountain man:

For a girl - subtract 5 inches from father's height and average with mother's height - for the Pie it would be about 5'7" (which would plot to between the 75th and 90th centile for an adult female).

For a boy - add 5 inches to mother's height and average with father's height.  If you have a son, it would be 6 foot which also plots to between the 75th and 90th centile for an adult male.

These are estimations and include 1.5 Standard deviations (approximately 3 inches) on either side of the target.

If the mountain man wanted taller kids, he should have married taller!

This snarky comment led to a long conversation about parents seeking endocrine treatment  for children who are not growth hormone deficient, but who are small compared to peers, (which is commonly just a result of smaller parents) and that parents often demand human growth hormone (HGH) treatment.

The increase in the incidence of HGH treatment in our children is creating a new smallest percentile of children – those who were small but not the smallest, who are not seeking treatment, now find themselves to be the smallest as the formerly smallest kids receive HGH. The kicker in all this, though, is that the children of those children now receiving HGH treatment will not benefit from the artificial growth of their parents – they will receive the genetic potential growth that their parents were born with. The perceived need for growth hormone treatment will be perpetuated into the next generation when artificially tall parents are seeking treatment for their “unexpectedly” small children, who are really just a product of their genetic height potential. Growth hormone, while it *may* increase height in children, will not change the height genes that a parent passes to his or her offspring.

But basically the charts do not take genetic potential height into consideration – that must be done for each individual child compared against his or her parents, and must be interpreted by the clinician.

Mama: Do most doctors pay more attention to baby’s weight-for-age or the baby’s weight-for-height? For instance, the Pie always charts above the 75th percentile, and has charted at the 99th percentile for weight-for-age for girls. However when you compare that with her height-for-age, she is more like the 25th to 50th percentile. So which number to doctors usually pay attention to?

Dr. Dad: Most providers look at the individual weight or length and not the weight-for-length or BMI (because that requires flipping the chart over and placing another dot on the chart).

Yikes, that’s a little scary – that doctors are looking at single points, out of context. So parents are being given their child’s weight-for-age percentile (and this may often be the only number they’re given) and told that the child may be unhealthy. They weight may be quite low, or quite high, and this information is meaningless without the context of the child’s height and the parents’ heights. If a child comes from a family with some meat (and fat and gristle) on their bones, they may measure as quite overweight, but actually is OK in the context of the healthy range for a heavy child of healthy parents. Additionally, if a child plots below “average” in weight-for-age, but the child’s height is significantly above “average” for age, the child is quite lean and possibly underweight.

Here’s more information about growth charts, how they work, and what they tell us:


Mama: When should a parent be concerned about the information on their child’s growth charts?

Dr. Dad: The question of when parents should be concerned really is difficult to answer.  Crossing centile channels during the first year is quite
common.  67% will cross at least one channel. 

The other concern is when a child is underweight for length or overweight for length.  The first is usually either due to poor nutritional intake, chronic illness (such as a malabsorption, chronic inflammatory or hypermetabolic condition). The second may be due to an endocrine problem, especially if small in length. 

Again, taking the dots in the context of each other is important. Breastfed babies tend to be “overweight” to six months, perhaps because breastmilk is ideally formulated to grow babies. After six months though, a formula-fed baby, and especially a schedule-fed formula-fed baby may far exceed the weight on the charts of a breastfed (and especially an on-demand breastfed baby) because parents often encourage the baby to finish that last ounce of formula (or milk, for bottle-fed breastmilk), thus leading to additional weight gain. Even a child’s source of food and feeding techniques must be taken into consideration!

It seems to me, after talking with Dr. Dad, doing some research, and getting a little more information, that the charts are best used to track the child’s growth against him or herself, NOT against other babies. Every child will grow at his or her own rate, in accordance with the genetics passed down from parents. If a child has a dramatic change in the chart, or if there is little growth occurring, it may be time to consider seeing a specialist, but as Dr. Dad said, most children will not stay at the same percentiles throughout the first year.

Mamas, please stop using this as a tool to compare your kid to your friend’s kid. Stop using this as a competitive tool or a way to ensure future success. It’s not. Above-average charting does not equal an above-average potential for success in life. (Thusly, a below-average charting child is not going to necessarily be below-average in life.) It’s a tool for us to use, in concert with other things, to ensure that a child is nourished, growing, and healthy – as compared to herself.

Some moms boast that their child is “above average” like this growth percentile will have an impact on her SAT percentile standing. Or they are proud that their child has always ridden the 50th percentile line – which just means the kid is growing as expected, with no leaps or drops, along a curve designed by measuring lots of different kids. Some moms are concerned about a baby girl’s weight moving about the 50th percentile line, for fear she will be overweight and unattractive, or fear of a boy’s weight falling below the 50th, for fear he will be too lean, and not muscular as a teen or adult.

Mamas, relax. If you and your partner are average height, your child will not be in the 7+ foot range. (At least probably not) You don’t need HGH just because your child is shorter than the others in his or her class. If everyone in your family is lean, or tall, or overweight, or short, your baby will probably follow those same trends.

In order for there to be an average, and in order for there to be genetic diversity, there have to be outliers. Someone has to be below the 5th percentile, someone has to be over the 95th. If a baby is growing steadily, and is not making huge drops or jumps, and is otherwise healthy, he is OK. There is no need to worry, to compete, to be concerned. For the vast majority of children, they will grow to the height they’re genetically going to grow to. And growth hormone (and the field of endocrinology) is not a toy. It’s not something to be used for aesthetic improvements.

And I am not just saying this because my Pie will likely be tall – I was short, often the shortest in my class and I often wished to be taller. I was normal for my genetics, though. I am more than two inches taller than my mom and less than 7 inches shorter than my dad. And in the equation that my dad gave above to estimate adult height – I subtract 5” from my dad’s height and average that with my mom’s height, and I should end up about 5’3” plus or minus 3 inches – and I’m 5’ 3.75”. So totally on target – but I could have been anywhere from 5’ to 5’6'” and still be completely normal; in fact I have a sister who is closer to 5’ and one who is closer to 5’6”.

The chart can certainly be an indication that something is amiss – a baby isn’t growing, although being fed, sleeping well, and having average sized parents. Or maybe a baby is growing too much – but in both cases, it’s not about one dot on the chart, but the curve that those dots create- your child’s personal curve, and against the background information of what is considered “average”. It’s comparative, not diagnostic. Change is what indicates a problem, not a single dot. When there seems to be a problematic trend, most pediatricians will refer you to a specialist.

If a doctor tells you what your child’s weight percentile and her height percentile are, and especially if he or she shows concern because of these numbers, ask what the weight-for-height percentile is, and if there is a steady pattern of growth for your child following the shape of the average curve, and if it is within the expected range for parents’ statures. This is what’s important.

If you think that your child’s growth is abnormal for his or her genetic potential, or if you notice a significant jump or drop on the charts, or if the curve is not roughly following the shape of the average curve (regardless of the percentile markers for your child) talk to your pediatrician, and ask questions until you understand.

Don’t compete with charts and percentile stats for other kids. This isn’t a contest.


Are there other questions you have for Dr. Dad? If there is a popular question, I’ll do another Dr. Dad post!


  1. "I also believe that a lot of medical procedures are unnecessary, and that when parents feel that there is something wrong that they should be insistent about getting more information or seeing a specialist."......or just rub some dirt in it, because Oma inevitably has a worse condition.

    Also, could you ask Dr. Dad if HGH will make my buttcrack taller? Thanks.

  2. oh boy, the unusual requests are pouring in. Sorry, I don't think HGH works on specific parts of the body, and I think it would work on your entire body proportionally, so as your buttcrack grew, so would you. (if of course it had any increased growth for you)

    and about the dirt - we should probably try using rubbing alcohol or vinegar.

  3. I just got an interesting email about this post - a mom asked how the estimated adult heights that I talked about could be related to her 9-month-old's current height. So here's what you can do - estimate, with the equation for adult height based on parents' heights, the height range for your child as an adult. Then find out what the percentile for adults of the same gender that height range represents. Your child will probably follow a childhood curve somewhat similar to that percentile - and even if there are small variations, that's OK. If there's huge variation (like you estimate an adult height at 75th percentile of the adult population for the gender of your child, and your child is below the 25th percentile) there could be a problem. Bring your concerns up at an appointment, but if your doctor is not too concerned, you probably don't need to be, either.


Related Posts Plugin for WordPress, Blogger...