by Emily Oster
What is less clear is whether this really has much actual health impact. Relatively few studies have looked at the actual outcomes associated with vitamin D, like bone growth. In two that did—very small randomized trials of supplementation—there were no impacts on bone growth or bone health, even though supplementation did increase the concentrations of vitamin D in babies.18
This isn’t to say you shouldn’t use vitamin D supplements. And certainly rickets does occur, primarily in developing countries with serious nutritional limits. But it does suggest that if you miss a day here or there, you shouldn’t panic.
If you are very uncomfortable directly supplementing your baby, there is evidence that if you are breastfeeding, high levels of supplementation for Mom will increase her vitamin D concentration and accomplish a similar goal.19
Breastfed infants are also sometimes iron deficient, which can cause anemia. Breast milk is low in iron. Iron supplementation is not commonly recommended, unless the infant actually shows signs of anemia, and iron is present in rice cereal, so once your kid starts eating, this problem diminishes. Also, anemia rates are improved by delayed cord cutting (see part 1), which is a lot easier than supplementation.
All this supplementation applies to breastfed infants. Formula contains iron and vitamin D, along with the rest of the vitamins. So if you use formula even some of the time, your child is unlikely to have these issues.
The Bottom Line
Early exposure to allergens reduces incidences of food allergies.
Kids take time to get used to new flavors, so it is valuable to keep trying a food even if they reject it at first, and early exposure to varying flavors increases acceptance.
There is not much evidence behind the traditional food-introduction recommendations; no need to do rice cereal first if you do not want to.
Baby-led weaning doesn’t have magical properties (at least not based on what we know now), but there is also no reason not to do it if you want to.
Vitamin D supplementation is reasonable, but don’t freak out about missing a day here and there.
PART THREE
From Baby to Toddler
Babies are exhausting in many ways—they don’t sleep, they can’t tell you what they want, they eat all the time on an unpredictable schedule. When you have an infant or a four-month-old, you may look forward to the time when your child can eat dinner at the table and tell you what they want.
Once realized, though, this is not always all it’s cracked up to be. Take the battle of the socks. With a baby, it can be hard to find socks that do not fall off. But it’s easy to put the socks on! They are happy to have them; it’s easy to manipulate them. With a baby, rarely do you spend time thinking about getting ready for the day early so as to have time for socks.
Not so with a toddler. “Time for socks and shoes!” you say, eleven minutes before you need to leave the house. “NO! I don’t WANT socks! I don’t WANT them.” Foot stamping, face scrunched up. Arms may be folded in anger pose.
“Let’s put on your socks.” Wrestling.
“AHHHHH!!!! NOOOO!!!!!”
“If you don’t let me put on your socks, I’ll have to get Dad to come help.”
“NO SOCKS. NOOOO SOCKS!!!!!”
“Sweetie, can you help me with him?” Second parent arrives, holds kid still.
Socks are on. Great! You go looking for shoes. Return. Child has taken off socks, is wearing no socks, just an evil grin. Has also removed pants.
Toddlers are a new ball game. They are funny, playful, exciting to be around. But they also bring resistance. And at the same time, there are more things you are trying to accomplish, things that you need their help with. Sleep training, vaccination—you can do these without your child’s cooperation. Potty training, not so much. You can set up a system, you can have stickers, M&M’s, a special potty video. But ultimately, your child will have to decide to use the toilet. It’s just a fact: you cannot force someone to poop.
Parenting a toddler also seems somehow more consequential than parenting a baby. As you see your child’s personality come through, you also start to see what they will struggle with. And you, all of a sudden, face choices—like screen time, or what kind of preschool to send them to—that seem like they may follow your child forever. On top of this, add the issue of discipline, which, suddenly, you have to think about, and it adds up to a much more complex parenting problem.
As your child ages, evidence-based approaches to parenting become more challenging. The more variation across children, the more difficult it is to pull strong conclusions out of data. Heterogeneity across kids means that what works for one kid might not work for another, and if you estimate an effect of some approach on average, you may get nothing, even if it works really well for some kids.
There are, however, some general principles to learn. In this part of the book, I’ll also talk a bit about milestones—physical, which you’ll see some of in the first year, and language-oriented, which come later. Most of us worry, at least sometimes, whether our kids are developing normally. Why isn’t my daughter crawling or walking or running? Why does my sixteen-month-old just use “da-da” for everything? There aren’t likely any decisions about this, but knowing something about the data can relax even the most neurotic of us.
Unfortunately, I have found nothing in the data to address the sock problem. I am holding out hope for technological progress that will produce a sock you can lock onto your child’s leg. Stay tuned.
13
Early Walking, Late Walking: Physical Milestones
My friend Jane’s son was born three months after Penelope. Once they got a bit older—five, six, seven—you’d never notice this age difference at all, but early on, it was hard to believe that was true. When Benjamin was born, Penelope seemed like a giant. When he was a floppy six-week-old infant, she was four and a half months old, well on her way toward being a real, solid baby.
But then came walking. At a year, like the average kid, Benjamin got up and started toddling around. Not Penelope. By the time he was walking, she was fifteen months old and seemed to show no inclination. It is sometimes easy to ignore the way your children differ from the average, but it’s made much harder if you see the average all the time.
At Penelope’s fifteen-month well-child visit, the ever-calming Dr. Li told me not to worry that she wasn’t walking. “If she’s not walking by eighteen months,” she said, “we’ll call in early intervention. But don’t worry! She’ll figure it out.” Early intervention is an excellent government program designed to intervene at young ages to help kids with developmental delays—physical or mental. This is a hugely valuable program to have access to, but still, I did not like the suggestion that we were approaching it.
I tried to explain to Penelope how to walk; she didn’t care. I tried to provide incentives, which really was going off the deep end.
And then, about two weeks after the doctor visit, Penelope walked. Just like it was no big deal. Perhaps because she was so old by the time she learned, she never fell down much, either, just went from crawling around to walking normally in a day or two. And then I promptly forgot about my fear that she would never walk and moved on to other neuroses. (There are always more neuroses around the corner when you’re parenting.)
I don’t think my experience was unique. In the moment, physical milestones—sitting, crawling, walking, running—take on an outsize importance. I have many notes from the first months of Penelope’s life about her rolling ability (very early rolling to the left, but poor rolling to the right). Things like head control are among the first means we have to evaluate how our kids are doing.
Failure to achieve these milestones at the time we expect, therefore, tends to worry parents. I think part of the issue is the focus on average ages—as in, “Most children walk around one year.” This is true, but it misses the fac
t that there is a wide distribution in what is typical.
We are used to thinking about these distributions in, say, our child’s weight. The average one-year-old weighs 23 pounds, but there are some much smaller and some larger. When you go to the pediatrician for your one-year visit, they’ll actually tell you something like, “Your child is at the twenty-fifth percentile for weight.”
In the case of milestones—physical, and language development—we don’t really talk about distributions. I’m not sure why not; could be lack of data or an unwillingness to assign percentiles in these areas. But whether we discuss them, these distributions are there. And even just knowing this may relax you a bit. It’s true that the average age of walking is a year, but having a kid who walks somewhat earlier or later than this one-year average is also totally fine in the same way it’s totally fine for your child to be at the 25th (or the 75th) percentile of weight.
So why do we pay attention to this at all? Why do pediatricians evaluate motor skills? There is good reason to do it, but the goal is to detect children who are outside the normal range of the distribution. In particular, pediatricians are looking for kids who are very delayed. Children who are very delayed on early milestones—head control, rolling over—are more likely (not very likely, just more likely) to have serious developmental issues.
Some of these issues will also manifest in cognitive or behavioral problems, but we do not see evidence of delays in these areas until kids are much older. There is some literature showing that children with serious early motor delays also show some lower spatial skills in later childhood,1 and perhaps even have lower reading test scores as middle-age adults.2 For this reason, detecting early motor delays is a pediatric focus.3
There are also some particular diseases or conditions that motor delays can signal.
The primary one is cerebral palsy (CP), which, broadly, is a term for developmental problems caused by very early damage to the nervous system. This affects 1.5 to 3 children in 1,000, meaning it is rare but common enough to be something many pediatricians see in normal practice (these rates are much lower for full-term babies with nontraumatic births). In the past it was believed that CP was exclusively a result of injuries at birth, but more recent evidence suggests prenatal conditions may also have an effect on whether a child is born with CP.4
Cerebral palsy isn’t a disease—like a virus or cancer—or a genetic defect. It’s a term to describe motor issues that result from nervous system injury. The issues resulting from CP vary widely—it can affect different limbs or body parts, and be more or less severe. At birth, doctors are likely to know if babies are at higher risk for CP—due to birth trauma, prematurity, or other risk factors—but a definitive diagnosis typically cannot be made at birth. Instead, CP is typically recognized later when motor development is abnormal. More severe cases can be detected early—at four to six months—but less severe cases may take a year or more to become apparent. Careful evaluation of babies for motor delays is helpful in increasing the chance of early detection, which can in turn lead to earlier intervention.
The other group of conditions that may be detected this way are progressive neurological diseases. These are extremely rare. Muscular dystrophy is the most common, but it affects just 0.2 in 1,000 births. The others are even less common. Given their progressive nature, these are also more difficult to detect early on; still, they are one of the things pediatricians are looking for.
Motor delays are also common in some conditions that you’d know about at birth. Spina bifida (a birth defect in which the body fails to close over the spinal cord), for example, or a genetic condition like Down syndrome. Motor development is carefully monitored for children in this group, but we do not expect these conditions to be detected by motor development alone.
When you see your pediatrician for a well visit (which will happen many, many times in the first three years), they’ll be looking for signs of these serious motor delays. But what signs, exactly, and how?
First, at any visit, your doctor will poke around at the baby, see about their muscle development, do various baby manipulations (your baby will not like this). They’ll look for good reflexes, for good movement “quality.” This is an important part of the evaluation, although pretty hard to quantify (and extremely difficult to evaluate on your own).
In addition, doctors will look for some basic developmental milestones at each visit. Here are some examples from the 9-, 18-, and 30- or 36-month visits.
Visit
Milestones
9 months
Rolling both sides, sitting with support, motor symmetry, grasping and transferring objects between hands.
18 months
Sitting, standing, and walking independently; grasping and manipulating small objects.
30 months
Subtle gross motor errors, looking for loss of previous skills (marker of progressive disease).
The 9- and 18-month milestones are the most crucial here; by 30 months, most major issues have been well identified, and doctors are looking for smaller things.
Nearly all children will have achieved these milestones by these points. Typically, developing babies roll over between 3 and 5 months; if they have not rolled over by 9 months, that is definitely outside the normal. Similarly, although typical development calls for walking between 8 and 17 months—with an average of 12 months—looking at 18 months catches children who are outside the norm.5
Setting up formal assessment times is valuable to make sure children with delays are not missed, but a good pediatrician will be evaluating your child’s motor development at all visits, and they’ll be looking for places where your child is out of the normal range on any particular milestone, or especially on two or more.
What are these normal ranges? For that, we can go to the data. The World Health Organization, using data from six countries, calculated the range of the 1st percentile to 99th percentile for each of a variety of outcomes among healthy children. The children they studied do not have diagnosed motor issues, so their argument is that this can be seen as the range of normal development.6
Milestone
Range
Sitting without support
3.8 months to 9.2 months
Standing with assistance
4.8 months to 11.4 months
Crawling (5% of kids never do)
5.2 months to 13.5 months
Walking with help
5.9 months to 13.7 months
Standing alone
6.9 months to 16.9 months
Walking alone
8.2 months to 17.6 months
From this data, we see the logic for Dr. Li’s suggestion that we wait for 18 months before panicking about walking, and we see the very wide normal ranges on almost all of these. Standing alone, for example, occurs any time between 7 and 17 months. This is an eternity in baby time!
Your doctor will be very focused—correctly—on the upper ends of these ranges. But what if your kid is walking really early—like, at 7 months? Does this mean they are going to be an amazing athlete? And what if they’re at the older end of the normal range—doomed to being picked last for the kickball team?
There is, in fact, very little evidence on the long-term impacts of late walking. Virtually all children—indeed, even the vast majority of those who are delayed—do end up walking and running. If you ask, “Does early walking predict walking?” the answer will be, “No, everyone walks.”
; When it comes to being an elite athlete, there is just nothing. I don’t know if it is just that researchers are not interested in predicting elite athletic performance. Perhaps the issue is that even if there were some relationship, the outcome is so unlikely, we’d never see it in the data. The Olympics, we find, are just not a realistic goal for most people. Thanks, data.
There is simply nothing in the data that would make us think that earlier walking or standing or rolling or head raising is associated with any later outcomes. Looking for delays is a good idea; looking for exceptionalism, or worrying about a child who is at the end of the normal range, is probably not.
ILLNESS
Although not technically a milestone, baby’s first cold is definitely a moment for a parent. A bad one. Then there is baby’s second cold, baby’s third cold, and on and on.
As the parent of a young child, you will spend the period from October to April drowning in a lake of snot. To many of us, it may seem that our child has a cold, or possibly some other illness, literally all the time. If you have two children or, god forbid, more than two, the winter months are a haze of repeated illnesses: you, kid 1, kid 2, your partner, back to kid 2, now kid 1 again. Usually there’s a dose of stomach flu somewhere in the middle (you all get that, obviously).
This can naturally leave you wondering, Is this normal? Is everyone else spending their life savings on tissues with lotion, too?
Basically, yes.
Kids younger than school age get an average of six to eight colds a year, most of them between September and April.7 This works out to about one a month. These colds last on average fourteen days.8 A month is thirty days. So in the winter, on average, your kid will have a cold 50 percent of the time. On top of this, most kids end their cold with a cough that can last additional weeks. It adds up.