-->

Tuesday, September 30, 2008

Our Pediatrician Needs a Crash Course in Statistics and Inference

Chris took Gioia to the doctor today to discuss a little rash on her face. No biggie there, but they took her vitals for her records (as they always do when we come). Turns out she has gained 0.4 kilos since her 15 month visit, which was a month ago. So the doctor says, "She has gained too much weight this month. You need to watch what she eats. No more rice or noodles. Only fruit and veggies and meat."

Um, what? Before we dissect Gioia's diet, let's take a step back and see if the conclusion that Gioia is gaining too much weight holds up to scrutiny.

Data Review #1 - Peer Group Comparison

First, let's look at how she compares other 16 month old females.
  • Height 32.5 inches/ 83.6 cm (88th percentile)
  • Weight 11.5 kilos/ 25.3 pounds (75th percentile)

So the data confirm what we can tell by just looking at her: Gioia is big for her age. But not fat. She is very tall and of an appropriate weight fer her height. Taller than 88 percent of her peers, and heavier than 75 percent of them.

If she were in the 75th percentage in weight, but 40th percentage in height, then we might have an issue.

Data Review #2 - The Historical Record

Second, looking at Gioia's historical weight data, we see that she has been between the 68th and 81st percentile in weight since we brought her back to Taiwan.


Data Review #3 - The Height Perspective

Third and finally, the same pattern holds in her height data: she has been in the 82nd to 94th percentile since February.


So she has been tall (OK, really tall) for a long time. Again, if she were heavy and short, that would be one thing. But she is not.

To top it all off, here is a nice little internet tidbit: "Toddlers aged between six and eighteen months can have widely fluctuating growth curves because the rate of growth is higher in this period. So the doctor will especially be looking for changes in the percentile ratings to see if the relative size of your baby is stable."

Conclusion: Dr. X's* assessment that a one month weight gain of 0.4 kilos (taken in isolation) is cause for a diet was not consistent with the data. I have a sister-in-law who is a pediatric nurse and a brother who is an internal medicine resident, so I am sure that if I am wrong here someone will correct me. Please do. But for now, I stake my claim that my daughter's doctor needs a refresher in statistical inference.

She eats a balanced diet - no junk food - with lots of veggies, fruits, and other healthy things. She is fine.

For those of you who have made it through this boring rant, I leave you with a sneaky night shot. Note the position of her beloved bunny. Today, we fed bunny lunch and tried to put shoes on him, all at Gioia's request.


*Not his (or her?) real name. Ha.

2 comments:

Leslie said...

SHe looks so comfy! I wish I slept that good. She is so beauitful. That doctor must be crazy. She looks perfect for a 16 month old.

popo said...

She sure looks perfect to me. You are doing a perfect job of mothering and nuturing her too.