Building Better Growth Curves
By William E Bennett
Current standards for assessing growth in infants and children often raise unwarranted concerns. Better models could improve care.
This Article From Issue
July-August 2023
Volume 111, Number 4
Page 216
I first met Michael years ago, when he was just over a year old, the first and only child of a young Amish couple. He was admitted to our children’s hospital to figure out why he was growing so slowly. I was the supervising gastroenterologist that week, but he had already seen several of my colleagues, including other specialists. His diagnosis was something doctors call failure to thrive, a term plucked out of Victorian-era medicine that has no clear consensus definition and is simply what we say when a child isn’t growing as we expect.
QUICK TAKE
- Growth curves are a standard screening tool in pediatric clinics around the world. However, those using these curves end up flagging healthy kids as having potential problems.
- Failing to understand the patient’s context when they are not growing as expected can lead to unnecessary testing, expense, time, parental anxiety, and even risks to the patient.
- Improving growth curves is no easy task, requiring the use of longitudinal data and more diverse datasets. Personalizing growth curves is theoretically possible with machine learning.
American Scientist Comments and Discussion
To discuss our articles or comment on them, please share them and tag American Scientist on social media platforms. Here are links to our profiles on Twitter, Facebook, and LinkedIn.
If we re-share your post, we will moderate comments/discussion following our comments policy.