Hofstetter's age-amplitude formula is one of the first things every optometry student memorizes: 18.5 minus 0.3 times age. It's fast, it's simple, and it's been the default reference for expected amplitude of accommodation since 1950. You probably haven't questioned it since your board exams, and I know I certainly see it in infographics shared here on LinkedIn.
Here is what none of you learned: Hofstetter's formulas weren't discovered using children.
Munsamy et al., in their 2023 meta-analysis, report that the original dataset was overwhelmingly comprised of adults. Of the pediatric data folded in, only 33 children were included at all, and every one of them was over the age of 8. There was no group of 5-year-olds, no group of 7-year-olds, sitting in a room having their amplitudes measured. The pediatric predictions came from statistically extrapolating an adult model downward. That was a reasonable thing to do with the tools available in 1950, but it is not the same thing as truly measuring children, and as such, it should have been left behind in the last century.
Sit with that for a second, because it's easy to read past it. Every time you've referenced "amplitude should be around X for a 6-year-old," you were referencing a number that was never actually derived from a 6-year-old. It was derived from adults, run through a formula, and handed down as pediatric gospel for seventy-plus years. Nobody checked it against real children at scale until recently, because, well....
What the data shows
Munsamy et al. pooled five studies spanning 6,276 children across five continents. All children were measured with the same push-up technique Hofstetter used, so the comparison would be apples to apples. They then ran the numbers against Hofstetter's predicted average for each age group.
At six years old, measured amplitude came in 3.4D below what Hofstetter predicts. At nine, the gap was 4.1D. At ten, 4.6D. At eleven, 5.2D. All four differences were statistically significant.
That's not noise. That's four separate age groups, five independent studies, and one consistent direction: the formulas you were taught are predicting more accommodative capacity than modern children actually have.
Think about what that means in the exam chair. If a 10-year-old walks in and measures 12D of amplitude, Hofstetter's formula says they should be closer to 15.5D, and you might flag that as a meaningful deficit worth chasing. But if the real-world expected value for a modern 10-year-old is closer to 13D, that same child is sitting almost exactly where they should be. You're not looking at the same child differently. You're looking at the same child through two different lenses and only one of them was ever tested on someone their age.
Why this isn't a fluke of one study
The Munsamy paper isn't operating in isolation. Sterner et al. found a 3.5D average reduction in 6-10 year olds using the same method. Ikaunieks et al. found children measuring roughly 3.0D below Hofstetter's expected values across ages 7-15. Hashemi et al., studying over 5,000 Iranian schoolchildren, found the same pattern and went as far as proposing an entirely revised formula for their population: 16.59 minus 0.23 times age, a flatter curve than Hofstetter's, suggesting the decline in amplitude with age isn't as steep as we were taught either.
Different countries, different research teams, different decades within the same 21st century window, and they keep landing on the same conclusion: kids today are measuring lower than a formula built on 1950s data says they should. This is a pattern that keeps reappearing every time someone bothers to look into the information.
It's also worth noting what the researchers themselves consider once they see this gap. Munsamy et al. raise a real possibility: some of what looks like reduced amplitude compared to Hofstetter's predictions might reflect an actual increase in accommodative lag among modern children, not a true drop in maximum capacity. In other words, the gap between formula and finding isn't just a math correction. It might be pointing at something clinically real happening in how children's visual systems are functioning under today's demands.
What this means clinically
The authors of the meta-analysis put it plainly: practitioners should be cautious about expecting young children to hit Hofstetter's numbers, and should lean on complementary testing rather than a single formula to avoid under-diagnosing real accommodative dysfunction. That's not a radical position. That's just what happens when you let the data update the model instead of the other way around.
Hofstetter wasn't wrong, and this isn't exactly a call to burn down his formulas. He built the best model he could with the population and the tools he had access to in 1950. But his studied population never opened a laptop and stared at it with bad posture for seven hours a day. This piece should serve as a wake up call to re-evaluate what you are doing clinically. Think about what else we might have inherited from that era, taught as collective wisdom, that hasn't actually been tested since, and that needs to be revamped around what modern visual science has since made clear.