The MAHA Report: A Generation at Risk?
Part II of Reading the MAHA Report so You Don't Have to
On Feb 13 2025, President Trump issued an Executive Order to create the Make America Healthy Again Commission, and directed them report on the “childhood chronic disease crisis”. The Commission was given 100 days and on May 22, their MAHA Report was released to the public.
At 73 pages long, most people won’t read the whole report. And even if they do, it’s not easy to parse. Many claims are misrepresentations of real studies, and others are backed up only by fake citations to studies that never existed. But this is exactly what E is for Epi is here for! To help you make sense of public health claims.
Which is why I’m reading the MAHA Report; so you don’t have to. Yesterday, I posted Part I of my reactions1. Here is Part II. Enjoy!
P.S. Don’t forget to Subscribe so you don’t miss out on Part III.
The Chronic Disease Crisis: A Generation at Risk
We’re now getting into the meat of the report, with a section that (presumably) reviews the evidence for a chronic disease crisis in American children. This is going to be a bit longer than yesterday’s article, but it’s also gonna be juicy!
This section of the Report kicks off with some actual data on the scope of the crisis (finally!):
Over 40% of the roughly 73 million children (aged 0-17) in the United States have at least one chronic health condition, according to the CDC, such as asthma, allergies, obesity, autoimmune diseases, or behavioral disorders.
Over 40% is a big number, especially given the article they cited in the Introduction which showed that less than 30% of American children had a chronic condition in 2018. An increase from less than 30% to over 40% in less than 10 years would certainly be alarming! But is it true?
First, let’s take a look at the source they cite for this statistic:
Right away I notice 3 things:
The citation says the data is from 2018-2019.
Even though the text claims that the “over 40%” statistic is “according to the CDC”, this citation is not to the CDC. Instead it’s from “ChildHealthData.org”; but if you look that up, you’ll find that it’s an interactive data system for a survey conducted by the US Department of Health and Human Services. So it is government data.
For reasons inexplicable to me, you, and anyone else, they accessed this data on February 24, 2021. Over 4 years ago. Why? Did no one think to check if there were more current data available while writing this report? Did they just reuse this text from somewhere else? Your guess is as good as mine.2
These raise some red flags about the validity of this statistic. Which means my next step is to visit the website they claim is the source of the data and see if the data are there.
This is clearly not the actual data, since it’s just the landing page, but on the left-hand side of the page we can see a link to the data query system for the National Survey of Children’s Health. Interestingly, the default settings on the data query are for the 2022-2023 Nationwide data. If the Report authors had bothered to check, they could have had the 2023 data instead of 2018-2019 data. Let’s see what those 2022-2023 data say, shall we?
When we click on “continue” we’re given some additional data query options. For transparency, I chose: “Child and Family Health Measures” → “Physical, Oral Health, and Functional Status” →”Indicator 1.9: One or more health conditions”. Here’s the result:
The table says that 59.4% of respondents had no “current or lifelong health conditions”, 20.2% had “1 current or lifelong health conditions”, and 20.4% had “2 or more current or lifelong health conditions”.
Well, that does look like (just) over 40% reporting a health condition, but if that’s the case, then the prevalence of “current or lifelong health conditions” didn’t jump from less than 30% to over 40% in a decade. Instead, the prevalence seems to have stayed about constant. Note that if we check 2018-2019 (the time period they cite) in the same data query system we get 19.0% with 1 condition and 18.7 with 2 or more conditions which by my calculation is actually NOT larger than 40%, but it is close.
Since we have access to the data here, it’s also worth asking what health conditions are included. According to the information button, a child has a current or lifelong condition if their parent answered yes to any of the following 25 questions in 2022 or 24 questions in 20233. The list of conditions is:
Allergies (food, drug, insect, seasonal, or other) - ever and currently?
Asthma - ever and currently?
Autoimmune disease (such as Type 1 diabetes, celiac, or juvenile idiopathic arthritis)?
Blood disorders (such as sickle cell disease, thalassemia, or hemophilia), ever?
Cerebral palsy, ever?
Cystic fibrosis, ever?
Type 2 diabetes - ever and currently?
Down Syndrome, ever?
Fetal Alcohol Spectrum Disorder (FASD), ever?
Epilepsy or seizure disorder - ever and currently?
Any other genetic or inherited condition (2022 only)?
Heart condition - ever and currently?
Frequent or severe headaches, including migraine - ever and currently?
Tourette Syndrome - ever and currently?
Anxiety problems - ever and currently?
Depression - ever and currently?
Behavioral or Conduct Problems - ever and currently?
Developmental delay - ever and currently?
Intellectual disability “(formerly known as Mental Retardation)” - ever and currently?
Speech or other language disorder - ever and currently?
Learning disability - ever and currently?
“Autism or Autism Spectrum Disorder (ASD), including diagnoses of Asperger’s Disorder or Pervasive Developmental Disorder (PDD)” - ever and currently?
“Attention Deficit Disorder or Attention-Deficit/Hyperactivity Disorder, that is, ADD or ADHD” - ever and currently?
Blindness or problems with seeing, even when wearing glasses?
Deafness or problems with hearing?
This is a pretty long list of conditions, and both the potential severity and potential causes of the various things on this list range very widely.
And it turns out that we can actually get the frequencies for each of these conditions individually from the data query system.4 When we do that, we see that 22.0% of kids in 2022-23 had some type of allergy alone. And while allergies can be really severe and even life-threatening, it’s worth noting that the allergy question here asks about any type of allergy at all including seasonal allergies. And in fact, only 8.7% of kids overall were reported by their parents as having moderate or severe allergies.
Something else on this list which stands out to me is “Behavioral or conduct problems” as identified by “a doctor, other health care provider or educator” including teachers and school nurses. This isn’t what I personally would think to add to a list of chronic health conditions, but since it’s on the list, I wanted to see what the frequency was. Turns out that 9% of children ages 3 -17 have ever or currently been told they have behavioral or conduct problems. That’s a significant chunk of our 40% of kids with chronic health problems!
I also think it’s worth noting what isn’t on this list. For example, cancer. This, I think, is something most parents would be concerned about when worrying about children’s health. It’s also worth noting that obesity isn’t on this list, which seems like it’s something the Report authors would have chosen to include if they had analyzed these data themselves (which, incidentally, they could have, and you could too, because the data are freely accessible!).
Okay, let’s stop looking at this data (even though data are super fun to look at!), and return to the report. Remember, we were only on the very first sentence of the section entitled The Chronic Disease Crisis. The next sentence tells us:
Although estimates vary depending on the conditions included, all studies show an alarming increase over time.
The authors offer 2 citations for this “alarming increase”. One is the Wisk & Sharma study we looked at yesterday, where, if we’re being honest with ourselves, there was only a gradual increase over time. The other is a study from 2011, and while this too is a real study, it is not a study of child health over time. Instead, it’s a snapshot of the National Survey of Children’s Health from 2007 with comparisons across states and across insurance status. Not an appropriate citation for an “alarming increase over time”.
So far, not so good. What do the authors say next? Much like in the Introduction, they pivot from asserting the scope of the problem to telling us why they think we should care: because, apparently, “[o]ver 75% of American youth (aged 17-24) are ineligible for military service — primarily due to obesity, poor physical fitness, and/or mental health challenges.”
This is accompanied by a terribly low-resolution image of a pie chart of what is presumably the reasons for military service ineligibility. I say “presumably” because the graphic is so fuzzy that I can barely read the caption, and even that doesn’t actually seem to mention military service. I’m not going to waste your time reposting this ridiculously poor-quality graphic here. The citation does link us to a 2020 military news article, but this article does not include the graph.
But I do want to point out two things. First, in this pie chart, the top reasons for ineligibility (assuming, that is what this graph is about) are “multiple reasons” (44%), overweight (11%), drug abuse (8%), and medical/physical health (7%). Poor physical fitness isn’t actually even on the pie chart, despite the Report text claiming it’s a top 3 reason. And mental health alone accounts for only 4% of the pie; not a top 3 reason.
For those of you who are counting, we’re at 0 of 3 on correct statements so far on this page alone.
Childhood Obesity is a Worsening Health Crisis
On to the second page of The Chronic Disease Crisis. This page focuses on childhood obesity, and includes 4 bullet pointed factoids plus one graph.
IMO it’s an interesting choice to focus first on childhood obesity, given that this is NOT one of the health conditions included in their estimate that 40% of American children have a chronic condition, but to each their own. Let’s look at the 5 claims they make here:
Today in the U.S. more than 1 in 5 children over 6 years old are obese. This is a more than 270% increase compared to the 1970s, when less than one in twenty children over 6 were obese.
Rates of severe obesity increased by over 500% in the same period.
The U.S. obesity rate is, on average, more than double that of its G7 peers.
Approximately 80% of obese teens will be obese into adulthood.
Around 70% of youth with obesity already have at least one risk factor for heart disease.
Where do these claims come from, and are any of them true? Let’s find out.
The authors give two citations in support of the first claim. What’s peculiar about these two citations is that they are actually both to the same document: a CDC summary document on the prevalence of overweight, obesity, and severe obesity in American children over time. This document does say that 19.3% of children surveyed had obesity in 2017-2018 versus 5.2% in 1971-1974.
And yeah, that is technically a 270% or 2.7-fold increase (you calculate that as: 100*(0.193-0.052)/(0.052)=271), although you could also describe it as a 14 percentage point increase. The rates of severe obesity also match the citation going from 1% in 1971-74 to 6.1% in 2017-2018 (again, technically a 500% or 5-fold increase, but also a 5 percentage point increase).
But, like before, I can’t help but wonder why they aren’t using more recent data? If this is a crisis, where’s the urgency?!
The next claim is that the situation in the US is worse than in other G7 countries. This is not given a reference, but there is a graph showing obesity prevalence over time in G7 countries. The graph apparently comes from the WHO and it does show US obesity rates in children and adolescents in 2022 are roughly twice as high as in the other G7 countries, but there’s also quite a range between those countries: from somewhere around 4% to maybe 12%, while the 2022 US rate looks like it’s about 21%.
But also, IDK about you but this is not a graph which depicts a run-away increase in obesity over time to me. In fact, it actually looks like obesity prevalence plateaued back in 2006-7, and then maybe even started to decrease a little.
The third claim is that approximately 80% of obese teens will be obese into adulthood. This claim comes from a 2016 systematic review and meta-analysis, and while I’m not going to go into that whole paper, I will note that this paper also says that 70% of obese adults were NOT obese in childhood or adolescence, and that childhood obesity may not actually be a good predictor of adult obesity.
The last claim in this section is that 70% of youth with obesity have at least one risk factor for heart disease. This statistic comes from a 2007 study using data from a cohort study based in a rural community in Louisiana, and for which the children were surveyed between 1973 and 1994. The only risk factors assessed were triglycerides, LDL and HDL cholesterol, insulin levels, and blood pressure. Is this really the best data on children’s cardiovascular health the Commission could find?
So of the 4 obesity-related factoids, one is misleading and out-of-date, the second is just misleading, and the last two are selectively reported. But, at least none of them are outright fakes?
Substack is warning me that I’m near the email limit already, and I don’t want to exhaust you (or me), so I think I’m going to stop here for today even though we’re only on page 10 of this Report.
The tl;dr so far:
We haven’t actually encountered any fake citations yet.
But, most of the claims made by the report so far are either out-of-date, exaggerations, incorrect, or a combination of the three.
What will we find tomorrow? Stay tuned!
As an independent epidemiologist, I make my living entirely through reader support. Readers like you make E is for Epi possible. Thank you to all the paid subscribers. If you find value in this work, consider becoming a paid subscriber too!
Reminder: I downloaded the MAHA Report from the government’s website on June 1 at 10:49am. They occasionally revise and repost the Report, so, there’s a possibility things I discuss here will have been changed or removed from any version of the Report you go download later.
Where indicated, a parent had to answer yes to both “ever?” and “currently”; parent recollections were not verified against medical records. For Questions 1-16: “Has a doctor or other health provider … told you that this child has"; Questions 17-23 : “Has a doctor, other health care provider or educator ever told you… Examples of educators are teachers and school nurses.”; Questions 24-25: “Does this child have any of the following”
Fantastic reporting Ellie. Thank you. I learn not just about the report, but also about assessing research (and I have done a lot of research myself). A question, is it an epi thing that a fold increase is the same as a percentage increase? I would read the data as a 2.7 fold increase or 270% increase, not 270 fold increase.
So according to that list I would have been a child with multiple health problems because I was allergic to dust and required a short period of therapy to learn to say my R's. I can't say either of those affected my quality of life much at all other than sneezing a lot when I had to help clean the basement.
As a scientist myself the sheer lack of quality in this report is like one of those movies that's so bad it's almost good again. Imagine having the confidence to put out something that inept! Do the mystery authors think they did a good job, or do they know it's garbage? I really want to know.