Fairly recent guidance from the American Academy of Pediatrics is to routinely screen female adolescents and menstruating individuals one after menarche and by age 14 both for anemia and a test of iron stores called ferritin. Cut off for action includes no anemia and ferritin level below 30.
Here’s the problem: something like 70% of menstruating adolescents are below 30 of ferritin.
Logic of recommendation includes exactly that it is very common in that specific demographic and that reporting low iron stores can reduce the fairly low risk of developing anemia in the future. Identifying those with low ferritin can offer an opportunity to discuss better nutrition and possibly to investigate excessive menstrual losses. And to supplement selectively at doses sufficient to replenish which a regular MVI with iron won’t accomplish, plus iron supplements are not side effect free or always safe. And there may be some decrease in risk of anxiety when ferritin is over 30. FWIW we aim for 50 for endurance athletes. And if symptoms like fatigue.
Still the vast majority of menstruating adolescents without anemia or symptoms and with isolated mildly low ferritin would be just fine. As this guidance gets more universally implemented the majority of menstruating adolescents will fall in that group: now identified as iron deficient and getting treated for that as a problem. Patholgized. And labeling something as a pathology is well established to increase anxiety.
I suspect many of this crowd is past have teens in their household but hopefully can still remember?
We are starting to implement the guideline more consistently in our group. Is my concern about the harms of pathologizing a result that statistically is normal justified. I’m personally wanting to be very careful to phrase the reporting of the result not a an abnormal number, because it actually isn’t, not as a pathology, not a problem, but an identified area we can better optimize. Worth doing; not disputing the guideline.
And I am concerned that more broadly it will be presented as a pathology.
I do have a teen in that age range. I’m also conversant in statistics and scientifically minded, so my reaction may be different than other parents. The doctor could say several things, and I’ll have different reactions
Most teens are deficient in these numbers, and that by itself isn’t a problem, but it means we should pay attention and watch for problems
What are the signs of a problem, what should I watch for?
Why don’t you just test for the problem?
Teen: can we get steak after this
Most teens are deficient in these numbers, and we should treat it even though it is common issue
What are the downsides of treatment?
Is this some moral panic blaming phone screens for iron levels, is there any real evidence behiind this?
Teen: can we get steak after this
Most teens are deficient in these numbers, and as a doctor I’m telling you not to worry about it
Do you know more than the panel of experts pushing this?
I’m being serious and not just making an appeal to authority
I trust you as a doctor, but why are you even recommending the tests if it doesn’t test for something that you think is an actual problem
Teen: can we get steak after this
Most teens are deficient in these numbers, and it is a moral failing in today’s youth
Opens insurance app to find new GP
Anyway, how is this different than when the same doctor as my kid visits put me on amladopine and rosuvastatin? Most, many? people of my age have high blood pressure and high cholesterol. No aggressive pathologizing by the doctor (“getting these numbers down will help prevent stroke and heart disease”), just told me that if I get any side effects from the drugs we could try other ones, and the benefits greatly outweigh any downsides.
A good analogy but includes that many feel that starting a statin is having failed to achieve control by nurtrition plan and exercise. And sometimes is!
My eldest now 40 indeed saw his starting a statin as a problem, decline, getting old. And he very well appreciates our family history and the benefits of having more optimal LDL. It is still pathologized.
And I do wonder the same as an adult about high blood pressure, where statistically it is less common to not have hypertension than to have it, by age 60 anyway. But doing something about it still statistically improves outcomes.
For now my phrasing is that “there is no anemia but the iron stores are below what we are wanting as optimal so let’s look at what can contribute to that.”
(And why is a bit higher than this more optimal? Lower risk of future anemia and maybe less risk of anxiety, which is pretty endemic right now!)
I still am not thrilled about calling most menstruating teens results as abnormal.
More the sense that everyone has pathology today as define the statistically normal as such. And it may be good medicine but it still bugs me?
This is a major issue in medicine. When they calculate things like NNT or NNH (number needed to treat, or number needed to harm, in case anyone wonders), they don’t factor in all the psychological trauma of pathologizing everything, or the excess spending.
For breast cancer, you may need to screen 2000 people to prevent 1 death. But of those 2000 people, its possible more than 1 will die from complications of medical care. Far more than 1 will be harmed from complications of the medical care, and far far more will experience psychological trauma and excess spending.
According to the Cochrane analysis, mammography decreases the risk of death from breast cancer by about 15% in relative terms, or 0.05% in absolute terms.
I really don’t understand the math of when a medical intervention does more harm than good, but it feels like they really underestimate not just the deaths from medical intervention, but also the serious injuries, psychological trauma and money spent.
I’ve developed multiple serious health problems due to properly administered medical intervention (not medical mistakes), I’ve known people who developed serious near fatal health problems from proper medical treatment. But it would require someone who understands statistics far better than me to determine where the line should be.
But I do think there is an overall institutional mentality to exaggerate the NNT and minimize/ignore the NNH.
“Abnormal” and other medical terms are statistically based and don’t sound neutral. “These results are pretty common, but not in the healthiest range.”
If anyone had looked at these levels when I was a teen, it might have been identified that I had terribly heavy periods (my report was insufficient to convince anyone), and perhaps would have identified low ferratin and anemia much earlier in my life.
As a scientist, I feel like it’s best to both know the facts, and to know the accurate statistical implications of those facts. But I also recognize that most people don’t think that way, and that something I regard as mere data to be analyzed in the context of all of the other data I have, might be a source of anxiety and distress for most people.
That is exactly the rationale for the guideline recommendation. It gets the specific history attended to: seven or more days of bleeding, five or more pads a day … then check for certain bleeding disorders, simple Endo screen, iron supplement, consider hormonal intervention … so on.
I know most of us will do okay at phasing it in terms of “healthiest” and “optimal” and … a majority of menstruating teens - it’s a lot of follow up after lab results come back conversations with some that are sure to not be heard exactly how we want them to be heard, some that will just be MyChart communications.
Oh, and “abnormal” is definitely the wrong term to use, here. Even aside from the negative connotations, something that happens to 70% of the relevant population by definition isn’t “abnormal”. The solution might be to give the accurate information, but to be careful to use the right language to describe it: “These ferritin levels are normal, but below the optimal levels. I recommend eating an iron-rich diet and watching for these possible symptoms of anemia”, or something like that.
Long out-of-practice Med Tech here. Two things come to mind: normal population ranges and algorithms. The normal ranges we reported with the test result(s) was based on the result average +/- 2sd for the population seen at that facility. Given the broad population treated, sometimes the normal ranges were inappropriate for a given individual. I wonder if the normal ranges for iron and ferritin need to be recalculated for the cohort of, say, ages 14-16 post menarche, then 16-18, then 18-24, etc.
I was also involved with PGX testing, and my lab developed an algorithm that assesed the patient’s metabolic capacity for classes of drugs based on an array of CYP450 gene analysis results. So, could a similar algorithm be developed for the cohort of interest that incorporates iron, ferritin. RBC count, MCH, MCHC, age, etc.?
Never blanket worship doctors, they can have their heads up their arses just like anybody. I fired my cardiologist for 2 main reasons - never managed to get my blood pressure under control in over a year, my primary managed to do it in 3 months. When I went back for my quarterly appointment, with the ‘fixed’ blood pressure issue, he then started leaning on me because I wasn’t on a statin. My numbers were text book good, and Yale Medical were using my echocardiogram in classes [at the time 55 years old, NO placque, no adhesions, perfect blood chemistry - the cardiovascular of a 20 year old] when I mentioned that my primary had taken me off statins [mainly because I love grapefruit and my cholesterol balance was wonderful] He mandated that “he used more stringent numbers” so I told him to do the test, and see what the results were, and if they were fine, to bugger off. Oddly, textbook perfect blood work and I told him to close my files and forward them to my primary, I was done driving all the way down to Yale New Haven to see a doctor that wasn’t doing anything for me.