The striking detail here is that the patient's post-shift numbers wouldn't have tripped the static alarms we currently use; a resting heart rate moving from 44 to 49 bpm is still "athletic" by population standards. What mattered was the synchronized shift in several metrics relative to his personal baseline.
Perhaps because statistics education attunes us to average over individual differences, any mention of "data" unfortunately makes people instinctively jump to this 'cross-sectional' corner... textbooks explain how the typical human body behaves, and standard education and medical practice respectively emphasize averaging and population-level guidelines.
I think this approach is limited in some fundamental ways, and forces a dilemma: either accept wide and unactionable bands for what counts as normal, or use narrow bands that over-pathologize most individuals as diseased. The natural consequence of this is to start "fixing" deviations, and this is the territory of dangerous over-steering. This is why the medical community is afraid of "incidentalomas" and over-intervention. Tests which are point-in-time and compare one human to others are problematic in this respect. This is especially acute when the body is measured along many axes: like a full body scan. Much like the US Air Force Study of pilots discovered that there is no single pilot compatible with the "average" pilot measurements, comparing metrics from an individual's full body scan to a cohort average is likely to drastically overestimate problems.
In some ways, learned systems (like our physiology which is constantly adapting to our individual environment and experiences) follow the opposite of the Anna Karenina principle: every healthy individual is healthy in their own way. So the real diagnostic leverage isn't comparing an individual to an aggregate human template, but measuring change against their own homeostasis, and this requires high temporal resolution, not wider population comparisons.
I'm currently exploring how to bring this kind of longitudinal tracking to higher-fidelity modalities like wearable imaging. If any clinicians or researchers here are thinking about how to operationalize personal baselines in practice, I'd love to chat (email in profile).
I can't believe so many people think these wearables are "accurate"; then they wonder why they aren't losing or gaining weight when they go by the expenditure estimates on these.
A 2020 systematic review by Fuller et al. found that consumer wearables often perform poorly at estimating energy expenditure, even when they measure steps and heart rate reasonably well. Across brands like Apple, Garmin, Polar, and Withings, calorie burn estimates frequently showed substantial error, often in the 15–40% range and sometimes exceeding 50%. Wearables can still be useful for tracking trends, but not for precise calorie calculations.
protip: if you actually want to do better, look into "adaptive TDEE".
This comment confuses me. Why would we expect calorie burn to be related to steps and heart rate? How could a wearable measure energy expenditure? What if someone was wearing a heavy backpack? The energy to move the backpack has to come from somewhere but the wearable can't tell that the person has one.
Is it really that confusing? Heart rate is correlated with oxygen consumption which is correlated with metabolic activity. The wearable would see the increased heart rate when a person is moving with a heavy backpack versus moving without one.
Now whether we can do that accurately with today’s wearables and without calibration to each individual is up in the air as far as I’m concerned. But it seems strange to me that you’re confused as to how we could even directionally infer these things.
The article is a case report. They’re almost always N=1. They’re just meant to contribute to the sum total of general knowledge of practicing medicine.
Why is it so surprising that continuous measurement is likely to catch a failing system? The more surprising belief is that sporadic random measurements at random times give you good enough predictive powers. Like all the blood work and vitals we do at the doctor’s office - I’m sure it captures some problems but more subtle deviations from baseline are much more difficult to suss out.
The striking detail here is that the patient's post-shift numbers wouldn't have tripped the static alarms we currently use; a resting heart rate moving from 44 to 49 bpm is still "athletic" by population standards. What mattered was the synchronized shift in several metrics relative to his personal baseline.
Perhaps because statistics education attunes us to average over individual differences, any mention of "data" unfortunately makes people instinctively jump to this 'cross-sectional' corner... textbooks explain how the typical human body behaves, and standard education and medical practice respectively emphasize averaging and population-level guidelines.
I think this approach is limited in some fundamental ways, and forces a dilemma: either accept wide and unactionable bands for what counts as normal, or use narrow bands that over-pathologize most individuals as diseased. The natural consequence of this is to start "fixing" deviations, and this is the territory of dangerous over-steering. This is why the medical community is afraid of "incidentalomas" and over-intervention. Tests which are point-in-time and compare one human to others are problematic in this respect. This is especially acute when the body is measured along many axes: like a full body scan. Much like the US Air Force Study of pilots discovered that there is no single pilot compatible with the "average" pilot measurements, comparing metrics from an individual's full body scan to a cohort average is likely to drastically overestimate problems.
In some ways, learned systems (like our physiology which is constantly adapting to our individual environment and experiences) follow the opposite of the Anna Karenina principle: every healthy individual is healthy in their own way. So the real diagnostic leverage isn't comparing an individual to an aggregate human template, but measuring change against their own homeostasis, and this requires high temporal resolution, not wider population comparisons.
I'm currently exploring how to bring this kind of longitudinal tracking to higher-fidelity modalities like wearable imaging. If any clinicians or researchers here are thinking about how to operationalize personal baselines in practice, I'd love to chat (email in profile).
I can't believe so many people think these wearables are "accurate"; then they wonder why they aren't losing or gaining weight when they go by the expenditure estimates on these.
A 2020 systematic review by Fuller et al. found that consumer wearables often perform poorly at estimating energy expenditure, even when they measure steps and heart rate reasonably well. Across brands like Apple, Garmin, Polar, and Withings, calorie burn estimates frequently showed substantial error, often in the 15–40% range and sometimes exceeding 50%. Wearables can still be useful for tracking trends, but not for precise calorie calculations.
protip: if you actually want to do better, look into "adaptive TDEE".
This comment confuses me. Why would we expect calorie burn to be related to steps and heart rate? How could a wearable measure energy expenditure? What if someone was wearing a heavy backpack? The energy to move the backpack has to come from somewhere but the wearable can't tell that the person has one.
Is it really that confusing? Heart rate is correlated with oxygen consumption which is correlated with metabolic activity. The wearable would see the increased heart rate when a person is moving with a heavy backpack versus moving without one.
Now whether we can do that accurately with today’s wearables and without calibration to each individual is up in the air as far as I’m concerned. But it seems strange to me that you’re confused as to how we could even directionally infer these things.
Isn’t that in line with the comment, that they aren’t accurate?
N = 1?
The article is a case report. They’re almost always N=1. They’re just meant to contribute to the sum total of general knowledge of practicing medicine.
Why is it so surprising that continuous measurement is likely to catch a failing system? The more surprising belief is that sporadic random measurements at random times give you good enough predictive powers. Like all the blood work and vitals we do at the doctor’s office - I’m sure it captures some problems but more subtle deviations from baseline are much more difficult to suss out.
Yes, this is a „case study“.