The offsides analogy captures something we don't discuss enough: the rule didn't change, but the category did, because the instrument measuring it became more sensitive than the phenomenon it was designed to detect. That's a more precise way of describing diagnostic expansion than simply saying "we're finding more disease."
Neurology has its own version: TIA. Before diffusion-weighted MRI, the definition was purely time-based. Once MRI became routine, many patients with transient symptoms were found to have infarction, and the same clinical event could now be labeled either a TIA or a minor stroke depending on which definition was used. The disease didn't change. The camera did.
The difference from soccer is that in medicine the consequences of crossing the line are much greater. A more sensitive threshold doesn't just change the call—it changes who acquires a diagnosis, who gets treated, and who carries that label for life.
But medicine is not entertainment. Increased precision and more data are generally beneficial in science, and I would argue in medicine as well. Let’s improve data integration and decision support rather than resign ourselves to low resolution inputs.
Not to forgot the VAR is an non-perfect interpolated measure too: mismatch in measurements from the ball sensor (sampling rate around 500 Hz) and the tracking camera (50-100 fps) means limb positions must be interpolated (with an approx 10-20 ms gap). A foot or leg (when sprinting or kicking, at 15-20 m/s as example) can travel 17-35 cm in that window, with measurement uncertainty to 28-40 cm with some cases.
Another specific example wpuld be the move from cardiac troponin (cTn-I/T) to high sensitivity troponin (hs-cTn) assays probably (where before if there was a genuine but smaller lesion this would probably not be detected or be labeled as unstable angina)
Sportsball will continue to adapt as new technology is introduced. However, unlike in medicine, as the athletes, coaches, referees and the media utilize new technologies, the changes will be absorbed into the highest level of play - likely for the better. Does anyone here miss "the chalk flew up" tantrums of a bully like John McEnroe back in the days before Wimbledon had implemented technology as the line judge? I, for one, do NOT! I recommend Suzanne O'Sullivan's book "The Age of Overdiagnosis" which introduces us to the harms and waste associated with the trinity of "overs" including overdiagnosis, overmedicalization and overdetection that seems to serve the purpose of the business of medicine without the goal of improving lives for individuals or the community - all things she had witnessed first hand in her decades long career as a neurologist. In medicine, unlike in sports, these wasted resources could be put to meaningful use.
To be pedantic, the offside rule isn't fixed. It has changed several times in my lifetime. But more seriously, a test that is more specific is universally a good thing because not treating people without disease is goog. However, more sensitive tests (your brain MRI example) pick up cases never picked up before and therefore never studied before. It would be foolish to think that therapies that work on people detected by the old test will necessarily be effective in this new group.
Just a note that, technically, an "elbow past the defender" is not considered offsides, since offsides is only judged by body parts you can legally score with.
The main issue I have with sports technology is its pretense of precision. There’s no evidence that an offsides system can reliably measure to 1mm or less even in a theoretical test. Making some questionable assumptions like the playing surface and the ball being perfectly smooth, fixed, incompressible objects, perfectly calibrated cameras with no wind vibration or thermal expansion of the objects they’re mounted to, you can get maybe 1cm. Add real world conditions with 22 players, camera obstruction and we are talking 10cm, maybe even 15cm. I expect wild exaggeration from the companies that manufacture the systems but I don’t really understand why the sports officials are so credulous.
I agree (and to my surprise similar technology has also metastasized to baseball!) This philosophy should instead be applied to enhance critical appraisal so that journals and editorials become more selective to high quality studies.
So, like purified water still has coliforms, the touted disease free periods in oncology are misnomers? Birds Eye view vs satellite imagery. Mammograms with CAD vs without. Our technology advances and resets the ranges, remember never use the word “ normal “, making it more challenging to interpret and categorize whats happening because we all want black and white only, greys need not apply.
The offsides analogy captures something we don't discuss enough: the rule didn't change, but the category did, because the instrument measuring it became more sensitive than the phenomenon it was designed to detect. That's a more precise way of describing diagnostic expansion than simply saying "we're finding more disease."
Neurology has its own version: TIA. Before diffusion-weighted MRI, the definition was purely time-based. Once MRI became routine, many patients with transient symptoms were found to have infarction, and the same clinical event could now be labeled either a TIA or a minor stroke depending on which definition was used. The disease didn't change. The camera did.
The difference from soccer is that in medicine the consequences of crossing the line are much greater. A more sensitive threshold doesn't just change the call—it changes who acquires a diagnosis, who gets treated, and who carries that label for life.
Catering to the few who demand more accuracy often comes at the expense of why something is done in the first place.
But medicine is not entertainment. Increased precision and more data are generally beneficial in science, and I would argue in medicine as well. Let’s improve data integration and decision support rather than resign ourselves to low resolution inputs.
Not to forgot the VAR is an non-perfect interpolated measure too: mismatch in measurements from the ball sensor (sampling rate around 500 Hz) and the tracking camera (50-100 fps) means limb positions must be interpolated (with an approx 10-20 ms gap). A foot or leg (when sprinting or kicking, at 15-20 m/s as example) can travel 17-35 cm in that window, with measurement uncertainty to 28-40 cm with some cases.
Another specific example wpuld be the move from cardiac troponin (cTn-I/T) to high sensitivity troponin (hs-cTn) assays probably (where before if there was a genuine but smaller lesion this would probably not be detected or be labeled as unstable angina)
Sportsball will continue to adapt as new technology is introduced. However, unlike in medicine, as the athletes, coaches, referees and the media utilize new technologies, the changes will be absorbed into the highest level of play - likely for the better. Does anyone here miss "the chalk flew up" tantrums of a bully like John McEnroe back in the days before Wimbledon had implemented technology as the line judge? I, for one, do NOT! I recommend Suzanne O'Sullivan's book "The Age of Overdiagnosis" which introduces us to the harms and waste associated with the trinity of "overs" including overdiagnosis, overmedicalization and overdetection that seems to serve the purpose of the business of medicine without the goal of improving lives for individuals or the community - all things she had witnessed first hand in her decades long career as a neurologist. In medicine, unlike in sports, these wasted resources could be put to meaningful use.
Love this post!
Sports analogies are the best analogies.
To be pedantic, the offside rule isn't fixed. It has changed several times in my lifetime. But more seriously, a test that is more specific is universally a good thing because not treating people without disease is goog. However, more sensitive tests (your brain MRI example) pick up cases never picked up before and therefore never studied before. It would be foolish to think that therapies that work on people detected by the old test will necessarily be effective in this new group.
Just a note that, technically, an "elbow past the defender" is not considered offsides, since offsides is only judged by body parts you can legally score with.
The main issue I have with sports technology is its pretense of precision. There’s no evidence that an offsides system can reliably measure to 1mm or less even in a theoretical test. Making some questionable assumptions like the playing surface and the ball being perfectly smooth, fixed, incompressible objects, perfectly calibrated cameras with no wind vibration or thermal expansion of the objects they’re mounted to, you can get maybe 1cm. Add real world conditions with 22 players, camera obstruction and we are talking 10cm, maybe even 15cm. I expect wild exaggeration from the companies that manufacture the systems but I don’t really understand why the sports officials are so credulous.
Excellent analogy. Thank you
I love the play btw philosophy and medicine!! So good. Thank you.
Great analogy.
Is it “offsides” or “offside”? I always thought it was the latter, but during this World Cup have heard both.
If you are American, it is offsides. If not, then offside. For Americans that comes from American Football. For the rest, it comes from football.
I agree (and to my surprise similar technology has also metastasized to baseball!) This philosophy should instead be applied to enhance critical appraisal so that journals and editorials become more selective to high quality studies.
So, like purified water still has coliforms, the touted disease free periods in oncology are misnomers? Birds Eye view vs satellite imagery. Mammograms with CAD vs without. Our technology advances and resets the ranges, remember never use the word “ normal “, making it more challenging to interpret and categorize whats happening because we all want black and white only, greys need not apply.
I say, "Touché!" to Vinay's brilliant comment! — :)