To be a “good doctor” I must order microalbumins on every diabetic once a year, lipid panels on everyone once a year now (is a company contract thing) even if the patient swears they will nerver take a statin, etc etc
So many metrics that tell me to do tests not always needed by the individual sitting on front of me
This is so important and I ask myself this question every time I’m asked to get a test by someone on the healthcare team. And I always ask myself this question every day. Thank you Dr. Cifu for this post!!!!
Great article. Applicable in so many ways. I’m curious that you didn’t bring up liability and the abundance of attorneys in this discussion. That seems to be a common thread in similar discussions.
Agree in general, but disagree about micro-albumen testing. If persistently positive, add SGLT2 even if a1c is controlled, maybe GLP1 or finerenone. Yes, there is significant cost involved so some cant do it. Also, use it to calculate the 5 year risk of esrd (Kidney Failure Risk Equation), and if >5% get renal involved.
I have some patients that refuse colonoscopy and ask about cologuard or FIT. I then ask will you do the colonoscopy if its positive. If the answer is no, I tell them there's no point (usually they know someone who is the 1:1000 perforation). Still shake my head at the patient with Crohn's that won't do colonoscopy, even though she's high risk, but did to the grail gallery test. Not sure who ordered it, wasn't me.
Absolutely HATE the patient that comes to me to interpret the pages and pages of results that their "Naturopath" ordered. Yes, you have antibodies to Candida, EVERYBODY does.
Respectfully, (and with the audacity that only a resident can have in challenging someone who has seen much more than they have) I'd like to modify your statement- "Will the result change my management OR the patient's life?" Often it is true that a test result is unlikely to change the orders that I make for my patients (either because my pre-test probability is so high or so low, or because the test's LR isn't that strong), but I fear we sometimes take a physician-centric approach that thinks only about what WE will change rather than how the test itself impacts the patient.
For the patient with an elevated ASCVD score on a statin, will I change anything from a CAC? Likely not (though as a nerdy aside, those with CAC scores >300 should be managed as secondary prevention with aspirin per the NLA). But will the patient change their diet once they see evidence of calcification in their own coronaries? I believe I have seen them do so.
For the patient already receiving broad spectrum antibiotics, will following up their CXR consolidation with a CT chest to determine if they have a true pneumonia change anything I prescribe? Not likely, but it really does make a difference when that patient tells their family what is going on to have an answer as to why they feel short of breath.
For the patient whose father had Huntington's, will genetic testing change what I do? No, since there are no disease-modifying therapies. But boy will it change how they live the rest of their life, and I think they have a right to know how many decades they have left.
I'm not saying we should just order whatever the patient wants (gosh knows I've said no to a lot of lumbar MRIs) but I fear we look only through our own lens, and not that of the patient. There is a lot of value to them in diagnostic certainty, even if it doesn't change the stroke of the physician's pen.
In current emergency department practice, in my experience as a radiologist, I would bet thousands of dollars that this question is either never asked, or if it is asked the answer is either ‘I don’t know’ (because it is an inexperienced midlevel provider) or ‘I don’t care because this imaging gives me liability coverage, administration wants me to do it because it generates revenue, and it will get the patient out of here faster than me thinking and talking to the patient ‘. There is zero chance of reversing this trend at all unless the administration and the physician are both incentivized to spend less on diagnostic workups, whether by capitation +tort reform, or a combination of means.
I have experienced every single example you gave and feel exactly the same. Luckily, many patients will agree with not "checking things" if you explain it clearly. Some people just don't feel comfortable unless we "check". For many people, tests (of almost any type) are reassuring and make them feel better. Though many of these are elderly/older people, so maybe past attitudes in medicine reinforced that.
I will say that I don't think this question is asked as much anymore!
And in the grand scheme of things in US medicine, the unnecessary tests that I've been able to avoid probably don't occupy even a drop in the bucket of rampant "testing" that goes on. Sigh.
I’m curious if there are certain labs you order routinely as a pcp. I just placed orders today for CBC/BMP/LFT for an “annual” in a middle aged healthy patient because I know her and I know it’s what she wants, but man, you do that 20 times a day 5 days a week and suddenly you’re chasing down alk phoses and explaining why the MCHC doesn’t matter…
Excellent article. I can think of just one situation in which ordering a test did not affect management but still had a positive outcome on a patient’s healthcare. He was quite anxious about his risk of MI even though he had minimal cardiac risk factors, and presented to the emergency room multiple times with anginal type chest discomfort. Each time he underwent a nuclear cardiac scan, all normal. Even after multiple attempts by me and his cardiologist to reassure him, he remained convinced that the nuclear scans might have missed a significant coronary lesion.
I finally talked his cardiologist into performing a coronary angiogram, simply to reassure him that his coronaries were normal. As expected, his coronaries were completely clean. After that, his chest pain never reappeared, his ER visits stopped, and he stopped being exposed to so much irradiation.
Interesting; I guess that just shows that the literature may not apply to all of our patients, because the angio certainly reduced my patient's ER visits.
Yes! So common! It's the patient with headaches that keeps complaining that everything is getting worse, several ER visits. She even tells you she wants an MRI. When you finally give in and get the MRI, headaches miraculously stop and never mentioned again.
An excellent summary. When I was in the last year's of my ED carrier, I worked exclusively in the 'fast track' (an ED-based urgent care). This was usually staffed with PA's - ours were better than average in this role. Still, I could see ~30% more patients in a shift than the PA's in large part because I ordered fewer tests.
They are a 2 year program. You have thousands of patient encounters ahead of them, multiple years of practice/education, and many years of experience to know the difference. Teaching them what you know is your contribution, not the moving of the meat. Not a contest of numbers. Not being offensive, just truthful.
I never thought of 'moving meat'! Getting more patients (people needing my care) seen in less time (and less cost, less potential harm from unnecessary tests, etc) meant less time in the ED for these people who were happy to get out of this noisy place, and more patients/people seen more quickly so they might suffer less.
And I did try to teach those PAs what I could, but I had very little opportunity for this since we were rarely in the ED at the same time...
Hi Adam - Important aspect of medicine and insightful questions posed. Reminds me of the times I've asked my residents why they didn't get a serum porcelain (to add to the myriad of tests ordered), and some have looked remorsefully at me, trying to think of why they didn't!
I learned from my dog's veterinarian. "We can run This or that Test, but it does not change the treatment".
One word: Metrics
To be a “good doctor” I must order microalbumins on every diabetic once a year, lipid panels on everyone once a year now (is a company contract thing) even if the patient swears they will nerver take a statin, etc etc
So many metrics that tell me to do tests not always needed by the individual sitting on front of me
This is so important and I ask myself this question every time I’m asked to get a test by someone on the healthcare team. And I always ask myself this question every day. Thank you Dr. Cifu for this post!!!!
Great article. Applicable in so many ways. I’m curious that you didn’t bring up liability and the abundance of attorneys in this discussion. That seems to be a common thread in similar discussions.
Your link is to a restricted article. :-(
Agree in general, but disagree about micro-albumen testing. If persistently positive, add SGLT2 even if a1c is controlled, maybe GLP1 or finerenone. Yes, there is significant cost involved so some cant do it. Also, use it to calculate the 5 year risk of esrd (Kidney Failure Risk Equation), and if >5% get renal involved.
I have some patients that refuse colonoscopy and ask about cologuard or FIT. I then ask will you do the colonoscopy if its positive. If the answer is no, I tell them there's no point (usually they know someone who is the 1:1000 perforation). Still shake my head at the patient with Crohn's that won't do colonoscopy, even though she's high risk, but did to the grail gallery test. Not sure who ordered it, wasn't me.
Absolutely HATE the patient that comes to me to interpret the pages and pages of results that their "Naturopath" ordered. Yes, you have antibodies to Candida, EVERYBODY does.
Respectfully, (and with the audacity that only a resident can have in challenging someone who has seen much more than they have) I'd like to modify your statement- "Will the result change my management OR the patient's life?" Often it is true that a test result is unlikely to change the orders that I make for my patients (either because my pre-test probability is so high or so low, or because the test's LR isn't that strong), but I fear we sometimes take a physician-centric approach that thinks only about what WE will change rather than how the test itself impacts the patient.
For the patient with an elevated ASCVD score on a statin, will I change anything from a CAC? Likely not (though as a nerdy aside, those with CAC scores >300 should be managed as secondary prevention with aspirin per the NLA). But will the patient change their diet once they see evidence of calcification in their own coronaries? I believe I have seen them do so.
For the patient already receiving broad spectrum antibiotics, will following up their CXR consolidation with a CT chest to determine if they have a true pneumonia change anything I prescribe? Not likely, but it really does make a difference when that patient tells their family what is going on to have an answer as to why they feel short of breath.
For the patient whose father had Huntington's, will genetic testing change what I do? No, since there are no disease-modifying therapies. But boy will it change how they live the rest of their life, and I think they have a right to know how many decades they have left.
I'm not saying we should just order whatever the patient wants (gosh knows I've said no to a lot of lumbar MRIs) but I fear we look only through our own lens, and not that of the patient. There is a lot of value to them in diagnostic certainty, even if it doesn't change the stroke of the physician's pen.
Great piece.
‘Just because we can, doesn’t mean we should.’
Those are words I try to practice by. And it applies both to diagnostics and therapeutics.
The blind-faith adherence to “guidelines” is another pet peeve of mine.
I say "just because we can, doesn't mean we should" to patients at least several times a week :). It's one of my favorites.
In current emergency department practice, in my experience as a radiologist, I would bet thousands of dollars that this question is either never asked, or if it is asked the answer is either ‘I don’t know’ (because it is an inexperienced midlevel provider) or ‘I don’t care because this imaging gives me liability coverage, administration wants me to do it because it generates revenue, and it will get the patient out of here faster than me thinking and talking to the patient ‘. There is zero chance of reversing this trend at all unless the administration and the physician are both incentivized to spend less on diagnostic workups, whether by capitation +tort reform, or a combination of means.
I have experienced every single example you gave and feel exactly the same. Luckily, many patients will agree with not "checking things" if you explain it clearly. Some people just don't feel comfortable unless we "check". For many people, tests (of almost any type) are reassuring and make them feel better. Though many of these are elderly/older people, so maybe past attitudes in medicine reinforced that.
I will say that I don't think this question is asked as much anymore!
And in the grand scheme of things in US medicine, the unnecessary tests that I've been able to avoid probably don't occupy even a drop in the bucket of rampant "testing" that goes on. Sigh.
I’m curious if there are certain labs you order routinely as a pcp. I just placed orders today for CBC/BMP/LFT for an “annual” in a middle aged healthy patient because I know her and I know it’s what she wants, but man, you do that 20 times a day 5 days a week and suddenly you’re chasing down alk phoses and explaining why the MCHC doesn’t matter…
Yes, the longer you practice and the more worn out you get, you realize that ordering less unnecessary tests makes your life a whole lot easier.
You packed a lot of wisdom into this article. Learning in medicine never stops.
🙏
Excellent article. I can think of just one situation in which ordering a test did not affect management but still had a positive outcome on a patient’s healthcare. He was quite anxious about his risk of MI even though he had minimal cardiac risk factors, and presented to the emergency room multiple times with anginal type chest discomfort. Each time he underwent a nuclear cardiac scan, all normal. Even after multiple attempts by me and his cardiologist to reassure him, he remained convinced that the nuclear scans might have missed a significant coronary lesion.
I finally talked his cardiologist into performing a coronary angiogram, simply to reassure him that his coronaries were normal. As expected, his coronaries were completely clean. After that, his chest pain never reappeared, his ER visits stopped, and he stopped being exposed to so much irradiation.
There’s a long literature on normal coronary angiograms not reducing ER visits for chest pain.
Interesting; I guess that just shows that the literature may not apply to all of our patients, because the angio certainly reduced my patient's ER visits.
Yes! So common! It's the patient with headaches that keeps complaining that everything is getting worse, several ER visits. She even tells you she wants an MRI. When you finally give in and get the MRI, headaches miraculously stop and never mentioned again.
I'm with you. Occasionally, an unnecessary test is necessary for reassurance.
“Lose 10-20 pounds and you just might not need any medicine nor a test.”
Great article Adam.
An excellent summary. When I was in the last year's of my ED carrier, I worked exclusively in the 'fast track' (an ED-based urgent care). This was usually staffed with PA's - ours were better than average in this role. Still, I could see ~30% more patients in a shift than the PA's in large part because I ordered fewer tests.
They are a 2 year program. You have thousands of patient encounters ahead of them, multiple years of practice/education, and many years of experience to know the difference. Teaching them what you know is your contribution, not the moving of the meat. Not a contest of numbers. Not being offensive, just truthful.
Yes - but...
I never thought of 'moving meat'! Getting more patients (people needing my care) seen in less time (and less cost, less potential harm from unnecessary tests, etc) meant less time in the ED for these people who were happy to get out of this noisy place, and more patients/people seen more quickly so they might suffer less.
And I did try to teach those PAs what I could, but I had very little opportunity for this since we were rarely in the ED at the same time...
Unfortunately, now it is neatly described as metrics and not patient welfare. Don’t drink the koolaid.
After nearly 50 years of working in a wide variety of healthcare activities, I can assure you - I don't drink Koolaid and I don't need lectures.
Retirement would be good. 👍
Retired several years ago - and yes, it has been good.
Have a nice day...!
Hi Adam - Important aspect of medicine and insightful questions posed. Reminds me of the times I've asked my residents why they didn't get a serum porcelain (to add to the myriad of tests ordered), and some have looked remorsefully at me, trying to think of why they didn't!
The serum rhubarb and standing stool velocity are other critical tests!