The 11th Reason Doctors order unnecessary tests
I liked this list of reasons why doctors order tests. It's based on medical practice in the US but most apply to doctors in the UK too. I'd go so far as to add another - temporizing. It's really an extension of reason 1, with a bit of 2,3 and to some extent 5 as well.
Time can be an excellent way of finding out what the natural history of a disease process is, of gaining new information, etc, so ordering a few tests while watchfully observing your patient is often reasonable or even very good practice. However there's definitely a trap that many doctors fall into where they have a patient in want of a diagnosis or definitive plan, who doesn't readily fit into a disease paradigm, and they'll keep on ordering tests until they get bored. The problem with this sequential over-testing is it allows the doctor to stop thinking. All you need to do is fire off a few tests, then you don't need to think until they all come back negative. What to do? Order another test that takes a few days! And again, and again…
Although this could result in the diagnosis coming to light, either by eventually finding the 'right' test, or by the disease revealing itself more clearly (or just resolving), the unfortunate side effect of the process is that instead of being watchful and considering possible diagnoses for a time, the doctor disengages brain for all but the 30 seconds it takes to think up another few tests - thus while thinking he's exemplifying the considerate, watchful doctor, he becomes the exact opposite of that, sometimes for weeks on end.
However, I'd add just a tiny critique of Dr Rangel's underlying rationale for critiquing over-testing. Not that I disagree with him, because the behaviours he describes are absolutely not good medicine and should all be avoided. But why are they not good? In criticizing the lazy physician who can't be bothered to formulate a diagnosis using clinical skills, he says:
Yes, that's not very impressive doctoring. But the problem with the 'cookie cutter' approach is not that it's intellectually lazy, although it is. It's that it doesn't work - it has a terrible signal to noise ratio, and it results in patients being exposed to risks from the original investigation and from subsequent investigations or procedures relating to incidentalomas. However, if we had some amazing new body scan that could accurately predict the natural history and effects of every 'abnormality', at £1 per scan, then ordering a GIANT WIZZBANG SCAN of EVERYTHING might be very good for patients, even though any lazy idiot could order the scan. I'd be out of a job, but people would probably be healthier.
Despite what a few mail-order scanning companies would like to tell you, that scan doesn't exist, and is very unlikely to any time soon, so us good doctors who use clinical skill and judgement can rest safe in our paycheques. But it's important to remember what the point of our jobs is - being a 'good doctor' (which includes using investigations judiciously) improves the health and lives of our patients. It's not an end in itself.
As a medical teacher, I can't teach my students / juniors about every situation where they should or shouldn't order a particular test. But if I can teach them an underlying throught process or behaviour pattern relating to how to approach diagnostic situations - with the outcome for the patient paramount - then I shouldn't need to tell them how to avoid each of the 10 bad reasons for ordering tests. They should be able to work that out for themselves.
Time can be an excellent way of finding out what the natural history of a disease process is, of gaining new information, etc, so ordering a few tests while watchfully observing your patient is often reasonable or even very good practice. However there's definitely a trap that many doctors fall into where they have a patient in want of a diagnosis or definitive plan, who doesn't readily fit into a disease paradigm, and they'll keep on ordering tests until they get bored. The problem with this sequential over-testing is it allows the doctor to stop thinking. All you need to do is fire off a few tests, then you don't need to think until they all come back negative. What to do? Order another test that takes a few days! And again, and again…
Although this could result in the diagnosis coming to light, either by eventually finding the 'right' test, or by the disease revealing itself more clearly (or just resolving), the unfortunate side effect of the process is that instead of being watchful and considering possible diagnoses for a time, the doctor disengages brain for all but the 30 seconds it takes to think up another few tests - thus while thinking he's exemplifying the considerate, watchful doctor, he becomes the exact opposite of that, sometimes for weeks on end.
However, I'd add just a tiny critique of Dr Rangel's underlying rationale for critiquing over-testing. Not that I disagree with him, because the behaviours he describes are absolutely not good medicine and should all be avoided. But why are they not good? In criticizing the lazy physician who can't be bothered to formulate a diagnosis using clinical skills, he says:
"It takes time to listen to and sort through a patient’s symptoms and to do a proper and directed physical exam. But if you have 55 patients to see today and you want to make it home on time then you can just order a GIANT MRI SCAN of EVERYTHING that’s all but guaranteed to detect any and every abnormality. Wrong. That’s not practicing medicine. That’s the cookie cutter approach. My dog can do that."
Yes, that's not very impressive doctoring. But the problem with the 'cookie cutter' approach is not that it's intellectually lazy, although it is. It's that it doesn't work - it has a terrible signal to noise ratio, and it results in patients being exposed to risks from the original investigation and from subsequent investigations or procedures relating to incidentalomas. However, if we had some amazing new body scan that could accurately predict the natural history and effects of every 'abnormality', at £1 per scan, then ordering a GIANT WIZZBANG SCAN of EVERYTHING might be very good for patients, even though any lazy idiot could order the scan. I'd be out of a job, but people would probably be healthier.
Despite what a few mail-order scanning companies would like to tell you, that scan doesn't exist, and is very unlikely to any time soon, so us good doctors who use clinical skill and judgement can rest safe in our paycheques. But it's important to remember what the point of our jobs is - being a 'good doctor' (which includes using investigations judiciously) improves the health and lives of our patients. It's not an end in itself.
As a medical teacher, I can't teach my students / juniors about every situation where they should or shouldn't order a particular test. But if I can teach them an underlying throught process or behaviour pattern relating to how to approach diagnostic situations - with the outcome for the patient paramount - then I shouldn't need to tell them how to avoid each of the 10 bad reasons for ordering tests. They should be able to work that out for themselves.
No comments:
Post a Comment