1) The idea that in-office imaging raises costs is because of the way Medicare is set up right now. If physicians could separately bill for reading images within the same visit, than the entire scenario you describe doesn't hold water any more.
You are blending two issues into one. Let me separate them.
The first is: "Should we be encouraging in-office imaging?" I would argue that the answer to this question is an emphatic "No", because numerous studies have shown that in-office imaging leads to significant (2-8x) increase in imaging volume. The reason is very simple, and it's purely financial. The imaging equipment becomes a revenue-generating device (and is a liability if it doesn't cover its own overhead), and therefore is treated like any other billable procedure. ie. let's maximize it and get as much money as we can by ordering scans as often as possible.
As I've previously mentioned, the bulk of the profit from in-office imaging comes not from reading the scan, but rather simply performing it. The fee earned by performing the scan is many times greater than the one for interpreting it, and it also doesn't require the physician's input; the exam is performed by your tech, coded by your office assistant, and you pick up the check from the insurance company. It's like passive income. In-office imaging is a highly abusable resource, with questionable benefit to the patient (it's extraordinarily rare that the patient actually gets the scan done and the images reviewed on the same day).
The second issue is: should clinicians be interpreting their own studies? Many clinicians already are. Their efficacy with doing so is something not well studied, for the same reasons that comparing efficacy and accuracy rates across radiologists are also not well studied. It's simply VERY difficult to evaluate this in any sort of objective and fair manner. Anecdotally however, we see clinicians come down to the reading room all the time, and these individuals are very often biased either by their H&P (and therefore don't look for other possibilities within the radiologic differential diagnosis), or simply aren't even looking for, or wouldn't recognize that other pathology because it is significantly outside the scope of their practice.
If I or a family member had a condition requiring a CT scan, I can assuredly tell you that I'd want a radiologist interpreting the study, while being supplied good clinical history by the ordering physician, either on the order requisition, or else in person in the reading room. I think this is the interpretation scenario most likely to lead to a complete and accurate diagnosis for the patient.
Like I said above, just recently I've caught two missed PE's, and what looks like metastatic ca in front of clinicians who were diligent enough to actually come down and review findings with me. They had already looked through the images before talking to me. Any radiology resident or attending will relate that these sorts of stories occur on a weekly or daily basis.
2) The idea that radiology teaches one to be able to correlate lesions across imaging modalities can be easily extended further. Clinical training teaches one to correlate findings across everything: exam, history, pathology, imaging, surgical findings, electrophysiology, etc. For every subtle sinusitis missed on head CT by a neurology, there is a hypoattenuation in the left cerebellum missed by the radiologist. Strength of clinical suspicion is key in interpretating imaging data and the radiologist cannot simply blame the clinician for not sending this information.
Yes, they can, and should. Your job as a clinician is to evaluate the clinical differential diagnosis for the patient. My job as a radiology resident is to evaluate the radiologic differential diagnosis given both the imaging and supplied history. You are supposedly the expert in auscultating the crackles over your patients right lower lobe. You do that 40 times per day.
Since I'm reading 40 chest x-rays per day, I'll be the one correlating whether your patient being febrile or this being an acute illness means that this is more likely to be a consolidation from pneumonia, versus a bronchogenic carcinoma. If you don't supply me that history, or come down to discuss the film with me, then I will give you both within my report. I am a consultant trying to answer your clinical question. It's within your responsibility to supply that information to your consultants (otherwise, why did you even bother obtaining it?).
The crappy histories that we get as radiology residents limits the usefulness of the report we can generate. If you truly are trying to get to the bottom of the study, it would be extremely helpful to supply your consultant with the information required in order for him/her to answer your clinical question.
You'd never ask a surgeon to go see a patient, with an order that states: "Pain" as the indication. You'd never ask the pathologist to interpret a microscopic slide without telling him/her what organ you chopped that tissue out of. And by the same token, giving your radiologist colleague a crummy history virtually guarantees that the report will be much vaguer and less accurate than one that actually supplies a clinical history and asks for confirmation/exclusion of a given set of disease entities.
I'm frankly surprised with all the medicolegal CYA garbage entangling our system that clinicians have been able to get away with ordering potentially invasive tests with such weak indications as: "Pain", or "Dizziness", or "Fall". Have you ever seen a patient get an anaphylactic reaction to contrast? I have. Not to mention the number of patients within the PACS system who have 20-40 CT scans to their credit. I read a patient yesterday who had 80 prior CT scans. That's a hell of a lot of radiation.
3) There are advantages to radiology training, yes. But my entire argument is not that radiologists are useless. It is that they block clinicians from being able to bill for reading images themselves when there are obvious advantages to this also.
In the office setting, clinicians can bill for their own image interpretation all they want. A significant percentage of medicare billing claims are generated by non-radiologists. This includes orthopods doing MRI, ENT's doing sinus CT's, cardiologists with their echo's and Nucs studies, and OB/GYN with prenatal ultrasounds. In the hospital setting, radiologists often have an exclusive contract with the hospital. This is no different than a set of surgeons defending their OR time, or vascular surgery trying to block cardiology from gaining privileges to do carotid stents, or a closed ICU where intensivists run the show.
If you don't like the exclusionary contract, go work at a hospital that doesn't support them. Then don't come bitching when the PM&R's are doing all your EMG's, the cardiologists and vascular surgeons are doing the carotid dopplers, the neurosurgeons decide that they'll take over the "interventional neurology" service (whatever that is), the intensivists are running your Neuro-ICU, and a competitor neurology group or radiology group underbids you and gains priveleges to read the head CT's and MRI's on your personal patients.
Exclusionary contracts are all over the place in medicine, whether it's the group that holds the contract to supply the cafeteria food, or the scrubs you wear on call, or the local nurses union, or the pathologists who are interpreting your slides, or the anesthesiologists pushing the propofol in the OR. They certainly aren't unique to radiologists. Every specialty and every profession protects its own turf.
The best clinicians I know regularly spend time reading images to help care for their patients. So why shouldn't clinicians be reimbursed for this?
As I've mentioned above, this already happens. If you are willing to accept the medicolegal liability for it, also can get your malpractice insurer to cover it, and finally can convince your local insurance company to pay for it, go right ahead. Nowhere within that system is the radiologist present to block it.
Especially in the scenario where you still require the clinician to get additional training (i.e. 6 months to 1 year) in imaging a particular area of the body. Please address this specifically, because this is the argument on the table. (I still fail to see how learning to read MRI's of the knee will help a neurologist in reading head CT's in the future.)
The next time you see a calvarial lesion on a head MR, or a head CT for that matter, and blow it off as a venous lake or an odd cranial suture, it will become very clear why a strong knowledge of bone lesions is important. In all probability, you've already missed a calvarial lesion, at whatever your current training level is. Just hope that it's a benign one.
As I've mentioned earlier, lesions and diseases often don't limit themselves to a single organ system, and clinicians who have never been exposed to the imaging outside of their clinical interest (in other words, those 10-20 diseases they've subspecialised into), are going to miss findings.
An orthopedic surgeon will catch the fibrous dysplasia in the facial bones because he/she has seen many cases of it in the femur. The average ENT surgeon will as well, because he/she has seen many cases of it in the maxilla. A neurologist is likely to blow by it, but might catch the basilar tip aneurysm. A radiologist is far more likely to catch both, by virtue of that cross-training.
That's why we do it. If it wasn't an integral part of being an imaging specialist (which is what we are), we wouldn't be doing it, and would instead be cranking out radiologists through a 3 year residency so we could make more bucks.
