For our health care system to remain sustainable, scarce resources must be managed effectively both at the system level and on the front lines. In health care, decisions by doctors have a major impact on resource management in front-line care. But our health care system does not provide doctors with all of the information they need in order to manage front-line resources effectively. Specifically, our system does not provide doctors with information about the cost of medical care. We believe that providing physicians with accurate information about costs, and particularly the costs of diagnostic tests, is an important piece of the sustainability puzzle.
The first thing to realize is that doctors do not know how much diagnostic tests cost the health care system. A Canadian study of emergency room physicians revealed that they have limited knowledge of the costs of the diagnostic tests they order every day, and tend to significantly underestimate the costs of diagnostic imaging. A systematic review of physician knowledge of diagnostic testing costs found the accuracy of physician estimates to be low, with fully 50% of the estimates ranging from 50% to 200% of true costs. The review included studies of physicians in a number of different specialties and at different levels of training, including many who had completed their training. It didn’t matter how much they had trained or what specialty they were in – estimates of true costs were wildly inaccurate.
Okay, so the research shows that doctors don’t know how much tests cost. So what? Well the research also shows that when doctors are educated about the costs of tests they order every day, they start ordering fewer tests. A study in Israel examined laboratory testing habits of physicians at a single hospital for one year following an education course on costs of diagnostic testing. The study showed an overall reduction of 19% in laboratory tests performed at the hospital in the year following the educational program. Another Israeli study involved a system for ordering tests for an internal medicine department, in which notes were placed on the order form about the laboratory costs and diagnostic accuracy for blood testing. The study showed significant reductions in tests per visit, with no difference in the readmission rate or in the number of diagnoses of conditions based primarily on blood tests, between the study group and the control groups. A Swedish study showed similar results in a primary care setting.
We know you’re thinking “I’m not sure I want my doctor to know how much tests cost… what if they didn’t order an important test and my care suffered as a result?” But there’s a very important detail about those studies: they show a modest but significant reduction in tests with no measured negative effect on patient care.[i] This research suggests that when doctors are informed about the costs of tests, they still order all the important tests; they just stop ordering the ones they know won’t tell them anything important.
Now, maybe you’re thinking this wouldn’t really save the system that much money. Not so fast. Canada spent about $200 billion on healthcare in 2011. Diagnostic tests account for about 10% of health care spending in hospitals. Assuming that trend is roughly the same in community care, that means Canada spent about $20 billion last year just on tests. If we can figure out how to bring that number down by even 10%, significantly less than the amounts seen in those studies from Israel and Sweden, then we could be talking about saving the Canadian health care system billions every year. That should be enough to get our attention.
Of course, the challenge is how to provide this information to doctors in a way that is both accessible and useful. Doctors already have to keep pace with breaking developments in medicine and there’s not much room left in the medical school curriculum for robust education about costs. And doctors are under so much pressure for their time that they’re not going to be able to look up the cost of a test they don’t order very often.
The answer, we think, is to put this information at doctors’ fingertips: build it right into the electronic medical record system they use to order tests. When doctors click the box to order a test on their computer screen, they would see the cost of the test right next to the box (and why not information about test accuracy and clinical guidelines while we are at it?). They instantly get the information they need, right when they need it. This doesn’t stop them from ordering the tests they want to; it just provides them with all of the information they need to act as both clinicians and responsible front-line resource managers.
This isn’t a completely new idea – a group of American researchers attempted something similar using clinical practice guidelines instead of test costs. They incorporated practice guidelines for routine laboratory and chest radiographic testing directly into the electronic ordering system of a cardiac care unit, and provided education on these guidelines to house staff and nurses. Their research showed an estimated reduction in expenditures for “routine” blood tests and chest radiographs of 17%, with no significant changes in health outcomes. This suggests that putting information at doctors’ fingertips right when they need it can promote more effective, more efficient care.
Of course, as with any innovation there are challenges associated with this idea. Many physicians do not currently use EMRs, so this idea could only spread as fast as EMRs are adopted across the system. As a result, not all of the potential cost savings could be realized right away. It’s also important to acknowledge that costs in health care can be difficult to estimate accurately, can vary across regions, and can change as technology advances. Care would have to be taken in developing accurate estimates and keeping these estimates current.
We don’t want to suggest that this is some kind of silver bullet for sustainability. But what is exciting about this idea is that we might not need endless committees to start reducing unnecessary tests, we might just need to empower individual clinicians by giving them the information they need to choose wisely.
[i] It should be noted that these studies were relatively small (and thus underpowered), so until larger studies are conducted or a systematic review is performed these results should be treated with some caution. However, these initial results are promising enough that we think they indicate larger Canadian studies are warranted.
The comments section is closed.
This is a great piece and an issue rarely addressed. To add onto what docs should know the price of – drugs! Working mostly with people on social assistance, I order what is covered on the ODB formulary, which has been determined based on cost-effectiveness. If I have patients not on ODB, I use RxFiles which is a wonderful resource that includes the cost of medications on the side. This notion though that if a patient has drug coverage, we don’t need to care what we order is a bad attitude. Again building the cost into EMR next to the electronic Rx tool would be a really practical way of informing on this issue. The next question though is how you get cost-effectiveness taken better into account in clinical guidelines, which it often isn’t.
Hi Jeremy,
I fully agree with you that more often than not, the tests are ordered as a matter of routine and consideration of clinical needs of patients is a MUST. However these do give a clue to associated sub-clinical conditions in a patient and perhaps a physician would want to err towards a positive side.
I am from India. I would not know why the routine laboratory investigations should cost as it does in UK/USA. We have the same instruments, same reagents/calibrators/controls and EQAS expenditure in compliance with ISO15189.and some of the commercial accredited labs are doing all profiles(a total of 47 +tests) for under INR 2000.00, that is under 50$ and mind you these laboratories are profit making. In any way, the KFT/LFT/Cardiac enzymes including troponin/Lipids/Haematocrit/thyroid profiles and host of others would not cost more than INR 6000.00, that is $125. Why the disparity? It leads to a perception to non-optimising the resource expenditure or even an attempt to accrue unusually large profits. Surely the developed countries know the answer. It should not be too difficult to bring down the costs.
Tests should to be valued in terms of the potential clinical benefit associated with performing them. (e.g. 10% mortality on average if not testing vs. 2% mortality on average if using the test). Seeing the listed price of a test likely just reminded physicians to consider the clinical value of that test. The price is irrelevant to clinicians if the potential clinical benefit is justified.
That being said, certain tests of minimal clinical value should not be funded in a single payer system with a fixed budget…but those aren’t individual physician-level decision.
I wholeheartedly agree that efficiencies need to be found, and that unecessary test ordering should be addressed. However, I am not convinced that listing costs beside each tick box for a test is the answer.
What is the underlying core problem? It’s not because physicians do not know the exact costs of investigations they are ordering; they don’t randomly ask for tests nilly-willy simply because they don’t feel the true financial cost. The real reason doctors over-investigate in certain circumstances is because they think/feel that those investigations may in fact be helpful in the care of their patients….when in fact evidence has shown that they are not.
You already addressed this in your article with several studies looking at implementing clinical guidelines for routine test ordering (although you took a leap of faith to use it in support of listing dollar-values instead): doctors order less irrelevant tests when they know the most up-to-date clinical evidence. We need to get physicians to practice evidence-based medicine that tells us which tests are important and which ones are useless. That is the true answer to curbing unecessary over-investigations.
Listing $ values on the frontlines will only introduce another level of bias into a clinician’s decision making … something we do not need when as a profession we are already struggling to tackle the issue of medical errors. A physician dealing with a dying patient in the ER should not be wasting time contemplating which is the most cost effective investigation to get; he/she should be figuring out which investigations are the most clinically effective in providing the best care for the patient.
Thank you for your comment, Edmund. I agree that including guidelines is important – indeed, I think EMRs provide quite a few untapped opportunities to provide clinicians with helpful information at the moment is needed (as opposed to having to look it up in a book or through a smartphone).
However, test costs are an important part of the equation: one of the reasons too many tests get ordered is the prevailing view that health care in Canada is ‘free’. Adding to this is the fact that many patients equate more tests with higher quality of care, which together can result in ‘fishing trips’… going down the list and just ticking off all the boxes that might apply, no matter how unlikely the tests are to yield valuable information. An intervention like this reminds practitioners that there are costs (sometimes significant costs) associated with the tests they order, and that part of their role as a physician is to manage scarce resources effectively. And we can’t expect clinical guidelines to do all the work, particularly when they don’t factor in cost
Of course I agree that physicians should follow the evidence, and I think this should extend to our system managers and policy makers too. In this case the evidence suggests that knowing test costs, and not just guidelines, are part of how we can reduce the volume of unnecessary tests, without compromising patient care.