Improving the appropriateness of diagnostic tests
The number of diagnostic imaging tests performed on Canadian patients has been increasing rapidly in recent years.
Although the cost of each test is small, the aggregate costs add up quickly. The costs have become sufficiently large that both doctors and the government have raised questions about whether all the tests that are being performed are beneficial to patients.
At the same time, there has been growing awareness that diagnostic tests can also cause harm. Some tests cause harm directly – for example by increasing the risk of cancer by a small amount. Other tests can cause harm indirectly, by leading to additional tests and procedures with their own costs and risks. A closer examination of the benefits and harms of testing has begun at all levels of the health system, with the goal of more appropriate use of tests.
In this article, we focus on tests that are done in response to a patient’s symptoms such as pain, tiredness or shortness of breath, and not on testing done to screen for cancer. Healthydebate has previously published several articles and blogs on the controversy about mammographic screening for breast cancer, which you can view here.
Benefits of testing
CT, MRI, and echocardiography are among the “miracles” of modern medicine. They have helped physicians make diagnoses more quickly and more accurately. They have almost completely eliminated the need for diagnostic surgery – surgery that was done simply to find out what was wrong with a patient rather than to fix a problem. Many young physicians today have never seen a patient whose abdomen was cut open solely to make a diagnosis. In contrast, 50 years ago this was common practice.
The ability to rapidly determine the cause of a patient’s symptoms is at the heart of what medicine is about. Patients can now quickly learn what they do or do not have, and subsequently receive an appropriate treatment plan for their illnesses.
In many instances, an early, accurate diagnosis leads to swifter treatment and better outcomes, such as the rapid treatment of internal bleeding after trauma detected through CT scanning or ultrasound.
Diagnostic tests can also be useful to help determine if a patient is responding to treatment, or to look for recurrence of disease. One example is echocardiography, which can be used to monitor heart failure and other forms of heart disease.
There is also important value in the reassurance that a normal CT scan can bring to patients such as those who are concerned they might have a brain tumour because they are suffering from severe headaches.
Harms of testing
It is also important to consider the potential harms that accompany testing.
No test is perfect. Some test results are “false positives”: results that suggest an abnormality even when no abnormality is actually present. False positive tests can lead to unnecessary stress and anxiety for patients.
Modern tests are not always definitive. For example, about one-quarter of CT scans of the abdomen and chest in Ontario find an abnormality which can’t be definitively diagnosed, and results in the radiologist recommending another test in an attempt to come to a diagnosis. For example, if a CT scan shows an abnormality, an ultrasound may be recommended to characterize it better, and if a question remains an MRI might be ordered. Sometimes a more invasive test such as endoscopy, biopsy or rarely even surgery is recommended. Endoscopy, biopsies and surgery all carry small risks, which can include bleeding, pain, and infection. If all of these test lead to the definitive diagnosis of an abnormality , then all these tests were worth it. If it turns out that the original abnormality really was a false positive and was nothing to worry about, then the patient underwent a series of tests, and was exposed to their inconvenience and small risks, for no real benefit.
Recently, doctors have become more concerned about the long-term effects of radiation from CT scanning, which expose patients to low dose radiation. In younger people who undergo numerous CT scans, the amount of radiation is enough to cause a small increase in the risk of cancer decades after the scans are complete.
Costs of testing
In recent years there has been a marked increase in the use (and related costs) of several types of diagnostic tests, which cannot be accounted for by the aging of the population or new indications for testing.
One test that has seen dramatic increases is echocardiography, which is an ultrasound test of the heart. According to data Healthydebate has acquired from the Institute for Clinical Evaluative Sciences, 652,714 echocardiograms were done by cardiologists in Ontario in 2010/11, which is about one echocardiogram for every 20 Ontarians in one year. That compares with only 361,440 echocardiograms 5 years earlier – an 81 percent increase.
The Ministry of Health and Long-Term Care spent $68 million on physicians’ fees alone for echocardiograms in 2010/11. (The Ministry pays “technical fees” as well, which are not included in this figure).
The use of many other tests, such as CT and MRI scans, has increased dramatically in Ontario as well. The increase in CT and MRI scans has been partly driven by a government focus on decreasing wait times for CT and MRI. Despite a marked increase in MRI availability in Ontario, many patients continue to wait an inordinately long time for an MRI .
No studies have calculated the costs of all diagnostic tests in Ontario, or the magnitude of the rise in costs, but the costs are large enough that the cash strapped Ministry of Health and Long-Term Care has taken notice. Earlier this year, the Ministry appointed an Expert Panel to provide “advice regarding the appropriate utilization of diagnostic and imaging studies.”
Decreasing “unnecessary” testing is not easy
The panel’s interim report states that “….echocardiographic studies carried out for inappropriate indications [should not be] paid by the Ministry.”, and suggests that a recent report from the Cardiac Care Network should be used to determine the appropriateness of echocardiography.
The CCN report lists 107 appropriate indications for echocardiography, many of which are imprecise, and lists no inappropriate indications. In many ways, this is not surprising since it is almost impossible to list every possible patient symptom for which a test like an echocardiogram is either appropriate or inappropriate.However, it is not clear how this approach will increase appropriateness.
Many tests fall into the “grey zone” of medical care, meaning that they are often done “just in case” something has been missed even though the likelihood of the patient actually having an ailment is quite low. For example, less than 2% of CT scans done for headaches in Ontario find an abnormality that may actually explain the reason for the headaches.
At the same time, some tests are not used as frequently as they should be, or at least not in patients who need them. For example, Andrea Gershon, a respirologist at Sunnybrook Health Sciences Centre, has shown that fewer than half of adults in Ontario treated for asthma by their doctors received the appropriate test to confirm the diagnosis. The extent to which physicians make other diagnoses without ordering the confirmatory diagnostic test is largely unknown.
Can the public become more informed consumers of diagnostic tests?
According to Wendy Levinson, chair of the Department of Medicine at the University of Toronto, improving the appropriateness of testing will require changing the culture of “more is better” that is deeply embedded in health care.
Patients generally believe that more tests will benefit them, and are often not aware of the potential harms that can be associated with testing. Physicians may also be concerned that they could be held legally liable if they do not “leave no stone unturned” by ordering every test to rule out a potential problem. One approach to addressing the “more is better” culture is the Choosing Wisely initiative, which is a partnership between nine physician groups in the United States and Consumer Reports.
“The goal of Choosing Wisely is to get both patients and doctors thinking about values and risks,” says Levinson, who has been involved in launching the programme in the United States. The hope is to improve understanding among both groups of when procedures and tests are necessary and when they are not.
Levinson believes the key to the initiative is that it targets both doctors and patients. “For patients, this is about becoming informed consumers,” she says.
To help patients, Consumer Reports has created summaries of research that are easy to understand, like this one: “Many people who experience severe headaches want a CT scan or MRI to see if they’re caused by a brain tumor or other serious problem. But most of the time neither test is necessary.”
These reports also detail the times when the tests are appropriate. In the case of CT scans for headaches, if the headaches are: sudden or explosive, different from other headaches you’ve had in the past (especially if you’re 50 or older), brought on by exertion, of if accompanied by fever, seizure, vomiting, a loss of coordination, or a change in vision, speech, or alertness.
With this kind of information, Levinson hopes patients will be able to better judge whether they should be asking their doctor for a particular test (or whether to decline a test that may be offered).
For doctors, the nine medical specialties involved have created evidence-based lists of the top five areas where unnecessary tests and procedures are likely to occur, along with a transparent methodology of how these lists were generated.
There is a combination of reasons that low yield tests are done, and decreasing their frequency is certainly no small task. While Levinson is hopeful that Choosing Wisely can help improve the appropriateness of diagnostic tests, this has not yet been demonstrated. She cautions that the culture of “more is better” did not start overnight – it has been built over decades – and so change will take time.
Diagnostic tests are both more expensive and used more frequently in the United States than in Canada. It is also generally agreed that the frequency of inappropriate testing is greater south of the border. So, it is not at all clear what impact a Choosing Wisely approach would have in Canada, especially considering the evidence that in Canada we may be testing too infrequently in some instances.
Increasing the appropriateness of testing will require many approaches, including establishing clear criteria about when tests are likely to be appropriate and when they are not, changes to the way physicians are paid so that unnecessary testing is discouraged, better oversight of diagnostic test quality to ensure that tests are of high quality and correctly interpreted (thus decreasing repeat tests), mechanisms to electronically transfer imaging studies and results from one centre to another (also decreasing the need for repeat testing), a greater emphasis in medical training that “more is not always better” and that the costs and harms of low-yield tests are considerable, as well as a greater understanding by the public about the value, risks and costs of diagnostic tests.
Although it is clear that we have a lot of work to do to optimize appropriateness, small changes – for example a decrease in unnecessary MRIs for back pain – could start to free up capacity and possibly even decrease wait times.