Too much of a good thing
Too much of a good thing is a blog series that is published as a collaboration between Wiser Healthcare and Croakey.org. The series investigates how to reduce overdiagnosis and overtreatment in Australia and globally. The articles are also available for republication by public interest organisations, upon request.
Diagnostic imaging scans for back pain: three things they can do, three things they can’t
It’s time for patients and clinicians to have frank discussions about what diagnostic imaging scans can do for low back pain, and what they cannot, according to research published recently in BMJ Open.
Three of the authors, Adrian Traeger, Swee Sharma and Chris Maher, all from the Institute for Musculoskeletal Health at the University of Sydney and Sydney Local Health District, summarise their work below.
It is published as part of the TOO MUCH of a Good Thing series, which is investigating how to reduce overdiagnosis and overtreatment in Australia and globally, and is published as a collaboration between Wiser Healthcare and Croakey.
To follow the series, bookmark this link, and follow #WiserHealthcare on Twitter.
Adrian Traeger, Swee Sharma and Chris Maher write:
Adrian Traeger
Many Australians will have experienced low back pain. When it strikes, back pain can be surprising, disabling, and
almost always worrying. For some, the pain subsides quickly and so too does the worry. With the help of little more than a heat pack, they can return to normal activity within days or weeks. For others, the pain persists much longer, upturns what was once a happy and healthy life, and triggers a long and unsatisfying search for answers.
Finding out what is wrong
Regardless of how long the pain has been there, people with low back pain want to know what is wrong.( Clinicians
Swee Sharma
can diagnose most conditions that cause low back pain by asking specific questions and performing an examination (e.g. they may press your spine, ask you to bend in different directions, or lift your leg into the air). If a clinician suspects a serious cause of low back pain, they might send you for a diagnostic imaging scan (x-Ray, CT or MRI).
The problem with too much imaging
There is increasing evidence that it is not just those who clinicians suspect have a serious condition that are sent for imaging.
Chris Maher
One recent study of over nine million consultations found that clinicians referred around one quarter of people with
low back pain for a diagnostic imaging scan. If we expect that around 1 per cent of patients who see a GP have a condition that needs urgent imaging, at least on the surface, these numbers seem too high. Some concerned clinicians, the back pain research community, and more recently international initiatives such as Choosing Wisely®, have held lumbar imaging for low back pain firmly in their sights as a “test to question”.
One reason for the concern about too much imaging is the risk for a problem called overdiagnosis. Overdiagnosis occurs when a person receives a diagnosis that brings them no benefit or causes harm. That is, a diagnosis that people would be better off not knowing.
When it comes to diagnostic imaging for low back pain, imaging reports often show age-related changes such as “degenerative disease” “disc bulges” and “arthritis” that are very common in people without low back pain. Such findings increase worry and may lead people to pursue interventions to try to fix these problems; but they may not be the cause of their pain. This is an example of overdiagnosis.
What did we do in this new research?
We wanted to understand how clinicians and patients viewed – in their own words – the role of imaging tests for low back pain. Understanding these views could help uncover reasons why imaging happens too often. We compiled all of the available studies that had interviewed patients and clinicians about diagnostic imaging for low back pain.
We located 69 studies from across Europe, Australia, North America, and Asia, which included 1747 participants. This was the first time such a large volume of diverse diagnostic imaging studies had been combined to understand the problem of too much imaging.
What did we find?
Although many participants had accurate views of the role of diagnostic imaging, we were surprised to find several misconceptions expressed in interviews with patients and clinicians. We could organise some of these views into key facts and key misconceptions about what diagnostic imaging can and cannot do.
- Diagnostic imaging can locate the source of typical low back pain FALSE
The most commonly expressed view by patients and clinicians was that imaging is useful to locate the source of low back pain. This surprised us because we know that for the majority (around 90-95 per cent)of people) who have typical or ‘garden variety’ low back pain–sometimes referred to as “non-specific” low back pain – accurately locating the source of the pain is not currently possible. In other words, even the most sophisticated diagnostic imaging scan cannot tell a clinician with any certainty whether someone’s back pain is due to a joint, muscle, or disc problem.
- Diagnostic imaging can make patients happier TRUE
Our review found many patients associated a referral for imaging with better clinical care. In other studies where imaging is provided to one group of patients with low back pain, and withheld from another, there was evidence that a referral for imaging could increase satisfaction with the consultation.
- Diagnostic imaging can only harm you via exposure to radiation FALSE
Participants in our review rarely mentioned potential harms of unnecessary imaging. When they did, understanding of potential harms appeared to be limited to exposure to radiation. There are several others. Documented harms of imaging include imaging findings that increase worry, worsen recovery, and risk use of misdirected, invasive interventions such as injections or surgery.
- Diagnostic imaging can rule out serious problems with the spine e.g. cancer TRUE
A key reason both patients and clinicians valued imaging was to rule serious problems such as cancer, fracture, or infection. Tests such as MRI do form a critical part of the diagnosis of these problems.
- Diagnostic imaging can provide evidence that persistent pain is real FALSE
In our review patients expressed a desire for imaging results that would demonstrate to friends and family that their ‘invisible’ pain was real. Unfortunately, it is not possible using current spinal imaging techniques to determine whether a person is in pain or not.
- Diagnostic imaging can validate the pain experience TRUE
Many patients expressed relief when their imaging findings showed what appeared to be a damaged spine. Previous research has also shown that imaging results which appear to ‘fit’ with a patient’s symptoms could make them feel as though their pain experience was validated.
Why do these findings matter?
Diagnostic imaging scans are an essential test for only a small number of people who experience low back pain. Although they can rule out serious problems, so too can skilled clinical questioning with less hassle for the patient and less potential for harm.
Scans cannot provide visual evidence of pain; but patients clearly want this. Clinicians need tools to provide reassurance and validation for an experience that in most cases cannot be adequately explained by even the best medical tests.
Our findings suggest it is time for patients and clinicians to have frank discussions about what diagnostic imaging scans can do for low back pain, and what they cannot.
This article is part of an ongoing series that is published as a collaboration between Wiser Healthcare and Croakey.org.
The series investigates how to reduce overdiagnosis and overtreatment in Australia and globally. The articles are also available for republication by public interest organisations, upon request.
Bookmark this link and follow #WiserHealthcare on Twitter.
Weighing up the potential benefits and harms of comprehensive full body health checks
Most people are familiar with the concept of undergoing general health checks. This might mean that your doctor talks to you about your medical history and lifestyle including diet, physical activity, alcohol intake and smoking history. This is part of providing holistic care to identify and address risk factors to prevent disease. Health checks are also performed as part of national screening programs, such as the national cervical cancer screening program, which aim to detect and treat disease early.
However, a less familiar form is the general or ‘full body’ health check that not only involves a comprehensive clinical assessment, but also includes performing a whole battery of laboratory, genetic and imaging tests in a person who does not feel ill and has not sought care. General health checks are a common element of health care in some countries and purport to be able to detect disease early, prevent disease from developing, and provide reassurance. In Australia this type of comprehensive full body health check-up may be offered to company executives or directly advertised to the general public, targeting the ‘worried well’.
But do these full body check-ups truly deliver on their stated aims and are there any potential downsides?
In the article below, Dr Romi Haas and Professor Rachelle Buchbinder describe how testing asymptomatic people to ‘screen’ for a range of diseases as part of these types of general health checks can be dangerous. They outline what people need to know before consenting to such checks and suggest improved regulation of public advertising could help in ensuring the judicious and informed use of screening.
This article is published as part of the TOO MUCH of a Good Thing series, which is investigating how to reduce overdiagnosis and overtreatment in Australia and globally, and is published as a collaboration between Wiser Healthcare and Croakey.
To follow the series, bookmark this link, and follow #WiserHealthcare on Twitter.
Romi Haas and Rachelle Buchbinder write:
Romi Haas
One-page advertorials are appearing in Australian and regional newspapers for something called “HealthScreen”, with the headline “The future of medicine is here now”. HealthScreen is a direct-to-consumer service that conducts a variety of assessments that search or ‘screen’ for signs of disease in people who are not displaying any signs or symptoms (asymptomatic).
These general health checks offered by HealthScreen cost AU$2,000 and include Magnetic Resonance Imaging (MRI) scans for 20 of the most common cancers, coronary heart disease and stroke risk assessments, genetic screening, laboratory tests and sleep health evaluation. The $2,000 cost is not covered by Medicare or private health insurance. It is also only an initial cost: it does not include the associated cost of the inevitable additional tests as well as treatments that will follow for many people.
A quick Google search has revealed that HealthScreen is not alone in offering comprehensive general health checks. In Melbourne alone, similar services are also offered by Epworth HealthCheck, Life First, Men’s Health Melbourne and National Institute of Integrative Medicine.
Rachelle Buchbinder
HealthScreen’s promotions say they aim to identify medical conditions that “are likely to reduce life expectancy before the onset of symptoms or any indication of a problem”. They promise their clients will know they “are doing everything possible to take control of your health and longevity”. Framed this way, parting with $2,000 to future-proof your life might sound quite appealing to those who can afford the cost. But can it truly be this easy or is it too good to be true?
This type of screening is highly likely to lead to further tests. This is because screening tests like these are not designed to be diagnostic, but rather to identify people who are at higher risk of developing a disease so they can have further diagnostic testing. Positive screening tests require confirmatory tests to definitively rule in or rule out disease.
In fact, in December 2019, HealthScreen reported that “so far the doctors have found hidden problems in every patient”.
So what is the evidence that finding ‘hidden problems’ is beneficial in future-proofing your life?
Potential for more harm than good
This is not simply a question of cost. Before consenting to undergo such general health checks, it is important to be fully informed about both the potential benefits and the potential harms of undergoing such a comprehensive check-up.
The best available evidence evaluating the benefits and harms of these types of health checks indicates they are unlikely to be beneficial. A recent update of a Cochrane review included 17 trials, 15 of which reported outcome data from more than 250,000 people in total. As well as a comprehensive clinical assessment, the trials assessed various combinations of blood, urine and lung function tests, electrocardiograms, cancer screening, and vision and hearing assessments.
It found high-certainty evidence that general health checks have little or no effect on either overall mortality or death from cancer, and moderate-certainty evidence that they probably have little or no effect on fatal or non-fatal heart attacks or strokes. General health checks offered by various organisations are therefore not evidence-based; they have not been shown to increase life expectancy.
As well as being unlikely to be beneficial, full body general health checks in asymptomatic people can potentially be harmful. The main harms are overdiagnosis, detrimental psychological effects, negative effects on health behaviours (for example, failure to quit smoking due to reassurance of good health), complications related to follow-up tests, and unnecessary treatments.
No screening test is one hundred percent accurate. There is always a trade-off between sensitivity (correctly identifying when you do have a problem) and specificity (correctly identifying when you don’t have a problem). Tests with high sensitivity but low specificity have a higher chance of false positive results (saying you have the disease when you don’t).
A positive screening test always requires further confirmatory or diagnostic tests. A false positive result cannot help you because you do not have the disease but it can harm you from unfounded worry and stress. Tests with high specificity but low sensitivity have a higher chance of false negative results (missing people who have the disease). False negative results can provide false reassurance, delay detection of disease, lead to legal action and reduce public confidence in screening programs.
Overdiagnosis is another harm that should be considered when weighing up the potential harms and benefits of full body general health checks. This occurs when a test leads to a diagnosis that would never have caused any symptoms or problems within a person’s lifetime.
In cancer testing, for example, this can happen because a tumour may grow so slowly that a person dies of something else before it starts to cause symptoms, a tumour may not grow at all or a tumour may even disappear without treatment. In each of these cases the diagnosis has no direct benefit since no treatment is needed.
However, the diagnosis can cause harm through unnecessary psychological distress, adverse effects associated with unnecessary further testing (overtesting) and unnecessary treatment including surgery (overtreatment), and unnecessary medical costs. A recent study has estimated that in Australia nearly one in five cancers diagnosed in men and one in four cancers diagnosed in women are overdiagnosed cancers.
HealthScreen uses “state-of-the-art” medical imaging such as MRI and CT scans to look for ‘hidden problems’ in people. While these types of tests were not evaluated in the trials included in the Cochrane review, paradoxically, the use of such sensitive testing technology has been identified as a major source of overdiagnosis.
Because they are so sensitive, they can detect minute tissue changes that may not ever cause any symptoms. For example, there is evidence to suggest that very small thyroid papillary tumours are being overdiagnosed and overtreated in people who have no symptoms but have been tested for some other reason with a CT scan, MRI scan or ultrasound.
These scans also detect degenerative findings that commonly occur with age. For example, degenerative changes in the low back, knee and shoulder are seen with ageing in people with and without symptoms, and are mostly benign. If full body checks find these changes in people without symptoms, it can lead to unwarranted diagnoses, cause worry and lead to unnecessary treatment.
There is also concern that screening the genes of healthy people may cause a whole new wave of unnecessary diagnoses. This is because genetic testing is often unable to determine if a person will show symptoms or whether the condition will progress over time.
One example of harm that may arise from full body general health checks is treating high levels of uric acid detected in the blood in people without symptoms. While allopurinol is the mainstay of treatment for gout (a form of arthritis causing painful joints), prescribing this treatment solely on the basis of a high uric acid is not recommended.
In a large population-based study, people taking allopurinol were 10 times more likely to end up in hospital because of a severe skin reaction than in those who did not take allopurinol. And two out of every 10,000 of these people died from the reaction.
Regulating public advertising
It seems HealthScreen is not alone in advertising the benefits of their services without adequate consideration of the potential harms.
A recent project by Professor Ken Harvey at the Monash University School of Public Health and Preventive Medicine known as the “whack-a-mole” project has resulted in a considerable number of individual complaints against claims made about therapeutic goods and services being upheld. This means that the manufacturer or service provider had made claims for a product or service which were not supported by evidence.
Greater regulation of direct-to-the-public advertising for whole body health checks in Australia may be needed. At the very least, companies such as HealthScreen should ensure their advertising includes outline of the potential harms as well as the potential benefits of participating in their services.
Healthscreen’s Medical Director, Dr David Badov, was invited to address the concerns laid out in this article. His response can be found here.
TOO MUCH of a good thing: should the Queen go for breast screening?
In the article below, University of Sydney Public Health Professor Alexandra Barratt explores the contention around the AgeX trial, the biggest randomised trial ever, which is considering how long should older women continue to be screened for breast cancer.
It is published as part of the TOO MUCH of a Good Thing series, which is investigating how to reduce overdiagnosis and overtreatment in Australia and globally, and is published as a collaboration between Wiser Healthcare and Croakey.
To follow the series, bookmark this link, and follow #WiserHealthcare on Twitter.
Alexandra Barratt writes:
As you read over the Queen’s Birthday Honours list this weekend, and perhaps admire the longest reigning British monarch ever, might you wonder whether Her Majesty has regular mammograms for breast screening?
If you did, the answer is ‘probably not’ as the Queen is aged over 90 years and the United Kingdom National Health Service breast screening program offers standard screening to women between the ages of 50 and 70 years.
But what if she were in her 70s or even her 80s? Should she still have a screen? If something was found, would she live long enough to see the benefit of early detection, or would she just experience the more immediate traumas of surgery, radiotherapy and hormone therapy?
This is the question which the AgeX trial, the biggest randomised trial the world has ever seen, is trying to answer: How long should older women keep going for breast screening?
Originally planned to offer just one more screen to English women over 70 years of age, AgeX is now inviting women for up to 3 more screenings through to age 79. That’s five years longer than the current Australian breast screening program which invites women from 50 to 74 years of age.
With a planned sample size of “at least 6,000,000” (yes you read that right, at least SIX MILLION women) this cluster randomised trial “will assess reliably the risks or benefits” of extending screening past 70 years of age. The trial was set up in 2009-2010 to inform screening policy because of uncertainty about the advantages and disadvantages of screening beyond 70 years, amid growing interest in screening longer as life expectancy increased for English women.
A gift to women?
The AgeX trial is a rare and laudable example of rigorous policy testing. It stands in stark contrast to the 2013 Australian extension of the upper age of breast screening from 69 years to 74 years without setting up any comparable study to assess the impact of the policy change. AgeX should be a gift to women the world over by providing much needed evidence about screening in the modern world.
But the AgeX trial has been repeatedly criticised over its design, cost, potential harm to participants, and, most vehemently, over its ethics.
One of the key grounds for concern is the harm that breast screening causes older women through overdiagnosis and overtreatment. Overdiagnosis happens when indolent (i.e. inactive) cancers are detected by screening, but they wouldn’t have caused any harm within the woman’s remaining lifetime had they been left undetected and untreated.
The UK’s own independent review in 2012 found the risk of overdiagnosis and overtreatment was the major downside of screening.
Older women are less able to tolerate surgery and other cancer treatments than younger women, so overdiagnosis and overtreatment have a greater impact on their quality of life. Their higher risk of having, and dying from, other conditions means they’re also more likely to be overdiagnosed in the first place, as well as being less likely to live long enough to benefit from early cancer detection.
A question of consent
Yet, despite older women’s greater risk of being harmed by screening and lower chance of benefiting from it, AgeX is including women (and their medical data) in the trial without their informed consent.
Instead, women in the control arm simply stop receiving invitations after their 71st birthday, while women randomised to the extended screening arm continue to be sent invitations for up to 3 more screens (brief information about the trial is mailed out along with these invitations, but there is no consent process).
So it’s likely that many women are unaware of the risk of overdiagnosis and overtreatment they’re taking by continuing to be screened, and also unaware that they’re participating in a giant experiment. Critics argue that isn’t good enough when the trial itself was designed precisely because the advantages and disadvantages of screening women past 70 years of age are uncertain. More detailed information and an explicit informed consent process is needed, they say.
Complaints are also directed at the analysis. Although the information for women reads “This trial will assess the risks of screening in particular, the chances of being diagnosed and treated for a non-life-threatening cancer”, overdiagnosis isn’t mentioned in the protocol and estimating overdiagnosis isn’t part of the analysis plan.
The primary outcome is breast cancer mortality. All-cause mortality – the outcome that’s crucial for being certain that harms don’t outweigh the benefits of screening – will be reported, but the investigators warn there probably won’t be sufficient statistical power to detect a significant effect as most deaths will be from causes other than breast cancer.
So it’s anything but clear how AgeX is going to measure reliably the risks of screening past 70 years of age.
While the cost of AgeX isn’t given in the protocol documents, funding is being allocated by the Department of Health for the additional screenings. The ever-increasing demand for screening is putting the UK breast screening workforce into crisis: “The number and complexity of breast imaging examinations is rising and services cannot cope” says the Royal College of Radiologists.
The additional screenings have flow-on effects too – more women diagnosed with breast cancer means more surgeries, more radiotherapies and other cancer treatments. As well as extra costs, this has put extra load on surgical and oncology staff.
Coronavirus has changed all that, as breast screening – and AgeX – have been put on hold. This could mean that the results of AgeX – which had already been pushed back from the mid to late 2020s – could be delayed further.
No wonder this trial keeps on whipping up a storm. And it doesn’t look like there will be evidence-based information to guide older women, and possibly Her Majesty, in deciding when to stop screening anytime soon.
Alexandra Barratt (MBBS, MPH, PhD) is a Professor of Public Health in the School of Public Health, University of Sydney. She has a background in epidemiological research spanning clinical epidemiology and public health epidemiology. She is recognised internationally for her research to quantify the benefits and harms (including overdiagnosis) of cancer screening, particularly breast cancer screening. She is a lead investigator on Wiser Healthcare, an NHMRC funded research collaboration to reduce overdiagnosis and overtreatment in healthcare. Overdiagnosis is one of the biggest drivers of iatrogenic harm, waste and opportunity cost in healthcare and is a serious challenge for citizens, patients and healthcare services around the world.
This article is part of an ongoing series that is published as a collaboration between Wiser Healthcare and Croakey.org.
The series investigates how to reduce overdiagnosis and overtreatment in Australia and globally. The articles are also available for republication by public interest organisations, upon request.
Bookmark this link and follow #WiserHealthcare on Twitter.