Selective screening: eliminating unnecessary imaging

28 February 2014



Estimates suggest that 30–50% of imaging exams are medically unnecessary, with troubling repercussions for patients and providers alike. However, as cuts continue to bite, consideration of the appropriateness of tests performed may soon be a legal requirement for US physicians. In this special report, Abi Millar talks to Dr Nehmat Houssami and Cindy Moran about the issues at stake and explores the efforts under way to eliminate unnecessary imaging.


It's a typical day in the radiology department, some time in 2018. To all intents and purposes, little has changed: same pressures, same panics, same fast-moving influx of patients. However, if our notional time traveller is particularly observant, he or she might notice a subtle difference. Rather than simply reading exams, radiologists will need to ensure the ordering physician has consulted appropriateness criteria. If that hasn't happened, they must view the guidelines themselves, and if they don't then their pay is held at risk.

This change may sound small - an exercise in box-ticking - but its repercussions are potentially enormous. The new requirement, which could become US law by 2017 under physician payment reforms, is designed to tackle the issue of imaging overuse. It would mean that exams only take place when medically necessary, and could provide a nuanced approach to cutting costs.

The proposal forms part of a new draft policy document, which is intended to replace the sustainable growth rate (SGR) formula in the US. Pending at the time of writing, it has been welcomed by the American College of Radiology (ACR) among others. According to Paul Ellenborgen, chair of the ACR board of chancellors, it would "help medicine transition from volume-based to quality-based care without interfering in the doctor-patient relationship".

The Medical Imaging & Technology Alliance (MITA) has also endorsed the changes. In the words of executive director Gail Rodriguez, "Appropriateness criteria, rather than the blunt instrument of across-the-board imaging cuts, is how to root out inappropriate use of imaging." Rarely is a policy draft met with such strong and unanimous approval.

But why is the move necessary, and to what extent is inappropriate imaging a problem? To answer these questions, we must survey the last two decades, when the growth of advanced imaging services began to pick up pace.

When considering this period, the phrase 'too much of a good thing' may spring to mind. The late 1990s and early 2000s were a golden era for the market, with technologies developing fast and usage following suit. Compared with other physician-ordered Medicare services, medical imaging advanced at a prodigious rate. Take CT for example, which grew 14.3% year on year between 2000 and 2005.

The boom brought clear benefits to patients - improved diagnostics and better-tailored services among them - but growth could not continue unchecked. Since 2006, the figures have stalled or even declined. A 2012 study in Health Affairs attributed the slowdown to decreased Medicare reimbursements, pre-authorisation requirements and increasing concerns about radiation exposure.

New-found caution

It is clear that today's end users do not see medical imaging as an unalloyed good. Cindy Moran, assistant executive director of government relations and health policy at the ACR, believes payers and patients alike are keen to play it safe.
"From the middle 1990s to the middle 2000s, there was such an enormous volume of studies - there was a CAT scan on every corner," she recalls. "The radiologist in that period was doing their very best just to keep up with reading and interpreting the studies. But the utilisation rate of diagnostic imaging studies has flattened if not diminished in the last few years, and people are becoming smarter consumers of medical care."

This new-found caution has set the tone for concerns about imaging appropriateness. While estimates vary, it is believed that 30-50% of imaging exams may not be medically necessary, a troubling statistic on several counts. Firstly, there's the raw economics - all those extra scans are wasteful. Secondly, there's the issue of patient safety. Any medical imaging exam carries risks, however negligible, and it can be difficult to justify those risks when the scan serves no clear end.

One high-profile example is the use of routine preoperative MRI for breast cancer. This has long been a contentious and emotive subject, with supporters and detractors on either side. However, in recent years, the evidence has stacked up in one direction. According to a meta-analysis, published online in the Journal of Clinical Oncology in January 2014, the procedure is entirely redundant.

"If patients are provided with the correct information based on the best available evidence, then they are able to make an informed decision about whether or not to have a test."

"We used individual data from four studies to investigate whether there was an association between having preoperative MRI and breast cancer recurrence at follow-up," explains lead study author Dr Nehmat Houssami, an associate professor at the University of Sydney, Australia. "Our work showed that there was no significant difference in local or distant recurrence-free survival between the women who had MRI and the women who only had conventional preoperative breast imaging."

The study was the first meta-analysis to examine preoperative MRI in relation to longer-term clinical outcomes. It concluded that, in this scenario, MRI might actually have negative effects.

"It could lead to unnecessary wider surgical resection, and therefore potentially worse cosmetic outcomes," says Houssami. "There are significantly higher odds of mastectomy, without improved prognosis, and delay to definitive surgery. There is also the issue of increased costs, such as the cost of the MRI and the related needle biopsy."

Because MRI is beneficial for women with newly diagnosed cancer, it might seem to follow that presurgical scanning would carry these benefits too. Unfortunately, this conclusion now appears erroneous, and practices may well need to change.

"I hope this study will be used to inform evidence-based breast cancer and imaging guidelines," says Houssami. "There is always room for more research, but this is the best available evidence to inform practice in 2014."

An educational tool

If the new SGR-replacement legislation is passed, then evidence-based guidelines will form the crux. No medical imaging exams can be undertaken without at least consulting the available information. And, while physicians will not be bound by what they read - they will still have free rein to follow their judgement or honour a patient's wishes - they can no longer plead unawareness of the facts.

This should have a tremendous effect on imaging use. While there are many possible reasons for unnecessary imaging - financial incentives, defensive medicine and self-referral among them - it is clear that missed educational opportunities have a significant role to play.

"A lot of it is ignorance," confirms Moran. "This legislation will address that aspect - hopefully it will be used as a tool to educate the ordering physician and also educate the patient about what the appropriate studies should be."

It might mean, for instance, that breast cancer patients are told about Houssami's latest work. "If patients are provided with the correct information based on the best available evidence, then they are able to make an informed decision about whether or not to have a test," she says. "It is essential to tell these women that the test has not been shown to improve breast cancer treatment or prognosis and that it increases the likelihood of mastectomy without related benefit."

The question is really one of practicalities - as best practice changes in line with the evidence, how might policymakers go about ensuring up-to-date guidelines are used?

"It requires a real commitment of time and money to develop the criteria," confirms Moran. "They are constantly being re-evaluated, and, as the evidence changes, it affects the appropriateness guidelines. So this is a dynamic product that is always evolving. The criteria will need to be developed by national physician professional specialty societies, but, ultimately, it will be up to the secretary of the Department of Health & Human Services (HHS) to determine which ones meet the standards that Congress lays forth."

Here, the ACR will be in a good position to help. Its own appropriateness criteria, considered the industry gold standard, have been in existence for 20 years. They are publicly available and continuously updated, and, when used to inform clinical decision support systems, have been shown to slash unnecessary imaging. Whether these criteria are adopted wholesale is up to the HHS.

At present, the SGR replacement bill is awaiting consideration by the House of Representatives and Senate. It is hoped that this provision, along with the rest of the bill, will be signed into law in the early part of this year, but all depends on budget considerations. The policy, after all, comes with a hefty price tag - a minimum of $150 billion over the next ten years - and Congress has yet to determine where it will find the funds.

"In a sense, we're held hostage by the negotiations over the larger bill," says Moran. "The question is: is that vehicle going to be able to get to the floors of the house of the Senate and get signed into law? - and there are some big steps to take before that can happen. But if they can agree how they're going to come up with the money, there's a very strong change that the overarching bill will pass, and we will then go to work to get it implemented."

In the meantime, there is surely scope for optimism. These proposed changes should reinstate the radiologist as the manager for diagnostic imaging services for patients, shaking any perceptions that theirs is a passive role. Through checking the appropriateness of every study ordered, they can better target their services, prioritising quality over quantity and evidence over convention. A true patient-centric model of healthcare may not be too far afield.

Routine use of preoperative MRI for breast cancer is an example of potential imaging overuse.
Dr Nehmat Houssami is a breast and public health physician who has worked in breast cancer services for over 20 years. Her clinical practice and research focus on breast imaging and screening, and much of her research is aimed at supporting evidence-based practice.
As assistant executive director of government relations and health policy at the ACR, Cindy Moran leads the college’s advocacy efforts and economics and health policy initiatives. Moran previously ran the Washington DC government affairs office of the American Academy of Ophthalmology.


Privacy Policy
We have updated our privacy policy. In the latest update it explains what cookies are and how we use them on our site. To learn more about cookies and their benefits, please view our privacy policy. Please be aware that parts of this site will not function correctly if you disable cookies. By continuing to use this site, you consent to our use of cookies in accordance with our privacy policy unless you have disabled them.