Advances in medical technology are often seen as a key reason for improvements in the nation’s healthcare. Against this backdrop, policymakers and the public see growing medical expenditures and the lack of affordable health insurance as major concerns, especially in this election year. The United States spends more of its gross domestic product on health than any other Western nation—roughly 16% in 2007, up from 5% in 1960, according to the Centers for Medicare & Medicaid Services’ National Health Expenditure Data.

Because of its rapid growth, medical imaging technology faces increased scrutiny from policymakers anxious to reduce costs. Of relevance to the radiology community, the Deficit Reduction Act of 2005 proposed a cut of $2.8 billion from Medicare reimbursements for medical imaging services over the 2007 to 2010 period, according to a Congressional Budget Office cost estimate. MedPAC, a congressional advisory committee, identified the need for cost-effectiveness studies on justifying spending on expensive technologies. Experts have also highlighted the need for evidence-based approaches to healthcare policy and decision making, including a role for cost-effectiveness analysis (Am J Manag Care. 2008;14(4):185-188).

This article provides an overview of economic evaluation as it pertains to medical imaging technologies, including opportunities and challenges in measuring value. This evaluation can help shape the policy debate by reining in benefits, as well as costs. We hope that this brief review will provide radiologists and other imaging professionals with a better understanding of economic and policy trends with respect to medical imaging procedures in light of increasing cost-containment pressures in the U.S. healthcare system.

Value in Medical Imaging
What does cost-effectiveness really mean? Imaging modalities may pay for themselves if they yield cost reductions (such as the avoidance of unnecessary hospitalizations) that more than offset the expense of the imaging test. Such cases of cost savings certainly represent a cost-effective use of healthcare resources. Alternatively, there may be circumstances for which imaging is associated with an increase in overall costs.

But nonetheless, the test provides reasonable value for money because the higher costs are worth the expense in terms of improved clinical outcomes produced. This can also represent a cost-effective use of resources. Of course, some imaging may lead to higher costs and marginal clinical benefits and thus would not be considered cost-effective.

Economists attempt to capture value in a broad sense, accounting for all costs and benefits associated with a health service, including information on the accuracy and costs of tests and the consequences for follow-up treatment. This broader definition of value is consistent with a trend toward patient-centered outcomes, namely the consequences of a patient’s test results in terms of treatment decisions, morbidity, and mortality. This clearly involves taking a longer term perspective incorporating patient preferences and quality of life in addition to costs. So what are the health consequences of an imaging test and its results?

Potentially positive consequences include the following:

• More timely and effective treatment: Imaging information can lead to improved quality of life by helping doctors start the proper treatment sooner and potentially avoid unnecessary procedures.

• Reassurance value: The value of knowing is difficult to quantify, but efforts are underway to measure the value of medical information.

Potentially negative consequences include the following:

• Incorrect information: False-positive results may raise fear with patients and lead to unnecessary treatment.

• Potential side effects: Many imaging tests expose patients to ionizing radiation, and there are also the risks of reactions to injected drugs or contrast agents.

The ultimate value of any given imaging procedure depends on how the positive and negative consequences play out relative to other clinical strategies, such as no imaging or an alternative test.

Quantifying Outcomes
Economic evaluations often use a single index of value to measure the impact of medical technology on life expectancy and the quality of those subsequent years. The most commonly used measure is the quality-adjusted life-year, a measure that can be compared across disease areas. The difference between the ratios of costs to quality-adjusted life-years of two treatments (the incremental cost-effectiveness ratio) is used to inform reimbursement decisions in the United Kingdom.

To evaluate consequences over the long term, researchers develop decision-analytic models to attempt to predict outcomes following a test or sequence of tests. For example, Jager and colleagues reported that MRIs may yield insights into early cancer, which may progress, influencing life expectancy and quality of life (Radiology. 2000;215(2):445-451).

As noted, medical imaging may improve outcomes but also increase overall costs. Whether a procedure is cost-effective in any particular disease area depends on several factors, including disease prevalence, test accuracy, test frequency, and costs for both testing and treatment. One prominent example is the guaiac stool test, which has been shown to be reasonably cost-effective in certain patient groups when administered once a year but not so when repeatedly administered, as described in a well-known paper by Neuhauser (N Engl J Med. 1975;293(5):226-288).

Several published reviews have evaluated the cost-effectiveness of radiologic procedures, and the results show variability, depending on disease area and type of test. Most of the reviews have evaluated studies that included costs and either quality-adjusted life-years or some measure of health-related quality of life.

Other studies examine only costs, assuming no comparative benefit, but they do account for the possibility of avoiding costly healthcare services. Some of these analyses have demonstrated instances in which imaging may ultimately save money. Examples include conventional and multidetector CT, diffusion-weighted and conventional MRI, and bone densitometry. Examples of selected studies in specific patient groups follow.

Other economic evaluations have shown that some uses of imaging may not save money or be cost-effective under conventional standards of what constitutes reasonable value for money. A review of some cost-saving examples is instructive in highlighting opportunities to demonstrate that imaging can bring good health and economic value and illustrates some of the measurement challenges that arise with this kind of research.

Abdominal CT scans in appendicitis: A prospective, nonrandomized study published in The New England Journal of Medicine followed 100 consecutive patients presenting to an emergency department (ED) with suspected appendicitis who were admitted to the hospital (1998; 338(3):141-146). The researchers found that routine use of abdominal CT scans prevented 13 unneeded appendectomies (for a savings of $47,281) and 50 unneeded inpatient days (for a savings of $20,250). After the costs of CT were accounted for, the total savings overall was $447 per patient.

Another study by Morse and colleagues retrospectively analyzed data on patients who received appendectomies and found cost savings associated with abdominal CT scans of $1,412 per patient for women aged 14 to 69 (Am Surg. 2007;73(6):580-584).

Emergency evaluation of low-risk presentations of chest pain: A randomized trial of 203 patients by Goldstein and colleagues was recently conducted to assess the safety, diagnostic efficacy, and efficiency of multidetector CT (MDCT) vs. standard of care in patients with low-risk acute chest pain presenting to hospital EDs (J Am Coll Cardiol. 2007;49(8):863-71). In order to avoid overlap (see below), the indication of chest pain did not include patients with known heart disease. Although both methods were deemed safe, MDCT reportedly reduced diagnostic time compared with standard of care and significantly lowered ED costs (including imaging tests) by nearly $300 per patient due to a reduction in hospitalization time.

Coronary CT angiography (CTA) in heart disease: Coronary CTA for heart disease patients led to an average cost savings of $1,454 per patient (accounting for the cost of the procedure), according to a single-center, prospective study published last year in the Journal of Cardiovascular Computed Tomography (2007;1(1):21-26). The study examined the cost savings of selective catheterization in patients with mildly abnormal or equivocal myocardial perfusion scans.

CT for mild head injury: A Swedish study examining the effects of imaging on mild head injury modeled the costs of acute CT and home care vs. inpatient observation for mild head injuries and concluded that CT would be a less costly option (between 8% and 54% less) than the alternative (Emerg Med. 2004;21(1):54-58). On average, costs were one third lower with CT and home care than with hospitalization. A switch to the CT strategy was projected to save $62 per patient scanned.

CTA-CT perfusion for stroke: A retrospective database study by Gleason et al was conducted to compare the scenario of a switch from standard medical management to CTA-CT perfusion protocol for patients with suspected ischemic stroke admitted to a hospital (Acad Radiol. 2001;8(10):955-64). The findings showed a total savings of $3,568 per patient with small-vessel disease and an estimated mean cost savings across all other subtypes of stroke patients of $1,695 per patient, even after accounting for imaging costs.

Diagnostic radiology in general inpatient cases: In a study that examined the cost-savings associated with general inpatient care, Beinfeld and Gazelle reported that CT and MRI were not responsible for the increases in total costs incurred by hospitals from 1996 to 2002 (Radiology. 2005;235(3):934-939). They reported that for every $100 spent on diagnostic imaging tests, hospital stays are shortened by 0.25 days at an estimated cost savings of $293 per patient.

The Way Forward
Pressures to contain costs will lead to uninformed restrictions in coverage or reimbursement unless value concepts are brought into the discussions. Therefore, establishing an agenda for measuring value is an important task. One approach may be a comparative effectiveness center, which has been proposed in recent federal legislation. While the merits of this proposal have been debated, a movement toward evidence-based medicine is entirely consistent with the notion of demonstrating value.

Another trend is to link evidence of value to payment arrangements. Examples include pay-for-performance programs for physicians and noncoverage of the costs of avoidable complications, which has now become part of the Medicare Part A payment system. Medicare will not pay extra for hospitalizations complicated by several conditions, such as decubitus ulcers and Staphylococcus aureus infections.

We are not trying to debate whether these efforts should be pursued; our point is that they are being pursued, and it is prudent to be out in front of these challenges.