Flatlined

Unnecessary Imaging

Outpatient services such as MRI and CT scans, colonoscopy, and outpatient surgeries account for nearly ten percent of health care spending. Medical imaging accounted for about one-quarter of outpatient spending in 2005 and was the programs fastest growing cost, escalating at 9 percent per year.(1)(2) The volume of CT scanning is so high that the radiation exposure from it could increase cancer risk. While the risk of radiation-induced cancer from CT scanning is minimal for a single individual, it is increased for the entire US population.(3)

There are five reasons for the rapid growth in the volume of medical imaging lack of standards for a diagnostic work-up, lack of systems of care, improvements in the technology, defensive medicine, and the ownership of diagnostic facilities and equipment by physicians who refer their patients to themselves.

Improved Management

MRI and CT scanning have become so sophisticated that they permit doctors to detect problems that only a few years ago would have been invisible or only guessed at. In many conditions such as following the course of a patient with cancer, such imaging permits doctors to sensitively determine of chemotherapy is working by comparing the size of tumors before and after treatment.

In my clinic, if a patient did not improve with spine surgery, as about 15 percent do not, we could use simple, non-invasive imaging studies to determine why. Usually we would see scar tissue which is only treatable by pain medication. Rarely, we would pick up a fluid collection, a cyst that had recurred on the spine, a ruptured disc that had come back or something that needed further surgical treatment. Without imaging procedures such patients problems would either never have been detected or only detected at a follow-up operation performed because there was no other way to tell what was going on.

I practiced long enough to remember when neurosurgeons did not have CT scans and MRIs. Modern imaging transformed neurosurgery. Some of the other reasons that imaging is growing so fast are not so attractive, however.

Lack of Systems of Care

I saw about thirty-five patients every Wednesday in clinic and nearly every patient came with an MRI under his or her arm. Two-thirds of the studies were of such marginal quality that I simply ordered a new MRI or a study called a myelogram in which dye is injected into the spinal fluid by a spinal tap before making a surgical decision. The MRIs that came with the patients were usually of poor quality because the referring doctor had ordered the study on the nearest equipment, often cheap units housed in mobile trucks that visited their clinic's parking lot once a week. My patients were often from small towns where such units seem to be ubiquitous. The MRI scanners had insufficient power to detect the anatomic distinctions needed to make a surgical decision. The cost to Medicare and to commercial insurers for a good quality MRI is exactly the same as the cost of a poor MRI.

Sometimes the studies were of good quality but I could not decipher the labeling of the images because there are no standards for labeling. The only convention agreed upon in radiology is that right and left are standard on all films. If the labeling of the images was foreign to me and I could not decipher it, I could not afford to guess. So I never thought twice about ordering duplicate studies. The cost of the images I reordered in each clinic was about $24,000 per clinic or about $1,152,000 per year in duplicated studies. Over the fifteen years I ran that clinic I was responsible for over $17 million in studies that from a systems perspective were unnecessary. I can assure you that I was not alone in this behavior.

Ideally, I might instead have called the referring physician and recommended that they not image the patients they sent me at all, so I could order my own. I did not always know who the referring doctor was. The doctors probably would not have desisted if I had called anyway. A doctor in a small town uses such imaging studies to determine whether to trouble the patient with a referral to Houston—the patients expect it. The problem here is that the practice of medicine is disorganized, without any systems and to aggravate that situation, there was absolutely no incentive for me or any other doctor to be more judicious about the use of imaging studies.

Physician-owned Facilities

Proven overuse of services in certain physician-owned facilities resulted in passage of the Stark Law in 1989. This law prohibits physicians from referring Medicare patients to facilities in which they or family members have ownership if those facilities provide laboratory services, physical or occupational therapy, radiology, or radiation therapy. There are few other restrictions on physician ownership of facilities because at the time these laws were passed there was only data relating physician-ownership with over-utilization for these specific services. Quite a bit has changed since 1989, though the law has not.

I interviewed two radiologists from a western state who told me what had happened to their practices. Orthopedic surgeons are now getting into the radiology business. Although it is illegal for doctors to refer their patients to outside radiology facilities that they own, it is legal for a doctor to install imaging in his office and refer his patients for studies there. One of the radiologists told me about a large orthopedic group in her community. “The orthopedic surgeons buy cheap MRIs (which also produce poor quality images) and then image their patients in their office. Now my husband is an orthopedic surgeon and he will not image anyone with a painful knee unless the pain persists for at least three weeks. This group of orthopods immediately images anyone who comes to their office with a sore knee. They send the images to us to be read after collecting the technical fee. The images are so poor that if they sent a patient to us and we gave them an image like that, they would be howling. Our imaging volume is down because so many surgeons are doing this.”

I have heard numerous stories of how doctors have found ways around the Stark Laws, however. One means is for a group of doctors backed by someone with a lot of money to buy several imaging facilities in a community. The doctors split ownership of the facilities so that each doctor refers to the other doctors' facilities. I have heard of one innovative group that shifted ownership of a facility on different days.

You can be sure that a doctor with a lot of money to make or lose from such a business will find many more indications for imaging than a doctor who does not.

Medical Malpractice and Defensive Medicine

When doctors are accused of wasting resources, they blame medical malpractice litigation. The public repeats what they have heard doctors say. But the data do not provide any support for the position that defensive medicine is the main driver of health care cost. Defensive medicine is the ordering of images, laboratory tests, and the performance of invasive testing procedures (such as biopsies) for no other reason than to reduce the doctor's risk of being sued or of losing a malpractice suit. The most common defensive act is to use excessive diagnostic techniques in order not to miss a low probability condition, for example cancer or a heart attack.

No one can say how much of the outpatient imaging is appropriate utilization of services and how much is over-utilization. In only a few areas, such as cardiac catheterization studies for the heart, is there even a yardstick to determine which patient should undergo diagnostic imaging.

In highly litigious states almost all specialists say they practice defensive medicine. (4)Three-quarters of US physicians say that they order unnecessary tests and make unnecessary referrals, and half suggest unnecessary biopsies as a means of protecting themselves. Almost all physicians, nurses, and hospital administrators believe that defensive medicine is a significant contributor to health care costs.(5) I can assure you that when a patient requested an imaging study that I did not think was needed, that I ordered it with no questions asked. In the unlikely event that my failing to order a study missed something that the patient thought I should have found, I would have been set up for a law suit.

When Texas passed a law that capped punitive damages (the kind that a jury can levy because they think you did something wrong or are mad at you) the number of lawsuits decreased precipitously. States with caps on punitive damages consistently have fewer medical malpractice suits so their cost of care should be lower than states without such caps.

However, when health care costs (adjusted for other factors) in the twenty-eight states that have enacted limits on medical malpractice payments were compared with states without such caps, the study's investigators concluded that laws limiting malpractice payments lower state health care costs by no more than three to four percent. (6)Another similar study but older and more limited in scope, found an increase in hospital spending of five to nine percent in states without caps.(7) The other cost of malpractice litigation, malpractice premiums, accounted for 0.46 percent of total health spending in 2001.(8)

Lack of Standards for Diagnostic Work-ups

An example of what commonly happens with medical technology in general and imaging in particular is the story of Coronary CT angiography. To definitively determine whether a patient has heart disease, the gold standard is a procedure called a cardiac catheterization with coronary angiography. This is a catheter procedure just like the one done to open blockages in the coronary arteries except dye is injected into the arteries of the heart (coronary arteries) through the catheter rather than using the catheter to dilate vessels. In practice, if a coronary angiogram is positive, the patient is likely to undergo a procedure to dilate the blockage at the same time.

Cardiologist have recently developed a CT scan method of imaging the coronary arteries that only requires injection of dye in a vein, much less invasive than a catheter threaded through a blood vessel in the groin to the heart. The question is who needs it. The study seems to be accurate if it is negative but it may not be accurate enough for most treatment purposes if it detects blockages. There is not enough information to be certain what the appropriate use of the study is. The worry is that it will just be layered on top of other studies and drive cost with little or no benefit.

The one thing that everyone seems to agree on is that it should not be used to detect blockages of the coronary arteries in people without symptoms. In fact, the American College of Cardiology developed “appropriateness criteria” that state this.(9)

Based upon this judgment and the relatively unproven indications for the procedure in symptomatic patients, the Medicare program limited coverage for the procedure. Their plan was not to pay for the procedure in patients without symptoms and to pay for it for those with symptoms only if they were part of a clinical trial.(10)

The resulting outcry from cardiologists resulted in letters to Medicare from 79 members of the House of Representatives and a dozen Senators calling the coverage decision into question. The Medicare program reversed a correct decision in response to political pressure from cardiologists applied through the US Congress. According to a number of cardiologists quoted in a New York Times report, more information is needed before it is known in which clinical circumstances this procedure is of benefit. (11) As long as there is insufficient information to determine when images are useful and when they are not and no political will to find out, unnecessary imaging will flourish.

A Multi-faceted Problem

I have thought about what would have changed my behavior so that I took responsibility for ordering fewer images. Two policies would have changed my clinic practice. A regular report about how far I had deviated from the standard would have kept the matter on my mind, but that would not have been enough. Surgeons mostly endure long clinics because it is the way they find people who need surgery. Operating is what surgeons enjoy. I used the screening MRIs done by the referring doctors as much as they did. I would not allow a patient to make an appointment with me without an imaging study in hand because I did not want my clinic clogged up with non-surgical patients.

In this circumstance neither I nor the referring doctors would have paid attention to reports of our imaging practices unless we paid a financial price for them. For instance, taking the cost of excess studies out of our Medicare payments at the end of the year would have quickly generated some policy changes in our clinics.

Standards for diagnostic work ups could serve another purpose as well. Doctors don't perform major procedures, especially marginally-indicated ones, to avoid lawsuits; rather they over-order images, blood tests, and diagnostic procedures such as biopsies. One-third of malpractice cases are for failure to diagnose.(12) A doctor that adhered to a standard protocol for a diagnostic workup could be held harmless for failure to diagnose. Such a law combined with a financial penalty for excessive ordering would be a powerful inducement for appropriate imaging.

  1. A Data Book: Healthcare Spending and the Medicare Program, June 2007, Medicare Payment Advisory Commission (MedPAC). Accessed online 2007: http://www.medpac.gov/documents/Jun07DataBook_Entire_report.pdf.
  2. David Kashihara and Kelly Carper, “National Health Care Expenses in the US Civilian Noninstitutionalized Population, 2003,” Medical Expenditure Panel Survey (MEPS) Statistical Brief #103, November 2005.
  3. David J. Brenner and Eric J. Hall, Computed Tomography — An Increasing Source of Radiation Exposure, New England Journal of Medicine, 2007; 357:2277-84.
  4. David M. Studdert, Michelle M. Mello, William M. Sage, et al., “Defensive Medicine Among High-Risk Specialist Physicians in a Volatile Malpractice Environment,” JAMA 293, no. 21 (June 1, 2005): 2609-2617.
  5. “Most Doctors Report Fear of Malpractice Liability Has Harmed Their Ability to Provide Quality Care,” Editors, Humphrey Taylor and Robert Leitman, Health Care News, Harris Interactive 2, no. 10 (May 16, 2002).
  6. Fred J. Hellinger and William E. Encinosa, “The Impact of State Laws Limiting Malpractice Damage Awards on Health Care Expenditures,” American Journal of Public Health 96, no. 8, (August 2006).
  7. Daniel P. Kessler and Mark McClellan, “Do Doctors Practice Defensive Medicine?” NBER Working Paper Series #5466, February 1996.
  8. Anderson, et al., “Health Spending in the United States and the Rest of the Industrialized World,” (see note 17, chapter 7).
  9. David A. Bluemke, Stephan Achenbach, Matthew Budoff, Thomas C. Gerber, Bernard Gersh, L. David Hillis, W. Gregory Hundley, Warren J. Manning, Beth Feller Printz, Matthias Stuber and Pamela K. Woodard From the American Heart Association Committee on Cardiovascular Imaging Multidetector Computed Tomography Angiography. A Scientific Statement: Noninvasive Coronary Artery Imaging. Magnetic Resonance Angiography and the Councils on Clinical Cardiology and Cardiovascular Disease in the and Intervention of the Council on Cardiovascular Radiology and Intervention in the Young, Circulation published online Jun 27, 2008; DOI: 10.1161/CIRCULATIONAHA.108.189695
  10. “Medicare Reverses Decision to Limit Coverage of Cardiac CT Heart Scans (updated)”, Angioplasty.org, Imaging News, March 15, 2008.
  11. Alex Berenson and Reed Abelson, Weighing the Cost of a CT Scan's Look Inside the Heart: The Evidence Gap, June 29, 2008, New York Times
  12. Studdert, et al., “Claims, Errors, and Compensation Practices in Medical Malpractice Litigation” (see note 16, chapter 9).