For a sidebar on how this story came together, click here.
Dual-energy x-ray absorptiometry (DXA) has long been the gold standard for osteoporosis screening. But now, thousands of patients may be inappropriately undergoing bone mineral density studies via quantitative computed tomography (QCT) -- which may be less accurate and more expensive, with far higher radiation exposure.
Experts told 51˶ QCT can exaggerate fracture risk in the lumbar spine by as much as 1.5 standard deviations, can be almost three times as expensive, and delivers between 1,000 and 3,000 times the radiation of DXA, according to some estimates.
Use of QCT is on the rise, according to a 51˶ review of the Medicare claims . From 2015 to 2019, providers submitted 32,694 claims for QCT bone mineral density studies performed for fee-for-service beneficiaries. The number grew each year, from 5,574 in 2015 to 6,473 in 2017 to 7,453 in 2019, the latest year for which data are available.
The database also revealed inexplicably wide variation across states, with 22 states -- including some very large ones like Pennsylvania and New Jersey -- not submitting any claims for QCT services in 2019 while smaller states like Alabama, Arkansas, Iowa, Mississippi, and Oklahoma submitted hundreds.
The issue is cause for concern because, even though osteoporosis is due to low adherence to screening recommendations, a misdiagnosis can prompt clinicians to put their patients on drugs such as bisphosphonates or newer, more expensive injectable drugs. Medication side effects can range from broken bones or fractures, jaw bone problems, infections, skin problems, and muscle pain.
Bart Clarke, MD, a bone and mineral metabolism specialist at the Mayo Clinic in Rochester, Minnesota, who helped write the for the American Association of Clinical Endocrinology, has seen QCT's inaccuracy up front.
He told 51˶ that in neighboring Iowa, 40 miles to the south, a lot of patients -- for unclear reasons -- get their bone mineral density screenings by QCT rather than DXA.
"And, of course, when they do them, most often they're told you've got terrible osteoporosis and you've got to go on drugs. Well, they come here [to the Mayo Clinic] because they don't want to go on drugs and we say, 'Okay, let's get a DXA to see. And almost always the DXA T-scores are 1.0 to 1.5 higher than, less negative [better] than, the QCT."
T-scores are a common measurement in bone mineral density imaging. They compare the patient's bone to that of a young adult. Z-scores, also usually given, compare the patient to others in her age group.
The problems with using QCT to measure bone density in the lumbar spine aren't just the radiation, the cost, and the accuracy. There's also a lack of reliable comparisons that allow a patient's study to be compared with the average for the U.S. population, Clarke said: "There are a lot of technical details where the QCT is felt to be less accurate."
The Cortical Bone's Role
More specifically, Clarke explained, the QCT focuses on density changes in the inner, trabecular bone in the spine. It does not include the cortical shell on the outside of the vertebrae. The DXA includes both, and is a better way to measure bone density, he continued, because cortical bone plays an important role in bone density measurement.
Clarke said QCT is more likely to concur with DXA on hip bone density because for the part of the hip that is measured, there is not much trabecular bone; it's mostly cortical bone.
The Medicare database shows that DXA overwhelmingly has been the technology of choice for bone health. In 2019, 23,967 clinicians submitted more than 2.66 million claims under 77080, the CPT code for DXA.
Thomas Link, MD, PhD, chief of musculoskeletal imaging at the University of California San Francisco School of Medicine where the QCT technology was first tested in the 1970s, noted that the QCT has some limited uses. It sometimes offers more detail for patients who are either very small or obese, and may assist with diagnosing patients with severe degenerative disease of the spine.
But today, Link said, QCT is mostly used for research or solving problem cases when results of a DXA are inconsistent.
He agreed with Clarke, however, that the QCT exaggerates bone mineral density loss in the spine if the T-score is used. If it is, more patients will be diagnosed incorrectly with osteoporosis with a QCT than with a DXA, he said.
For example, if a T-score given by a DXA of the spine says -2.0, then there is a likelihood that the QCT T-score on the same patient would be -2.7, showing more bone density loss than actually exists "and that's why you shouldn't use T-scores with QCT of the spine," he said.
That's why, he continued, the American College of Radiology's recommend using an absolute value of <80 mg/cm3, and not a T-score, for a more accurate assessment.
The U.S. Preventive Services Task Force recommendation for osteoporosis screening for women starting at 65 years of age does not mention QCT among the tools used.
'I Never Order Them'
Nearly all experts interviewed for this story prefer a DXA as a first choice to screen for osteoporosis.
Cleveland Clinic musculoskeletal radiologist Naveen Subhas, MD, who chairs the American College of Radiology's committee for imaging, said that while the QCT gives different values for bone mineral density for the spine than DXA, the QCT is not inaccurate nor does it lead to a false diagnosis if appropriately interpreted using specific established ranges for QCT.
As for the differences between the two, he continued, QCT's ability to measure bone loss in the trabecular region of the spine is important because that's where bone loss first appears, so the QCT may be more sensitive to detecting early change.
But for clinical purposes, he said, "DXA should be the first line."
The bottom line is that it's rare that Cleveland Clinic doctors order a QCT. "We just feel that the higher costs and higher radiation is usually not needed," he said.
The Medicare claims database showed that no Ohio provider submitted a claim for a QCT in 2018 or 2019.
"QCT is not used in clinical practice very often," said Suzanne Jan De Beur, MD, an endocrinologist and immediate past president of the American Society for Bone and Mineral Research (ASBMR) and associate director of the Johns Hopkins Bayview Clinical Research Unit in Baltimore.
Besides, she said, QCT results can be artificially altered by aortic calcification or arthritis.
If people took a QCT at face value and equated a T-score on a QCT with a T-score on a DXA, patients could be put on drugs who don't need to be, she acknowledged.
"In my mind, there's not a lot of utility for studies other than DXA," she said. "I never order them."
The Medicare database showed that just one physician in the state of Maryland submitted a claim for 13 QCT study reads in 2019.
Likewise, Janet Rubin, MD, an endocrinologist at the University of North Carolina who receives federal funding to study metabolic bone disease, remodeling, and osteoporosis, said, "I don't order them and don't see a need to."
Radiation exposure is a key concern in the use of QCT, physicians said.
American College of Radiology spokesman Shawn Farley told 51˶ that the QCT, with 1-3 millisieverts, delivers radiation that is approximately 1,000 to 3,000 times greater than a and is roughly equivalent to between 4 months and a year of background radiation exposure.
Additionally, patients 65 and older are advised to return every 2 years for repeat studies using the same imaging technology, adding another dose to their exposure.
Huge Variation
It's a mystery why QCT Medicare claims seem to be clustered in a few states and why no claims were submitted from 22 states, including some of the largest states like Pennsylvania and New Jersey, raising questions about the reasons for the wide variation. (See infographic above.)
Nearly half of the 7,453 claims in 2019, in fact, came from clinicians who ran QCT studies in just four states: 21 in California submitted 1,888 claims, most of them from the Los Angeles area; six in Alabama submitted 749 claims; 17 in Mississippi submitted 565 claims; and 17 from Iowa submitted 493 claims.
In Iowa, population 3.9 million, three locations submitted the 493 claims for QCT screening, and two, Iowa Ortho in Des Moines and Iowa Radiology in Clive submitted 431. 51˶ called and emailed them to ask about why so many claims were submitted by those addresses, but no one responded.
Several clinics in other states such as Oklahoma, where claims were submitted for 338 QCT studies, were called and though a return call was promised, none was received.
One possible explanation for the steadily rising number of QCTs is its higher payment in office settings and imaging centers, compared with DXA.
According to the American College of Radiology and the Medicare Physician Fee Lookup Tool, Medicare pays almost three times more for a QCT in a physician's imaging center (code 77078) than for a DXA (code 77080), with a global payment of $111.09 compared with $38.07. (Under the hospital outpatient prospective payment system, Medicare pays more, $120.78 for a DXA versus $94.82 for a QCT, but this isn't unexpected because payers understand regulatory requirements for hospitals are usually more expensive, and hospitals generally treat sicker patients.)
Reimbursement rates for DXA were drastically cut in 2007 under the Deficit Reduction Act, and except for a 2-year period, have not gone back up. Reimbursement rates for other radiologic services also saw dramatic decreases.
Some experts noted that over the past several years, clinicians unhappy with the low DXA reimbursement have decommissioned their DXA scanners, especially those in office practices that took up space.
"You could have a scanner in your office and get a decent reimbursement for it. But then reimbursement dropped down to $40 and a bunch of doctors got rid of their DXA," said Jan De Beur. "By the time you paid a tech and the upkeep for the machine, it was just a money loser. A ton of DXAs have been money losers."
It was easier for many to just have their area imaging centers use their CTs, she said. "QCT is just a CT scanner with a different software package, and a CT can be used for many different imaging studies. But a DXA can only be used for one thing."
Link, the imaging expert at UCSF, said that maintaining DXA services has been a struggle in Northern California, where several providers went bankrupt because of its "incredibly low reimbursement." UCSF has maintained its five machines because they're an important screening tool and "a public service; we're not making any real money on it," he said.
Improving DXA Reimbursement
Douglas Fesler, executive director of the ASBMR, said that DXA's low reimbursement has compounded a growing problem of underdiagnosis and undertreatment of osteoporosis. The number of people getting screened has been declining over the last 15 to 17 years, he said.
Today, he added, "most people, frankly, don't know they have osteoporosis until they break something. And that's usually, unfortunately, too late."
Asked why efforts to restore DXA reimbursement rates to reflect their costs have failed over the years, Fesler said that among those who set payment, "there's a perception that not providing this is not harming anyone -- just making it inconvenient for some, especially patients in rural areas" who have to drive farther to get a scan.
"However, the data is alarming," he continued. "Only 9% of female Medicare beneficiaries received a DXA within six months of an osteoporotic fracture. There are many reasons why folks are not getting DXAs but access is certainly one. The more follow-up appointments an individual needs to make to get a test, the less likely they are to get the test."
The of 61 radiology, orthopedic, cancer, women's health, and other organizations has been pleading with the U.S. and to pass Senate bill 1943, "Increasing Access to Osteoporosis Testing for Medicare Beneficiaries Act of 2021," saying that declines in screening because of reduced payment for DXA has cost Medicare $3 billion "to treat hip fractures alone."
"Despite being recommended by the Centers for Medicare and Medicaid Services as a critical preventive test in the 'Welcome to Medicare' exam, the reimbursement rate for the DXA administered in a doctor's office has declined from $140 in 2007 to only $39 in 2021, while hospital reimbursements have increased to $119 for the same procedure," they wrote.
That's resulted in a 44% decline in DXA office providers since 2008, with an estimated 1.65 million patients with osteoporosis who went undiagnosed and thus, untreated, which "caused 71,775 additional costly hip fractures due to reduced screenings, and unfortunately, 15,647 unnecessary deaths from additional hip fractures," the letters said.
Minority women were hit especially hard, the coalition continued, with 29% fewer Black women and 32% fewer Hispanic women tested than white women.
Bone mineral experts hope that lawmakers will order higher reimbursement for DXA to stop providers from decommissioning their machines. Said the ASBMR's Fesler: "I believe we had a good chance in finally getting the DXA bill passed in the last session of Congress if it were not for COVID. And we are doing everything we can in this session to finally realize it by the end of the year."