Medicare claims for a high-tech form of radiation therapy to treat breast cancer that's twice as expensive as simpler techniques increased tenfold in recent years, helping to drive up the average cost of breast radiation by 33 percent, according to a new study by M.D. Anderson researchers.

The jump in billing for the technique, intensity-modulated radiation therapy, could be driven in part by reimbursement policy, the researchers argue in the study, released Friday in the Journal of the National Cancer Institute.

They also suggest that local Medicare policy has a big effect on use of the technology.

IMRT uses 3-D treatment planning and dynamic multileaf collimators to shape the radiation beam to deliver a more even dose to the tumor and a smaller dose to surrounding, healthy tissue, such as the skin, the other breast and possibly the heart.

The authors, led by Dr. Benjamin D. Smith, a radiation oncologist with M.D. Anderson in Houston, believe that the growth of the procedure in recent years comes in part from patients getting drawn to the technology because of clinical evidence that it can reduce skin injuries, lower toxicity to healthy tissue and improve cosmetic outcomes.

But they also believe higher payments might have helped. According to the researchers, the mean cost for non-IMRT radiation treatments is $7,179; for IMRT, it's $15,230.

To study the growth of IMRT, the researchers examined a huge Medicare records source, the Surveillance, Epidemiology and End Results, or SEER database, where they examined records of 26,163 women 66 years or older with non-metastatic breast cancer, who were treated with surgery and radiotherapy between 2001 and 2005.

They found that IMRT billing jumped tenfold in the study period, moving from less than 1 percent of patients diagnosed in 2001 to over 11 percent in 2005. The resulting adoption was linked to the increase in the mean cost of breast radiation, which grew from $6,334 in 2001 to $8,473 in 2005.

Interestingly, local coverage determination by Medicare also influenced adoption, with regions with more favorably disposed Medicare Carriers seeing higher IMRT billing rates. However, less favorable regions saw 28 percent lower radiation therapy and total health care costs for the women within one year of diagnosis.

They also discovered that IMRT billing was 36 percent higher for patients treated in freestanding centers than in hospital-based clinics.

In the authors' view, this shows that under Medicare's current structure, some regions concentrate on controlling cost, such as those with less favorable IMRT stances, while others look to give patients the most advanced care, such as those with favorable IMRT billing practices.

"Our analysis demonstrates a real need for novel reimbursement strategies that simultaneously incentivize the implementation of such clinically important treatments while still promoting cost-effectiveness," Smith said in a statement.

The researchers also found regional variations in the definition of a related radiotherapy treatment might have influenced costs.

In the paper, they noted that field-in-field forward planning, a radiotherapy technique that's more straightforward and requires less planning time than IMRT but has potentially comparable clinical results, is generally not billable as IMRT delivery by most Medicare Carriers. As a result, in those regions where it's not billable as IMRT, it would be only slightly more expensive than non-IMRT treatments. However, in areas where it does meet IMRT billing criteria, it could costs thousands of dollars more.

"This type of geographic variation in Medicare payments has been previously cited, both in the lay press and in health policy circles, as a potential source of waste within the Medicare system," the authors write. "Our data suggest that with respect to breast radiation therapy much of the variation in cost can be directly attributed to inconsistent treatment definitions and reimbursement rates authorized by Medicare and its intermediaries."

Nonetheless, the study had a number of limitations. It only looked only at older women, and the authors concede that younger women might receive IMRT more often because of concerns over radiation's future toxic effects. Also, the study only captured IMRT billing data; it was not validated by a chart review, the "gold standard" for this type of study. And because it used Medicare data, it does not capture patient co-payments and so might not reflect the cost for patients with private insurance or the true cost of providing radiation therapy.