A three-year program spearheaded by UnitedHealthcare held down costs of cancer care but leaves researchers wondering exactly how it did so. “The source of the cost savings is enigmatic,” say the authors of the study, “Changing Physician Incentives for Affordable, Quality Cancer Care: Results of an Episode Payment Model,” published online in the Journal of Oncology Practice on July 8.
That’s because while overall costs of treatment dropped by a third when compared with patients who were not part of the study, the cost of cancer medications soared. This, despite the fact that one of the primary factors in the three-year study was to remove financial incentives for doctors prescribing oncologic medications.
UnitedHealthcare tracked 810 cancer patients in five large oncology practices between October 2009 and December 2012. The practices, which volunteered to participate in the bundled payment pilot, are in Ft. Worth, Marietta, Ga., Dayton, Ohio, Memphis, and Kansas City, Mo.
“Medical oncologists were paid a single fee, in lieu of any drug margin, to treat their patients,” the study states. “Chemotherapy medications were reimbursed at the average sales price, a proxy for actual cost.”
The predicted medical cost for the group was about $98 million, but came in at about $65 million instead — a remarkable savings of about $33 million.
However, researchers had predicted that about $8 million would be spent on cancer drugs. The actual amount was about $14 million.
Researchers cited several incentives to curtail cancer drug costs. According to the study, they were:
If the selection for a chemotherapy regimen yielded a lower drug margin than the UnitedHealthcare national average for the episode, the group’s episode payment was raised to the national average, providing an incentive to select low-cost regimens when appropriate.
The oncology practices did not realize any gains by switching to higher-priced drugs.
The metastatic episode payments continued every four months, even if the patient was no longer receiving chemotherapy. This policy was intended to compensate the oncologist for the additional work of palliative care.
The groups met twice during the study period to review whether they had hit benchmarks for 60 measures of cost, quality, and use. “They had not been exposed to performance data about their practice from any source before joining this project,” the study states. That might have triggered the Hawthorne effect, whereby a person or group does better when it knows it’s being observed.