The study was conducted by doctors at Memorial Sloan-Kettering Cancer Center in New York and published in Thursday's New England Journal of Medicine.
"The data is coming from so many sources that it is hard even with the best of conscience not to think that race is the factor, be it conscious or unconscious," said Dr. Walter W. Shervington of New Orleans, president of the National Medical Association, an organization of 20,000 black doctors.
Early lung cancer has few symptoms, so many patients do not learn they have it until it has spread and it is too late to operate. Even with surgery, the chances are poor.
About 34,000 people a year are diagnosed with non-small-cell lung cancer, the most common kind, early enough for surgery to make a difference.
The study found that the overall five-year survival rate among elderly whites was 34.1 percent vs. 26.4 percent among elderly blacks.
It was found, however, that the black patients were almost 13 percent less likely to have such surgery than the whites.
Among those who had the cancer cut out before it spread, about 26 percent of the blacks and 34 percent of the whites lived at least five years after the diagnosis.
Without the operation, only 4 percent of the blacks and 5 percent of the whites made it that long.
"We hear about these inequities every few months, [and you wonder,] is there racism in healthcare?" Healy said. "If there, it should not be there. It's a betrayal of the very fundamentals of medicine. This study tells us we have to go the next level below."
Doctors have long known that blacks with lung cancer are more likely than whites with the disease to die from it. But the reasons have been unclear.
This is evidence that different medical treatment is at least part of the reason. It is at least the second major study this year to find that blacks get different medical treatment than whites.
A study earlier this year found black women were particularly more likely than whites to die of heart disease.
The Memorial-Kettering study looked at 10,984 patients in a National Cancer Institute database of people in five states and five cities outside those states. To remove the availability of insurance as a factor, the researchers looked only at people at least 65 years old, and thus covered by Medicare.
Since poverty makes it harder to get good care, the researchers also tried to look at the poorest patients, black and white. They looked for people living in an area where the average income was in the nation's bottom quarter. There was still a black/white difference of nearly 9 percentage points.
But Dr. Garrett Walsh, a thoracic surgeon at M.D. Anderson Cancer Center n Houston, said the racial gap may be an oversimplification.
There are other factors which could have prevented them from being treated, Dr. Healy said.
For instance, there was no distinction was made between the kinds of doctors treating these patients.
Individual records for the 300 black patients and 2,400 white patients were not examined to see whether they were treated by specialists or general practitioners. Additionally, other health problems may have prevented them from being good candidates for surgery, Dr. Healy says.