ATLANTA — A long-awaited government study of cancer among Vietnam veterans found no evidence linking the disease with exposure to the herbicide Agent Orange, officials said Thursday.

The study by the U.S. Centers for Disease Control, which looked only indirectly at Agent Orange, did find veterans at increased risk of a relatively rare cancer called non-Hodgkin’s lymphoma. But researchers noted the risk was greater among veterans who served on ships than those who served on land, where the herbicide was used.

In Washington, Veterans Secretary Edward J. Derwinski quickly said he would order the Department of Veterans Affairs to pay disability compensation to vets suffering from non-Hodgkin’s lymphoma.

The study was immediately attacked by veterans groups, which complained it failed to sufficiently examine the cancer risks among the thousands of soldiers who actually worked with the dangerous herbicide during the war.

“It is not an Agent Orange study,” said Mary R. Stout, president of the Vietnam Veterans of America.

“This is only a look at Vietnam veterans” who had cancer, added John Hansen, spokesman for the American Legion.

And Sen. Tom Daschle, D-S.D., and member of the Senate Veterans Affairs Committee, said: “This action has been a long time coming. However, it is not the final word. We cannot forget the many other veterans who are also suffering and dying from Agent Orange-related diseases and who have just as much evidence supporting their claims.”

Agent Orange, sprayed by the U.S. military to remove jungle cover and crops in Vietnam, contained dioxin, a highly toxic chemical which some studies have found to increase cancer risks.

Criticism of the study focused on its methods, which involved comparing cancer patients of Vietnam-veteran age to healthy counterparts.

The study “only indirectly evaluated” Agent Orange exposure, because of the technical difficulties in determining two decades later how much a patient may have been exposed to the herbicide, the CDC said.

Specific analysis of dioxin exposure was impractical, partly because researchers don’t know how cancer may skew the results of blood tests for dioxin levels, said Dr. Daniel A. Hoffman, assistant director for science at the CDC’s Center for Environmental Health.

“Our study was designed to look at Vietnam experience and history, and secondly to ascertain the likelihood of exposure,” Hoffman said.

The study, conducted over five years, surveyed 2,067 cancer patients who were of approximate Vietnam service age — between 15 and 39 in 1968, at the height of U.S. troop involvement. They were compared to 1,776 men of similar age who had no history of cancer.

Both groups were studied for risk of six kinds of cancer, including soft tissue cancer and similar sarcomas, a group of cancers that the CDC concedes has been “of great concern among Vietnam veterans.”

Vietnam vets were found to have a roughly 50-percent increased risk for non-Hodgkin’s lymphoma.