CHICAGO (AP) â€” The five-year survival rate â€” the standard measurement of a cancer treatment's success â€” may be misleading because people are being diagnosed earlier and therefore living longer with detected cancer before they die, a study says.
The findings, published in Wednesday's Journal of the American Medical Association, are based on a National Cancer Institute effort to collect and analyze cancer data in the United States.
The five-year survival rate is the number of cancer patients who are still alive five years after diagnosis.
The analysis looked at the 20 major types of cancer and the rates of five-year survival from 1950 to 1995, then looked at deaths and the incidence of the disease. In some cancers, more people are living five years after being diagnosed, but more people are also developing the disease and more people are ultimately dying from it.
For example, the five-year survival rate for prostate cancer has increased from 43 percent in 1954 to 93 percent in 1995. However, the mortality rate â€” the number of patients per 1,000 who die from the disease â€” has gone up 10 percent.
Dr. H. Gilbert Welch, lead author of the study, said the increase in the survival rates is mostly influenced by earlier diagnoses of cancer, not advances in treatment.
The survival rate is computed from the time of diagnosis, so it appears people are living longer, when in reality many of them are only learning they have cancer earlier, he said.
For example, if a man is diagnosed with prostate cancer at 75 and dies at 78, the five-year survival rate is 0; if another man is diagnosed at 73 and dies at 78, the rate is 100 percent. Both men may have developed cancer at the same time, but one was diagnosed earlier and appeared to have a better five-year survival rate, even though the end result â€” the age of death â€” was the same.
``We know that the five-year survival rate always goes up when we find cancer earlier in patients' lives. Whether or not these patients have their deaths postponed is a different matter altogether,'' said Welch, a doctor at the Department of Veterans Affairs Hospital in White River Junction, Vt., and a professor at Dartmouth Medical School.
Welch said the findings are not meant to imply there have not been advances in cancer treatment or to discourage people from getting screening tests. But he said the public and the media should be more skeptical of findings based solely on the survival rate.
Dr. William Hoskins of the disease management department at Memorial Sloan-Kettering Cancer Center in New York City said the five-year figure is an important measurement because after that amount of time, a cancer patient usually is considered cured.
He agreed that early diagnosis has probably boosted the survival rate, but he said at least some of those cases probably were people who benefited from early treatment and were cured, not just people who were simply diagnosed earlier and died in the same amount of time.
On the Net: the National Cancer Institute statistics: http://www.seer.cancer.gov