Careers  |  Sign In  |  Register  |   Twitter

Cancer diagnosis quickly hikes risk of suicide, cardiovascular death

NEW YORK (Reuters Health) – Cancer can be deadly, but just being told you have cancer can be deadly as well.

Researchers examining data collected on 6.1 million Swedes between 1991 and 2006 have concluded that when people are diagnosed with cancer they are 12 times more likely to commit suicide during the following week and nearly six times more likely to die of a heart attack immediately after receiving the bad news.

The effect was seen among both men and women, and the people who had the cancers that are the hardest to cure faced the highest risk.

However, while the added statistical risk was much higher, the actual danger to an individual was small. Among the 534,154 people diagnosed with cancer, only 29 committed suicide during that first week.

“The incidence is low,” coauthor Dr. Unnur Valdimarsdottir of the University of Iceland told Reuters Health in a telephone interview. “But you wouldn’t expect to see more than two or three” suicides during that period in the general population.

“There’s a critical window of opportunity that, we hope, gives health authorities opportunity for intervention,” she said.

Although it has been widely documented that the risk of suicide and heart-related problems increases after a person is diagnosed with cancer, the new study in the April 5 New England Journal of Medicine was different because it looked at the risk of death immediately after a cancer diagnosis.

It was part of an effort to explore the question of whether the cancer treatments themselves or the burden of living with a tumor might explain the earlier research.

“I’m sure all those mechanisms exist,” said Dr. Valdimarsdottir. “But nobody has really understood that this may be due to the stress of diagnosis.”

Of the 14,070 suicides that were logged, 786 were among people who had received a cancer diagnosis; 29 were in the first week of diagnosis, or 2.5 per 1,000 people per year.

That translates to a risk that’s 12.6 times normal.

During the first 12 weeks, the rate declined to 4.8 times normal.

Over the first year, the suicide risk remained three times higher than people in the population who had not been diagnosed with cancer.

“The incidence rates of suicide were higher among patients with preexisting psychiatric or cardiovascular diseases, regardless of whether a patient had a recent cancer diagnosis. However, the relative risks of suicide after a cancer diagnosis were stronger among patients without these coexisting illnesses than among patients with those illnesses,” the researchers said.

They also found that the hardest-to-treat tumors produced the highest suicide rates.

During the first 12 weeks after diagnosis, the rate was 12.3 times higher than the rest of the population for those with lung cancer and 16.0 times the normal rate among those who had either esophageal, liver, or pancreatic cancer. Skin and blood cancers, in contrast, did not correlate with a significant increase in suicide during those first three months.

Among the 592,135 deaths from cardiovascular causes, 48,991 were from people who developed cancer; 1,318 came in the first week of diagnosis, for a risk 5.6 above normal.

“Here we are also seeing the largest effects very close to the diagnostic message being delivered,” said Dr. Valdimarsdottir. “This decreases rapidly after the first month after diagnosis.”

The only type of cancer than didn’t elevate the risk of cardiac death was skin cancer.

The researchers couldn’t precisely pinpoint the day in which each patient received their cancer diagnosis. Instead, they had to use the date in which the pathology report on a suspected tumor was completed. It may have taken a few extra days for a patient to get the news.

“If we had the actual date when diagnosis was given, the relative risk would be even steeper,” Dr. Valdimarsdottir said.

“We believe our results are just demonstrating the tip of the iceberg because we chose to study the most severe of many outcomes you could expect during this brief time period,” she said. “We hope this may lead to improvements in not only how the message is delivered to patients, but improved monitoring of psychological help.”


N Engl J Med 2012; 366:1310-1318.