James C. Coyne Ph.D.

The Skeptical Sleuth

Do Important Goals Prolong Life?

Can dying persons prolong their lives until they achieve an important goal?

Posted Feb 07, 2011


Just as I shall select my ship when I am about to go on a voyage, or my house when I propose to take a residence, so I shall choose my death when I am about to depart from life."--Seneca

David Spiegel's recent commentary, Mind Matters in Cancer Survival in JAMA: Journal of the American Medical Association suggests that people tend to die after their birthdays or holidays, rather than before. He offers this as evidence that death can indeed be postponed if a goal is in sight, at least for a short while.  He speculates that there might be some psychosomatic factors at work here that could be harnessed to extend the lives of cancer patients.

I informally polled Facebook friends about this question, not expecting hard evidence pro or con, but to elicit their beliefs and any relevant stories that they might have. Friends whose family members had died, as well as healthcare workers taking care of terminally ill cancer and hospice patients responded. There was a consensus that people somehow could prolong their lives.  A number of vivid anecdotes were provided, giving the appearance that patients had held on when there was a reason such as the anticipated arrival of a relative whom the dying person wanted to see before they died.

One person wrote

"My father was diagnosed with brain cancer and given 3 months to live...One thing that was very important to my dad was to see his family name carry on. He found out my sister-in-law was pregnant about 2 months after his diagnosis and he consistently amazed the doctors with how well he was still doing. Nine months later my sister-in-law brought my newborn nephew to the house to see my dad. We have video of how his eyes lit up and he reached for the baby even though he couldn't speak at that point. He died that night."

We cannot dispute the details of this example, but we can question its implied interpretation. Of course, it is notoriously hard for physicians to be make accurate prognoses, particularly for whether death will occur in a short time. But maybe the physician made the best estimate and the father somehow proved him wrong. We are not going to resolve this issue with one anecdote. But let's move on to the larger body of evidence.

David Spiegel's commentary cited a fascinating 1990 study by David Phillips and Daniel Smith. These authors reported that deaths among Chinese-Americans dropped by more a third the week before the Chinese Harvest Moon Festival, a drop that did not occur among various comparison groups, including American Jews who presumably pay no attention to the timing of the Harvest Moon Festival.

This study and others by Philips and Smith sparked considerable scientific interest in whether death dates can be shown to vary with birthdates, holidays, or other ceremonial occasions, with at least 18 such studies accumulated since the early 1970s. Yet, there have been numerous negative studies, and questions have been raised about the validity of most positive studies.  Simple changes in how relevant observation periods were defined could make some of the positive findings disappear and other positive findings seem to depend upon arbitrary choice of statistics. Still other positive findings simply could not be replicated, even when tested with similar important occasions in similar populations. And what if investigators found that their hypothesis did not hold, did they report this, or move on to another holiday without comment? And was it easier to get positive findings published?

A subsequent careful review of this literature concluded that "research over the past three decades has failed provide convincing evidence that psychological phenomenon such as 'giving up' or 'holding on' can influence the timing of death."

So, is it really possible for dying persons to control the timing of their death? In the very last days of life, there are often decisions to be made by patients and caregivers about nutrition, hydration, and whether to increase medication to control pain that could potentially interfere with breathing and thus cause death. There is often some leeway in exercising these choices, but it does not take some mysterious psychosomatic influence on lining or dying to explain: something could be done to forestall or hasten death, at least in the short term.

But beyond such possibilities, we should skeptical about expectations that dying persons will take charge and "hold on."  And we should be careful about imposing unrealistic demands on dying persons to "hang in there" until some anticipated event or blame them or ourselves when they cannot.