Being in ill health was associated with unhappiness (being sick is not fun), so the researchers excluded women who had already had cancer, heart disease, strokes, or chronic obstructive-airway disease. That way, they could see if it was unhappiness alone that made people more likely to die.

It did not.

This chart shows the risk of mortality based on how the women who started in “good or excellent health” answered questions about happiness, feeling in control, relaxation, and stress.

The line at 1.0 is “sort of average” with regard to mortality risk, says Richard Peto, a co-author on the study and a professor of medical statistics at Oxford University. As is clear from the chart, there was no difference in risk of death based on how people answered these questions.