Caution: Quality of Life Studies Can Be Detrimental to Your Quality of Life

Andrew J. Vickers, PhD


September 28, 2010

I recently read a paper that concluded, "The intervention reduced fatigue and improved vitality, aerobic capacity, muscular strength... physical and functional activity, and emotional wellbeing but not quality of life."[1]

This sent my own personal quality of life into a bit of a tailspin. I have spent years researching treatments to treat symptoms such as fatigue, but this was clearly a waste of time. Apparently, all your symptoms can improve, but you won't feel any better.

The authors' conclusions illustrate 2 major statistical errors. The first is what statisticians call "accepting the null hypothesis." A statistical test is a bit like a court case. If you go to a criminal trial, you are either "guilty" or "not guilty"; there is no verdict of "innocent." Similarly, if you go to a clinical trial, you start with a null hypothesis -- typically that the treatment is ineffective -- and then declare this hypothesis "rejected" or "not rejected." A high P value does not mean "the intervention did not improve quality of life"; it means "we did not reject the hypothesis that the intervention has no effect." Absence of evidence is not evidence of absence.

The second error is to assume that all types of outcome measure are equally sensitive to changes in patient outcome. For a good example, look no further than Money magazine, which every year publishes a list of best places to live. The quality of life of each town in America is rated by Money in terms of things like income, house prices, and commute times. Now I have no doubt that my quality of life would go up if I earned more, paid less on my mortgage, and had a shorter trip to work. But a patient cured of back pain would also experience improved quality of life, even if he stayed at the same minimum-wage job that was a 2-hour drive from his overpriced apartment.

His cure would also likely not change his response to the question: "Have you accomplished less than you would like at work due to emotional problems?" And if he had developed angina at around the same time, he would probably give a negative answer to: "I am as healthy as anyone I know." These 2 questions, about emotional problems and overall health, are typical of those found on quality of life questionnaires. In short, you might cure a man of back pain but not change some of the scores on a quality of life questionnaire. But, of course, this doesn't mean you haven't changed his quality of life. To confuse "quality of life" with "quality of life questionnaire score" is little more than mistaking the map for the territory.

So if you are one of those people who believe that a patient can experience improvements in fatigue, vitality, function, and well-being, but not quality of life, let me show you the New York subway guide: it isn't dark, or loud.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: