From chapter 12, Notice the Good Grass

The real size of a gratitude list

This note is from Running with Llamas. It is general education, not care. The chapter rates the practice weak.

Gratitude exercises, in the trials, are usually a short list of good things or a letter of thanks. They are not a personality transplant, and they are not a demand that a hard thing smile for the camera.

One meta-analysis compared those exercises with different control groups. Against doing nothing in particular, gratitude did a little better for psychological well-being (d = 0.31). Against an alternative activity that already asked something of the mind, gratitude itself was d = 0.46 and well-being was d = 0.17. Anxiety was d = 0.11, and the confidence interval included zero. Against an already psychologically active task, well-being was d = −0.03. That is a modest good. It is also why a journal should not be sold as the thing that will lift a clinical depression.

A later meta-analysis looked directly at symptoms of depression and anxiety: 27 studies, 3,675 people. The university summary is plain. The benefits were limited. There was a difference, and it was small. A line in that summary says it would not be something you would recommend as a treatment. Telling people who are depressed or anxious to be more grateful is unlikely, on these results, to bring the reduction in symptoms one would want. The same summary refuses a flat no. Gratitude can have a worth of its own. "Gratitude can't fix everything."

Expressive writing, a few private sessions about something stressful or meaningful, is the other drawer. A meta-analysis of 146 randomized studies found an average effect of r = 0.075. Studies tended to do a little better with more than one session and with sessions of at least fifteen minutes. Some people feel worse, especially if the material is traumatic, ongoing, or unsafe to touch alone. It is not trauma therapy.