Does answering survey questions about mental health cause distress?

Self-report surveys measuring mental health constructs such as anxiety, depression and trauma symptoms are regularly used in psychological research. Researchers and ethics committees are often concerned about the possibility that completing surveys on mental health might itself provoke negative emotional reactions. While prior studies have not found strong evidence of such impacts, the existing literature is limited by a lack of preregistered studies, and a lack of studies applying statistical methods that can establish the absence of an effect. In Study 1, we therefore conducted a preregistered within-subjects experiment (N = 337) testing the effects of completing measures of anxiety and depression on state distress, tiredness, happiness, and anxiety. In Study 2, we conducted a between-subjects experiment (N = 871) testing the effects of answering questions about trauma symptoms on state distress, perceived meaningfulness of participation, and unwanted thoughts. In both studies, equivalence tests indicated that the effects of the manipulations on state distress were significantly within our preregistered equivalence bounds of d ± 0.2, albeit that in Study 1 the estimated effect (d = 0.10) was significantly greater than zero. Our findings suggest that answering surveys about mental health topics may provoke some distress. However, these effects appear to be very small, even with outcomes measured immediately after an experimental manipulation.