Measures of Daily Activities Associated With Mental Health (Things You Do Questionnaire): Development of a Preliminary Psychometric Study and Replication Study

Abstract

Background A large body of research has identified modifiable cognitions and behaviors (actions) associated with psychological health. However, little is known regarding the actions that are most strongly associated with psychological health or the frequency with which they should be performed. Objective This paper described 2 studies that used survey methodology to create the Things You Do Questionnaire (TYDQ), which aims to identify and rank actions (items) and domains of actions (factors) most strongly associated with psychological health. Methods We used digital marketing strategies to recruit Australian adult participants, who were asked to complete 2 web-based surveys comprising versions of the TYDQ; validated measures of depression, anxiety, and satisfaction with life; and demographic questions. In study 1, a total of 3040 participants rated how often they performed each of the 96 items comprising the TYDQ. This design was replicated in study 2, in which a 59-item version of the TYDQ was completed by 3160 participants. In both studies, the factor structure and validity were examined, as were the associations between individual TYDQ items and 3 mental health outcomes: depression, anxiety, and satisfaction with life. Results In study 1, factor analyses revealed that a 5-factor model comprising 27 items achieved an optimum balance between brevity and variance and accounted for 38.1%, 31.4%, and 33.2% of the variance in scores on measures of depression, anxiety, and satisfaction with life, respectively. The factors were interpreted as realistic thinking, meaningful activities, goals and plans, healthy habits, and social connections. These 5 factors were more strongly associated with psychological health than those such as practicing kindness, exercising gratitude, and practicing spirituality. This pattern of results was replicated across gender, age groups, and depression severity. The 5-factor solution found in study 1 was replicated in study 2. Analyses revealed that a 21-item version accounted for 46.8%, 38.2%, and 38.1% of the variance in scores on measures of depression, anxiety, and satisfaction with life, respectively. Conclusions These findings indicate that some actions are more strongly associated with psychological health than others and that these activities fall within 5 broad domains, which represent skills often taught in psychological treatments. Subsequent studies are planned to explore the reliability of these items and results in other samples and to examine patterns of change in scores during treatment for anxiety and depression. If replicated, these efforts will assist in the development of new psychological interventions and provide an evidence base for public mental health campaigns designed to promote good mental health and prevent the emergence of common mental disorders.