Present and future time perspectives and health behavior

Abstract

<p><i>Background and purpose: </i>Lifestyle diseases are the leading cause of life-time disability and death in modern society. Knowledge of the underlying causes and mechanisms of health behavior choices is crucially important when developing health-promoting campaigns and planning health behavior interventions. This dissertation aimed to advance our understanding of the relationship between time perspective (TP) and health behavior. The six major objectives were: <ul> <li>to establish the discriminant validity between the present and future dimensions of the two most frequently used operationalizations of TP: the Consideration of Future Consequences Scale (CFCS) and Zimbardo Time Perspective Inventory (ZTPI), as well as to compare their relationships with health behavior.</li> <li>to establish the discriminant validity between the consideration of immediate and future consequences (CFC-I and CFC-F), and to test whether they differentially predict healthy and unhealthy behaviors.</li> <li>to explore CFCS-based, ZTPI-present/future based, and a combined CFCS and ZTPI-present/future based temporal profiles of the Norwegian population.</li> <li>to test whether the domain-specific CFC-health would be more strongly related to health behaviors than the general CFC.</li> <li>to study the direct effect of perceived connectedness to the future self on health behavior, as well as its moderating effect on the relationship between CFCS and health behaviors.</li> <li>to test the moderating effect of age on the relationship between TP and smoking behavior.</li> </ul> <p><i>Design/methodology/approach:</i> This dissertation utilized a quantitative survey design and a survey experiment. The data are analyzed with t-tests, confirmatory factor analyses, structural equation modeling (SEM), and cluster analyses. <p><i>Results:</i> Paper 1 established the discriminant validity of the present and future time perspective constructs of CFCS and ZTPI, as well as the discriminant validity of the present and future constructs of both scales. The present TP – but not future TP – of both scales was related to smoking. The ZTPI model had stronger predictive power than the CFCS model. <p>Paper 2 revealed present and future temporal profiles in the general Norwegian population. CFCS-based profiles were more predictive of health behaviors than ZTPI-based profiles. However, the profiles based on the combination of both CFCS and ZTPI variables were more predictive of smoking, exercising, and health intentions. <p>Paper 3 established the discriminant validity between CFC-I and CFC-F that differentially predicted health behaviors. CFC-I was associated with unhealthy behaviors, whereas CFC-F was associated with healthy behaviors. PCFS had a direct negative effect on healthy behaviors, strengthened the positive effect of CFC-I and weakened the negative effect of CFC-F on unhealthy behaviors. <p>Paper 4 established in a randomized experimental design that a domain-specific CFC–health was a better predictor of self-reported eating and exercising behaviors than a general CFC. Moreover, the discriminant validity between health-specific CFC-I and CFC-F was established. Health-specific CFC-F was a stronger predictor of exercise behavior than CFC-I. <p><i>Conclusions and practical implications:</i>The results of the study indicate that the present and the future TPs are related but distinct constructs that differentially predict healthy and unhealthy behaviors. The relationship between CFC and health behavior is likely influenced by regulatory focus. In western societies, where the promotion regulatory orientation dominates, individuals are more likely to be motivated by approach goals rather than avoidance goals. The findings suggest that future orientation was associated with healthy behaviors, but not with unhealthy behaviors. At the same time, present-oriented people were more engaged in unhealthy behaviors, but were not less engaged in healthy behaviors. The findings suggest that emphasizing the future health benefits of healthy choices and reducing the immediate attractiveness of unhealthy choices might be more effective in the Norwegian population. <p>Another finding was that CFCS-based population profiles were more predictive of health behaviors than ZTPI-based profiles; however, the profiles based on a combination of CFCS and ZTPI differed in their health behaviors the most. These findings could be helpful when choosing the basis on which to segment the population during health intervention programs. <p>We demonstrated that a domain-specific CFC-health was more effective at predicting health behaviors than a general CFC. The usage of CFC-health in future studies might increase effect sizes and the overall predictive power of the models and, thus, decrease the ambiguity of the findings. <p>Finally, we showed that lack of connection with the future self influenced health behaviors both directly and indirectly. This finding emphasizes the importance of the programs aiming to increase individual connectedness to the future self.