Date of Award:


Document Type:


Degree Name:

Doctor of Philosophy (PhD)



Committee Chair(s)

Kerstin E. E. Schroder


Kerstin E. E. Schroder


Scott Bates


M. Scott DeBerard


Christopher Johnson


Edward M. Heath


The current study used a randomized control design to determine if an implementation intention intervention based on the Health Action Process Approach would increase exercise behavior in volunteer firefighters. One hundred forty-seven male and female volunteer firefighters from nine fire departments participated. The treatment group received a survey that prompted them to create exercise implementation intentions by describing "where," "when," and "how" they will exercise, while the control group survey included a general exercise message. We hypothesized that the implementation intervention would increase exercise behavior while the general exercise message would have no effects on exercise, that earlier stage-of-change exercisers would show greater increases in exercise than late-stage exercisers and that there would be an interaction between the intervention and stage membership such that the intervention would be more effective among firefighters in earlier-stage exercisers. We also hypothesized that the intervention would indirectly decrease sedentary behavior. Negative binomial regression showed that the intervention (p = .03) was a statistically significant predictor of exercise minutes, supporting the first hypothesis. However, using exercise sessions and leisure time as dependent variables, no significant effects were found for the intervention. Thus, hypothesis one was only partially supported by the results. A stronger group-administered implementation intention formation intervention may be more effective in increasing exercise behavior in volunteer firefighters. Participants who self-identified as early-stage exercisers were more likely to increase exercise behavior; however, the effects of stage on behavior change were not significant in any of the analyses, leading to the conclusion that hypothesis two had to be rejected. Further, there was no significant stage-by-intervention interaction in any of the analyses, lending no support to hypothesis three. However, a tendency in the expected direction was detected, suggesting that a low-intensity implementation formation intervention may increase exercise time in early-exercise stage volunteer firefighters more than among late-stage members. In conclusion, this research suggests implementation intention formation influences increased exercise behavior in volunteer firefighters who do not exercise or are not regular exercisers. Future research with a stronger intervention is warranted.




This work made publicly available electronically on October 1, 2010.