UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Running to quit? : exploring predictors of attendance in an exercise and smoking cessation intervention Wunderlich, Kelly


Run to Quit (RTQ) is a national smoking cessation and learn to run program with promising cessation and physical activity outcomes. However, attrition was high with only 41.1% of participants completing the program. Determining predictors of attendance could help to improve attendance and program effectiveness in future iterations. Given that the program was offered in a group setting, the purposes of this study were to explore predictors of attendance and examine whether including group-related variables added to the prediction of attendance beyond individual variables. Blocked multiple regression analysis was used, with mean substitution for missing data (n=335). Individual predictors included in block 1 were middle aged adults, older adults, gender, home ownership, quit self-efficacy, run self-efficacy, baseline nicotine dependence (FTND), and baseline moderate-vigorous physical activity (MVPA). Group-related predictors added in block 2 were group cohesion subscales (attraction to group-task, attraction to group-social (ATG-S), group integration-task, group integration-social), transformational leadership (TL) of the coach, belonging, perceived similarity, and group size (control variable). When only individual predictors were included, the model was statistically significant and explained 4.8% of the variance in attendance (adjusted R²=.048, F(8,326)=3.111, p=.002). Both baseline MVPA (β=-.135, p=.013) and FTND (β=-.135, p=.015) were statistically significant predictors of attendance. Once group-related predictors were added, the overall model identified additional individual and group-related predictors of attendance. Both individual and group predictors were significant and adding group-related variables explained an additional 4.2% of the variance. Overall, the final model explained 9.0% of the variance (adjusted R²=.090, F(16,318)=3.067, p<.001), with being an older adult (β=.140, p=.023), male (β=-.118, p=.032), having lower FTND (β=-.145, p=.009), lower MVPA (β=-.120, p=.025), higher ATG-S (β=.189, p=.011), higher belonging (β=.183, p=.006), and lower TL (β=-.160, p=.018) significantly predicting higher attendance. This evaluation identified both individual and group-related predictors of attendance, however, the model explained a modest amount of variance suggesting that there are additional factors that predict attendance, such as logistical and personal reasons, and further exploration is needed. Future RTQ programs may benefit from promoting group-related aspects such as cohesion and sense of belonging.

Item Media

Item Citations and Data


Attribution-NonCommercial-NoDerivatives 4.0 International