Testing hypothesized psychosocial mediators: lessons learned in the MassBUILT study.

Only a few of the interventions that target blue-collar workers have conducted formal analysis to evaluate the specific attributes of their intervention that are associated with success or failure. This study examined the role of dual hazard and decisional balance in the MassBUILT smoking cessation intervention. The authors conducted sets of multivariable linear and logistic regressions that examined if (a) the intervention was associated with changes in the psychosocial variables and (b) increase in psychosocial variables was associated with increase in smoking cessation. As hypothesized by the theoretical basis of the study, higher scores on both of the psychosocial variables were significantly associated with smoking cessation. However, the intervention did not change decisional balance and decreased dual hazard. The variables examined were important but were not mediators of the MassBUILT intervention and this could have contributed to the significant relapse in smoking among study participants.