Who Are At Risk in Experimenting with Smoking?: A Latent Profile Analysis of Smoking Outcome Expectancies in Nonsmoker Adolescents

Introduction:

Smoking outcome expectancies are defined as beliefs about the possible effects associated with smoking on behavior, cognition, moods, and emotions. Expectancies referring to negative consequences (NC), positive reinforcement (PR), negative reinforcement (NR) and appetite and weight control (AWC) may play a role in determining smoking behaviors; furthermore expectancies may develop before experimentation with smoking. We hypothesized that smoking outcome expectancies in nonsmoker adolescents can help us to understand and probably predict who is going to experiment with smoking.

The goal of this report was to establish latent classes of nonsmoker adolescents based on their smoking-related expectancies in order to identify those groups which are at higher risk to start smoking later.

 Method:

Participants involved a subsample of a large school-based cohort study located in Budapest Hungary, and included 877 students who declared that they have not tried smoking at the beginning of data collection (49% girls). The mean age was 15.2 (SD=0.50). We collected data semi-annually four times post-baseline.

 Measures included the Hungarian version of a smoking consequences questionnaire which included four subscales to measure negative consequences (NC), positive reinforcement (PR), negative reinforcement (NR) and appetite and weight control (AWC) expectancies, smoking-related questions, a vulnerability of smoking scale, sensation seeking scale and CESD depression scale.

Results:

Applying latent profile analysis at the baseline, we identified four groups: Class 1 (N=573) high NC, low PR, low NR, low AWC; Class 2 (N=173) high NC, low PR, high NR, high AWC; Class 3 (N=75) low NC, low PR, low NR, low AWC; Class 4 (N=56) high NC, high PR, high NR, high AWC.

Self-reported vulnerability of smoking uptake was significantly higher in Class 2 and Class 4 than in Class 1 and Class 3. However, during the four waves follow-up, only Class 1 showed significant lower cumulative proportion of students who tried smoking (13%, 26%, 33%, 46% ) compared to Class 2 (21%, 38%, 46%, 54%), Class 4 (32%, 43%, 48%, 65%), and Class 3 (31%, 47%, 51%, 59%).

Comparing classes in sensation seeking and depressive symptoms, Class 1 and Class 3 had a significantly lower sensation seeking mean scores than the other classes, and Class 1 had a significantly lower depressive symptoms mean score than the other three classes.

 Conclusions:

Smoking outcome expectancies develop before the experimentation with smoking and can direct the uptake of smoking. The combination of high negative consequences and low positive and negative reinforcement expectancies could prevent the experimentation with smoking. Smoking outcome expectancies could be important targets in prevention programs for adolescents.