Three VTCs were selected as intervention sites and five as controls. Of the 2,197 students in these VTCs who were in a two-year program and in classes with more than 10 students, 297 were either absent on the day of the visit by the research assistant administering the questionnaire or chose not to respond to the questionnaire. Thus, 1,900 students completed the baseline questionnaire, 86 of whom were below 15 or above 20 years old. In the end, 1,814 (82.4%) questionnaires were usable (Figure ). Of these students, 770 were included in the intervention group and 1,044 in the control group. The mean age of students was 16.9 (SD = 1.0) years, and 84.7% were males (81.8% in the intervention group and 87.0% in the control group). Students were involved in three areas of vocational learning: building and public works (64.0%), the catering industry (22.7%) and personal services (13.3%). Of the 1814 students interviewed, 52.0% were smokers, 42.3% non-smokers and 5.7% ex-smokers. Thus, the study population for the primary objective comprised 943 smokers, representing 50.1% (n = 386) of the intervention group and 53.4% (n = 557) of the control group (Table ).
Flow chart illustrating the inclusion and follow-up of students in the TABADO study. TABADO study, Nancy, France, 2008-2009.
Comparison of baseline characteristics of students in the intervention and control groups
Smokers in the two groups were comparable (Table ), except for sex (80.3% males in the intervention group versus 88% in the control group, p = 0.001) and dependence score (HONC score; 6.4 ± 2.7 in the intervention group versus 5.9 ± 2.9 in the control group, p = 0.01). Also, students in the control group more frequently reported smoking or having smoked cannabis (p = 0.03).
Comparison of baseline characteristics of smokers in the intervention and control groups
Participation in the enhanced program
Of the 386 students who were smokers in the intervention group, 95 expressed a desire to participate in the enhanced program (EP) (24.6%) and 70 of those were included (18.1%). The 25 others presented non-inclusion criteria: 18 did not have parental consent and seven had medical contraindications (e.g. major depression). The mean baseline HONC score was higher for EP participants compared to non EP participants (7.1 vs. 6.2; p = 0.02 – Table ), as were intention to quit within six months (55.6% vs. 26.5%; p = 0.0004) and daily smoking (100% vs. 85.8%, p = 0.03). Sex and area of learning also differed between groups. These variables were identified as adjustment covariates in multivariate analysis.
Comparison between EP participants and non-EP participants smokers
Follow-up of students
Of the 1,814 students in the study (Figure ), 1,206 were questioned again at 12 months in both categories of VTCs (66.5%: 65.3% in the intervention group and 67.3% in the control group). The proportion of males was higher in non-respondents than in respondents (88.5% vs. 83.0%, p = 0.002), and non-respondents were older (mean 17.0 ± 1.1 vs. 16.8 ± 0.9 years, p <0.0001).
Primary assessment criteria
Among the baseline smokers, and considering those lost to follow-up as non-abstinent, 10.6% in the intervention group were abstinent at 12 months versus 7.4% in the control group (univariate p = 0.08; adjusted p (for age, sex, training course, initial cannabis consumption, HONC score, smoking consumption) = 0.03; OR 1.8; 95% CI = 1.05–3.0).
Among the baseline smokers, and considering those lost to follow-up as missing data, 17% in the intervention group were abstinent at 12 months versus 11.9% of smokers in the control group (univariate p = 0.08; adjusted p = 0.008; OR 2.1; 95% CI = 1.2–3.6) (Table ).
Rate of smoking abstinence in the intervention and control groups at 12 months
In the intervention group (Table ), the abstinence rate among EP participants was 5,7% versus 11.7% among non-EP participants (univariate p = 0.21; adjusted p = 0.37). There was no statistical difference between the 2 subgroups.
Rate of smoking abstinence among EP participants and non-EP participants at 12 months
Of the 70 EP participants, 33 (47.1%) received nicotine replacement therapy (NRT) (patch or gum) and used it. At 12 months, 11.5% of the NRT users were abstinent versus 5.6% among EP participants who didn’t needed NRT according to physician or/and student.
At 12 months, smoking prevalence was 50.9% (+2.1%) for the control group and 48.9% (+1%) for the intervention group (for evolution: crude p = 0.76; adjusted p = 0.75).
To study this effect more specifically, we conducted post hoc further analysis on the abstinence rate among non-EP participants in each class based on the number of EP participants in that class. The abstinence rate by class differed significantly depending on the number of EP participants: the abstinence rate among non-EP participants was 15.2% when there were fewer than two EP participants in the classroom versus 25.4% when there were at least two EP participants (p = 0.04 after adjustment for age, sex, baseline dependence and cannabis use).