The Relationship Between Tobacco Use,Psychological Behaviors And Diet In The Adolescents

Part one Reproducibility and Validity of a Adolescent Food Frequency QuestionnaireObjectiveTo determine and monitor adolescents'eating habits and nutritional status longitudinally in large populations, a self-administered food frequency questionnaire was designed for the adolescents. This study was conducted to evaluate its reproducibility and validity.Methods405 subjects entered the study by three-stratification cluster random sampling from 8th grade of 4 schools, 2 classes with each school, of which 190 were male and 215 were female. The study was carried out from December 2003 to November 2004. During this period, all subjects were asked to complete the FFQ once a season. Two 24-hr recalls data were collected respectively from the subjects of one class of each school in turn in four seasons. The FFQ was administered once again to this subset sample after 24-hr recalls. The interval of the first FFQ and the second FFQ was 2 to 3 weeks each season. Pearson correlation coefficients were calculated between estimates of nutrients and food group intake assessed by FFQ and 24-hr recalls.ResultsReproducibility correlations varied from 0.49 for vitamin D to 0.73 for vitamin B1. Energy-adjustment decreased the observed correlations for all nutrients, with a correlation range from 0.34 for retinol vitamin D to 0.60 for calcium. Spearman correlation coefficients between estimates of food group intake ranged from 0.43 for fish to 0.68 for vegetables. Validity correlations between the energy-unadjusted nutrients computed by the two methods ranged from 0.23 for total fat to 0.59 for calcium, the average correlation coefficient was 0.31. After energy-adjustment the range of correlation was from 0.18 for protein and total fat to 0.72 for calcium, with a mean value of 0.37. Most mean absolute nutrient intakes were within 25% of each other. With regard to classification, on average, 24.07% of subjects fell into the same quintile, 34.77% into adjacent category and 4.72% into extreme category.ConclusionA self-administered adolescent food frequency questionnaire has a reasonable ability to capture similar information as 24-hr recall. It might be an ideal dietary assessment instrument for epidemiological studies on adolescents'habitual diet and adult diseases in large populations.Part two Adolescent Tobacco Use and DietObjectiveTo describe the prevalence of tobacco use and the effect of tobacco use on nutrients and food intakes in the adolescents.Methods Survey data was derived from the longitudinal school-based smoking prevention trial in Wuhan. The subjects recruited for the baseline survey were sampled by stratum multi-steps random sampling. Two schools were selected from each of the 11 districts: one was selected randomly, and a second was selected to match the randomly selected school in size, academic rating and functional zone. Six classes in each of the selected schools were randomly selected for data collection: four seventh-grade classrooms were the samples for follow-up survey, two classrooms were from the other grades respectively and sampled randomly every year. The data collected in the 2nd wave survey were used for analysis, including the 2nd wave smoking questionnaire and the baseline food frequency questionnaire. There were 6612 middle-school students participating in the smoking survey while 6535 middle-school students were for the diet survey. Chi-square test, nonparametric ANOVA analysis and multiple linear regression analysis were used for smoking rates and association between tobacco use and diet practices.ResultsLifetime smoking rate, past 7-day smoking rate, past 30-day smoking rate, established smoking rate, regular smoking rate, susceptibility to smoking rate and passive smoking rate were respectively 39.02%,5.21%,13.88%,2.02%,1.12%,24.74%及19.91%. Boys were significantly more likely than girls ever to have tried smoking, to have smoked in the past 7 and 30 days, to be established smokers and regular smoker, and to be susceptible to smoking (p<0.01 for all). Girls were more likely than boys to be passive smokers. Urban students were less likely to have tried smoking, to have smoked in the past 7 and 30 days, and to be susceptible to smoking than rural students (p<0.01 or p<0.05), but more likely to be passive smokers than rural students (p<0.01). Moreover, lifetime smoking rate, past 7-day smoking rate and regular smoking rate increased with the increase of BMI (p<0.05 or p<0.01). Multiple linear regression analysis that the intakes of energy, protein, total fat, iron, sodium, zinc, vitamin B1, niacin, vitamin B12 and selenium were significantly, positively correlated with smoking levels while iodine andβ-carotene intakes were negatively associated with smoking levels (p<0.01 or p<0.05). Similarly, the consumption frequencies of meats, drinks, sweets, fast food and instant noodle were positively correlated with smoking levels whereas rice and vegetables consumption frequencies were negatively associated with smoking levels(P<0.01 or P<0.05).Conclusion The findings indicated that smoking was quite common in the adolescents and it had a significant effect on diet practice. The government and the school should pay more attention on the issue. Effective and comprehensive intervention measures involved nutrition education should be developed and implemented to control the tobacco use.Part three The Relationship between Psychology And Diet BehaviorsObjectiveTo explore the influence of psychological behaviors on nutrients and food intakes through psychological assessment and food frequency survey.MethodsSurvey data was derived from the Wuhan cohort study. The total sample was the same as the smoking survey. The data collected in the 2nd and 3rd wave surveys were used for analysis, including depression, stressful life events and food frequency questionnaire (FFQ) survey. Data collection for food survey was conducted for three times: baseline survey and two follow-up surveys. There were 2512 middle-school students participating in depression survey, 6619 for life events survey. 6535 students completed the baseline FFQs while 377 and 371 students completed the 2nd and 3rd FFQs respectively. Nonparametric ANOVA analysis, multiple linear regression analysis and ordinary logistic regression model were performed to test the association between depression, stressful life events and diet practices.ResultsStarch and iodine intakes of high score group for depression were significantly lower than that of low score group (P<0.01 or P<0.05). The consumption frequencies of rice and egg of high score group for depression were significantly lower than that of low score group whereas the frequencies of preserved vegetables, sweets and nuts were higher than that of low score group(P<0.01 or P<0.05). The multiple linear regression analysis showed that the intakes of calcium, sodium, copper and niacin were positively correlated with depression score(P<0.05). Likewise, the consumption frequencies of preserved vegetables, tea and coffee, sweets, fast food and nuts were positively associated with depression score(P<0.01 or P<0.05).The intakes of most nutrients for the students with more life events were significantly higher than that of the students with less life events. The multiple linear regression analysis indicated that the consumption of meats, vegetables sweets and nuts were positively correlated with the total score for life events (P<0.01 or P<0.05). Moreover, gender difference was detected for the effects of stress on diet. Boys with more life events was more likely to intake meats, drinks, tea and coffee, fast food than one with less life events while girls with more life events was more likely to intake nut than one with less life events (P<0.01 or P<0.05). Logistic regression analysis showed that meat intake was positively associated with smoking (OR=2.14), vegetable intake was negatively correlated with smoking (OR=0.58)and positively correlated with life events (OR=1.51), preserved vegetable intake was negatively correlated with smoking (OR=0.55), drink intake was positively correlated with life events (OR=1.57), alcohol use was positively correlated with smoking (OR=4.92), sweets intake was positively associated with life events (OR=1.79) and nut intake was positively correlated with life events (OR=1.64).ConclusionThe findings demonstrated that depression and stressful life events were associated with diet. It is necessary to add health-related curriculum on risk behaviors prevention into quality education. Public health agencies should carry out comprehensive behavior surveillance on psychology, nutrition and diet, and conduct early intervention in adolescent.