| Summary: | 碩士 === 國防醫學院 === 公共衛生學研究所 === 102 === Background: Questionnaire is often used as the research tool of smoking prevalence. However, the reliability and validity of the questionnaire could be influenced by the recall problem of participants, misclassification, and the situation when answering the question. By detecting the urine nicotine metabolites, the problems can be ruled out and the smoking condition of participants can be reflected precisely.
Objectives: To study the smoking behavior change of the recruits during initial entry training, and to test the reliability and validity with the questionnaire by detecting the urinary cotinine concentration.
Methods: We chose two companies 280 people as research subjects from initial entry training recruits through random sampling. Longitudinal study design was used while we collect questionnaire and 50 mL urine sample from the recruits at four timing periods. The second day of training, before the vacation, after three-day vacation and after six-day vacation. ELISA was used to analyze cotinine concentration of the collected urine sample, and the data from questionnaire was used SPSS 20.0 version for statistical analysis.
Results: According to the result of Cotinine concentration, the recruits smoke less during training and more during holidays. Besides, the reliability and validity with the questionnaire, the health investigation questionnaire and the three other daily exposure questionnaires are all good tools for smoking behavior measurement (AUC>0.95) except the daily exposure questionnaire on the second day of initial entry training (AUC=0.903).
Conclusions: The smoking condition of recruits is more severe during holidays then during training. Besides, we should avoid doing the questionnaire at the beginning of the initial entry training due to the less reliability and validity of the questionnaire on the second day of the training.
|
|---|
