Assessment of the validity of self-reported smoking status among schoolchildren in Sousse, Tunisia

Introduction: Tobacco use, which begins in adolescence and childhood and continues in later life, is the major avoidable risk for non-communicable diseases and death in the world. Self-reports have frequently been used to estimate smoking prevalence and health consequences. This study explores the validity of self-reports of smoking behavior among schoolchildren in Tunisia. Materials and methods: This study was conducted in March 2014 among a sample of 147 schoolchildren randomly selected. Data concerning the smoking habit were collected by a questionnaire designed for the purposes of this work. Then, exhaled CO, a biochemical marker of smoke exposure, was measured using piCO+ Smokerlyzer^® breath CO monitor among participants. Sensitivity and specificity of self-reports were calculated. Results: The prevalence of reported smoking was 9.5% with 16.7% and 1.7% respectively among boys and girls. Their mean age was 14.5±1.28 years old. When considering 4 ppm as the cut-off level of breath CO, sensitivity and specificity of self-reports were 100% and 93.7%, respectively. But at a breath CO cut-off of 3 ppm, self-reporting was 62.5% sensitive and 93.5% specific. Conclusion: According to our findings, we suggest that self-reports can be considered as a good tool to be used with a reasonable confidence to assess the smoking status.