目的探讨多水平发展模型在儿童被动吸烟干预性研究中的应用。方法应用多水平发展模型,以儿童尿样可丁宁水平为分析指标,评价被动吸烟干预效果。结果两组间在基线时的可丁宁均不相同,随时间变化趋势也不同。照料者吸烟支数对可丁宁随时间变化有影响。结论多水平发展模型可以有效处理干预研究的纵向数据资料。 Objective To explore the application of multi-level development model in children\'s passive smoking intervention study. Methods The multi-level development model was applied to evaluate the effect of passive smoking intervention on children\'s urine cotinine levels. Results Both groups were different at baseline, with varying trends over time. The number of caregivers smoking may have effects on cotinine over time. Conclusions Multi-level development model can effectively deal with longitudinal data of intervention study.