INTRODUCTION First-order Markov models assume future tobacco use behavior is dependent on current tobacco use and are often used to characterize patterns of tobacco use over time. Higher-order Markov models that assume future behavior is dependent on current and prior tobacco use may better estimate patterns of tobacco use. This study compared Markov models of different orders to examine whether incorporating information about tobacco use history improves model estimation of tobacco use and estimated tobacco use transition probabilities. METHODS We used data from four waves of the Population Assessment of Tobacco and Health Study. In each wave a participant was categorized into one of the following tobacco use states: never smoker, former smoker, menthol cigarette smoker, non-menthol cigarette smoker, or e-cigarette/dual user. We compared 1 st, 2 nd and 3 rd order Markov models using multinomial logistic regression and estimated transition probabilities between tobacco use states. RESULTS The 3 rd order model was the best fit to the data. The percentage of former smokers, menthol cigarette smokers, non-menthol cigarette smokers, and e-cigarette/dual users in Wave 3 that remained in their same tobacco use state in Wave 4 ranged from 63.4%-97.2%, 29.2%-89.8%, 34.8%-89.7%, and 20.5%-80.0%, respectively, dependent on tobacco use history. Individuals who were current tobacco users, but former smokers in the prior two years, were most likely to quit. CONCLUSIONS Transition probabilities between tobacco use states varied widely dependent on tobacco use history. Higher-order Markov models improve estimation of tobacco use over time and can inform understanding of trajectories of tobacco use behavior. IMPLICATIONS Findings from this study suggest that transition probabilities between tobacco use states vary widely dependent on tobacco use history. Tobacco product users (cigarette or e-cigarette/dual users) who were in their same tobacco use state in the prior two years were least likely to quit. Individuals who were current tobacco users, but former smokers in the prior two years, were most likely to quit. Quitting smoking for at least two years is an important milestone in the process of cessation.

中文翻译：

---

改进对烟草使用随时间变化的预测：烟草与健康研究人口评估第 1-4 波的发现。

简介 一阶马尔可夫模型假设未来的烟草使用行为取决于当前的烟草使用，并且通常用于描述随时间推移的烟草使用模式。假设未来行为取决于当前和之前的烟草使用的高阶马尔可夫模型可以更好地估计烟草使用模式。本研究比较了不同阶的马尔可夫模型，以检验纳入有关烟草使用历史的信息是否可以改善烟草使用的模型估计和估计的烟草使用转变概率。方法 我们使用了四波烟草与健康人口评估研究的数据。在每一波中，参与者被分为以下烟草使用状态之一：从不吸烟者、前吸烟者、薄荷醇香烟吸烟者、非薄荷醇香烟吸烟者或电子烟/双重使用者。我们使用多项逻辑回归和估计烟草使用状态之间的转移概率来比较一阶、二阶和三阶马尔可夫模型。结果 三阶模型最适合数据。第 3 波中的前吸烟者、薄荷醇卷烟吸烟者、非薄荷醇卷烟吸烟者和电子烟/双重使用者在第 4 波中保持相同烟草使用状态的百分比范围为 63.4%-97.2%、29.2%-89.8% 、34.8%-89.7% 和 20.5%-80.0% 分别取决于烟草使用史。目前吸烟但在前两年曾吸烟的人最有可能戒烟。结论 烟草使用状态之间的转变概率差异很大，具体取决于烟草使用史。高阶马尔可夫模型可以改进对烟草使用随时间的估计，并可以帮助理解烟草使用行为的轨迹。影响 这项研究的结果表明，烟草使用状态之间的转变概率很大程度上取决于烟草使用史。过去两年处于相同烟草使用状态的烟草产品使用者（香烟或电子烟/双重使用者）戒烟的可能性最小。目前吸烟但在前两年曾吸烟的人最有可能戒烟。戒烟至少两年是戒烟过程中的一个重要里程碑。