文献速递认知障碍与谵妄老年综合评估

基于边际结构模型的口腔健康不良与痴呆风险:考虑时变混杂的队列研究

Poor oral health and dementia risk under time-varying confounding: A cohort study based on marginal structural models.
PubMed 文献 · J Am Geriatr Soc · 2023-12-08

本研究基于日本老年学评价研究9年随访队列,纳入37,556名≥65岁独立老年人,采用边际结构模型处理认知功能作为时变混杂因素,评估自我报告的口腔健康不良与痴呆发病的关联。结果显示,剩余牙齿≤19颗、无牙、咀嚼困难和口干症与痴呆风险升高显著相关,而吞咽问题无显著关联。

要点

  • 剩余牙齿≤19颗、无牙、咀嚼困难、口干症与痴呆风险升高显著相关。
  • 吞咽问题与痴呆发病无显著关联。
  • 考虑认知功能时变混杂后,口腔健康不良仍与痴呆风险独立相关。

核心结论

  • 在考虑基线和随访期间认知功能作为时变混杂因素后,老年人口腔健康不良仍与痴呆风险升高显著相关。

研究设计与样本

  • 为期九年的前瞻性队列研究,目标人群为参加日本老年学评价研究的≥65岁独立老年人。
  • 暴露变量为2010年和2013年评估的自我报告口腔健康不良(剩余牙齿≤19颗、无牙、咀嚼困难、吞咽问题、口干症)。
  • 结局变量为2013年至2019年期间从市政行政数据库获取的痴呆发病率。
  • 采用边际结构模型处理认知障碍作为时变混杂因素,并使用稳定逆概率加权的Cox比例风险模型估计风险比(HR)和95%置信区间(CI)。

主要发现

  • 在37,556名参与者(平均年龄72.8岁[1 SD=5.5],男性46.8%)中,痴呆发病率为2.2/100人年。
  • 剩余牙齿≤19颗、无牙、咀嚼困难、吞咽问题、口干症的比例分别为61.3%、11.0%、24.4%、14.7%、19.2%。
  • 基于边际结构模型的回归分析显示,剩余牙齿≤19颗(HR=1.12,95%CI=1.03-1.23)、无牙(HR=1.20,95%CI=1.09-1.32)、咀嚼困难(HR=1.11,95%CI=1.02-1.21)和口干症(HR=1.10,95%CI=1.01-1.20)与痴呆风险升高显著相关。
  • 吞咽问题与痴呆发病无显著关联(p>0.05)。

关键数值与推荐

  • 剩余牙齿≤19颗:HR=1.12,95%CI=1.03-1.23。
  • 无牙:HR=1.20,95%CI=1.09-1.32。
  • 咀嚼困难:HR=1.11,95%CI=1.02-1.21。
  • 口干症:HR=1.10,95%CI=1.01-1.20。
  • 吞咽问题与痴呆发病无显著关联(p>0.05)。

临床怎么用

  • 在老年综合评估中,可将剩余牙齿≤19颗、无牙、咀嚼困难和口干症作为痴呆风险升高的提示因素。
  • 对于存在上述口腔健康不良的老年人,应加强口腔健康管理并关注认知功能变化。

局限性

  • 口腔健康和认知障碍均基于Kihon checklist的自我报告项评估,可能存在测量偏倚。
  • 研究人群为日本独立老年人,结论外推至其他人群需谨慎。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BACKGROUND: Bidirectional association between oral health, including tooth loss and oral hypofunction, and cognitive impairment can induce time-varying confounding in association with dementia. This study aimed to investigate the association between oral health and the development of dementia among older adults, considering cognitive impairment as a time-varying confounder. METHODS: This nine-year follow-up cohort study targeted independent older adults aged ≥65 years who participated in the Japan Gerontological Evaluation Study. The exposure variables were self-reported poor oral health (≤19 remaining teeth, edentulousness, chewing difficulty, swallowing problems, and xerostomia), which were assessed in 2010 and 2013. The outcome variable was the incidence of dementia between 2013 and 2019, which was obtained from the municipalities' administrative database. Furthermore, in 2010 and 2013, we considered cognitive impairment as a time-varying confounder and treated it based on a marginal structural model (MSM), including possible baseline confounders. Oral health and cognitive impairment were assessed using the items of Kihon checklist. We employed a Cox proportional hazards model with a stabilized inverse probability weight and estimated hazard ratios (HRs) and 95% confidence intervals (CIs). RESULTS: Among 37,556 participants (mean age: 72.8 years [1 SD = 5.5], males: 46.8%), the dementia incidence rate was 2.2/100 person-year. The proportions of those with ≤19 remaining teeth, edentulousness, chewing difficulty, swallowing problems, and xerostomia were 61.3%, 11.0%, 24.4%, 14.7%, and 19.2%, respectively. From the regression analysis based on MSM, ≤19 remaining teeth (HR = 1.12, 95% CI = 1.03-1.23), edentulousness (HR = 1.20, 95% CI = 1.09-1.32), chewing difficulty (HR = 1.11, 95% CI = 1.02-1.21), and xerostomia (HR = 1.10, 95% CI = 1.01-1.20) were significantly associated with an increased risk of dementia; however, swallowing problems were not significantly associated with dementia onset (p > 0.05). CONCLUSIONS: Even after considering time-varying confounding by cognitive function at baseline and follow-up, we observed significant associations between poor oral health and increased risk of dementia among older adults.

来源:J Am Geriatr Soc,查看原文。版权归原作者所有。

引用

  1. Kusama T, Takeuchi K, Kiuchi S, et al. Poor oral health and dementia risk under time-varying confounding: A cohort study based on marginal structural models. J Am Geriatr Soc. 2023. PMID: 38064294. doi:10.1111/jgs.18707

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。