University of São Paulo Dental School, São Paulo, São Paulo, Brazil.
Australian Research Centre for Population Oral Health, The University of Adelaide, Adelaide, South Australia, Australia.
PLoS One. 2020 Jun 3;15(6):e0233972. doi: 10.1371/journal.pone.0233972. eCollection 2020.
The study of oral health literacy (OHL) is likely to gain new and interesting insights with the use of network analysis, a powerful analytical tool that allows the investigation of complex systems of relationships. Our aim was to investigate the relationships between oral health literacy and oral health-related factors in a sample of Indigenous Australian adults using a network analysis approach.
Data from 400 Indigenous Australian adults was used to estimate four regularised partial correlation networks. Initially, a network with the 14 items of the Health Literacy in Dentistry scale (HeLD-14) was estimated. In a second step, psychosocial, sociodemographic and oral health-related factors were included in the network. Finally, two networks were estimated for participants with high and low oral health literacy. Participants were categorised into 'high' or 'low' OHL networks based on a median split. Centrality measures, clustering coefficients, network stability, and edge accuracy were evaluated. A permutation-based test was used to test differences between networks.
Solid connections among HeLD-14 items followed the structure of theoretical domains across all networks. Oral health-related self-efficacy, sporting activities, and self-rated oral health status were the strongest positively associated nodes with items of the HeLD-14 scale. HeLD-14 items were the four most central nodes in both HeLD-14 + covariates network and high OHL network, but not in the low OHL network. Differences between high and low OHL models were observed in terms of overall network structure, edge weight, and clustering coefficient.
Network models captured the dynamic relationships between oral health literacy and psychosocial, sociodemographic and oral health-related factors. Discussion on the implications of these findings for informing the development of targeted interventions to improve oral health literacy is presented.
使用网络分析研究口腔健康素养（OHL）可能会有新的有趣发现，网络分析是一种强大的分析工具，可以调查复杂的关系系统。我们的目的是使用网络分析方法研究澳大利亚原住民成年人样本中口腔健康素养与口腔健康相关因素之间的关系。
使用 400 名澳大利亚原住民成年人的数据来估计四个正则化偏相关网络。最初，使用口腔健康素养量表（HeLD-14）的 14 个项目来估计网络。在第二步中，将社会心理、社会人口统计学和口腔健康相关因素纳入网络。最后，为口腔健康素养高和低的参与者分别估计两个网络。根据中位数分割将参与者分为“高”或“低”OHL 网络。评估中心性度量、聚类系数、网络稳定性和边缘准确性。使用置换检验来检验网络之间的差异。
在所有网络中，HeLD-14 项目之间的牢固连接遵循理论领域的结构。口腔健康相关自我效能、体育活动和自我报告的口腔健康状况是与 HeLD-14 量表项目最强的正相关节点。在 HeLD-14 +协变量网络和高 OHL 网络中，HeLD-14 项目都是四个最中心的节点，但在低 OHL 网络中不是。在整体网络结构、边缘权重和聚类系数方面，高 OHL 和低 OHL 模型之间存在差异。
网络模型捕捉了口腔健康素养与社会心理、社会人口统计学和口腔健康相关因素之间的动态关系。讨论了这些发现对告知开发有针对性的干预措施以提高口腔健康素养的意义。
