Hierarchical Data Structures and Multilevel Modeling

Dental Caries and Associated Factors in Tibetan School-Age Children in Gannan, China

OBJECTIVES

Data on the oral health status in the Qinghai-Tibetan Plateau are limited. This study aimed to investigate the dental caries status and identify associated risk factors among 8- to 12-year-old children in Gannan, a Tibetan Autonomous Prefecture in Northwest China.

METHODS

The study population was recruited using a multistage stratified random sampling procedure. Caries status was assessed by clinical oral examination and indicated by the decayed, missing, and filled teeth (DMFT/dmft) index. Information on sociodemographic characteristics, oral hygiene behaviours, and dietary habits was collected using a comprehensive structured questionnaire. Statistical analyses, including the chi-square test and a two-level logistic regression model, were conducted to examine the risk factors associated with dental caries.

RESULTS

A total of 1024 Tibetan children were examined, revealing a dental caries prevalence rate of 89.55%, with a mean DMFT/dmft index of 4.25 ± 3.27. The two-level logistic regression model showed that initiating tooth brushing at four to 5 years old (OR = 2.295, 95% CI 1.053-5.002) and frequent dessert and candy intake (OR = 2.655, 95% CI 1.216-5.794) were risk factors for dental caries. Conversely, being 11 years (OR = 0.399, 95% CI 0.177-0.904) or 12 years old (OR = 0.396, 95% CI 0.163-0.961) and if the father's education level was high school or above (OR = 0.311, 95% CI 0.098-0.986) were indicated to be protective factors against dental caries.

CONCLUSIONS

The prevalence of dental caries was high in Tibetan children aged 8 to 12year-old in Gannan. Therefore, it is imperative to prioritize comprehensive home- and school-based interventions to improve the oral health of this population.

Construction of multilevel statistical models in health research: Foundations and generalities

This topic review aims to present a global vision of multilevel analysis models’ applicability to health research, explaining its theoretical, methodological, and statistical foundations. We describe the basic steps to build these models and examples of their application according to the data hierarchical structure. It ir worth noticing that before using these models, researchers must have a rationale for needing them, and a statistical evaluation accounting for the variance percentage explained by the observations grouping effect. The requirements to conduct this type of analysis depends on special conditions such as the type of variables, the number of units per level, or the type of hierarchical structure. We conclude that multilevel analysis models are a useful tool to integrate information, considering the complexity of the relationships and interactions involved in most health conditions, including the loss of independence between observation units.