| dc.description.abstract | This research used a data science approach to explore multilevel interactions among key aspects of child and primary caregiver (PCG) psychosocial and physical health, as well as diet and lifestyle factors affecting dental problems in preschool children. Data were derived from the infant cohort of the Growing Up in Ireland (GUI) study and the the National Preschool Nutritional Survey (NPNS). Developing code for tools such as data mapping, association analysis and visualisation plots can be used to improve our understanding of the pattern of cariogenic food and free sugar intake.
Classification tree analysis (CTA) was used to classify variables and describe interactions between multiple variables including PCG reports of children s dental problem visits, general health, temperament, emotional and behavioural difficulties as well as their own general health, stress, depression, relationship and sociodemographic variables. The GUI survey used a short frequency questionnaire (SFQ) which provided a limited list of healthy and unhealthy foods whereas NPNS used a detailed 4 day weighed food diary. Data mapping protocols were developed to link dietary intake estimates for 3 year old children from NPNS and the second wave of the GUI infant cohort. The pattern of consumption of cariogenic food and drink (CF) intake was determined using two different methods. Finally, a free sugar (FS) algorithm was developed to determine the FS content of foods in the NPNS database.
Dental problems were reported among 2.7% of infants at 9 months of age and 5.0% at 3 years. CTA identified infant temperament as the primary predictor of dental problems at 9 months and child global health at 3 years. The prevalence of overweight (including obesity) in 3 year old children was ~23%. In assessing the role of diet and weight status, PCG ethnicity was the most significant predictor of dental problems followed by longstanding illness or disability, mother s BMI and household income. The highest prevalence of dental problems was among children who were obese or underweight with a longstanding illness and an overweight mother. Through mapping the food codes, it was possible to visualise the relative performance of the SFQ compared to the detailed food diary in capturing specific food types, e.g., high sugar foods. The SFQ did not capture a substantial portion of foods consumed by 3 year olds in Ireland. The estimation of CF intake, using two different methods, highlighted how presentation and reporting can affect the interpretation of CF consumption data. Bean plots and alluvial figures also illustrated the usefulness of visualising the overall distribution of intake when using different methods of estimation and comparing snacking and main meal consumption patterns of CF and non-cariogenic foods (NCF). Key findings indicated that all children consumed CF over the NPNS 4-day period and the GUI survey covered less than half of the CF items selected in NPNS. More than one-third of all eating occasions were described as snacks which were consumed twice per day, on average. Association analysis of the meals and snacks provided an insight into the combination of meal components and how CF was consumed with other NCF. Biscuits, squashes, cordials, fruit juice drinks and chocolate confectionary were the most commonly consumed CF snack items. Alluvial plots visualised the CF and NCF interactions of both snacks and main meals. The key sources of FS were similar to the items that do not contribute significantly to nutrient intake but were primary contributors to CF intake. Almost three-quarters of 3 year olds had FS intake greater than the WHO recommendation that FS intake is a maximum 10% of total energy intake. A crude simulation which excluded all FS, from snacks only, of soft drinks, confectionery, cakes, biscuits and sugar demonstrated that mean daily intake of FS would reduce from 14.1% to 11% of total energy intake. As a population level estimate, this would double the proportion of 3 year olds, from one-quarter to one-half, meeting the maximum WHO recommendation of 10% FS. Data analyses such as those adapted in this project should inform future public health approaches in dentistry in Ireland. | en |
