Dietary patterns associated with dental caries in adults in the United States.
Level 4 - case-series / case-control
Cross-sectional survey analysis
PubMed 31809561 · doi:10.1111/cdoe.12509
What was done
The authors analyzed cross-sectional data from 4,467 adults aged 18 and older from the 2013–2014 National Health and Nutrition Examination Survey (NHANES). Standard food categories from 24-hour dietary recalls were reduced into three dietary patterns using principal components analysis. Multivariable regression models were used to examine the association between dietary pattern score quartiles and both caries prevalence (decayed, missing, and filled teeth [DMFT] > 0) and caries severity (log-transformed DMFT score), adjusting for potential confounders. Analyses were stratified by age group: 18–30 years (n = 1,074) and >30 years (n = 3,393).
What was found
Caries prevalence (DMFT > 0) and median DMFT were 78.7% (95% CI: 76.1 to 81.3) and 4 (95% CI: 4 to 5) among participants aged 18–30, compared to 92.6% (95% CI: 91.4 to 93.7) and 12 (95% CI: 11 to 13) in those >30. In 18–30-year-olds, no dietary pattern was significantly associated with caries prevalence or severity. In adults >30, each subsequent quartile of a dietary pattern high in sugar-sweetened beverages and sandwiches was associated with a 2% higher prevalence of caries experience (adjusted prevalence ratio: 1.02, 95% CI: 1.001 to 1.03; 6% higher in quartile 4 vs. quartile 1) and a 1.98% higher DMFT score (95% CI: 0.15 to 3.85). Isolated analyses using the two strongest loading food groups from the components did not independently predict caries experience.
Why it matters
This study demonstrates that overall dietary patterns may capture caries risk better than isolated single food groups, though the observed associations in adults over 30 were very modest.
Limits
The cross-sectional design cannot establish temporality or causality, particularly because DMFT reflects lifetime cumulative disease while dietary data reflected only recent 24-hour recalls. Dietary intake was self-reported and prone to recall error. The effect sizes were very small, and potential unmeasured confounding from lifelong oral hygiene, professional dental care, and fluoride exposure remains possible.