Related Experiment Videos
Analytic approaches to longitudinal caries data in adults
J D Beck1, H P Lawrence, G G Koch
1Department of Dental Ecology, University of North Carolina, Chapel Hill 27599, USA.
Community Dentistry and Oral Epidemiology
|February 1, 1997
Summary
This study analyzes longitudinal caries data in adults using generalized estimating equations and survey regression models. The incidence density strategy is recommended for analyzing time-to-event data in older adults, accounting for tooth loss and remineralization.
Area of Science:
- Dental epidemiology
- Biostatistics
- Gerontology
Background:
- Longitudinal dental data requires specialized analysis due to repeated measures and clustering within subjects.
- Older adults present unique challenges in caries studies, including subject attrition and tooth loss.
Purpose of the Study:
- To review and illustrate methods for analyzing longitudinal caries data in adults.
- To compare statistical approaches for dental caries research.
- To address biases and misclassification in caries estimation.
Main Methods:
- Utilized data from the Piedmont dental study, a prospective cohort of older adults.
- Applied generalized estimating equations and survey regression models.
- Evaluated the incidence density (time-to-event) analytic strategy for longitudinal caries data.
Main Results:
- The incidence density strategy effectively handles censoring due to attrition and tooth loss in older adults.
- Identified biases in caries experience estimation for missing teeth and crowns.
- Proposed methods to adjust for misclassification in caries indices and account for lesion reversals.
Conclusions:
- The incidence density approach is optimal for longitudinal studies of older adults, maximizing data utilization.
- Adjustments for misclassification and remineralization improve the accuracy of caries assessment.
- Statistical modeling is crucial for robust analysis of complex longitudinal dental data.