Objectives: To re-evaluate analytical challenges and how they were related to oral epidemiological gold standard’ protocols in longitudinal geriatric studies.

Methods: Data and oral epidemiological gold standard’ protocols were used from two longitudinal studies in South Australia: the South Australian Dental Longitudinal Study (SADLS); and the Oral Health of Older Adults with Dementia.

Results: De Paola tables were constructed comparing surface-level coronal and root caries increments and reversals (e.g., baseline to 1-year Dementia study had 31968 coronal and 27648 root surfaces categorized). Components of these tables were highlighted in relation to Beck’s Adjusted Caries Increment (ACI) formula. Tables were also constructed to analyse surface-level calls’ by examiners over time. These tables identified mean percent of reversals for examiners (e.g., SADLS 5-11-year reversals were 6.5% (coronal) and 35.8% (roots)). Reversals differed by surface (e.g., SADLS 5-11-year coronal reversals ranged from 0.6% (occlusal) to 14.0% (buccal)). Threshold reversal percentages recommended for the use of the net caries increment versus the ACI were reviewed with comparisons of the results for each option analysed. Issues identified as influencing ACI were related to: (1) reversals – criteria for delineating the origin of a carious lesion or restoration; remineralization’ of exposed root surfaces; changes in examiners over time; access to interproximal root surfaces; (2) blinding of examiners to prior tooth status and tooth surface-level data; (3) missing teeth; (4) retained roots; (5) abundant plaque accumulation; (6) incisal caries on anterior teeth; and (7) high caries experience.

Conclusion: Caries analytical challenges were related to criteria used in oral epidemiological gold standard’ protocols, which require revision for future longitudinal studies in geriatric populations with increasing numbers of dentate older adults with high caries experience. A modified ACI method is proposed to overcome known anomalies reported in the literature. Supported by ARCPOH, NHMRC, and NIH R01-DE09588.

KD Carter*, J Chalmers, AJ Spencer, K Roberts-Thomson, WM Thomson

Presented at the 84th General Session and Exhibition of the IADR, 28 June – 1 July 2006, Brisbane Australia

Note: * indicates presenter