Risk assessment for severe clinical attachment loss in an adult population.

Background: This study was carried out to identify variables related to severe clinical attachment loss (CAL) in an adult French population.

Methods: This cross-sectional survey employed 2,132 subjects of the First National Periodontal and Systemic Examination Survey (NPASES I) aged 35 to 64 years, each with at least six teeth. A nationally representative sample was obtained from September 2002 to June 2003 by a quota method stratified on age, gender, socioeconomic status, and geographic areas. The subjects had a complete full-mouth periodontal examination of four sites per tooth, assessment of missing teeth, and a number of laboratory tests and questionnaires. The periodontal status of each subject was assessed by criteria based on the severity and extent of CAL. The data were analyzed by univariable and multivariable models using logistic regression analyses.

Results: Nineteen and seven-tenths percent (19.7%) of the subjects had CAL>5 mm. When dental variables were not included in the analysis (model 1), age (odds ratio [OR]=1.8), male gender (OR=1.7), body mass index (OR=1.2), and white blood cell count (OR=2.2) showed significant association with severe CAL. A significantly higher risk was also present in non-drinkers and regular drinkers compared to occasional drinkers (OR=1.6). Model 2, including dental variables in addition to model 1 variables, showed that a significantly higher risk for severe CAL was also present with age (OR=1.6) and in males (OR=1.7). The number of teeth (OR=1.1), and the mean gingival bleeding index (OR=1.7) were the dental variables significantly associated with severe CAL.

Conclusions: The results indicated that age and gender are powerful independent predictors of clinical attachment loss, as is the mean gingival bleeding index. To a lesser extent, the number of missing teeth was a good predictive variable. The patient profile for severe clinical attachment loss also included body mass index and white blood cell count. Occasional drinking may be associated with decreased severity of CAL.