Objectives: To describe: (1) the 24-month incidence of tooth loss in a diverse sample of dentate adults; and (2) the clinical, attitudinal, behavioral, and sociodemographic correlates of tooth loss incidence. Methods: The Florida Dental Care Study is a prospective longitudinal cohort study of persons who at baseline had at least one tooth, were 45 years or older, and who resided in north Florida. An in-person interview and clinical examination were conducted at baseline and 24-months after baseline, with 6-monthly telephone interviews between those times. A two-level hierarchical generalized linear regression (logit model) was used to quantify tooth-specific and person-level factors simultaneously. Results: Of the 739 persons who attended for a 24-month examination, 24% lost one or more teeth during follow-up. Tooth loss was more common in persons with dental disease at baseline, incident dental signs or symptoms, those with negative attitudes toward dental care and dental health, those with limited financial resources, older adults, blacks, females, and problem-oriented users of dental care (as distinct from regular attenders). Although disease presence at baseline was a major factor associated with incident tooth loss, most diseased teeth were in fact still present 24 months after baseline. Conclusions: Other than periodontal attachment loss, severe tooth mobility, and dental caries, no single factor was a dominant predictor of tooth loss; instead, numerous factors made statistically significant but small contributions to variation in tooth loss. Tooth loss apparently is the result of complex interactions among dental disease, incident dental signs and symptoms, tendency to use dental care in response to specific dental problems, dental attitudes, and ability to afford non-extraction treatment alternatives.