Objective: We examined the potential for research using administrative databases containing dentists' claims to identify both the type of health services research questions addressed and the strength of the evidence that is achieved in such studies. Methods: We searched Medline (1966 to March, 2003), retrieved additional reports from personal files, reviewed the literature cited in the relevant articles and conducted electronic searches on investigators' surnames. Information from relevant articles was abstracted into tables and the strength of the evidence for each was classified. Results: Thirty-eight studies met our inclusion criteria. Researchers have used administrative databases of dental records to examine provider practices, the longevity or consequences of dental interventions, the prevalence of dental conditions, and patient factors that determined care, and to establish quality assurance criteria or standards of care. The strongest designs were prospective or case-control (Level 11–2). Conclusion: Studies analyzing administrative databases have the advantage of size and economy but are subject to several threats to their validity and are seldom population-based. The strongest designs occurred with investigation of the longevity or consequences of care. Several studies demonstrated the benefit of linking the service data to patient or provider characteristics. The study of dentists' claims data appears under exploited, especially in the area of identifying and recommending changes in dental health care policies.