Studies concerning the prevalence of extractions prior to orthodontic treatment have been limited in scope. This quasi-experimental analysis from secondary data explores patient and provider variables as they relate to extractions prior to comprehensive orthodontic therapy in the mixed dentition. This national database contains 38,529 children who had at least one comprehensive orthodontic (mixed dentition) visit within a 27-month period (January 1987-March 1989). Because of the relatively small number of Class III malocclusion cases, an equal allocation, random sample method was used in choosing children from the three Angle malocclusion classifications and the seven NIDR regions. Of those selected 24.7% had one or more extractions prior to orthodontic treatment, with 56% occurring at either 11 or 12 years of age. There were slightly more extraction cases for the Class I malocclusion children (26.7%) than either Class II (23.1%) or Class III (24.1%). Those children who had an orthodontic extraction were slightly older (P < 0.05). There were no statistically significant differences relating to orthodontic extractions for the following patient and provider variables: gender, malocclusion classification, years since dental graduation, and type of dental practice. There were regional differences among extraction rates for pediatric dentists, with those from the NIDR Midwest region more likely to have children receiving one or more extractions.