Objectives: Randomized clinical trials provide evidence to support clinical decision-making. However, single centers are frequently unable to recruit large enough samples to detect meaningful differences. Efforts to increase the usefulness of trials include using external controls, so all patients at one center are allocated to one treatment and then compared to those undergoing different treatments elsewhere. Whether external controls can be considered equivalent to those derived by randomization to a group at a single center is an important question for orthodontic research. Methods: This study compares two groups of children, both randomly allocated to serve as the control/“observation only” groups for two distant (USA and Britain) but rather similar large trials concerning the possibility of orthodontic growth modification. The patients were selected using very similar selection criteria and managed under approximately equivalent protocols for 15 months. Only patients with distinct images of cervical vertebrae 2, 3 and 4 on baseline and 15 month cephalograms were included (USA n=42 and Britain n=46). The groups were compared for skeletal maturity using the cervical vertebrae, skeletal morphology, and growth rate. Results: Comparison of skeletal maturity showed no significant difference at either time point (Mann-Whitney test p > 0.05). There was no significant difference (ANOVA p > 0.05) between the two groups in skeletal jaw relationship (ANB angle) or growth rate, but there were significant differences (ANOVA p < 0.05) in initial jaw size. These differences were greater than the differences between control and treatment groups produced by randomization in the USA trial. Conclusions: We conclude, even when selection criteria and protocols are carefully matched, external controls can vary in important ways that may influence the interpretation of outcomes. Because such differences may not be readily apparent, external controls are limited in their usefulness as simple comparison groups.

Back to the Craniofacial Biology Program
Back to the IADR/AADR/CADR 80th General Session (March 6-9, 2002)

Top Level Search