Supervision ratios for dental hygiene therapy trainees and clinical error detection in Ugandan teaching hospitals: a multilevel variance decomposition

Clinical supervision ratios for dental hygiene therapy trainees in Ugandan teaching hospitals are set by convention rather than by evidence, and the assumption that lower ratios improve error detection has not been tested against the structure of variance in supervision outcomes. The indicator is treated as a proxy for the precision of routine health-event ascertainment in settings where dental hygiene therapy trainees are trained to detect clinical error. The mean of the indicator was 2.779 and the standard deviation was 0.959, with country-level values ranging from 1.94 in Angola to 4.26 in Lesotho. The framework partitions supervision-embedded detection variance into trainee, supervisor, hospital and country components, and shows that the ratio of supervisor-to-trainee variance, not the absolute ratio of supervisors to trainees, governs the marginal return to tighter supervision. Ugandan teaching hospitals should set ratios by the variance structure of the tasks being supervised rather than by a single national figure, and it specifies the data that a prospective decomposition would require.