Recruitment and standardization of a group of Australian dentists for a multipractice study on dental caries...

This introductory paper details the recruitment and standardization of a group of dentists participating in a clinical trial. The trial is being undertaken to determine the cost-effectiveness of a structured preventive programme compared to standard care within private dental practices. We recruited private dental practitioners from a variety of locations in New South Wales (NSW) and the Australian Capital Territory (ACT). We sought to quantify the diagnostic reliability of dentists involved, and to define, quantify, and analyse standard care.

This is a multi-centre, clustered randomized controlled trial, where dentists are allocated to an intervention preventive or control group. Recruitment was facilitated with the support of key stakeholders and included oral presentations at divisional meetings of the Australian Dental Association, NSW Branch (ADA). A detailed time-in-motion study of 426 dental procedures was undertaken in order to define the parameters of standard care. The reliability study involved each dentist reading a set of 12 pairs of bitewing radiographs that had been produced and reviewed under standardized conditions. The reliability analysis was undertaken blind to allocation status of the dentist.

Recruitment ceased three months into the planned six-month recruitment period, 31 practices having approached the researchers. Eight suburban, five Central Business District (CBD), five rural (in fluoridated communities), and four rural (in non-fluoridated communities) practices have been recruited. Standard care did not differ significantly between intervention and control practices (Mann-Whitney U: z = -0.50; P = 0.6). Diagnostic reliability was substantial (Kappa = 0.79 [range 0.73-0.811 and 0.78 [range 0.72-0.82]) in relation to the intervention and control practices, respectively; P = 0.6.

The involvement of private dental practices in research is feasible and well supported by the profession. Standard care does not differ significantly between intervention and control practices. Inter- and intra-observer reliability was substantial, and not statistically different between the two arms of the trial.