Delegation scope, required training, and staffing shortage: a preregistered secondary analysis of a national survey of Korean dentists on a proposed dental assistant (DA) role

This record contains Version 1.0 of a preregistered plan for two secondary analyses of a cross-sectional online survey of Korean Dental Association (KDA) members conducted from 11 to 15 January 2021. Of 2,291 respondents, 2,223 records remained after the original data cleaning. The survey was collected for a policy project commissioned by the KDA. Plan 1 examines whether dentists’ perceived difficulty recruiting dental hygienists is associated with the scope of ten proposed dental-assistant tasks they consider delegable and the level of training they consider necessary for dental assistants. Task and training analyses are restricted to the 1,964 respondents who supported creating a dental-assistant role, because these items were not asked of other respondents. The plan specifies an assessment of task measurement, ordinal and beta-binomial regression, a respondent-by-task generalized estimating equation, multiplicity control, and sensitivity analyses. Tasks are classified according to whether they are listed in Annex 1, item 6 of the Enforcement Decree of the Medical Service Technologists Act; injection is treated as a separate task type. Plan 2 prespecifies an analysis of the association between time to hire and closure consideration among 1,315 solo-practice dentists. Predefined go/no-go criteria determine whether the results warrant a separate report or should be presented as supplementary material to Plan 1. The protocol discloses prior use of the survey and the investigators’ relevant prior knowledge, including one previously seen cross-tabulation designated as exploratory. These observational analyses will be interpreted as associations and not as causal effects. Confirmatory exposure-outcome analyses will begin only after this preregistration is public and has been assigned a DOI. Survey data are not included because they cannot be shared publicly under the applicable ethics approval. No respondent-level data or personally identifying information are deposited. No separate grant or award number has been confirmed.