Prediction of intern attendance at a seminar-based training programme: a behavioural intention model.

Context The factors determining intern attendance at ‘resident rounds’ training programmes are not known. Aims This study aimed to use a behavioural model (the Triandis theory of social behaviour) to predict intern attendance at the resident round programme at a metropolitan teaching hospital. Subjects These included 109 interns in a large metropolitan teaching hospital, in the 1996 and 1997 cohorts. Methods Tabulated responses from a structured interview administered to 10 randomly selected interns were used to develop survey items related to the Triandis variables. Attendance was monitored using a sessional logbook and by self-report. Item analysis was conducted for each variable scale. Stepwise multiple regression models were constructed to predict attendance and intention. Results The average proportion of resident rounds attended was 43% (95% confidence intervals (CI) 40–47%) from logbook attendance, and 63% (95% CI 61–65%) by self-report. Cronbach’s alpha for the subscales ranged from 0·62 to 0·91. Intern attendance was predicted by the habit of attendance and negatively predicted by facilitating conditions (AdjR2=0·26, P < 0·0001). The intention to attend was only independently predicted by perceived consequences (AdjR2= 0·19, P < 0·0001). Conclusions The establishment of resident education programmes within teaching hospitals requires attention to and modification of facilitating conditions that may allow improvements in attendance and in quality of the programme. [ABSTRACT FROM AUTHOR]