Data Collection
This was a cross-sectional study conducted in 2018 within the Registrar Clinical Encounters in Clinical Training (ReCEnT) project, an ongoing multi-site cohort study described in detail elsewhere (20). Participants were general practice trainees from three Australian states, from training organizations delivering general practice vocational training to 44% of all Australian general practice trainees (21). These trainees complete ReCEnT data collection each 6-month training term as a part of their educational program and may also provide consent for the data to be used for research purposes. The hard-copy ReCEnT survey for this study was either completed by trainees at educational workshops or completed after workshops and returned by post.
This paper reports findings from survey items specifically concerning in-consultation help-seeking included in one round of ReCEnT data collection. The items were piloted with two trainees who did not participate in ReCEnT. Question 1 rated the frequency of use of five specific strategies for obtaining in-consultation help from supervisors, using a 5-point Likert scale. The five strategies were: 1) trainee, supervisor and patient face-to-face in the trainee’s consulting room, after supervisor interrupts their own consultation; 2) trainee, supervisor and patient face-to-face in the trainee’s consulting room, after supervisor completes their own consultation; 3) by phone between trainee and supervisor consulting rooms; 4) by phone or face-to-face outside the trainee’s patient’s hearing; and 5) using an internal electronic messaging system between trainee and supervisor rooms. These survey items are included as Appendix A.
Question 2 rated trainee perceptions of any change in patient impressions of trainee competence after they seek in-consultation help, using a 5-point Likert scale.
Question 3 rated relative trainee comfort presenting to their supervisor outside, compared to within, the patient’s hearing, using a 5-point Likert scale.
The full survey also elicited trainee and practice demographic details.
Analyses
Responses rating the frequency of using help-seeking strategies were dichotomised as Never/Rarely and Sometimes/Often/Always. Responses indicating perceptions of change in patient impressions of trainee competence after trainees sought in-consultation help were dichotomised as: ‘Decreases a lot’/’Decreases somewhat’ and ‘Does not change’/’Increases somewhat’/’Increases a lot’. Responses rating relative trainee comfort presenting outside, compared to within, the patient’s hearing were dichotomised as: ‘Much less comfortable’/’Somewhat less comfortable’/‘Neither more nor less comfortable’ and ‘Somewhat more comfortable’/’Much more comfortable’.
Analyses were programmed using STATA 13.1 and SAS V9.4. Descriptive statistics included frequencies and proportions for categorical variables and mean with SD for continuous variables. The frequencies of categorical variables were compared between outcome categories using a Chi-square test. For continuous variables, means were compared using a t-test.
The dichotomised responses to questions 2 and 3 were used as dependent variables in two logistic regression analyses, using trainee and practice demographic information as independent variables (see Table 1).
Table 1
Trainee and practice demographic information included as independent variables in multivariable regression analyses
Factor group | Variable | Class |
Trainee factors | Trainee gender | Male/Female |
| Trainee FT or PT | Part-time/Full-time |
| Term | Term 1/Term 2/Term 3 |
| Worked at practice previously | No/Yes |
| Qualified as doctor in Australia | No/Yes |
| Has previous health qualification | No/Yes |
| Has post-grad medical qualification | No/Yes |
| Has other regular medical work | No/Yes |
| Trainee age | mean (SD) |
| Years worked as doctor prior to GP training | mean (SD) |
Practice factors | Practice size* | Small/Large |
| Practice routinely bulk bills (charges patient no fee for service) | No/Yes |
| Rurality (ASGC-RA)** | Major city/Inner regional/Outer regional-remote |
| Regional Training Provider (RTP) | RTP 1/2/3/4/5/6/7 |
* Small ≤ 5 FTE GPs; large ≥ 6 FTE GPs |
**Increased rurality indicated by higher ASGC-RA, and more disadvantaged locations indicated by lower SEIFA |
Univariate and multivariable logistic regression analyses were undertaken to estimate the associations between outcomes for Question 2 (perceived change in patient impressions of trainee competence) and Question 3 (relative trainee comfort presenting outside the patient’s hearing) and the independent variables.
All covariates with a p-value < 0.20 in the univariate analyses were considered in the multiple regression model. Once the model with all significant covariates was fitted, model reduction assessed covariates with p > 0.20 in the multivariable model. These were tested for removal and if removal did not substantially change the model the covariate was removed from the final model. A substantive change to the model was defined as any covariate in the model having a change in the effect size (odds ratio) of greater than 10%. P values < 0.05 were considered statistically significant.
In post-hoc analyses, we examined differences in the use of help-seeking strategies, and trainee responses to Questions 2 and 3, across training (Terms 1, 2 and 3) using a chi square test for trend. A sub-group of participants who responded Never/Rarely to both trainee, supervisor and patient face-to-face strategies (strategies 1 and 2) was also identified. Associations between this sub-group and responses to Questions 2 and 3 were explored using a Chi square test.
The study has ethics approval from the University of Newcastle Human Research Ethics Committee Reference H-2009-0323.