This study is one of the first major studies presenting data on the effect and impact of a national PGME reform on the daily clinical training practice and learning environment in hospital departments 31/2 years after reform start. Overall, the results indicate that the reform had a small effect on some structural educational issues, but still only limited impact on daily clinical training practice and educational culture.
The impact of the reform was primarily found on two structural items, attendance to courses of learning and teaching and preparation of individual learning plans. The introduction of mandatory courses on learning and teaching for junior doctors in training was clearly reflected in the results with a large increase in number of trainees having attended such a course. However, course attendance among senior doctors decreased and corresponding to findings in other contexts, only around one quarter of the senior doctors had attended such courses after the reform [11
]. Considering that in the work-based context of PGME almost all senior doctors are involved in the training of the junior doctors, the fall in course participation among senior doctors is a problem. One might speculate that the mandatory course participation should have been extended to include the seniors as well. We have previously shown that when courses on clinical teaching involved all doctors in a clinical department, it had a positive and lasting effect on training practice and educational climate [10
]. This positive effect in part related to participation of all doctors from the departments, thereby reaching the "critical mass" necessary to change the educational culture and behaviour in a department. The findings of this study thus support the recommendation for a future strategy to make it mandatory for senior doctors to participate in courses on clinical training.
Regular appraisal meetings and preparation of personal learning plans became mandatory already in 1998. However, the present reform further emphasised the use of individual learning plans. The significant increase in preparation of individual learning plans following the reform might reflect that the reform had induced an increased adherence to these structural elements of the reform. However, it might also be the result of more junior doctors having attended courses on learning and teaching thus being aware of the usefulness of such plans as an essential tool for them in taking responsibility for their own training [12
The fact that items like 'To what extent are trainees' educational needs attended to in the organisation of daily clinical work routines?' and 'To what extent are clinical situations used for learning?' only had a mean score of five before as well as after the reform indicates that structural educational initiatives fails to be effective unless the entire work-based organisation accepts and prioritises the educational responsibility in the planning of the daily clinical work. Here focus should also be on increasing the daily interaction between seniors and junior doctors. Trainees are an important part of the clinical workforce [7
] and engagement in clinical work activities is pivotal for their learning [13
]; supervision and feedback from seniors are crucial elements for trainees' learning in practice [7
We found that high ratings regarding "To what extent are trainees' educational needs attended to in the organisation of daily clinical work routines?" and "How well do you know each other in the department?" were more prevalent at regional and general hospitals than at university hospitals. This finding is probably due to the smaller size of the departments. Supervision of others was higher at university hospitals, possibly due to the complexity of patients and easier access to specialists. It was apparently not related to educational culture, as university hospitals were poorer in incorporating educational needs in organisation of work routines, and trainees' ratings of receiving feedback and supervision were not related to type of hospital.
In surveys, others have found a discrepancy between consultants' and trainees' perception of the amount and quality of supervision and feedback in clinical settings [4
]. In part, this might be due to consultants' striving at balancing supervision and development of trainees' independence [15
]. Yet, we did not find that higher ratings were associated with being senior rather than junior trainee. Another explanation of this discrepancy might be different expectations and perceptions of what supervision is [15
]. Part of consultants' supervision is probably backstage clinical oversight that trainees are not aware of [16
]. However, matching expectations will be important in creating a well-balanced educational environment.
Interestingly, the laboratory specialties had higher ratings than the other specialities. This could be explained by the often small size of these departments fostering a better learning environment, similar to our findings regarding the smaller hospitals. Another reason could be that better training and supervision in laboratory specialties derive from the presence of requirements for accreditation standards regarding quality assurance of service [18
] in these specialties including proper training and assessment of staff before they are given responsibility for specific service tasks. A key component of medical education reforms is the move from process to outcome evaluation [2
] and a wide variety of assessment strategies have been suggested. It has been demonstrated that implementation of these strategies is feasible and reliable, but also that in-training assessment takes time [20
]. Yet, in order for outcome assessment to be meaningful in the work-based context of PGME there is probably a need for focusing on quality of service by using assessment as a licence for trainees to independently engage in practice of specific tasks and activities [21
]. Unfortunately, supervisors might have less focus on aspects important to the quality of care such as trainees' development of clinical skills, effective communication, and clinical decision-making in connection with cost-effective care [14
]. However, our data suggest that in the work-based context, issues of quality of care might be a driving force of optimising daily teaching practice and learning environment.
It is a strength of this study that the survey included a rather large study group representing all hospital doctors in one geographical area, approximately 3000 doctors. The response rate was high (75%) in 2003, a little lower (58%) but still acceptable in 2007. The lower response rate in 2007 might be due to difficulties of keeping the list of e-mail addresses updated, especially on trainees often changing training sites and a generally lower response rate in e-mail studies.
Although the study only included a limited number of items, we doubt that the overall result of this study is due to instrumentation bias. The items in our questionnaire corresponded well with items rated of high importance in a recently developed tool, Postgraduate Educational Environment Measure (PHEEM) [24
]. This international tool was subsequently validated in a Danish study [25
]. Unfortunately, this tool was not available when this study was conducted. Our questionnaire was used in another study [10
], successfully detecting changes in daily clinical practice. Hence, we feel confident of the validity of our instrument.
The challenge of implementing medical education reforms in the clinical context is probably not a local phenomenon, as similar findings have been reported in other contexts [26
]. The reform process was expected to be rather slow [27
] and maybe our results would have been different if the follow-up period had been longer. In order to implement reforms in the clinical work-based context wide change management strategies are necessary [26
]. This includes involvement of all clinicians in the reform process, from leaders to trainees, and not only designated educational supervisors [27
]. Our data suggest that the next step in implementation of the reform content must be a higher degree of involvement of senior trainers in improving the educational culture in clinical departments. In addition, accreditation bodies of postgraduate education and hospitals should increasingly focus on issues related to quality of daily clinical training practice as well as mandating courses on learning and teaching for consultants.
Future studies might explore which factors, processes and initiatives are important for successful implementation of changes in the clinical training and educational environment as well as improvements of the effectiveness of clinical trainers. Next step could be focus group interviews with senior trainers and junior trainees regarding which areas and items should be in focus in the next phase of the specialist reform.