Medical education programs in North America are often based on the adult education theory of self-directed learning (SDL). Many kinds of SDL have been introduced into preclerkship education, clerkships, residency training, and continuing medical education. The first goal of this paper was to review SDL in North American medical education. The second goal was to describe an example of community-based clerkship in which SDL was applied in the department of family and community medicine of the University of Toronto. A third goal was to give three recommendations for Japanese clinical clerkships. The first recommendation is the effect of the learning contract. The second is that the preceptor should give quick and frequent feedback to students and that a useful Japanese feedback device should be developed with information technology. The third recommendation is that a new curriculum combining community-based education for students with continuing medical education for doctors is necessary to improve Japanese medical education.

1) The concept of interprofessionality, i. e., cohesive and cooperative practice between professionals, is necessary in North America because health professionals specializing in various fields work together on medical services.
2) To introduce the concept of interprofessionality, interprofessional education, in which many kinds of health profes-sionals or students learn together, is extremely important. The Canadian government has been promoting and generously subsidizing interprofessional education as a way of improving public health and emphasizing patient-cen-tered medicine.
3) In 2006, the University of Toronto established a new certificate course for future educational leaders who promote in-terprofessional education. We participated in this course and recommend that a course of this type be introduced toJapan.

BACKGROUND: Japanese medical education has undergone dramatic changes over the last 5 years. Clinical exercises and ambulatory-care training are now stressed to prepare medical students and residents for work in primary and continuing-care settings. For comparative purposes, we conducted a review of the undergraduate and residency training programs for ambulatory care at the University of Toronto in Canada. This report will examine the problems of training programs for ambulatory care in Japan by comparing the Canadian and Japanese models.
METHOD: From December 2004 to March 2005, the first author observed the ambulatory training systems at the University of Toronto.
OUTLINE OF CANADIAN AMBULATORY TRAINING PROGRAMS: There are three typical types of ambulatory training programs in Canada: community-office based programs for undergraduate students in family and community medicine; hospital/clinic based programs for junior residents in internal medicine; and consultation service programs for senior residents in internal medicine. Undergraduate and residency training programs are largely consistent with each other. The current trend in medical education is towards increased consolidation and efficiency in teacher and student training systems, with a reduction in the number of teaching hospitals and integration of teaching staff and curricula. Moreover, team-based training for ambulatory care appears effective.
DISCUSSION: To improve the Japanese ambulatory training system, it is desirable to increase communication and contact between undergraduate-program educators and residency-training program educators in order to achieve integration and consistency between programs.

In Japan, awareness has increased in recent years of the importance of evaluating clinical educators. In Europe and North America, the Objective Structured Teaching Evaluation (OSTE), which employs standardized students, multiple stations, video recording, and scoring by multiple observers, is used to evaluate clinical educators. We report on the implementation of an OSTE in Japan.1) Ten clinician-educator physicians participated in the OSTE, which comprised 5 stations and included standardized residents. The stations were video-recorded, and the educators were assessed by 7 different evaluators.2) The educators were evaluated with a checklist and a 5-point scale. We assessed the reliability and validity of the checklist and analyzed the background characteristics of the clinician educators.3) The factors most closely associated with high ratings on the checklist and the 5-point scale were: having a history of attendance at a seminar for clinician-educators, having greater than 5 years experience as an educator, and not being an internist. There was no interobserver variability among the evaluators.4) The generalizability of the checklist was 0.81, and its reliability index was 0.83. The correlation coefficient between the total scale score and the checklist score was 0.8. 5) Although biases by participants were identified, our project suggests that the OSTE could be used in Japan to objectively evaluate the teaching skills of clinician-educators. Further research on the OSTE in Japan is warranted.