Context and setting Early clinical encounters during the first 2 years of medical school place students with limited knowledge and skill in contact with sick patients, often without the support of the team that traditionally accompanies inpatient clinical clerkships. Why the idea was necessary Course chairs struggle to keep in touch with students' experiences, an essential component of developing and maintaining a student-centred curriculum. Students, in turn, need support and guidance as they start their clinical encounters. Existing pre-clinical curricula lack the social support system that comes with the inpatient care team. Timely insights into students' experiences and struggles during their first exposure to sick, hospitalised patients may provide opportunities for course chairs to connect with and counsel students. What was done At Mayo Clinic College of Medicine, all Year 2 medical students, in groups of 12–14, are required to take an 8-week, on-campus internal medicine course (Med2), a pre-clinical course following the Year 1 Introduction to Clinical Medicine. During Med2, students complete 8 comprehensive histories and physical examinations on hospitalised patients, followed by written and oral case presentations, and participate in weekly physical examination rounds. Beginning in 2001, the students were required to write weekly e-mails to the course director. The students were given 5 questions to promote self-reflection, but were told that they might write freely. The course chair responded to the e-mails, allowing for sharing of experiences and reflections. The journals were not graded. In 2003, the students' attitudes towards the required e-mails were assessed by an anonymous survey at the end of the course. Evaluation of results and impact All students participated and wrote an average of 6 e-mails over 8 weeks. The course chair spent 1 hour per week reading and responding to 10–11 e-mails, each between half and 1 page in length. While the quality varied, the majority of the e-mails provided insight into the students' struggles with detecting abnormal physical examination findings, encountering issues of death and dying, and dealing with patient−learner relationships. There were also remarks of affirmation, teacher effectiveness and insights into the learning process. In 2003, all 39 students (of that year's class) completed the anonymous survey. Twenty-six (66%) considered the e-mails helpful, 3 (7%) found them somewhat helpful and 10 (25%) found them unhelpful. The majority (66%) indicated that the e-mail system was a good forum for expressing their struggles and concerns. Twenty-eight (72%) of the students thought they benefited from the course chair's responses to the e-mails. While candour was encouraged and the course was assessed on a pass/fail basis, an obstacle for honesty and disclosure in the e-mails for some students may have been the lack of anonymity. The benefits of individualised e-mail responses versus complete anonymity should be evaluated. Consideration should also be given to the frequency of e-mails as some students felt it became tedious. While our experience is that e-mailing has been beneficial, adaptations (such as e-mail facilitators) will need to be made to implement this at larger medical schools.