Problem based learning using ambulatory patients

Introduction: Problem based learning (PBL) tutorials based on ambulatory patients were conducted for year 5 medical students undertaking their General Practice rotation. The students, who had little prior experience of PBL, participated in two sets of PBL tutorials over the four-week course, in addition to traditional seminars, tutorials, and practice-based teaching. Cases were constructed around real patients with ongoing problems which were often evolving or incompletely explored. Working in a PBL format students explored dimensions of the case, with access to the patient and other resources including health workers involved in the case, the medical record, and a house call. At the end of the week students presented the case and their recommendations for management and summarized this in a letter which formed part of the medical record. Methods: Sixteen PBL tutorials with eight groups of students were formally evaluated by a combination of student and tutor questionnaires, direct observation, and interviews with patients and tutors. All tutorials were observed by an experienced independent rater, using a validated instrument. Students completed questionnaires at the completion of each problem and participated in a focus group at the end of the rotation. Tutors completed a Group Assessment Schedule and were interviewed by a research assistant after each tutorial. The same research assistant interviewed the patient and the clinician responsible for the patient's care, probing their perception of the PBL process, the students' learning, and any positive or negative impacts on the patient's health care. Results: Students enjoyed the group work, but had mixed feelings about the PBL format. Groups struggled with the new approach but many saw the value of exploring a problem widely and holistically. Group characteristics such as leadership and prior experience of PBL, along with tutor factors had a major influence on the outcome. Groups and tutors adopted a variety of strategies to overcome problems with the new approach. Most groups undertook an appropriate range of PBL-related tasks, although self-evaluation occurred rarely, and groups did not make full use of available resources. Students valued the real patient contact and authenticity of the case. The patient interview was seen as a key part of the process, providing students with rich information and immediate feedback. House calls, when undertaken, provided students with many insights which were often unexpected and only appreciated in retrospect. A standardized approach was developed to case design, tutor training, and implementation of tutorials. The most suitable cases were those that were evolving or incompletely explored, with a number of dimensions to investigate, and with a level of complexity appropriate to the students' prior knowledge. The use of real patients was seen as engaging, stimulating and contextually deep, providing immediate feedback and an opportunity to integrate and apply learning. Most students felt that this approach emphasised thinking, creativity and holistic care. The authenticity and opportunity to potentially contribute to patient care created an encouraging learning environment. Patients were willing to take part, with none reporting any adverse effects. Discussion: This approach appears to be consistent with the literature on PBL and the use of real patients, and extends the literature on clinical PBL by describing the added value of real patients and making specific recommendations on case selection, tutorial design and delivery and educational outcomes. There was an apparent mismatch between the educational innovation and student experiences, but tutors and groups worked together to resolve these issues. This approach is consistent with trends towards community-based education, teaching holistic care, and the use of the patient's voice. It appears suitable for clinical teaching in a variety of settings providing suitable patients matched to the learning objectives can be identified, although further work is needed to address this question