Remote, interactive, web-based blood cell morphology training

Conventional teaching of blood cell morphology entails an instructor experienced in the interpretation of morphological changes developing a process for teaching the understanding of their clinical significance. The goal of this project was to compare the effectiveness of live interactive teaching of haematological pathology compared to the traditional face to face method of teaching. To address this goal the aims of the study were to: 1. Perform a systematic review of web-based instruction in pathology 2. Compare face to face training versus web-based instruction in a randomized controlled manner. 3. Assess acceptability of web-based morphology teaching. 4. Perform a cost comparison for the two modalities. The utility of a live interactive environment for the initial components of blood cell morphology instruction was assessed by administering the didactic instruction either face to face or via a live web-based interactive environment. Remote, live interaction was achieved using computers to transmit blood film images on computer screens in the laboratories with telephones used for verbal communication, which has previously been termed ‘dynamic telepathology’. Evaluation of accuracy was assessed in terms of Descriptions and Diagnoses for each of the assessment slides, as per the Royal College of Pathologists Quality Assurance Program scoring system. Acceptability was assessed by means of a questionnaire. Costs were compared using a data sheet itemising the costs of the two modalities. 7 participants received web-based instruction and 2 received traditional face-to-face instruction. Statistical analysis of pre- and post-training results showed an improvement in scores post-training for the complete cohort. The small number of participants in the face-to-face group precluded direct comparison the two groups. However, when scores for individual assessment slides were compared, the Wilcoxon signed ranks test showed a significant difference in pre- versus post-training results for both Descriptions (z=2.0, p<0.003) and Diagnoses (z=4.5, p