main difference between the two methods was undoubtedly the different type of information offered to student. We did not distinguish between the effect of using multimedia and that of giving additional information on the analysis of lung sounds, as these two aspects are deeply intertwined. The concepts on which the current classification of lung sounds is founded are themselves multimedial in nature 3], and computer multimedia seemed to be just the best method available for their presentation. We suggest that the combination of acoustical, graphical and analytical representation of sounds provides a homogeneous set of information that is more easily fixed in memory and may match different learning styles feedback questionnaire, where the vast
majority of association
students reported that they found the
representation to be understanding lung sounds.
It is interesting to note that, although all the parameters of lung auscultation used to compute the LSDIS were slightly improved in the multimedia group compared to controls (table 1), most of the effect of multimedia presentation on the cumulative LSDIS is due to a reduction in the number of fabricated findings, especially undefined and abnormal breath sounds. We speculate that students who are beginners are biased toward an excessive expectancy of sound abnormalities, and that a better knowledge of the origin and of the acoustic properties of lung sounds after the multimedia presentation could have induced an attempt at more precise identification of sounds, thus reducing the number of fabricated sounds.
Our data confirm a previous study indicating that practicing with a multimedia computer program does improve the proficiency of medical students and residents in the recognition of recorded lung sounds study, we used the computer as an aid for a small-group seminar, rather than using a program for individual use. Although this approach partially conflicts with the philosophy of personal computing, based on unrestricted individual usage of computing resources, we found that is also presents significant advantages. The first is that setting and maintaining a multimedia computing classroom with enough machines to guarantee sufficient access to all the students, not to mention the need for periodically upgrading the hardware at the current fast development rate of technology, may be prohibitively expensive, and most academic sites, including our university, do not provide such a facility . Using only one machine is cheaper, and any hardware or software upgrade is immediately available for all the classroom.
The second advantage of our approach is that it does not require previous computer experience by part of the students, or additional learning of program instructions and commands. Despite the increasing popularity of personal computing, the growth of educational computing in the academic setting is relatively slow as compared to the business environment, and sufficient computer skill cannot be expected in all the medical students . It is noteworthy that, despite a general agreement on the computer approach on learning lung sounds, only 56% of our students reported that they would find a teaching program useful for individual use, mostly because they had limited experience with computers. Since the groups
attending the multimedia presentation were kept relatively small, it was still possible for the students to participate interactively in the presentation, partially overcoming one of the possible drawbacks of this method compared to the individual use of a computer. We found this approach to be more satisfactory than giving a single seminar to the whole class in a large lecture room, using the same computer connected to a large-screen video-projector, as we had done the previous year.
Our method of evaluation of lung auscultation skills was also different from that used in previous studies, where the ability to detect lung sounds correctly was assessed on sounds previously recorded on audio tape . Although that method allows a better standardization of the test sounds, we are not fully convinced of the quality and the realism of tape-recorded lung sound. We therefore preferred a more realistic test based on auscultation of actual patients, which also offers the opportunity of evaluating some aspects of lung sounds, such as timing and localization, that cannot be assessed on audiotape .
We certainly do not imply that a multimedia seminar can completely replace conventional bedside teaching in medical training, including teaching of auscultation skills. However, time and emphasis dedicated to physical examination in medical training has recently decreased, mostly because of the increasing amount of knowledge and technologies coming to medicine from basic sciences , whereas proper knowledge of lung auscultation is probably no less important today, when lung sounds have been rationally and reproducibly classified , and their pathogenetic mechanisms and clinical significance are increasingly known In this perspective, being able to shorten and ease the learning curve of chest auscultation, multimedia presentation of lung sounds may improve the proficiency of the students in subsequent clinical activities.
The authors are indebted to A.
Henderson for reviewing the manuscript.
1. Murphy R, Holford S, Knowler W. Visual lung-sound characteristics by time-expanded waveform analysis. N Engi J Med 1977; 296: 968—971. 2. Forgacs P. Lung Sounds. London, Ballière Tindall, 1978. 3. Loudon R, Murphy RLH. State of the Art: lung sounds. Am Rev Respir Dis 1984; 130: 663—673.
Earls J. Lung sounds. Thorax 1992; 47: 671—672.
Spence DPS, Bentley S, Evans DH, Morgan MDL. Effect of methacholine-
induced bronchoconstriction on the spectral characteristics of breath sounds in asthma. Thorax 1992; 47: 680—683. 6. Kaisla T. Sovijarvi A, Piirila P, Rajala H-M, Haitsonen
S, Rosqvist T. Validated method for automatic detection of lung sound crackles.