Abstract
Background: Auscultation involves listening and recognizing abnormal lung sounds. Previous study showed that students more accurately interpreted the auscultation sounds in the time-frequency patterns of different abnormal lung sounds, such as wheezing, stridor, rhonchi, fine crackles, and coarse crackles. This study investigated whether teaching with spectrograms to visualize time-frequency patterns could improve the accuracy of interpreting lung sounds among respiratory therapy students.
Methods: Students from a respiratory therapy program in sophomore to senior years were recruited and randomly allocated into two groups. The control group was taught using a tutorial video playing an explanation of lung sounds. The experimental group was taught using the tutorial video and spectrograms. An audio test was developed to assess knowledge of lung sound interpretation before and after the intervention. The test included 20 multiple-choice questions about six types of lung sounds: normal, stridor, rhonchi, wheezing, coarse crackles, and fine crackles. The questions were presented in a random order. The test was administered through the Surveycake online platform. The students’ scores were compared to the correct answers and graded into percentage (%). Mann-Whitney U test was used to compare scores between the two groups, and Wilcoxon singed-rank test for those within group. Fleiss’ kappa for within-group agreement. The study was approved by the Chang Gung Institutional Review Board.
Results: A total of 20 students were recruited. The score of pre-intervention shows no difference with the median (interquartile range) of 50.0% (37.5–61.3%) with clinical experience and 60.0% (51.2%–75.0%) without experience (P = .10). The table below shows the scores between the two groups among the different lung sounds, and the results illustrate that the use of spectrograms for lung sound interpretation teaching did not improve the test scores. The figure shows the agreement to correct answers for each individual before and after intervention. The experimental group showed a significant improvement (0.669 to 0.813; P = .002) on the interpretation of continuous lung sounds (wheeze, stridor, and rhonchi).
Conclusions: Clinical experience has no relationship with lung sound interpretation. The results indicate that incorporating spectrograms into the teaching materials did not improve interpretation accuracy but did improve the within-group agreement of interpretation of continuous lung sounds.
Table: Comparisons of test scores before and after intervention
Figure: Cohen's kappa analysis for agreement of each individual before and after test. Red circles highlight the improvement of agreement of individuales.
Footnotes
Commercial Relationships: Dr. Fu-Shun Hsu is a founder of Heroic-Faith Medical Science. We thank Heroic-Faith Medical Science for providing spectrograms of the lung sounds.
- Copyright © 2023 by Daedalus Enterprises
