The Effect of Noise Pollution on Lung Sound Recognition by Medical Students

The Effect of White Noise on the Ability of Second-year and Third-year Medical Students to Recognise Correctly Adventitious Lung Sound

Title: The Effect of White Noise on the Ability of a Medical Student to Recognise Correctly Adventitious Lung Sound. Methodology: Randomized, controlled trial, blind study. Study Duration: The estimated duration for the main protocol (e.g., from the start of screening to the last subject processed and finishing the study) is approximately 4 weeks. Study Centre: Lithuanian University of Health Sciences (LSMU) and Kaunas Hospital of LSMU. Objectives: Primary Objective: To evaluate the accuracy of second-year and third-year medical students in correctly identifying pathological and nonpathological lung sounds under different white noise levels. Number of Subjects: A group of 104 students will be exposed (EXP) to learning pathological and nonpathological sounds for 3 days. After 3 days students will take 3 tests with different levels of white noise pollution added ( signal-to-noise ratio (SNR) 20, 40 and none added). Diagnosis and Main Inclusion Criteria: Inclusion Criteria: male and female second and third-year LSMU students; 18-40 years old; in any distribution. Consent and compliance with all aspects of the study protocol, and methods, providing data during follow-up contact. Please, see the methods section for a full list of inclusion criteria. Exclusion Criteria: deafness; age over 40; conditions that prevent the student from using earphones. Please, see the methods section for a full list of exclusion criteria. Regimen: EXP group will receive 3-day training of 101 lung sounds. Statistical Methodology: Results will be analyzed with the SPSS (version 27) and Jamovi (version 2.3.21). A p-value < 0.05 will be considered statistically significant. The effect of 3 levels of white noise on the EXP group will be compared utilizing a one-way analysis of variance (ANOVA) statistical test. Though, if data does not adhere to a normal distribution then Kruskal - Wallis will be applied. McNemar's test will be used to compare if there was a statistical difference between second-year and third-year students.

Purpose To evaluate the accuracy of second-year and third-year students in correctly identifying pathological and nonpathological lung sounds under different white noise levels. Background Auscultation is one of the four pillars of patient examination [1, 2, 3], and is particularly during the pulmonary objective investigation. Today, the assessment of auscultation sounds remains subjective and depends on the qualifications and expertise of a specialist [4, 5]. Arts et al. [6] authors recommend leaving the practice of traditional auscultation to experienced specialists since sensitivity is as low as 37% when identifying between normal cases and several lung pathologies. Hospitals can have a lot of white noise which easily exceed WHO recommendation [7]. Though some research shows that physicians can adapt to extreme loudness [9]. There is a lack of research in clinical settings on noise levels' effect on lung sound recognition by healthcare workers. Hence our research is interested to investigate the effect of white noise (WN) at 3 different levels (SNR 20,40, none) on medical students' ability to correctly identify various lung sounds. Goals of the study To evaluate the specificity, sensitivity and accuracy of second-year and third-year medical to correctly identify lung sounds under three different WN levels. Duration of the Study The study is estimated to complete enrolment within 3 weeks from study initiation; however, enrolment will remain open until the study goal is met. The duration of this study for each subject will be a maximum of 4 weeks. Product Description Commercially available various earphones, mobile phones and PC will be used by the students. These products have been acquired by students before the research. Product Intended Use The students' mobile phones, laptops, and PCs will be used to access the SAITI website. Their earphones will be utilized to listen to lung sounds. Product Acquisition The product was acquired by students before the study with their finances for personal use. Potential Benefits and Risks to Students There are no potential risks to the students. A potential benefit for the student is their ability to learn how to identify pathological lung sounds earlier in their medical career as compared to their peers. Methods: Study design A single-arm study involving 35 students and 101 lung sound recordings. The lung sounds will be used for training students and then assessing their accuracy to identify pathological sounds correctly. Creation of a lung sound database for teaching and assessment purposes: a total of 84 patients & 504 lung sounds were recorded by an Internal Medicine physician. Permission was obtained for the study from Kaunas' Regional Bioethics committee (P1-BE-2-57/2021). All participating parties signed the bioethics agreement forms. The lung sounds were recorded in the wards of the LSMU Kaunas Hospital using a 3M Littmann model 3200 electronic stethoscope. Each of the 504 recordings was 15 seconds long. Once all lung sounds were recorded, a double-blind review was initiated. During this process, 101 lung sounds were selected by a family physician. The family physician had to listen to the lung sounds and judge if she heard crackles, wheezes, and no adventitious. The additional criterion in judging was if the lung sounds were "good enough" quality for teaching purposes or not. The cases were used for training purposes if both the Internal Medicine physician and the Family physician were in agreement. All students studying in LSMU in the 2nd- and 3rd-years will be invited to participate in the study. Study population and selection criteria: all aspects of the study and consent forms will be IRB-approved before implementation. All participants will require full informed consent, be willing and able to comply with all study requirements and will meet the following criteria; male or female aged 18-40 (inclusive), must be LSMU students studying in 2nd and 3rd year. Subjects will be excluded from the study based on the following criteria: Not a student at LSMU Not studying in 2nd and 3rd year of medicine Age over 40-year-old. Deaf The subject does not provide full consent Recruitment methods Subjects will be identified and contacted through the Student Scientific Society and further contacted via emails. The students will receive a test before training on the SAITI website. There will be two presentations. The first will be an introduction to the project and its explanation to the students. The second will be an introduction to lung sounds (normal lung sounds and adventitious ones). The duration of both presentations will be about 30 - 35 min each. The main topics will be to introduce the students to how the study is conducted, and what sort of pathological and nonpathological sounds they can expect during the learning and assessment process. The training will consist of 16 cases with 6-channel recordings for each patient and a single-channel individual lung 5 sounds (bronchovesicular, bronchial, vesicular, crackles, wheezes). The six-channel locations will be; 2 above the scapula, two between the shoulder blades and two below them. Each channel will be a single-point recording (15s) from an actual patient. The cases will consist of 4 patients with no pathologies, 4 cases with crackers, 3 cases with wheezes, 3 cases with mixed pathologies (wheezes and crackles) and 2 control cases but with sonic artifacts. The students will have access to the site for 3 days. They will be advised to spend 45 to 90 min on the website per day. The students after 3 days learning process will receive 3 exams with different levels of WN. The questions will be made of 9 cases: 3 cases without pathological lung sounds; 3 cases with wheezes; 3 cases with crackles. All questions will be randomised. The student will have to listen through all six channels and will have to choose the correct answer. The number of true positives, false positives, true negatives and false negatives will be recorded. Data will be collected on an excel spreadsheet system as the primary source of patient response collection. A dedicated hard-copy paper of the excel will be produced and kept as a backup. Results Results will be analyzed with the SPSS (version 27) and Jamovi (version 2.3.21). A p-value < 0.05 will be considered statistically significant. The effect of 3 levels of white noise on the EXP group will be compared utilising a one-way analysis of variance (ANOVA) statistical test. Though, if data does not adhere to a normal distribution then Kruskal - Wallis will be applied. McNemar's test will be used to compare if there was a statistical difference between second-year and third-year medical students. Data collection and reporting Data will be collected at the following points: Immediately after 4 days of training, on the fourth day, there will be 3 days where students will take 3 tests ( each test will have different levels of SNR 0,40,20). Data from the study will be maintained for two years after the date the investigation is completed, terminated or until the records are no longer required to support the protocol, whichever date is later. Expected outcomes The study has the potential to measure the ability of second and third-year students to correctly identify pathological lung sounds under different WN levels. Adverse reactions: none expected. The participants will use their equipment to listen to the lung sounds according to the parameters that manufacturers set up for the equipment. Reasons for withdrawal or termination A subject may be discontinued from the study at any time if the subject, Investigator feels that it is not in the subject's best interest to continue. The following is a list of possible reasons for study treatment discontinuation: screening Failure; subject withdrawal of consent; the subject is not compliant with study procedures; protocol violation requiring discontinuation. All subjects are free to withdraw from participation at any time, for any reason, specified or unspecified, and without prejudice. Reasonable attempts will be made by the Investigator to provide a reason for subject withdrawals. The reason for the subject's withdrawal from the study will be specified in the subject's source documents and the Case Report Form (CRF). Premature Termination or Suspension of Study This study may be suspended or prematurely terminated if there is sufficient reasonable cause. Written notification, documenting the reason for study suspension or termination, will be provided by the suspending, or terminating party to the investigator or the Institutional Review Board (IRB), as appropriate. If the study is prematurely terminated or suspended, the Investigator will promptly inform the IRB and will provide the reason(s) for the termination or suspension. Circumstances that may warrant termination or suspension include, but are not limited to: Insufficient compliance with protocol requirements; Data that are not sufficiently complete and/or evaluable; Data handling and record keeping Only study personnel will collect data onto Excel spreadsheets. All spreadsheets will be encrypted and accessed only via a password. Only de-identified data will be used for data analysis. Statistical analysis will be performed on de-identified collected data. Institutional review board The protocol, informed consent form(s), and all participant materials will be submitted to the IRB for review and approval. Approval of both the protocol and the consent form must be obtained before any participant is enrolled. Any amendment to the protocol will require review and approval by the IRB before the changes are implemented in the study. All changes to the consent form will be IRB approved; a determination will be made regarding whether previously consented participants need to be re-consented. Protocol deviation All protocol deviations/violations will be documented using the Protocol Deviations/Violations CRF and submitted to the IRB according to their reporting guidelines. Laws and regulations This clinical study will be conducted in compliance with all national laws and regulations of Lithuania in which the clinical trial is performed, as well as any applicable guidelines. The trial will be registered on www.clintrials.gov and other sites, as appropriate. Publication and data sharing policy The preparation and submittal for publication of manuscripts containing the study results shall be by a process determined by mutual written agreement among the participating institutions. All publicly shared data will be de-personalized according to Lithuanian bioethics community regulations

Auscultation, Lung sounds, Pathological lung sounds, Adventitious lung sounds, Simulation-based learning, White noise

LSMU second-year and third-year medical students will be invited to participate in a single-arm, diagnostic study. Students will be trained for 4 days to recognise nonpathological lung sounds. The training will be via two lectures and a simulation-based study's website. Following the 4-day learning interval, students will be assessed via 3 tests, each test having different levels of white noise.The number of true positives, false positives, true negatives and false negatives will be recorded with the goal to assess students' diagnostic accuracy in identifying adventitious lung sounds.

Exposed group (EXP) - 35 students. EXP group will learn pathological and nonpathological sounds for 3 days.

Exposed group (EXP) - 35 students. EXP group will learn pathological and nonpathological sounds for 3 days.

After 3 days students will take 3 tests with different levels of white noise pollution added ( signal-to-noise ratio (SNR) 20, 40 and none added)

Measures of medical students' specificity, sensitivity and accuracy in identifying pathological lung sounds will be assessed under 3 levels of white noise (WN). The WN is created by using a sinus wave. The sinus wave is added to the 15s lung sound recordings. The signal-to-noise ratio (SNR) levels are SNR-20; SNR-40 and no WN. The SNR-20 is perceived as the loudest WN by a human subject. Once each of the 9 questions (each containing six 15s lung sound recordings) is created the files are uploaded to the SAITI website. The result is 9 cases with no SNR; 9 cases with SNR-40; 9 cases with SNR-20. During the examination, a student is asked to listen to each of the six points for each case and then decide if pathology is present. The test results are tabulated for true positives (TP), true negatives (TN), false positives (FP), and false negatives (FN). The diagnostic values are compared at different levels of WN (SNR-20; SNR-40; no-SNR).

Inclusion Criteria: Male and female second and third-year LSMU students; 18 years and older; subject that provides full written consent to participate in the study. Exclusion Criteria: not a second-year or third-year at LSMU; deafness; age over 40; conditions that prevent students from using earphones; subject does not provide full consent.

Narula J, Chandrashekhar Y, Braunwald E. Time to Add a Fifth Pillar to Bedside Physical Examination: Inspection, Palpation, Percussion, Auscultation, and Insonation. JAMA Cardiol. 2018 Apr 1;3(4):346-350. doi: 10.1001/jamacardio.2018.0001.

D. Pereira, P. Gomes, C. Sa Couto, O. Costa, Z. Reis, S. Mattos, R. Cruz-Correia, and M. Coimbra. IS4Learning-A Multiplatform Simulation Technology to Teach and Evaluate Auscultation Skills. In Health Professionals' Education in the Age of Clinical Information Systems Mobile Computing and Social Networks, pages 401-420. Elsevier, 2017. doi:10.1016/b978-0-12-805362-1.00019-x.

Garvick S, Gillette C, Gao H, Bates N, Waynick J, Crandall S. Can cardiac auscultation accuracy be improved with an additional app-based learning tool? Clin Teach. 2022 Apr;19(2):112-120. doi: 10.1111/tct.13462. Epub 2022 Feb 8.

Hafke-Dys H, Breborowicz A, Kleka P, Kocinski J, Biniakowski A. The accuracy of lung auscultation in the practice of physicians and medical students. PLoS One. 2019 Aug 12;14(8):e0220606. doi: 10.1371/journal.pone.0220606. eCollection 2019.

Kim Y, Hyon Y, Jung SS, Lee S, Yoo G, Chung C, Ha T. Respiratory sound classification for crackles, wheezes, and rhonchi in the clinical field using deep learning. Sci Rep. 2021 Aug 25;11(1):17186. doi: 10.1038/s41598-021-96724-7.

Arts L, Lim EHT, van de Ven PM, Heunks L, Tuinman PR. The diagnostic accuracy of lung auscultation in adult patients with acute pulmonary pathologies: a meta-analysis. Sci Rep. 2020 Apr 30;10(1):7347. doi: 10.1038/s41598-020-64405-6.

Darbyshire JL, Young JD. An investigation of sound levels on intensive care units with reference to the WHO guidelines. Crit Care. 2013 Sep 3;17(5):R187. doi: 10.1186/cc12870.

Buderer NM. Statistical methodology: I. Incorporating the prevalence of disease into the sample size calculation for sensitivity and specificity. Acad Emerg Med. 1996 Sep;3(9):895-900. doi: 10.1111/j.1553-2712.1996.tb03538.x.

Zun LS, Downey L. The effect of noise in the emergency department. Acad Emerg Med. 2005 Jul;12(7):663-6. doi: 10.1197/j.aem.2005.03.533.