1. How activity type, time on the job and noise level on the job affect the hearing of the working population. Using Bayesian networks to predict the development of hypoacusia
- Author
-
Jesús P. Barrero, Susana García-Herrero, José María Gutiérrez, and M.A. Mariscal
- Subjects
media_common.quotation_subject ,Applied psychology ,Public Health, Environmental and Occupational Health ,Occupational disease ,Bayesian network ,Sample (statistics) ,Affect (psychology) ,medicine.disease ,03 medical and health sciences ,Noise ,0302 clinical medicine ,Perception ,medicine ,Working population ,030212 general & internal medicine ,Noise level ,030223 otorhinolaryngology ,Safety, Risk, Reliability and Quality ,Psychology ,Safety Research ,media_common - Abstract
In this research we identify the main factors believed to trigger occupational hypoacusia in an effort to increase our knowledge of how this occupational disease occurs and develops. With this goal in mind, we have gathered various demographic/personal, occupational and non-occupational data from a heterogeneous sample of 1418 workers. The data selected include the noise levels to which the individuals in the sample are exposed. This entailed taking measurements at their respective jobs, as well as doing an objective assessment of their hearing ability, which required administering medical hearing tests. Lastly, the workers completed a survey on various habits and other factors deemed to be influential, and on the respondents’ own perception of their hearing. Bayesian networks were used to obtain the conditioned probability of developing hypoacusia based on the data collected from the sample. Specifically, for this study we used the general network created by the relationships between all of the factors associated with developing hypoacusia in order to analyze the influence individually and by grouping three specific variables: activity sector, noise level and time on the job. This work yielded a considerable database that can be used to conduct a multitude of analyses intended to study and predict the hearing acuity of the working population under different scenarios. Specifically, in the case at hand, the Bayesian network obtained indicates that the three factors analyzed influence the hearing of the individuals, though to different extents. The least influential factor involves the sector of activity, followed by the noise level on the job, which varies noticeably in favor of better hearing for workers in jobs whose noise levels are rated as low. Finally, we deemed time on the job (which is also related to age), as the most influential factor as it exhibits the largest differences among its potential states, with workers whose time on the job is rated as low or medium exhibiting the best likelihood of having good hearing.
- Published
- 2018
- Full Text
- View/download PDF