Abstract:
This study evaluated prevalence of hearing loss (dependent variable) in relation to occupational noise levels (main independent variable) in 359 press parts workers at an automotive parts factory in Samut Prakan Province, Thailand, in 2010, 2011, and 2012. Other independent variables included sociodemographic characteristics, work history, and knowledge, attitude, and behavior related to noise and hearing loss. Data collection tools were hearing test results, measured noise levels at multiple sites in the factory, and an interviewer-administered standardized questionnaire. Independent variables were considered separately, by chi-square test for categorical variables and by logistic regression for continuous variables.
Prevalence of measured hearing loss in either ear varied widely, from 2.5% in 2010 to 31.2% in 2011. Means of annual average noise level varied only from 89.9 to 90.4 dB(A), and means of annual high noise level veried only from 94.6 to 95.5 dB(A). Observed associations between independent variables and measured hearing loss prevalence were not consistent from year to year. Also, hearing loss was never statistically significantly associated with measured noise level. In 2010, hearing loss was weakly positively associated with behavior score, and with no other independent variable. In 2011, hearing loss prevalence was significantly negatively associated with male gender (p-0.048), and was significantly lower in workers who reported being sensitive to noise than those who did not (p=0.015). Hearing loss was negatively associated with being married and with self-reporting of hearing loss, and positively associated with educational attainment (p=0.116, p=0.076, and p=0.148, respectively). In 2012, hearing loss was significantly positively associated with age, income, male gender, and being married (p<0.001, p=0.002, p=0.005, and p=0.038, respectively). Hearing loss was also significantly lower in the motorcycle parts pressing section than in the automotive parts pressing section or the fuel tank production section. (p=0.050). Hearing loss was also moderately positively associated with high noise level (p=0.122), but not with average noise level. Reasons for inconsistency of results are not clear, although quality of hearing tests was evidently better in 2012 than in other years. Overall, the 2012 results appear most reliable. Conceivably, noise levels at the factory were not high enough to produce appreciable hearing loss. Further research on this topic is needed. Further multivariable analysis of this study's results would also be desirable.
