RETROSPECTIVE COHORT STUDY OF PURE TONE AUDIOMETRY THRESHOLD SHIFTS FROM OTOTOXIC SUBSTANCE, CONTINUOUS NOISE, AND IMPULSE NOISE EXPOSURES AT TINKER AIR FORCE BASE FROM 2005 TO 2019
Abstract No:
1725
Abstract Type:
Student Poster
Authors:
M Blair1, J Slagley2, N Schaal3
Institutions:
1United States Marine Corps, Beavercreek Township, OH, 2Air Force Institute of Technology (AFIT/ENWL), Wright-Patterson AFB, OH, 3U.S. Navy, Dayton, OH
Presenter:
Marc Blair
United States Marine Corps
United States Marine Corps
Faculty Advisor(s):
Jeremy Slagley, CIH
Air Force Institute of Technology (AFIT/ENWL)
Air Force Institute of Technology (AFIT/ENWL)
Nicholas Schaal, Ph.D., CIH, CSP
U.S. Navy
U.S. Navy
Description:
This retrospective cohort epidemiology study sought to establish the comparative risks and potential indicators of hearing loss associated with combinations of ototoxic substances, impulse noise, and continuous noise exposure. Currently, there is not an existing model or methodology in the Department of Defense (DoD) that joins occupational exposure data and pure tone audiometric data. After developing an integrated database model for Tinker Air Force Base, the largest of three depot installations within Air Force Material Command, 2,372 individuals were grouped into eight combinations of exposure groups with a minimum three years exposure duration to hazards. The incidence rates and relative risk of hearing loss indicators were calculated with five different pure tone audiometry evaluation methods. With the NIOSH Significant Threshold Shift criteria, a significant increase in risk occurred in the left ear at 2,000 Hz for the Metal/Solvent/Continuous exposure group (RR=2.44 CI 1.24-4.83) compared to a continuous noise only reference group. Further descriptive and inferential statistical analysis confirmed a significant difference (Bonferroni adjusted p-value=0.023) in hearing threshold shifts in the left ear at 2,000 Hz between this exposure group and reference exposure group. In the presence of continuous noise exposure, ototoxic effects on hearing loss could only be observed in the 1,000 and 2,000 Hz frequencies. Due to data availability, researchers could not establish further confidence in results with descriptive statistical analysis or logistic regression. Results indicate the current DoD Hearing Conservation Program's significant threshold shift criteria potentially do not capture the increased risk of hearing changes from ototoxic substance exposure.
Situation/Problem:
Traditionally noise exposure, both continuous and impulse, is the primary factor associated with occupational hearing loss. However, growing research indicates that ototoxic substances commonly found in occupational settings could potentially affect hearing loss independently, additively, or synergistically when combined with noise exposures. In response to this research, the American Conference of Governmental Industrial Hygienists (ACGIH) adopted the "OTO" notation for potential ototoxic substances in the organization's 2019 Threshold Limit Values (TLV) publication and the United States Department of Defense directed services to evaluate ototoxic exposures to determine their relation to the risk of occupational hearing loss. Despite the growing body of knowledge, it is unclear what effect ototoxic substances have on hearing loss, there are no established occupational exposure limits (OEL) based on hearing loss risk, and DoD specific epidemiology studies are limited.
The lack of established ototoxic substance-specific occupational exposure limits and knowledge of combined effects from combinations with continuous or impulse noise exposure requires additional epidemiological research to focus limited government resources. Recent updates to the DoD HCP direct components to assess the interactive effects of noise and ototoxic substance exposure, but specific substances of concern or methodologies are not detailed. There is a need to inform future DoD efforts to maximize limited resources both in industrial hygiene sampling efforts and hearing conservation program assignments. Understanding the interactive effects of ototoxic substances and noise requires a model to match an individual's occupational exposures of interest to pure tone audiometric data to determine potentially casual relationships. Currently, occupational and audiometric records are available separately but not in an integrated model for focused research. A study is necessary to identify the optimal integration of databases for the evaluation of exposures and health outcomes.
The lack of established ototoxic substance-specific occupational exposure limits and knowledge of combined effects from combinations with continuous or impulse noise exposure requires additional epidemiological research to focus limited government resources. Recent updates to the DoD HCP direct components to assess the interactive effects of noise and ototoxic substance exposure, but specific substances of concern or methodologies are not detailed. There is a need to inform future DoD efforts to maximize limited resources both in industrial hygiene sampling efforts and hearing conservation program assignments. Understanding the interactive effects of ototoxic substances and noise requires a model to match an individual's occupational exposures of interest to pure tone audiometric data to determine potentially casual relationships. Currently, occupational and audiometric records are available separately but not in an integrated model for focused research. A study is necessary to identify the optimal integration of databases for the evaluation of exposures and health outcomes.
Methods:
Despite the regular usage of Defense Occupational and Environmental Readiness System – Industrial Hygiene (DOEHRS-IH) and Defense Occupational and Environmental Readiness System – Hearing Conservation (DOEHRS-HC), there is not a direct linkage between systems, and a model is required to determine exposures and health effects. Utilizing data from 2005 to 2019, individual records were constructed independently from each database, combined, and then grouped by combinations of exposure with Microsoft Access. After grouping by study exposure group, researchers determined relative risk utilizing multiple pure tone audiometry evaluation criteria, conducted a statistical analysis to determine differences between groups, and constructed regression models.
Initial record construction from DOEHRS-IH and DOEHR-HC required extensive programming efforts to create relevant individual records. Researchers created individual exposure records utilizing SEG exposure assessment evaluations and personnel assignments with DOEHRS-IH data. Determination of an individual's health outcome utilized criteria to select an individual's first audiogram record and final audiogram record to calculate a threshold shift record. Calculations for threshold shifts utilized both unadjusted and OSHA age-adjusted frequency threshold values to identify if age was a confounding factor. This research only considered an individual eligible for the cohort if they demonstrated normal hearing on the selected first audiogram record in the research sample. Using a database joining process that excluded DOEHRS-IH exposures outside of established DOEHRS-HC audiogram dates, researchers then created a single database with a single record for each qualifying individual.
Following the creation of a combined single data source, researchers evaluated the relative risk of hearing loss by study exposure groups utilizing individual or aggregated frequency threshold values and shifts. The threshold values and shifts utilized by researchers included DoD, OSHA, and NIOSH indicators of hearing loss. Additionally, relative risk comparisons included analysis of data using both unadjusted and OSHA age-adjusted thresholds. Study exposure group data was then exported for statistical analysis utilizing Python to qualitatively and quantitatively describe data. Based on the descriptive analysis, researchers determined if there were statistical differences across 500, 1000, 2000, 3000, 4000, and 6000 center band individual frequencies and aggregated frequencies between study exposure groups.
Limitations: Researchers assumed that the current DoD exposure assessment strategy, a process that maximizes limited resources to manage prioritized risks, sufficiently captured the actual exposure hazards in an occupational setting. Researchers utilized selected baseline audiogram and final audiogram records from DOEHRS-HC data to "fence" SEG exposures and disregard SEG assignments outside the selected period.
Initial record construction from DOEHRS-IH and DOEHR-HC required extensive programming efforts to create relevant individual records. Researchers created individual exposure records utilizing SEG exposure assessment evaluations and personnel assignments with DOEHRS-IH data. Determination of an individual's health outcome utilized criteria to select an individual's first audiogram record and final audiogram record to calculate a threshold shift record. Calculations for threshold shifts utilized both unadjusted and OSHA age-adjusted frequency threshold values to identify if age was a confounding factor. This research only considered an individual eligible for the cohort if they demonstrated normal hearing on the selected first audiogram record in the research sample. Using a database joining process that excluded DOEHRS-IH exposures outside of established DOEHRS-HC audiogram dates, researchers then created a single database with a single record for each qualifying individual.
Following the creation of a combined single data source, researchers evaluated the relative risk of hearing loss by study exposure groups utilizing individual or aggregated frequency threshold values and shifts. The threshold values and shifts utilized by researchers included DoD, OSHA, and NIOSH indicators of hearing loss. Additionally, relative risk comparisons included analysis of data using both unadjusted and OSHA age-adjusted thresholds. Study exposure group data was then exported for statistical analysis utilizing Python to qualitatively and quantitatively describe data. Based on the descriptive analysis, researchers determined if there were statistical differences across 500, 1000, 2000, 3000, 4000, and 6000 center band individual frequencies and aggregated frequencies between study exposure groups.
Limitations: Researchers assumed that the current DoD exposure assessment strategy, a process that maximizes limited resources to manage prioritized risks, sufficiently captured the actual exposure hazards in an occupational setting. Researchers utilized selected baseline audiogram and final audiogram records from DOEHRS-HC data to "fence" SEG exposures and disregard SEG assignments outside the selected period.
Results / Conclusions:
Results: Researchers identified combinations of ototoxic substances appeared to have slight combined effects in almost all modeling, with the exception of the DoD STS model where effects were reduced to an RR<1. The maximum observed interaction observed across all PTA evaluation models was the Metal/Solvent/Continuous/Impulse exposure group (RR=1.57) utilizing the NIOSH material hearing impairment criteria, but without enough confidence in combined effects (CI 0.83-2.97). These broad ranges of confidence are primarily a product of the small size of the study population and the low rates of hearing loss development.
Conclusions:Researchers established there are likely hearing loss effects from exposure to ototoxic substances at the 1,000 and 2,000 Hz frequencies. Without detailed statistical analysis, it appears the NISOH STS evaluation method is the most sensitive in observing these changes through the inclusion of all frequencies from 500 to 6,000 Hz and a lack of averaging functions. This research identified that the adopted audiometric record processing methods closely matched published rates in literature and provided a simple method for analysis of large volumes of data.
Conclusions:Researchers established there are likely hearing loss effects from exposure to ototoxic substances at the 1,000 and 2,000 Hz frequencies. Without detailed statistical analysis, it appears the NISOH STS evaluation method is the most sensitive in observing these changes through the inclusion of all frequencies from 500 to 6,000 Hz and a lack of averaging functions. This research identified that the adopted audiometric record processing methods closely matched published rates in literature and provided a simple method for analysis of large volumes of data.
Primary Topic:
Big Data: Data Management & Interpretation
Secondary Topics:
Noise
Occupational and Environmental Epidemiology
Co-Authors
Please add your co-authors below. Co-authors are listed for professional courtesy and will not be communicated with regarding the decision notification or any on-site logistics, if accepted. Only the primary presenter listed is expected to attend and present the content on-site.
N. Cody Schaal, PhD, LCDR USN
Acknowledgements and References
List any additional people who worked on the project or provided guidance and support along with details on the role they played in the research. (Please include first name, last name, organization, city, state and country).
USAF School of Aerospace Medicine Epidemiology Consult Service Division, Wright Patterson AFB, Dayton, OH, - provided audiometric data