To the Editor:
Hong et al1 claim to demonstrate “a considerable NIHL [noise-induced hearing loss] problem ... in firefighters.” For two important reasons, that claim cannot be considered valid. First, the paper does not control for age, which is the most important correlate of hearing loss in American adults. The authors focused on 4 and 6 kHz as “noise-sensitive frequencies,” but population-based audiometric surveys of American adults2,3 show that threshold elevations at those frequencies are common in men of all ages, increasing markedly for ages greater than 30 years. Hearing loss at those frequencies cannot be simply accepted as noise-induced. They could have compared their audiometric data with population-based data to determine whether for any age group, their firefighters (FFs) had worse hearing than men of the same age in the general population. Such a comparison would need to be “apples-to-apples” (eg, medians vs medians, better ear vs better ear).
Second, the sample of FFs in this study is flawed by selection bias of an unknown and probably severe degree. The article states a “response rate” of 59%, but this is the ratio of FFs completing the study to FFs who were recruited into the study. Another article by the same group4 describes their difficulties in recruiting subjects but never indicates how many FFs were approached or invited, so the true participation rate is unknown (and probably unknowable). Recruitment strategies included approaching FFs during shift changes, during occupational medicine clinic visits, during annual audiometric testing, at an annual union meeting, and at FF training sessions, and also included asking union stewards to solicit volunteers. To prevent selection bias, it is not enough to have a large sample; the sample must be representative, which requires a complete or random sample with a high participation rate. That participation rate should be the ratio of study subjects to FFs approached or contacted, not the ratio of completers to entrants. Firefighters with worse hearing would be expected to be more likely to volunteer, leading to falsely high estimates of hearing loss, not representative of FFs in general.
A secondary claim in the Hong study is that “firefighters who used HPDs [hearing protection devices] less were significantly more likely to have hearing loss.” This was a univariate statistical analysis (their Table 5) and did not account for other variables such as age, perceived levels of noise exposure, and barriers to the use of hearing protection. Another article by the same group5 based on the same sample of FFs found that when other variables were controlled, there was no relationship between audiometric thresholds and the use of HPDs.
Robert A. Dobie, MD
University of Texas Health Science Center
at San Antonio
1. Hong O, Chin DL, Samo DG. Hearing loss and use of hearing protection among career firefighters in the United States. J Occup Environ Med. 2013;55:960–965.
2. Hoffman HJ, Dobie RA, Ko C-W, Themann CL, Murphy WJ. Americans hear as well or better today compared with 40 years ago: hearing threshold levels in the unscreened adult population of the United States, 1959–1962 and 1999–2004. Ear Hear. 2010;31:725–734.
3. Hoffman HJ, Dobie RA, Ko C-W, Themann CL, Murphy WJ. Hearing threshold levels at age 70 years (65-74 years) in the unscreened older adult population of the United States. Ear Hear. 2012;33:437–440.
4. Hong O, Fiola LA, Feld J. Challenges and successes in recruiting firefighters for hearing loss prevention research. Workplace Health Saf. 2013;61:257–263.
5. Hong O, Chin DL, Ronis DL. Predictors of hearing protection behavior among firefighters in the United States. Int J Behav Med. 2013;20:121–130.