Noise-Induced Hearing Loss

To the Editor—

Niskar et al’s ¹ estimate of 5.2 million US children with noise-induced threshold shifts (NITS) was unexpectedly high for a condition known primarily as an occupational hazard. It was particularly surprising that 1 out of 12 of the youngest children, aged 6–11, already had NITS. Some other study findings were also remarkable. NITS generally occurs in both ears; yet 85% of the children with NITS were only affected in 1 ear. NITS typically occurs at 4 kHz, yet in this study only 24% of NITS involved 4 kHz, while >3 times as many cases (76%) involved 6 kHz. These unexpected findings (ie, the large prevalence estimate, the predominance of unilateral NITS, and the concentration of threshold shifts at 6 kHz) can be largely explained by a methodologic flaw.

Noise-induced threshold shift, as the term suggests, is a change in hearing level threshold (HLT) attributable to noise. Correspondingly, the standard operational definition entails a difference in HLT between baseline and follow-up audiograms covering a period with documented noise exposure. ¹ Niskar et al were unable to use this definition because their database included neither baseline audiograms nor noise exposure data. Given these limitations, the authors substituted the presence of a “notch” or “dip” on the audiogram at 3, 4, or 6 kHz as a proxy measure.

Serious questions have previously been raised about this choice. Sataloff ³ observed that a notch is not sufficient evidence to diagnose NITS because numerous other medical conditions also produce notches. Mostafapour et al ⁴ found that presence of a notch among students did not correlate with any source of noise exposure including personal listening devices, home stereos, or firearms. Another study found no association …