Perceived and measured hearing ability in construction laborers and farmers

Madeleine J. Kerr, Marjorie McCullagh, Kay Savik, Leah A. Dvorak

Research output: Contribution to journalArticlepeer-review

29 Scopus citations


Background: Construction and farming are characterized by small independently operated enterprises; few are included in hearing loss prevention programs. This study describes perceived and measured hearing ability among construction laborers and farmers in order to further the Healthy People 2010 goal to reduce noise-induced hearing loss (NIHL). Methods: Construction laborers (n = 147) and farmers (n = 150) had screening audiograms at their workplace and a farm show, respectively. The relationship of the audiograms with a one-item measure of perceived hearing ability was explored using non-parametric statistics. Results: At the 4,000 Hz frequency, indicative of NIHL, a high percentage of both groups exhibited hearing loss greater than 25 dB: laborers 53%, farmers 67%. Perceived hearing ability was significantly worse (P < 0.05) for workers with a measured hearing loss. The sensitivities of perceived hearing loss compared to audiometry ranged from 0.30 to 0.72 suggesting that perceived hearing loss was not a good indicator of actual hearing loss in this sample. Conclusions: The poor relationship between perceived and actual hearing loss in this study suggests that best practice requires inclusion of audiometric screening as part of a complete hearing loss prevention program. Future development of an improved instrument to measure perceived hearing ability may be useful in assessing this dimension of hearing.

Original languageEnglish (US)
Pages (from-to)431-437
Number of pages7
JournalAmerican Journal of Industrial Medicine
Issue number4
StatePublished - Oct 1 2003


  • Agriculture
  • Audiometric evaluation
  • Construction
  • Noise-induced hearing loss


Dive into the research topics of 'Perceived and measured hearing ability in construction laborers and farmers'. Together they form a unique fingerprint.

Cite this