Table 3. Multivariable Regression Models of Factors Associated With Speech-in-Noise Performance Measured With the Words-in-Noise (WIN) Test, and With Progression of Audiometrically Defined Hearing Loss.
| Factor | β (95% CI) | P value |
|---|---|---|
| Speech-in-noise performance (WIN) | ||
| Hypercholesterolemia | 0.88 (0.08 to 1.69) | .03 |
| Education | ||
| Not college graduate | 1 [Reference] | NA |
| College graduate | −1.43 (−2.29 to −0.56) | .002 |
| Postgraduate | −1.22 (−2.18 to −0.26) | .01 |
| Audiometrically measured hearinga | 0.07 (0.06 to 0.09) | <.001 |
| Temporal progression of audiometrically measured hearinga | ||
| Hypercholesterolemia | −4.38 (−7.42 to −1.34) | .01 |
| Age at TPS-1 (Assessment 1) | 0.33 (0.15 to 0.50) | <.001 |
| Time since baseline (TPS-1, Assessment 1) | 0.12 (0.02 to 0.23) | .02 |
| Hypertension | 1.96 (−1.15 to 5.07) | .22 |
Abbreviations: HL, hearing loss; NA, not applicable; TPS-1, The Platinum Study, initial baseline assessment; TPS-2, The Platinum Study, follow-up assessment.
a
Hearing variable defined as the average of both ears across the higher frequencies measured (ie, pure-tone average of 4, 6, 8, 10, and 12 kHz).