Table 1
One-way repeated-measures ANOVA indicated that different levels of PPE use did not significantly alter BKB sentence results in office or ED settings (p = 0.26 and p = 0.58 respectively) but showed a trend in ITU (p = 0.06). The statistical assumption of sphericity in the ITU setting was marginally violated; an effect perhaps due to the small sample size. If sphericity was assumed, the one-way repeated measured ANOVA test was statistically significant (F(2,8) = 6.64, p = 0.02, ηp2 = 0.73). Assumption of sphericity was not violated in the theatre setting results (χ2 (2) = 3.13, p = 0.21) and different levels of PPE use significantly altered BKB sentence results (F(2,8) = 17.16, p = 0.001, ηp2 = 0.81).
A Wilcoxon Signed-Rank test indicated that BKB sentence scores were significantly lower for subjects wearing PPE (median score 58) compared to those without PPE (median score 92) in a theatre simulated environment (Z = -2.02, p = 0.04). Increasing voice volume whilst wearing PPE significantly increased BKB sentence scores (median score 86) compared to normal speech volume when wearing PPE (median score 58; Z = 2.03, p = 0.04). There was no significant difference in BKB scores when wearing no PPE (median score 92) compared to when raising voice volume whilst wearing PPE (median score 86) (Z = -0.68, p = 0.50).
Our secondary outcome measure was mean change in volume of voice when wearing PPE. The increase in background noise rose by 25db from 45 dB (simulated office) to 70dB (simulated theatre). Our researcher elevated their natural voice by 13-20dB without PPE in response to simulated increasing sound levels which correlates with existing studies showing a natural shift to maintain signal-to-noise ratio in human speech18.
Mean voice volumes across all simulations tended to increase with PPE wearing and increase again with PPE wearing and raised voice. The mean SNR achieved are shown in Table 2 below: