Evaluating a smartphone digits-in-noise test as part of the audiometric test battery.
Jenni-Mari Potgieter, De Wet Swanepoel, Cas Smits
PMID 29781704WHAT IT FOUND
The smartphone digits-in-noise test matched pure-tone and speech-in-quiet scores well and could add information about hearing in noise.
It does not replace the standard audiometric test battery.
Key findings
01The smartphone DIN speech reception threshold was significantly correlated with best ear four-frequency pure-tone average (rs = 0.81; n = 120; p < 0.001) and with maximum speech recognition monaural performance score (r = 0.72; n = 111; p < 0.001).
02Using a cut-off of -9.5 dB SNR, best ear DIN SRT predicted abnormal best ear four-frequency pure-tone average with sensitivity 0.883, specificity 0.878 and AUROC 0.941.
03In the hearing aid subset, aided DIN SRT mean was -7.2 (2.1 SD; range -3.2 to -9.4) and unaided DIN SRT mean was -6.4 (2.6 SD; range -2 to -9.4), with a reported mean difference of 0.8 (1.5 SD).
STILL TO COME
How it was doneWhat they foundWhat it means for SLPs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
Participants were recruited by convenience sampling from selected audiometric practices, so they may not represent all adults who need hearing assessment. The study did not have 50% speech recognition scores for all subjects, and sites did not use the same methods for determining speech recognition scores. The aided and unaided hearing aid comparison included only nine subjects, so it cannot show how most hearing aid users benefit. Some reported analysis counts, n = 111 and n = 120, do not match the stated total of 109 subjects, and Table 1 group counts do not clearly match the stated normal-hearing and hearing-loss totals.
The easy way to misread this
Do not read this as proof that the smartphone test replaces pure-tone audiometry or speech recognition in quiet. The hearing aid comparison involved nine people and showed wide individual variation, so the reported mean change of 0.8 should not guide fitting decisions.