Table 2.

Characteristics of participants, methods of testing and results of studies of use of whisper test in adults

Criterion	Swan and Browning15	Macphee et al17	Eekhof et al16	Uhlmann et al14
Participants and setting
Number of participants (No of ears)	101 (202)	62 (124)	62 (124)	31 (62) (non-demented group only)
Age (years)	Mean 57, range 17-89	Mean 81, range 66-96	≥55	Mean 76, SD 5.8
Setting	Audiology Clinic	Geriatric Assessment Unit	ENT Outpatient Department	Otolaryngology Clinic
Whispered voice technique
Audiometry cut-off for hearing impairment	30 dB	30 dB	30 dB	40 dB
Distance	2 feet (arm's length)	6 inches and 2 feet	"Standard method" *	Began at 6 inches and move away at 6 inch increments
Masking	Occlusion and rubbing of external auditory canal	Occlusion and rubbing of external auditory canal	"Standard method" *	Pushing tragus over external auditory canal
Letters/numbers/words	Triplets of letters and numbers, eg 5B6	3 random numbers	"Standard method" *	Spondee words †
Number of repeats	1	1 or 2	"Standard method" *	At each distance until unable to hear
Threshold for hearing impairment	<50% correct	≤50% correct	≥2 combinations incorrect	<50% correct
Results of testing
Prevalence of hearing impairment (%)	43	61	59	26
Sensitivity (%; 95% CI)	100 (96-100)	100 (95-100)	90 (82-95)	90 ‡
Specificity (%; 95% CI)	87 (80-92)	84 (71-92)	80 (68-89)	70 ‡
Likelihood ratios:
Positive	7.7	6.4	4.6	Not done (insufficient data)
Negative	0	0	0.12	Not done (insufficient data)

^*"The whispered voice test was performed by a standard method (slightly modified)" -reference to Swan, 1985

^†Spondee words are two syllable words with equal stress on both syllables, for example "baseball" or "hotdog."

^‡Specificities were given for two sensitivities (80% and 90%)—unable to calculate confidence interval or likelihood ratio.