Heres an interesting abstract...
Jonathan B. Waugh, PhD; Dale W. Callahan, PhD; Wesley M. Granger, PhD; George A. Mathew, MS. University of Alabama at Birmingham, Birmingham, AL.
BACKGROUND: There are many brands of stethoscopes currently marketed. Little data exists to guide clinicians on the differences between models that often vary greatly in price. We tested eight high-quality stethoscopes to determine if there was a difference in average frequency response.
METHODS: Eight different models (Littmann Master Cardiology, Advanced Diagnostics Corp. Adscope 601, Heine Optotechnik Gamma20, Allen Medical Instruments Gemini 23, Welch Allyn Harvey DLX and Elite, Philips Rappaport-Sprague, Doctors Research Group Puretone Cardiology) of adult cardiology stethoscopes (8 had diaphragm chest pieces, only 7 had bell chest pieces) were measured using a pure tone (sine wave) sweep input from 50-3000 Hertz. Four copies of each model were tested and the data averaged for comparison by ANOVA. The average relative change in decibels (dB) (difference of stethoscope output signal from reference input signal) for each model over the measured frequency range was compared.
RESULTS: The ANOVA indicated a difference among models for the bell and diaphragm chest pieces (p<0.001). The stethoscope bell chest piece with the least negative average dB value (closer to zero dB the better) was the Welch Allyn DLX (post hoc analysis by Tukey-Kramer Test showed this model was significantly different [p<0.05] from all other models). The dB means for the bell measurements ranged from 7.80 to 14.20. The diaphragm chest piece with the least negative value was the Philips Rappaport-Sprague (post hoc analysis showed this model was significantly different from all other models except the Allen Medical Instruments Gemini 23). The dB means for the diaphragm measurements ranged from 11.10 to 13.73.
DISCUSSION: The results show that the best bell and diaphragm values did not occur with the same model of stethoscope. There was a greater range of values for the bell chest pieces than with the diaphragms chest pieces.
CONCLUSION: Bench testing using computerized signal processing methods shows statistical differences between the models tested. These results need to be compared to testing done with experienced human auscultators to see if these statistical differences have clinical significance.