>I have done this experiment hundreds of times and I occasionally
        >(1/20 times) get a spurious value. Otherwise, I consistently get
        >the same result within 0.2 deg C. Same results with my kids at home.

Hundreds of times?!

        >Are you using the same technique with an ear tug each time?

If some people do get consistent, reproducible results with tympanics, but
most people don't then it's only a reliable method for some.

        >These numbers are meaningless because the study design is flawed.
        >Asking the question "How often does ear temperature (or oral or
        >axillary) 'detect' a rectal fever?" is equivalent to aking "How
        >often does a toenail length of 1 cm 'detect' a fingernail length
        >of 1 cm?" The two are somewhat related but should not be used as
        >substitutes for each other. Rectal temperature is inaccurate and
        >is a poor diagnostic test, so why use it? Just because previous
        >authors used it for their studies of SBI, we should not perpetuate
        >the flawed logic...

I haven't read the study in question, however isn't it  most important to
decide on a standard thing being measured, then choose a method that
measures it accurately and reproducibly.  True "core temperature" has never
been the standard in the ambulatory setting.  True: Rectal temperature does
not mirror changing core temperature at any given time, therefore is not an
"accurate" measure of core temperature.  The trends are similar, allowing
for the time delay.  Rectal temp is, however, an accurate measure of rectal
temp.  It just depends what your standard is.

Isn't the primary use of measuring temperature via Swan-Ganz catheter to
estimate cardiac output by thermodilution?

In summary, who cares?
As it has been previously stated:
1. tympanic temperature probes don't fit into young infants' ear canals.
2. outside of the newborn period, it matters more whether there is fever or
not, than what the actual number is.

For more information, send mail to [log in to unmask] with the message: info PED-EM-L
The URL for the PED-EM-L Web Page is: