Interesting...
Here's a link to the full Schmidtmann et al. article (a Scottish/German research group; this does not appear to be a study sponsored by NIH, which provides access to research article abstracts through the PubMed search engine; NIH did not present conclusions in this case):
http://www.academia.edu/4471080/Intraoc ... _condition" target="_blank
Page 898 has suspect data analyses:
Correlation between blood pressure and intraocular pressure: 'small, but mostly non-significant correlations'
Correlations between data sets are either significant at the pre-selected (before the study even begins) alpha value or they aren't. There is no 'mostly' in correlation or any other statistical analysis. I would dismiss this article and would have rejected it as a reviewer for that brief 'analysis' alone. Studies that report specific
p values for individual statistical tests are poorly regarded. We simply report the significance threshold value and the result of the test (e.g.,
p < 0.05 or not). Experimental data sets cannot be more (as suggested misleadingly by reporting particularly low
p values) or less significant than others. That's not how statistics works. A difference is either significant or it's not.
They differentiate between high (trumpet; horn) and low (trombone; tuba) resistance brass instruments and present data suggesting that high resistance instruments produce higher intraocular pressures at 'middle' and 'high' frequencies, but don't indicate whether or not this is supported by a statistical analysis (though they do report mean +/- standard error), which is suspicious. They don't present all of the data for individual instruments, which is important. Also, what are the real (i.e.,
measurable) differences in resistance between trumpet and horn; trombone and tuba etc. and how does
measured resistance influence anatomy/physiology in real-time? I suspect that trumpet + horn (and trombone + tuba?) were combined to inflate sample sizes (
n). What's really strange is that they show individual means +/- standard error for the low resistance instruments (they are effectively the same but, which is for trombone and which is for tuba?) at middle/high frequencies even though they combined the sample according to the previous sentence.
There may very well be a connection between increased intraocular pressure and playing a brass instrument, but I would discount this research as the article has serious issues.
But... thanks for the reminder to call for my yearly eye exam appointment!