RT Journal
A1 Wrone DA, Sober AJ
T1 RAndomized trials and scientific methods
JF Archives of Dermatology
JO Archives of Dermatology
YR 1999
FD May 1
VO 135
IS 5
SP 602
OP 602
DO 10-1001/pubs.Arch Dermatol.-ISSN-0003-987x-135-5-dlt0599
UL http://dx.doi.org/10-1001/pubs.Arch Dermatol.-ISSN-0003-987x-135-5-dlt0599
AB 
Scientific methods do not necessarily require a randomized trial. We can argue that sometimes a randomized trial is harmful. When a trial is performed, a test of statistical significance is done, such as a t test. The standard in clinical medicine for a statistically significant result is a P value less than or equal to .05. Accepting a P value of .05 means that there is a 1 in 20 chance that a therapy without merit could be shown to be significant. One way to decrease the likelihood of getting a spurious result is to limit trials to topics that have a potentially medically explainable result. Is there even 1 scientifically defensible hypothesis for the therapeutic basis of homeopathy? Have any basic science experiments ever shown that the chemically pure water used in homeopathy is different from normal water?