RT Journal
A1 Beghi C
T1 STress tests
JF JAMA
JO JAMA
YR 1983
FD May 6
VO 249
IS 17
SP 2331
OP 2331
DO 10.1001/jama.1983.03330410029021
UL http://dx.doi.org/10.1001/jama.1983.03330410029021
AB To the Editor.— 
    I have been somewhat disturbed by the article by Philbrick and co-workers entitled "The Limited Spectrum of Patients Studied in Exercise Test Research," (1982; 248:2467). While I am grateful to the authors for pointing out the influence of the work-up bias on the calculations of sensitivity and specificity of a test, I am left with the impression that the use of a test less than perfect is permissible in situations that differ from those in which the test characteristics had originally been defined.To continue with the example of the exercise ECG, I believe that it would be incorrect to interpret test results obtained on patients treated with digitalis, β-blockers, or nitrates according to test characteristics derived from a patient population in which no drugs had been used. I think that, rather than caution the users of the test under these circumstances, this practice should be discouraged