Dear all,

I have a set of data with approximately 100 patiens treated for cancer in two groups in regards to increase in luteinising hormone. I have som problems understanding how the p-value can go below 0.05 when CI is 0.97 - 11.15.I know the p-value is sensitive to group size etc but I still cannot understand this result (pic below). Any ideas?

Best regards,
John Tapper
. stir elev_lh_3
failure _d: ttr
analysis time _t: (ttr_date-origin)/30.4
origin: time op_date
exit on or before: time op_date+(365*5)
id: patno
note: Exposed <-> elev_lh_3==1 and Unexposed <-> elev_lh_3==0
Luteinizing Hormone
increased 50+%
Exposed Unexposed Total
Failure 8 6 14
Time 971.2171 2322.796 3294.013
Incidence rate .0082371 .0025831 .0042501
Point estimate [95% Conf. Interval]
Inc. rate diff. .005654 -.0004166 .0117246
Inc. rate ratio 3.188846 .9702801 11.15018 (exact)
Attr. frac. ex. .6864069 -.0306303 .9103153 (exact)
Attr. frac. pop .3922325
(midp) Pr(k>=8) = 0.0179 (exact)
(midp) 2*Pr(k>=8) = 0.0357 (exact)