Evidence Based Calculator
References & support Lectures:
This program is released as Freeware and is a work in progress. As such some errors and bugs may exist in the code. The author will not take responsibility for any damages either direct or indirect resulting from its use. It is intended as a learning tool and not for clinical application. It is to be applied in an Evidence Based Morning Report and used by the house staff of a community teaching hospital. All results must be verified and checked by the user.
What follows is the only documentation offered for this software.
This calculator is designed to assist a medical practitioner in Evidence Based Medicine. It is based on a series of articles from the Journal of the American Medical Association and a multitude of Internet sources. (References are available upon request.) Kindly note my two lectures on Evidence Based Medicine above for further clarification (I may record these lectures someday again.)
This is a 2 by 2 table. The Sample Size is automatically calculated from the data you enter into a, b, c and d. The control group is on the top row and the Control Event Rate = a/(a+b). The Target Disorder is a dichotomous outcome which is either present or absent. For example Death is present (Dead) or absent (Alive). The lower row is the experimental group (eg. the one that receives the new therapy) and the Experimental Event Rate = c/(c+d). A positive study shows a better outcome in the experimental group than in the control group and the CER > EER.
The check box selects calculations only for Rx or therapy or prevention (and similarly for Dx and Harm). Right Clicking on CER, EER, RR, RRR, ARR, NNT and NNH show the formulae for each. (I haven't finished this for everything yet.) After all the data is entered in the 2x2 boxes hit Calculate (or Clear to reenter) and the %'s will be shown. Clicking (in all fields) displays the Worksheet For each topic which can be completed saved and printed. Be careful not to overwrite the original files which are stored in the application directory with no ".rtf" extension. This button: calls up ( and hides) an example screen:
The button only works under the Dx group and offers an alternate way of entering the data:
Note my second lecture for more details on the reasons. A Receive Operator Curve can be generated from the data and multiple 2x2 tables can be generated as well. This is exemplified by the PIOPED study.
I have not completed the Harm section of the program as of yet.
I may port this program to the Palm or Windows CE operating systems in the future.