FIELDS W, X, and Y 



Columns 68; 69 and 



70; and 71 



Field M, N, or P: 

 coding in Col. 46, 

 48, or 51 



Field Y: 



Coding in 

 Col. 71 



1 1 



2 2 



3 3 



4 4 



5 5 



6 6 



7 7 



8 8 



9 9 



The necessity for special interpretation of the code line is demonstrated by the following two 

 examples, designated as A and B: 



A. 



Field N 



100 mg 

 (Symbol 27) 



Field T 



Death 

 (Symbol 11) 



Field X 

 Criterion 20 



Field Y 



85 mg 

 (Symbol 27) 



Death 

 (Symbol 11) 



Criterion 21 



In Interpreting these, if the entry in Field X were ignored, Example A would be read as "the test 

 compound causes death at a dose between 81 and 243 mg (100 mg can not be coded exactly in the 

 dosage fields- -only ranges are represented by code symbols of those fields), which means that the 

 compound is not very toxic (a rating of 3 in Field Y).'" If the coding of Field X in Example B were 

 ignored, that line would be interpreted in the customary way, i. e. , similarly to the interpretation just 

 made for Example A: "the test compound causes death at 81 to 243 mg which means that it is quite 

 toxic (a rating of 7 in Field Y). " However, as has been pointed out, Criterion 21 lends a wholly 

 different interpretation to the line and, therefore, whenever retrieval is made from the CBCC file of 

 coded data and sorting is by the evaluation ratings in Field Y, Field X must be consulted for any IBM 

 cards coded with Symbol 21 and the cards interpreted accordingly. Thus, Example B above, interpreted 

 correctly, reads, "the test compound causes death, but at a dose higher than the dose administered 

 (it may or may not be larger than the highest dose of the range coded in Field N, as explained in the 

 following paragraph, and the compound is rated as being relatively safe as a therapeutic agent (a rating 

 of 7 in Field Y), as indicated by Criterion 21 in Field X. " 



The examples given here illustrate the fact that the procedure of coding doses (which means 

 coding, by a given symbol, a range of doses) in Fields M and N, rather than literally recording and 

 punching the dose itself has the disadvantage that the Biology Code Sheet must always be consulted 

 for the actual value, which is coded only as being somewhere within a specified range of values. In 

 the examples above, 85 mg did not cause the response while 100 mg did cause it; the coding for this 

 in Field N could just as well mean that 200 mg did not cause the response while 225 mg did cause it, 

 since the symbol coded in Field N represents a range of 81 through 243 mg; the Biology Code Sheet 

 must be consulted to learn the actual maximum tolerated dose or the threshold dose of this particular 

 compound, since the coder always writes the actual dose on the Code Sheet in addition to coding it. 

 The procedure of coding doses as is done in Fields M and N is based on the attitude that the literal 

 recording and IBM punching of an exact dose value, even in the case of maximum tolerated dose or 

 threshold dose, is less important than being able to sort out from the coded files data on all test 

 compounds causing responses at dosages below, above, or between certain levels, which purpose is 

 accomplished with adequate efficiency and with the minimum use of space on the IBM punched card 

 (a single column in each of Fields M, N, and P) by coding the dose range . 



A maximum tolerated dose level, like a minimum effective dose level (threshold level), may be 

 determined for each individual of a group of individuals, demonstrating the variation between individuals 

 for tolerating the test compound. (See Division 14, paragraph 8. ) In this case, Criterion 51, 52, or 

 53 is used in Field X and the rating for Field Y is found by plotting on the Grid the maximum tolerated 

 doses against the corresponding percentages of organisms responding at those dose levels. 



196 - 



