DKP levels for the feeding study were multiples of 100, 200 and 400 times the estimated human dose.  The levels in g DKP/kg body weight/day were 0,0.75, 1.5 and 3.0 for the control, low, medium and high treatment groups, respectively.  The doses were mixed in the diet as described in Calculating Diet Concentration and Blending of Treatment Mixtures.

Individual body weights were recorded weekly for the first four weeks, once every two weeks for the next eight weeks and once every four weeks thereafter.  The amount of food consumed was measured every week.  An automated weighing system was employed consisting of an Intec balance and a Teletype machine.  The Teletype produces a typewritten sheet and a machine-readable punched paper tape.  All the typewritten sheets for the study were available.  Xerox copies of these sheets were taken to the Division of Mathematics and technical Operations Staff of the form and calculated by a computer program designed by Dennis Wilson, Division of Mathematics.

In designing the computer program it was necessary to make certain assumptions on the handling of the data.  One assumption concerned missing data, e.g. the empty feed cups weights were missing for the "D" housing group at the 12th week.  Dr. George Clay, Group Leader, CNS Pharmacology, Searle and scientific co-ordinator for the FDA team, was unable to determine whether these animals were omitted from the food consumption calculations for that week, or whether the data for these animals from the 11th and 13th weeks were averaged and the average substituted for the missing data.  Employees of Searle's Math-Stat Department who had worked on the program for this experiment are no longer with the company.  Dr. Clay calculated a few of the figures from the 11th and 13th weeks and stated that it appeared that the data had been averaged.  For the FDA recalculation it was chosen to omit the animals with the missing weights from the calculations.  In several instances (For example, C group males, mid and high levels for the 13th week; A group males, high level for the 99th week) the dietary concentration shown on the weight sheets did not agree with the concentration listed for that level in the other housing groups.  Dr. Clay assured us that all the animals of the same sex in a given experimental group received the same dose for the same week on the experiment.  He also assured us that the Searle computer program did not pick up the doses from the weight sheets.  In the FDA program, the dietary concentrations were taken from the diet calculation sheets (Exhibit #34). Certain animals on the raw data sheets were marked with an asterisk.  Dr. Clay explained that the asterisk indicated spillage and such animals were omitted from the food consumption calculations.  This practice was followed in the FD computer program.  In calculating the food consumption (g food eaten/day/ kg body weight) and the dosage (mg test compound/day/kg body weight), the body weight used was the weight at the end of the period under consideration, i.e. the current weight.

In addition to the calculations which were included in the Searle submission, the FDA program included calculation of the actual amount of food ingested, i.e., the total amount of diet ingested minus the test compound, and of the food efficiency (g weight gained/100 g actual food eaten).  The food efficiency was calculated in order to determine whether the volume of DKP in the diet (which exceeded 7% of the diet for the high dose males at intervals during the study) was contributing to the body weight depression seen with DKP>  This explanation of the body weight depression was discussed by Dr. John H. Rust, a Searle consultant, in a memo dated April 5, l976 to Dr. R. McConnell; in a memo to the file dated September 30, l974 by Dr. McConnell; and in a memo to Dr. K. S. Rao dated August 29, l974 by Dr. G. L. Schoenhard.  (Exhibits #36-38).

The average body  weights and weight gain (% change/week) from the FDA analysis of the Searle raw data are presented in Table 1 (Exhibit #39) which corresponds to Table 3 of the Searle submission.