HOUSING
 GROUP             NO. OF RECORDS NOT SIGNED OR INITIALED

  A                6        (wks 80, 92, 84, 88, 93, 94)
  B                6        (wks 80, 84, 88, 92, 93, 96)
  C                5        (wks 80, 84, 88, 92, 96)
  D                7        (wks 79, 80, 84, 88, 92, 94, 96)
  E                6        (wks 76, 80, 84, 88, 92, 96)
  F                9        (wks 76, 77, 80, 81, 84, 88, 92, 93, 96)
  G                5        (wks 80, 84, 88, 92, 96)
  H               11       (wks 1,  68, 76, 80, 81, 83, 84, 88, 92, 95, 96)
  J                 6        (wks 76, 80, 84, 88, 92, 96)
  K                8        (wks 71, 76, 80, 84, 88, 92, 93, 96)
  L                 5        (wks 76, 80, 84, 92, 96)
  M                5        (wks 76, 80, 84, 92, 96)
       TOTAL      79 records not signed or initialed

In addition to the lack of signatures, it was noted that many of the records were not originals, but appeared to be xerox copies of the originals.  Surprisingly, some of the xerox copies had "original" initials.  It was obvious that the initials had been placed on these sheets sometime after the sheets had been filled out, and after they had been copied.

Some examples of discrepancies of this type are as follows:

1.)  In housing group A, 26 of the 39 observation records were xerox copies with original initials.

2.)  In housing group B, 27 of the 43 observation records were xerox copies with original initials.

3.)  The record for week 76 of housing group A was a xerox copy but the date, initials, an week are all original.

4.)  For week 96 of housing group K, both an original and xerox copy of the observation record are present.  The xerox copy has original initials and a "B" entered in the "tissue masses and lesions" column.  There is also an entry in the "notes" column for rat #K25CF.  The original record also has a "B" entered in the "notes" column.  All of the above entries had obviously been made sometime after their original record had been completed and the xerox copy made.

5.)  A record dated 4-27-73 for housing group M does not have the date entered.  The observations were made for this animal on sheets covering weeks 92, 96, 100 and 104, indicating that the animal was dead.  The record for week 108, however, shows that the animal is alive, with motor activity, appetite, and thirst. The record for week 112 again shows that the animal is dead.

In addition to the discrepancy noted above, there is also an obvious error in the dating of these records; the observation sheet for week 92 is dated June 13, l973, and the observation sheet for week 88 is dated July 16, l973.

Ante-mortem observations were also made on other types of records.  A volume entitled "Tissue Masses and Deaths" (exhibit #65) has a record of the date that each animal died during the study.  The deaths are recorded in two different ways in this volume.  One record has a chronological list of deaths, and another record has a list of deaths organized by housing group.  This volume also has a "palpation Record", which describes each tissue mass, and lists the date that it was initially detected.

We then pointed out that the chronological record of deaths also contained the date fixed "in toto" and the date autopsied, neither of which were found on the log organized by housing group.  If the data was transferred from one record to the other, we wondered where the "date fixed in toto" and "date autopsied" came from.  We posed this question to Searle personnel and we were finally told on 4-29-77 that the "fixed in toto" and "date autopsied" columns on the chronological record of deaths was considered to be primary data, although it was prepared simultaneously with the pathology sheets, which contained the same data.

Another source of ante-mortem data was the ophthalmoscopic examination record.  Dr. Youkilis performed the eye examinations at periodic intervals and recorded his findings on the ophthalmoscopic record sheets.  These eye exam sheets were eventually attached to the pathology records, and the results of the eye exam was incorporated into the Clinical History of the animal on the pathology sheet.

During our review of the pathology records we noted that there were no eye exam sheets for 15 animals, all of which had been included in the individual Ophthalmoscopic Findings in the submission to FDA (Vol. 1, table 1, pages 122-133).  All of these missing eye exam sheets were for animals that had died during the study.  On 6-30-77 we interviewed Donna Helms, who told us that she would make an attempt to find the missing records.  We advised her that Dr. R. Stejskal had told us on 6-29-77 that not all of the eye exam sheets were attached to the pathology records, and that there may be another file of eye exam sheets somewhere.

On 7-1-77 Donna Helms reported that she had found the missing eye exam sheets in the K-1 File Room in K-Building.  After reviewing these records we found that a few discrepancies still existed.  They are as follows:

1.)  It appears that animal J3CM on page 125, Vol. l of the submission to FDA is in error.  We could find no records to substantiate the listed corneal scar and haziness for this animals.  Also, the observation records indicate that J3CM died at week 78.  It appears that the correct animal on page 125 should be J2CM and not J3CM.

2.)  We found eye exam sheets for H26MF and J29CM, yet the findings were not reported in the submission to FDA.

3.)  There seems to be a discrepancy between G16CM and G12CM. The pathology sheets for both of these animals report the identical ophthalmoscopic finding, yet there is no eye exam sheet for G12CM, and only the finding for G16CM was reported in the submission to FDA (Vol. 1, p. 125).

When an animal spilled an excessive amount of food, this was noted on the observation records by means of an asterisk in the "appetite/thirst" column.  The asterisk was also used to denote food spillage on the Teletype sheets for body/feeder weight data.  The amount of food spillage was not quantitatively determined by the technicians assigned to observe, feed and weigh the animals, but we were told that they made an effort to return spilled food to the food cups whenever possible.  WE were also told that food consumption data for those erats marked with an asterisk on the body/feeder weight sheets was not used in Searle's statistical analysis of the data.