APPENDIX IV 
IX 
Appendix IV: Subject Database Methods 
In order to specifically identify the number of individuals involved in the 
research, and better understand what happened to them, the nearly 10,000 
pages from the Cutler Documents were read and analyzed with a particular 
focus on information about individual research subjects. Commission staff 
created a comprehensive database of individual subject information from 
these records. 
Creation of Subject Database 
A great deal of historical research was done to help interpret the Cutler Docu- 
ments. The Commission relied heavily on Modern Clinical Syphilology (1944) 
by John H. Stokes, Herman Beerman, and Norman R. Ingraham, which 
was considered the definitive text on syphilis at the time. In addition, the 
Commission consulted regularly with Dr. Jonathan Zenilman, Chief of the 
Infectious Diseases Division of the Johns Hopkins Bayview Medical Center. 
For data extraction and analysis purposes, the data sources were divided into 
two categories: research notebooks and additional archive documents. 
Research Notebooks 
The Cutler Documents include four research notebooks, two laboratory note- 
books (Notebooks 1 and 2), and two clinical notebooks (Notebooks 3 and 
4). The laboratory notebooks primarily contain laboratory test results. The 
clinical notebooks primarily contain research subject histories and clinical 
notes. All notebooks contain entries written in both English and Spanish. 
The primary data of interest included patient profile information (e.g., name, 
age, subject number, study population), inoculation data, treatment data, 
and information that independently raised ethical concerns (e.g., evidence of 
deceit on the part of the researchers or resistance on the part of the subjects). 
Based on a detailed reading of Notebook 1 and a review of Notebooks 2, 3, 
and 4, an initial coding scheme was developed to capture relevant informa- 
tion from the notebooks in an Excel database. Four coders then used the 
preliminary coding scheme to code a sample of about 10 pages from each of 
the four notebooks. Following each preliminary coding trial, the data were 
149 
