Record linkage, for compiling same-person records from various source files
, can improve the feasibility of epidemiological research using population-
based studies. The question is comply with the European legislation on data
privacy and data security. For example, a computerized record hash coding
and linkage procedure is described to link medical information within the f
ramework of epidemiological follow-up, Before their extraction, files are r
endered anonymous using a one-way hash coding based on the standard hash al
gorithm (SHA) function. Once rendered anonymous using using the software AN
ONYMAT, the linkage of patient information can be accomplished by means of
a mixture model, taking into account several identification variables. An a
pplication of this anonymous record linkage procedure was carried out in or
der to link medical files on cancer, from 3 hospitals of the French Rhone-A
lpes region, This application stresses how the use of the ANONYMAT software
allows compliance with the legislation on data confidentiality without ent
ailing problems oil data availability.