Most of the information retrieval models for documents are intended for SGML-like structured documents. In the context of medical informatics, the patient record document needs a looser structuration process as an a priori structure can hardly be defined. Thus we first propose an authoring tool that allows us to annotate embedded information, i.e. to give them a context, with qualifiers that are stored in a thesaurus rather than in SGML-like DTD. The retrieval process in the patient records collection takes into account the flexibility of the qualifying process while reformulating the queries.