A problem-oriented approach to journal selection for hospital libraries.

This paper describes a problem-oriented approach to journal selection (PAJS), including general methodology, theoretical terms, and a brief description of results when the system was applied in three different hospitals. The PAJS system relates the objective information which the MEDLARS data base offers about the universe of biomedical literature to objective, problem-oriented information supplied by the hospital's medical records. The results were manipulated quantitatively to determine (1) the relevance of various journals to each of the hospital's defined significant information problems and (2) the overall utility of each journal to the institution as a whole. The utility information was plotted on a graph to identify the collection of journal titles which would be most useful to the given hospital. Attempts made to verify certain aspects of the whole process are also described. The results suggest that the methodology is generally able to provide an effective library response. The system optimizes resources vis-a-vis information and can be used for both budget allocation and justification. It offers an algorithm to which operations researchers can apply any one of a variety of mathematical programming methods. Although originally intended for librarians in the community hospital environment, the PAJS system is generalizable and has application potential in a variety of special library settings.