In this paper we present the results of an intelligent analysis of osteoporosis database gathered during a longitudinal study in which a random sample of 100 women who had passed precautionary examinations for detection of osteoporosis over five years and were not referred for a definite medical diagnosis was pulled from the records. The intelligent data analysis using advanced methods for decision tree construction was used in order to try to find the main factors that can reduce the risk for development of osteoporosis in women. Most of the extracted knowledge confirmed known medical criteria that put women to risk for osteoporosis, however some new interesting patterns have also been shown.