Although corticosteroid therapy is associated with the development of osteopenia, it is unclear whether the cause of osteopenia in inflammatory bowel disease (Crohn's disease and ulcerative colitis) is related to corticosteroid therapy or other disease-related variables. Patients with Crohn's disease (a diffuse gastrointestinal disease) could have greater osteopenia than patients with ulcerative colitis because of small bowel disease and secondary malabsorption of calcium and vitamin D. A cross-sectional analysis of consecutive patients with Crohn's disease and ulcerative colitis was undertaken. Bone density was determined by measurements of the L2-L4 spine, the total hip, and Ward's triangle using dual energy X-ray absorptiometry (DXA). A number of clinical parameters were recorded prior to bone density evaluation and analyzed by univariate and subsequently multivariate analysis to determine possible predictors of osteopenia. Of the 26 patients with Crohn's disease, diminished bone density (a Z score of at least -1) was found at the hip in 64% and at the spine in 44%; and of the 23 patients with ulcerative colitis diminished bone density was found at the hip in 43% and at the spine in 48%. Among all the variables tested, only corticosteroid use was a statistically significant predictor of diminished bone density (p = 0.025 for the spine and hip and p = 0.005 for Ward's triangle). Disease diagnosis (Crohn's disease compared with ulcerative colitis) did not predict or correlate with diminished bone density. No obvious associations were seen between the measurements of any serum hormones or biochemistries and bone density, although the patients using corticosteroids had lower serum calcium levels than the nonusers.(ABSTRACT TRUNCATED AT 250 WORDS)