Dietary data from a prospective study were used to relate factors influencing calcium balance (estimates of dietary calcium intake, protein intake from nondairy animal sources (meat, fish, and eggs), and coffee consumption) to the incidence of hip fracture. During the years 1977-1983, women and men born between 1925 and 1940 and living in one of three Norwegian counties were invited to a cardiovascular screening that included a dietary survey. The attendance rate at screening was 91.1%, and 90.7% of these persons (19,752 women and 20,035 men) filled in and returned a semiquantitative dietary questionnaire. This cohort was followed for an average of 11.4 years (range, 0.01-13.8 years) with respect to hip fracture, defined as cervical or trochanteric fracture. During follow-up, 213 hip fractures were identified, excluding fractures associated with high-energy trauma and metastatic bone disease. There was no clear association between calcium intake or nondairy animal protein intake and hip fracture in this cohort. However, an elevated risk of fracture was found in women with a high intake of protein from nondairy animal sources in the presence of low calcium intake (relative risk = 1.96 (95% confidence interval 1.09-3.56) for the highest quarter of nondairy protein intake and the lowest quarter of calcium intake vs. the three lower quarters of protein intake and the three higher quarters of calcium intake). Women who drank nine or more cups of coffee per day also had an increased risk of fracture, while there was no association between coffee consumption and hip fracture in men. Although these findings do not necessarily imply causal relations, they suggest the presence of risk factors for hip fracture that act through a negative calcium balance in this population.