The relationship between parity, lactation and the occurrence of hip fracture was investigated in a case-control study of white women. The cases were patients (n = 174) aged 45 years and over with a radiologically confirmed first hip fracture sampled from among admissions to 30 hospitals in New York and Philadelphia between September 1987 and July 1989. Controls (n = 174) were selected from general surgical and orthopedic services during the same time period and were frequency-matched to cases by age and hospital. Ever having a live birth was associated with reduced odds of hip fracture, controlling for age and hospital of recruitment (odds ratio = 0.65; 95% confidence interval = 0.41-1.04). When body mass index was also controlled, each birth was associated on average with a 9% reduction in the odds of hip fracture (odds ratio = 0.92 [0.78-1.08]), although the trend was not statistically significant. After adjusting for number of births, lactation was not associated with hip fracture (adjusted odds ratio for 12 months of lactation or less = 0.80 [0.42-1.55]; adjusted odds ratio for more than 12 months = 1.08 [0.45-2.60]).