One out of five births in the United States in 1996 was by cesarean deliver
y. The cause for wide regional variation in cesarean delivery rates is not
well understood. The variation cannot be explained by simple demographic ri
sk factors and is likely due to differences in practice style. Adjustment o
f rates for case mix will lead to more valid comparisons of rates, which ma
y unveil opportunities for safe reduction of cesarean rates. Strategies for
reducing the number of cesarean births are described.