In Response

Although the 2 studies cited by Mr. Simonson1,2 in his letter3 are similar to our recent report4 in terms of data sources and methods, there are several notable differences. Whereas our study was designed to examine the incidence of and risk factors for obstetric anesthesia complications in a well-defined population, the studies by Simonson et al.1 and Needleman and Minnick2 were specifically designed to compare the quality of obstetric care between hospitals with different anesthesia staffing patterns, in particular, hospitals staffed by nurse anesthetists versus hospitals staffed by anesthesiologists. As a result, Simonson et al. used the hierarchical modeling technique to account for the multilevel feature of their data to address their specific aim. We did not have any hospital-level data. In our article,4, p 1180 we explicitly discussed this data limitation and its potential influence on the interpretation of our finding regarding the increased risk of obstetric anesthesia complications for women living in rural areas. The 2 studies1,2 also differed from ours in case definitions and study populations. First, spinal headache, accounting for about one-third of all obstetric anesthesia complications in our study, was not included in either the Simonson et al.1 study or the Needleman-Minnick study.2 Second, whereas our study included all deliveries occurring in New York hospitals during 2002 to 2005, the Simonson et al. study included only a select sample of hospitals in Washington State. Given these differences and the correlation between rural location and increased use of nurse anesthetists,1 it is no surprise that, after adjusting for staffing patterns and other hospital characteristics, Mr. Simonson and colleagues did not find any significant difference in the quality of obstetric care between rural and urban hospitals. Guohua Li, MD, DrPH Joanne E. Brady, SM Khadeen Cheesman, BS Pamela Flood, MD Department of Anesthesiology Columbia University New York, New York [email protected]