Question special
Study Author

There are a variety of data sources on perinatal outcomes, and each has its own strengths and limitations. Each is well equipped to answer certain questions, and will fall short in answering other questions. We analyzed birth certificate data in our study, which are well-suited to answer population-level questions about temporal trends, surveillance, and questions which require a census. In a US context, birth certificate data are uniquely well equipped to answer questions of the type: ‘what are the population risks and benefits, comparing across birth settings.’ Such a question requires a census, and also adequate numbers given the rareness of the exposure (in a US context), and the rareness of key outcomes of interest. For such population-level questions, a prospective cohort is generally not a feasible study design in a US context, given decreasing participation rates in prospective cohort studies (and inadequate funding to support such time and resource-intensive methods).