This paper critically appraised the validity and generalizability of the safety of the Stockholm Birth Center care study to determine if it can be relied on to answer the question, "Is primiparous labor and birth in a birth center as safe for babies as standard medical care?" The retrospective cohort study is summarized, and statistical and methodological aspects are evaluated. Errors that were identified include selection bias and two forms of performance bias, both involving the independent variable. Nondefinition and lack of control of the independent variable and minor statistical errors were also noted. More serious concerns relate to the validity of an intention-to-treat analysis. Some methodological problems reduced validity of the study and ability to generalize the findings to other birth centers. Birth center care is a desirable and established birth option. A more useful approach to improving maternity care provision could involve comparing multiple birth center sites with each other to find best practice so that it can be analyzed and duplicated.
Birth-Issues in Perinatal Care Vol. 32, no. 2, p. 145-150