Objective: To summarize existing data about the effectiveness of bed rest when used to improve various pregnancy outcomes and to determine how often bed rest is used and the cost associated with its use. Data sources: We used the MEDLINE data base to search for all English language papers evaluating the effectiveness of bed rest in pregnancy. We also reviewed a number of textbooks and the.1988 National Infant Mortality Survey. Methods of study selection: We reviewed these sources for recommendations about using bed rest in various obstetric conditions. We used the 1988 National Infant Mortality Survey to determine how often bed rest was used either to prevent or to treat various obstetric conditions and estimated the costs associated with its use. Data extraction and synthesis: Bed rest is used in nearly 20% of all pregnancies to prevent or treat a wide variety of conditions, including spontaneous abortion, preterm labor, fetal growth retardation, edema, chronic hypertension, and preeclampsia. There is little evidence of effectiveness. The estimated costs associated with bed rest, including hospitalization, lost wages, and lost domestic productivity, range from more than $250 million to billions of dollars per year. Conclusions: Bed rest is used extensively to treat a wide variety of pregnancy conditions, at substantial cost but with little proof of effectiveness. We recommend that because this intervention has failed the test of effectiveness, its use during pregnancy should be curtailed unless randomized trials demonstrate improvement in a specific outcome. (Obstet Gynecol 1994;84:131-6). © 1994 by The American College of Obstetricians and Gynecologists.