Babies who do not receive human milk are more likely to suffer health problems as newborns and later. Not all babies are able to feed at the breast because they are premature, ill or separated from their mothers; thus expressed milk is needed. Mothers may express milk for their own comfort, to increase supply, or to leave milk for their baby. This updated review includes 23 randomised controlled studies with 10 (632 mothers) providing data for analysis. It includes mothers of infants in neonatal units in the USA, UK, Malaysia, Brazil, Egypt, Kenya and Nigeria, as well as term infants in the USA and UK. A greater milk volume was associated with providing a relaxation tape (one study), and when the mothers made an earlier start after birth of expressing milk for an infant unable to breastfeed. In another study, a large electric pump provided a greater mean volume of milk than hand expression over six days in the first two weeks after birth, though a greater volume was not found in two other studies of the same pump looking at 12 to 36 hours postpartum or on day five. No difference in volume was evident between the manual and electric pumps studied or with single versus double pumping. Sodium concentration was found to be higher in hand expressed milk, which may be important for preterm infants. Any milk contamination was similar comparing hand-expressed and pumped milk, and the level of maternal breast or nipple pain was no different between methods. All studies were small and results may not apply to pumps other than those tested or in different situations. No study asked mothers if they had achieved their own goals for expressing. None of the studies examined costs involved with methods. Twelve of the 18 studies that evaluated pumps or products had support from the manufacturers. The available evidence indicates that low-cost measures such as early initiation of expressing for an infant unable to breastfeed, relaxation, hand expression, and lower cost pumps may be as effective, or more effective, than large electric pumps for some outcomes. Not all the studies mentioned whether basic supports were provided, particularly for mothers with hospitalised children, including access to food and fluid, a place to rest near their baby, and the availability of knowledgeable health workers. These supports could affect milk expression.