Achieving the goals of newborn screening is, as for any screening, a balancing act: getting the maximum benefit from screening while producing the minimum harm. The advent of "expanded" newborn screening, with a large number of disorders detectable using a single test, has also posed problems, not new, but now more obvious. One is the finding of many more cases by screening, the extra cases being largely patients who have attenuated phenotypes and may remain asymptomatic for many years, even throughout life. These may or may not require active management in the short term, but do need lifelong awareness. Additionally, disorders have been included that are now thought benign or largely so. Babies risk being unnecessarily medicalized. Assessing outcome has also proved difficult because of the rarity of some disorders and the impracticality of randomized controlled trials. The requirements for valid studies include the need for case definitions, comparable comparison groups and probably assessment on a whole-population basis. An Australia-wide study of tandem mass spectrometry newborn screening involving 2 million screened and unscreened babies has demonstrated benefits overall to screened patients at age 6 years. The study was too small to provide conclusions for individual disorders other than for medium-chain acyl-CoA dehydrogenase deficiency.