Building a Stronger Instrument in an Observational Study of Perinatal Care for Premature Infants

natal Care for Premature Infants Mike Baiocchi, Dylan S. Small, Scott Lorch, Paul R. Rosenbaum1 University of Pennsylvania, Philadelphia Abstract. An instrument is a random nudge towards acceptance of a treatment which a¤ects outcomes only to the extent that it a¤ects acceptance of the treatment. In settings in which treatment assignment is mostly deliberate and not random, there may nonetheless exist some essentially random nudges to accept treatment, so that use of an instrument may extract bits of random treatment assignment from a setting that is otherwise quite biased in its treatment assignments. An instrument is weak if the random nudges barely in‡uence treatment assignment or strong if they are often decisive in in‡uencing treatment assignment. Although one hopes that an ostensibly random instrument is perfectly random and not biased, it is not possible to be certain of this, so a typical concern is that even the instrument is biased to some degree. It is known from theoretical arguments that weak instruments are invariably sensitive to extremely small biases, so for this reason, strong instruments are preferred. The strength of an instrument is often taken as a given. It is not. In an evaluation of e¤ects of perinatal care on the mortality of premature infants, we show that it is possible to build a stronger instrument, we show how to do it, and we show that success in this task is critically important. Also, we develop methods of permutation inference for e¤ect ratios, a key component in an instrumental variable analysis.