An integrated review of the literature on demand feedings for preterm infants.

In the past 50 years, only 7 studies evaluating demand feedings in preterm infants have been published. The studies employed a variety of research methods including nonexperimental, quasi-experimental, and experimental designs. The earliest demand feeding studies are difficult to interpret due to inadequate sample sizes and incomplete descriptions of methodology. Trials in the 1980s and early 1990s were better described; however, they suffered from designs that produced findings that likely were strongly related to unit, nurse, or infant-specific interactions with the treatment, making it unlikely that findings could be replicated in other settings. The most recent studies are better conceived and described. Overall, the pattern of findings across the past 50 years indicates that preterm infants who were fed on demand showed some clinical improvement when compared to those infants offered other feeding patterns. Demand-fed infants exhibited more hunger cues and had improved behavior state organization, shorter duration of hospitalization, and gained weight at the same rate or faster than infants not fed on demand. Although each of the studies reviewed had some methodologic weaknesses, the consistency of findings across differing researchers, settings, empirical indicators, and years supports the cautious contention that demand feeding might prove to be the feeding approach of choice for most healthy preterm infants. Further studies in this area are warranted to add to current knowledge regarding the safety and efficacy of feeding various preterm infant populations. Research to address subpopulation differences (i.e., infants with congenital heart disease or orofacial lesions) in timing, method, and approaches to preterm infant feeding care is also needed.