Skip Navigation U.S. Department of Health and Human Services
Agency for Healthcare Research Quality
Archive print banner

This information is for reference purposes only. It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work. Persons with disabilities having difficulty accessing this information should contact us at: Let us know the nature of the problem, the Web address of what you want, and your contact information.

Please go to for current information.

Children's Health

Researchers examine neonatal ICU practices, triage, improvements in care, and the use of narcotics

The rapid decline in neonatal deaths in the past four decades has been attributed to improvements in the care provided in neonatal intensive care units (NICUs). These range from increased use of surfactant to improve immature lung functioning to newer modes of mechanical ventilation of low birthweight babies.

Triage of newborn infants is important for determining which infants need care in the NICU. This high-volume, low-intensity process accounts for nearly 10 percent of total NICU costs. These issues were addressed in two recent studies. Another study examined the seemingly common practice of using narcotics in the NICU for very-low-birthweight (VLBW) newborn infants who have not undergone surgery. Finally, it has been well documented that outcomes vary substantially from one NICU to another. These variations have led to increased interest in illness severity measures to adjust for case mix in the NICU. A fourth study examined progress made to date in developing and testing neonatal risk scoring systems for use in the NICU.

All four studies, which are summarized here, were supported by the Agency for Health Care Policy and Research (HS06123 and HS07015), and led by Douglas K. Richardson, M.D., M.B.A., of Harvard Medical School.

Richardson, D.K., Gray, J.E., Gortmaker, S.L., and others (1998, October). "Declining severity adjusted mortality: Evidence of improving neonatal intensive care." Pediatrics 102(4), pp. 893-899.

According to this study, neonatal deaths among infants weighing less than 3.3 pounds (1,500 grams) decreased nearly 50 percent from 1990 to 1995 in the same two NICUs. NICU deaths declined from 17.1 percent to 9.5 percent, and total deaths among these infants declined from 31.6 percent to 18.4 percent. This improved survival occurred despite inclusion of smaller, sicker infants at the border of viability.

One-third of the decline was attributable to improved condition of the infants on NICU admission (improved birthweight, higher Apgar scores, and better physiologic stability) that reflected improving obstetric and delivery room care during the 5-year period. Two-thirds of the decline was attributable to more effective newborn intensive care and aggressive respiratory and cardiovascular treatments, such as the use of surfactant, conventional and high-frequency mechanical ventilation, and continuous positive airway pressure.

According to the authors, this pattern of improved neonatal survival suggests a powerful effect of intensive care technologies both before and after delivery. They used models to estimate the changing odds of death between two groups of infants, in the same two NICUs, 5 years apart (1989 to 1990 and 1994 to 1995). They calculated odds of death based on birthweight and illness severity upon NICU admission and measured therapeutic intensity.

Zupancic, J.A., and Richardson, D.K. (1998). "Characterization of the triage process in neonatal intensive care." Pediatrics 102, pp. 1432-1436.

Neonatal triage involves the evaluation and short-term management of infants after delivery. The goal is rapid diagnosis and assignment of the infant to the appropriate level of neonatal care (regular nursery, intermediate care, or intensive care) based on triage findings. Triage is a time-intensive process that accounts for nearly 10 percent of total NICU costs, according to this study. The researchers collected data on 2,486 newborn infants who were admitted to two NICUs for less than 24 hours and subsequently discharged to routine care. They estimated daily NICU workload based on the number and labor intensity of NICU admissions. The researchers also correlated length of stay and costs for triage with diagnoses, perinatal descriptors, severity of illness, and markers of concurrent NICU workload.

The mean birthweight for triage infants was about 7 pounds, and the mean gestational age was 39 weeks. The major reasons for evaluation were exclusion of sepsis (34 percent); birth complications, including meconium aspiration, perinatal depression, and trauma (24 percent); and transitional respiratory distress (23 percent). Severity of illness was minimal. The most frequent forms of resource use were antibiotic administration (34 percent), placement of a peripheral intravenous line (40 percent), cardiac monitoring (53 percent), and external warming (26 percent). The median NICU cost was $870.

About 16 percent of all newborns undergo this type of triage, and they usually have low illness severity. Length of NICU stay and costs of care for the newborns in this study were affected not only by the infants' medical characteristics but also by nonmedical factors—such as workload of NICU staff or assignment of junior residents to make triage decisions—which may be amenable to change.

Kahn, D.J., Richardson, D.K., Gray, J.E., and others (1998). "Variation among neonatal intensive care units in narcotic administration." Archives of Pediatric and Adolescent Medicine 152, pp. 844-851.

NICUs vary 29-fold in their use of narcotics for very-low-birthweight infants who have not undergone surgery, a variation that cannot be explained by birthweight or by illness severity. Furthermore, little is known about any major advantages or disadvantages to using narcotics for these infants. The lack of a common standard of care for administering narcotics to high-risk neonates is probably related to differing impressions about the hazards (for example, drops in heart rate and blood pressure) and benefits (for example, improved synchrony with the ventilator) of narcotic use.

The researchers examined the medical charts of neonates weighing less than 1,500 g (3.3 pounds) admitted to six major regional NICUs in New England and found no differences in weight gain or need for mechanical ventilatory support at 14 and 28 days after birth in infants NICU practices receiving and not receiving narcotics. Narcotic use was not associated with differences in blood pressure or heart rate or with increased length of hospital stay. However, it was associated with more than 33 g of fluid retention, a higher direct bilirubin level on day 3, and delayed resolution of illness. There were no associations between narcotic treatment and cardiovascular instability, chronic lung disease, growth, or discharge timing.

One reason for the variation in narcotic use may lie in the difficulty of recognizing pain in neonates that might lead to overprescription or underprescription of narcotics. Another possible explanation is that some clinicians view narcotics as the treatment of choice to sedate neonates breathing asynchronously against the ventilator. Others prefer benzodiazepines or phenobarbital, simply stroking or talking to the infants, or adjusting ventilatory rates.

Richardson, D., Tarnow-Mordi, W.O., and Escobar, G.J. (1998). "Neonatal risk scoring systems: Can they predict mortality and morbidity?" (AHCPR grant HS07015). Clinics in Perinatology 25(3), pp. 591-611.

In this paper, Dr. Richardson and his colleagues evaluate progress in measuring and applying two neonatal risk scoring systems: the Score for Neonatal Acute Physiology (SNAP) and the Clinical Risk Index for Babies (CRIB), and discuss the future research and operational challenges facing the field. The authors suggest that widespread routine use of severity measures will occur only when data collection and scoring can be automated. The development, validation, and implementation of the next generation of severity scores will require several years of development. Entirely new approaches to severity scoring will become possible, using online digital data analogous to electronic fetal monitoring but extending to respiratory and electroence-phalogram patterns. Such data-intensive measures of physiologic status may substantially increase the sensitivity and specificity of severity indexes. As the authors note, physiology-based illness severity scores are proving their value through a wide variety of practical applications. While it has been shown that these scores can predict some morbidity and some mortality, this function is much less important than their application as a means of improving quality and lowering cost. Nevertheless, future development will depend to a large extent on commercially viable applications, concludes Dr. Richardson.

Return to Contents
Proceed to Next Article

The information on this page is archived and provided for reference purposes only.


AHRQ Advancing Excellence in Health Care