Their study, published May 16 in The New England Journal of Medicine, documents a wide regional disparity in the availability of newborn specialists and intensive care nursery beds that correlates poorly with improved survival. It lays the groundwork for reassessing the allocation of resources to improve newborn health and prevent disease and death, and raises questions about the continued growth of specialty medical services in the United States.
The advent of neonatal intensive care services—the neonatologists who care for premature and sick newborns and the intensive care nurseries where they are hospitalized—has dramatically improved the outlook for infants once considered too tiny or ill to survive. The growth in neonatal intensive care, driven by technological and clinical success over the past three decades, continues unabated, while its distribution across the country is more varied than other medical care resources.
Now health analysts and planners are beginning to reassess the ramifications of this growth in medical resources. In particular, they ask: are neonatal resources located where the sick babies are; does a greater supply translate into better outcomes for newborns; will further growth in supply continue to be beneficial?
The answers so far, says David Goodman, MD, associate professor of pediatrics, who led the DMS team, paint a disturbing picture of the landscape of neonatal intensive care. "Supply of neonatologists and intensive care beds is almost idiosyncratic in the way they are located; greater numbers are not found in the regions with more high risk newborns. A few regions of the country appear to be underserved in neonatologists, in that neonatal mortality rates are worse in the areas of very lowest supply. However, beyond the numbers found in the low supply areas, there are no further benefits of more neonatologists."
Goodman and his colleagues hypothesize that the "law of diminishing returns" applies to health care. "There comes a point," he says, " when adding more physicians can’t lead to any further gains in outcomes." In the case of neonatal care, the fundamental determinant of neonatal outcomes is whether a baby is born premature, a factor that neonatal specialists do not directly influence. And once a sick baby is born, survival may be more related to organization of existing resources and their effective application rather than the sheer quantity of physicians or other resources, Goodman explains.
The team studied the almost 3.9 million infants born in 1995 with a birth weight of at least 500 grams (1.1 pound). The overall mortality rate (measured by death within the first 27 days of life) was 3.4 per 1000 live births. Regions in the United States were grouped based upon the supply of neonatologists per newborn—very low, low, medium, high and very high. After adjusting for factors associated with greater risk of death, such as birth weight or prenatal care, investigators found no difference in the mortality once the supply of neonatologists exceeded the "low" supply category.
The death rate was 7 percent less in regions with 4.3 neonatologists per 10,000 births than in those with 2.7 neonatologists per 10,000, but further increases in neonatologists made no difference. The low supply regions—with 4.3 neonatologists/10,000 births—had the same death rate as the highest supply regions—those with 11.6/neonatolgoists/10,000 births. The research team also looked at the supply of neonatal intensive care beds and found no reduction in mortality as the number of beds increased. The article’s co-authors, who include neonatologists, epidemiologists and statisticians, are: Elliott S. Fisher, MD, MPH; George A. Little, MD; Therese A. Stukel, PhD; and Chiang-Hua Chang, MS, and Kenneth S. Schoendorf, MD. The research was funded by the Robert Wood Johnson Foundation.
"The saga of neonatology is emblematic of how a market driven health care system with inadequate public planning produced too much of a good thing," writes Kevin Grumbach of the University of California, San Francisco, in an editorial for the issue. Noting the importance of the study, he concludes that it "underscores the need for a more critical appraisal of the distribution, costs and outcomes of existing specialty services before the supply of expensive specialists and intensive care units is further expanded." Beyond neonatal mortality, whether higher numbers of physicians and intensive care beds improve newborn outcomes remains unknown. It is possible, for example, that more neonatologists could lead to faster illness resolution, while not affecting the chance of death. "The other possibility," Goodman suggests, "is that with more neonatologists or more beds, babies who would otherwise be cared for by their mothers on obstetrical units are instead admitted to intensive care units and receive more diagnostic and therapeutic interventions. Does this occur? If so, is that good or not? These are important questions for future research."
An intriguing phenomenon is that the growth of neonatal intensive care, like most specialty care, usually occurs without any explicit public planning, through private planning or institutional interest. "Strong market currents are shaping the care of our most vulnerable patients—newborns—at the beginning of life; these market forces are not currently balanced by public accountability," Goodman says.
Economic implications also make it important to determine if one is helping the right populations with the right resources, he continues. "Neonatologists and neonatal intensive care beds are expensive resources at the same time that the United States has one of the highest rates of unintended pregnancies and low birth weight of any industrialized nation." The Dartmouth study builds on earlier work documenting the geographic disparities in the neonatal workforce. The team used 1996 American Medical Association and American Osteopathic Association physician master files, and 1998 and 1999 surveys of all US neonatal intensive care units and neonatologists, as well as 1995 US vital records.
For more information, contact David Goodman: David.C.Goodman@Dartmouth.edu.
DMS news is on the web at www.dartmouth.edu/dms/news.
Journal
New England Journal of Medicine