Moen A, Rønnestad A, Stensvold H, Uleberg B, Olsen F, Byhring S, Vonen B. (Editor) The Norwegian Neonatal Healthcare Atlas, 2009-2014. Tromsø, Norway: Senter for klinisk dokumentasjon og evaluering (SKDE), Helse Nord. December 2016. 2. Howell EM, Richardson D, Ginsburg P, Foot B. Deregionalization of neonatal intensive care in urban areas. Am J Public Health. 2002;92(1):119-124. 3. Horbar JD, Edwards EM, Greenberg LT, Morrow KA, Soll RF, Buus-Frank ME, Buzas JS. Variation in performance of neonatal intensive care units in the United States. JAMA Pediatr. 2017;171(3). 4. Richardson DK. A woman with an extremely premature newborn. JAMA. 2001;286(12):1498-1505. 5. Payne NR, Finkelstein MJ, Liu M, Kaempf JW, Sharek PJ, Olsen S. NICU practices and outcomes associated with 9 years of quality improvement collaboratives. Pediatrics. 2010;125(3):437-446. 6. Carroll AE. The concern for supply-sensitive neonatal intensive care unit care: if you build them, they will come. JAMA Pediatr. 2015;169(9):812-813. 7. Bourque SL, Hwang SS. Underuse versus overuse of neonatal inten Perinatal health (e.g., low birth weight, neonatal mortality rate) and its antenatal determinants, such as poverty and intendedness of pregnancy,1 are routinely measured across the populations of states and counties in the United States through the Vital Records system and other perinatal surveys. Unfortunately, the U.S. and most other countries lack systematic population-based measures of newborn care. Health care statistics of this type are widely available for the majority of the Medicare population (i.e., the aged) through annual research datasets that have been available to researchers for more than 25 years. The resulting studies have accelerated the pace of achieving quality standards and have stimulated numerous new care and financing models. The data have also enabled public reporting of both quality and patient experience metrics for Medicare beneficiaries, initially by Dartmouth researchers and now routinely by numerous public entities, including the Centers for Medicare and Medicaid Services (CMS). Evaluating the medical care and outcomes of newborns lags far behind these initiatives because there is no comparable national medical claims dataset for the under-65 population. Nevertheless, in recent years, a few investigators have made progress in national population-based studies of newborns that have the potential to provide invaluable information for improving the quality and value of perinatal services. This section reports on studies conducted at Dartmouth that provide a national perspective on newborn care over time and across regions. Trends in the proportion of newborns admitted to neonatal intensive care units (NICUs) are presented first, followed by an examination of the supply of NICU beds across neonatal intensive care regions (NICRs), the chances of newborn NICU admissions, and the association between the two. Both of these analyses draw from the entire U.S. birth cohort as reported in the national natality file from the National Center for Health Statistics of the Centers for Disease Control (CDC). This file has very accurate information about maternal and newborn characteristics useful for accurate estimation of newborn health risk. From this, newborns can be grouped into categories of risk based on attributes such as birth weight or gestational age, and measures of their care can be further adjusted for other factors. In the second half of the section, the utilization experience of Anthem Blue Cross Blue Shield-insured newborns is reported across regions for those enrolled in commercial and Medicaid managed care plans. Anthem market penetration across regions is not uniformly high enough to report newborn measures for the entire country, but a large majority of NICRs are included. The strength of the Anthem database is that it includes utilization information (e.g., number of special care days and imaging procedures) not found in the national natality file, and it allows for a comparison of two different insurance groups. A relative weakness is that it has far less information related to maternal and newborn health risk. These analyses are limited to two groups that can be reliably identified in claims: very low birth weight newborns (< 1,500 grams) and those at low risk for complications (newborns with an absence of serious diagnoses). A Report of the Dartmouth Atlas Project 16 DARTMOUTH ATLAS OF NEONATAL INTENSIVE CARE 13.8% 16.1% 46.6% 42.2% 0% 10% 20% 30% 40% 50% 2007 2008 2009 2010 2011 2012 500-1499 grams 1500-2499 grams 2500-3999 grams ≥ 4000 grams All U.S. Newborns Trends in Neonatal Intensive Care Unit Admissions2 The expansion of neonatal intensive care units (NICUs) and beds in recent decades has been associated with changes in the newborn population receiving NICU care. In its early years of development, NICU care was in short supply, and admission was primarily limited to critically ill premature newborns. Many newborns who might have benefitted from the medical and nursing care of a specialized unit were not cared for in hospitals with NICUs. As the number of units increased and regional perinatal networks were developed, this unintended rationing declined, but simultaneously resulted in higher use of NICUs for less ill newborns. By 2007, only