When should newborns be discharged?

When should newborns be discharged? By Sharisse Arnold, MD, and Henry H. Bernstein, DO The question of when to discharge mothers and newborns from the hospital has been the subject of much change, debate, and controversy over the years. The authors examine the issues, review available data, and offer valuable insights for obstetricians. During the past several decades, the length of hospital stay after childbirth has become the target of much public attention and policy debate. It has also become an increasingly popular topic for research (Figure 1). How did the controversy evolve? Has it affected medical care for the millions of infants born each year? Do longer lengths of stay really equate with improved quality of care? What can an individual clinician do to provide quality care in the setting of limited lengths of stay? In an attempt to answer these questions, we will review the literature on newborn discharge, including the current guidelines, and propose practical solutions for use in the clinical setting. We will consider where we have come from, where we are today, and where we might be going in the future. The historical perspective Between 1970 and 1992 the average length of hospital stay for deliveries decreased by 37%, from 4.1 to 2.6 days for all deliveries and from 3.9 to 2.1 days after vaginal births.1 During the second half of the past decade, the average length of stay decreased further and then increased slightly to 2.4 days nationwide in 1997.1 Appreciating the historical context behind this gradual shift provides insight into today's clinical practice. Childbirth in the hospital really began in the 20th century to safeguard the health and well-being of mothers while providing the necessary practical and educational environment for physicians.2 The idea of early discharge began to evolve in the 1950s as a response to a shortage of hospital beds resulting from increasing numbers of deliveries. In the 1960s and 1970s, the trend toward early discharge continued but was driven more by consumer-initiated decisions. Women's desire to gain control over and "demedicalize" the childbirth process culminated in more "natural" births, the presence of fathers in the delivery room, and decreased lengths of stay.2,3 Social, rather than financial, issues were the driving forces since early discharge was originally intended as an alternative to home deliveries for low-risk, educated mothers with strong support systems at home. By the late 1980s and early 1990s, escalating health-care costs and the advent of managed care organizations changed compensation policies. Insurers began limiting reimbursement for stays extending beyond 12 to 24 hours after vaginal deliveries, putting pressure on physicians and hospitals to discharge women and newborns as quickly as possible. Since childbirth is the most common reason for hospitalization in this country, with approximately 4 million births per year, decreasing the length of hospital stays could potentially save billions of health-care dollars. What began as a socially driven practice soon emerged as a financially motivated policy by third-party payors, instituted without scientific information that the change in practice was safe. Concern that shorter hospital stays might impair the quality of health care provided to newborns and mothers prompted an outcry from the public, medical community, and policymakers. As a result, state and federal legislation to protect mothers and children emerged in the mid-1990s. The legislative arena In May 1995, Maryland became the first state to enact legislation to prevent early discharge practices that were inconsistent with the Guidelines for Perinatal Care prepared by the American Academy of Pediatrics (AAP) and The American College of Obstetricians and Gynecologists.4 Most other states followed Maryland's lead. Loopholes existed within individual state laws, however. Under the federal Employment Retirement Income Security Act, states do not have the legal authority to require longer postpartum hospital stays for women and newborns than the stays covered by the employee group health plans of corporations that are self-insured. To address this issue, Congress passed the Newborns' and Mothers' Health Protection Act of 1996. The law, which became effective January 1, 1998, mandates insurance coverage for mothers and newborns for a minimum of 48 hours after vaginal deliveries and 96 hours after cesarean births. Any decision to discharge the mother or newborn earlier can be made only by the attending physician in consultation with the mother, a provision intended to protect the patient from external financial pressures. While the federal legislation is valuable, laws alone cannot ensure that shortened hospital stays allow quality health care for mothers and newborns. By exaggerating the value of hospitalization without providing for educational initiatives within the hospital and specifics on follow-up services, such laws can deflect the focus from quality of care to length of stay. Legislation may actually stifle initiatives to develop alternative means of postpartum health care. Focusing primarily on the first 48 hours of life may cause neglect of subsequent postpartum days, when many newborn problems, such as jaundice and feeding difficulties, occur. So what is early discharge? The timing of newborn discharge has varied widely over the past several decades, ranging from 5 to 7 days to less than 24 hours. In 1977, the AAP Committee on Fetus and Newborn (COFN) first defined the timing of newborn discharge as 72 to 96 hours, but earlier if stable.3 In 1980, COFN outlined criteria for early discharge, including:5 prenatal education on infant care; a supporting person at home; a physician-directed source of follow-up care with an appointment at 2 to 3 days of age; an uncomplicated antepartum, intrapartum, and postpartum course; a full-term, appropriate-for-gestational-age infant with a normal physical exam prior to discharge; and a minimum of 6 hours of hospitalization and ability of the mother to demonstrate that she knows how to appropriately assess, feed, and care for the infant. When the first edition of Guidelines for Perinatal Care was published in 1983, the usual postpartum hospital stay was 48 to 72 hours with most healthy neonates remaining hospitalized for at least 24 hours. The 1988 edition of the guidelines defined the usual time of discharge as 48 hours for vaginal deliveries and 96 hours for C/S. Today's definition of early discharge evolved from the third edition of the guidelines, in 1992, which defined early discharge as less than 48 hours for healthy infants born by vaginal delivery and less than 96 hours for babies delivered by C/S. "Very early" discharge is defined as less than 24 hours. In 1995, COFN revised and expanded its 1980 policy statement on the length of hospital stay for healthy term newborns. The revision, reflected in the 1997 edition of Guidelines for Perinatal Care, states that hospitalization should be "long enough to allow identification of early problems and to ensure that the family is able and prepared to care for the baby at home."4 The exact length of stay should be "based on the unique characteristics of each mother-infant dyad including the health of the mother, the health and stability of the baby, the ability and confidence of the mother to care for her baby, the adequacy of support systems at home, and access to appropriate follow-up care." The guidelines also define minimum discharge criteria (Table 1). They emphasize that the criteria are unlikely to be fulfilled in less than 48 hours and that the physician rather than a third-party payor should make the decision regarding the time of discharge.6 Table 1 American Academy of Pediatrics criteria for early newborn discharge Uncomplicated antepartum, intrapartum, and postpartum course for baby and mother Single birth Vaginal delivery Term gestation (38 to 42 wk) with birthweight appropriate for gestational age by appropriate growth curve, with normal exam Baby shows normal, stable vital signs for 12 hr preceding discharge Thermal homeostasis (axillary temperature 36.1°C to 37°C in open crib) Respiratory rate Heart rate 100 to 160 beats/min Baby has voided and stooled; if circumcised, no excess bleeding for 2 hr Baby has fed at least twice, demonstrating coordinated suck, swallow, and breathing No significant jaundice in first 24 hr of life Mother has adequate knowledge to care for her newborn, documented by training in: Breastfeeding or bottle feeding (trained staff should assess breastfeeding) Knowledge of normal urine and stool frequency Care of cord, skin, genitals Recognition of common signs of infant illness, especially jaundice Infant safety issues (car safety seats, sleep positioning) Laboratory data obtained and reviewed as normal or negative, including: Maternal testing for syphilis and hepatitis B Cord/infant blood type, direct Coombs test if indicated Support (family members or health-care providers) available to mother and baby for first few days after discharge. Supporters must be: Familiar with newborn care and knowledgeable about lactation Able to recognize jaundice and dehydration Newborn screening sample obtained before 24 hr of milk feedings; follow-up testing is assured Initial hepatitis B vaccine given or scheduled within first week Physician-directed source of continuing care identified If discharge at Examiner competent in newborn assessment, reports to physician on day of visit Family, environmental, and social risk factors assessed, including but not limited to: Substance abuse, history of child abuse or neglect, no fixed home, teen mother, lack of social support (especially for single, first-time mother), domestic violence Mental illness in a parent in the home If risk factors are present, delay discharge until a plan to safeguard the baby is in place In addition, the guidelines recommend that follow up occur within 48 hours after early discharge and that the purpose of the follow-up visit be specifically outlined (Table 2). If timely follow up cannot be assured, discharge should be deferred until appropriate arrangements can be made. Table 2 Purpose of a follow-up visit Assess newborns health, hydration, and degree of jaundice Review feeding techniques and patterns Review adequacy of stooling and urination patterns Assess mother-baby interaction and newborn behavior Reinforce education on neonatal care Review laboratory results from discharge Perform appropriate tests (screening in accordance with state regulations) Prepare a health maintenance plan including preventive care and immunizations, periodic evaluations and physical examinations, necessary screening, and methods for obtaining emergency services The guidelines lack specific directives, but do make a plea for local hospitals, in collaboration with their professional staff and community agencies, to develop practice parameters most suited to their unique settings. Although they do not address follow up for infants discharged after 48 hours (which could be as early as 49 hours!), recommend financial solutions for all individuals, or contain much detail about follow-up services, the guidelines do provide a framework from which future efforts can be advanced. The point/counterpoint of early discharge Opponents of early discharge argue that because the immediate postpartum period is one of biologic and social transition, shorter lengths of stay diminish the opportunity of medical personnel to supervise and respond to problems occurring during this time of rapid change. Longer hospitalizations allow for appropriately timed newborn screening tests, postpartum assessments of mother and child, in-hospital education about family planning and infant care, the encouragement and teaching needed to establish successful breastfeeding, a maternal rest period, and transition time for parental and familial psychosocial adjustments. The first 2 to 3 days after birth are a time of many physiologic changes in the newborn that require close observation. Since hyperbilirubinemia peaks after the third day of life, lack of observation during this time can lead to increased incidence of kernicterus. Likewise, neonatal infections, gastrointestinal problems, and ductal-dependent cardiac lesions often do not appear until after the third day of life. Improper timing of the newborn screen may prevent prompt diagnosis of metabolic disorders. Delay in recognizing or treating some of these potentially curable conditions can increase neonatal morbidity and mortality. Breast milk does not usually come in until 2 to 4 days postpartum. Hence, early discharge may discourage mothers from starting to breastfeed or cause them to stop prematurely. Feeding problems associated with maternal inexperience or lack of support may lead to dehydration or complicate jaundice in infants. Proponents of limited hospital stays believe that early discharge promotes bonding and attachment, enhances family interactions, improves patient satisfaction, decreases exposure to nosocomial infections, and provides a safe alternative to home deliveries. They argue further that normal variations in newborn physiology are overly assessed in the hospital, increasing costs and iatrogenic risks to the newborn.7 The financial savings resulting from early discharge can be reallocated to more beneficial outpatient health-care resources.8 Proper follow-up services avoid the theoretical concerns about newborn safety. What does the literature tell us? Beyond speculation and anecdotal reports, what concrete informa-tion does the scientific literature provide? Unfortunately, current research fails to offer definitive answers regarding the safety of "early" discharge or the efficacy of prolonged hospital stays.7-10 We practice in an era of evidence-based medicine, yet financial implications have a far-reaching impact on clinical practice when clear scientific justification is lacking. Current studies are fraught with methodologic and design flaws. Many are retrospective and lack the sample size needed to generate statistical power. Selection bias resulting from strict eligibility criteria, lack of randomization, or voluntary enrollment in early discharge programs by low-risk women is a frequent criticism. Furthermore, interventions that are effective in a controlled setting do not always work in the community. Research that focuses on a homogeneous population, for example, is difficult to generalize to multiple socioeconomic groups in different geographic locations. The lack of a comparison group, insufficient follow-up intervals and follow-up data, and a limited range of measured outcomes are other common pitfalls. The definition of early discharge is variable and has changed over time, which makes comparisons between studies difficult. What is defined as early in one study is late in another. The quality and quantity of outpatient follow up by telephone consultation, office visit, urgent care, or home visit can dramatically affect outcomes. Postpartum follow-up practices are not always considered when interpreting the success or failure of early discharge. In a critical review of the nursery discharge literature, Braveman and colleagues stratified studies into three groups: early discharge without routine follow-up services, early discharge with office or clinic follow up 1 to 3 days after discharge, and early discharge combined with postpartum home visits.7 They concluded that studies of early discharge programs that did not include follow-up services could not validate the safety of discharge before 48 hours. Moreover, no reliable data exist regarding the health outcomes of early discharge when the mother is instructed to return for an office or clinic visit within 1 to 3 days. Low-income patients have a high "no-show" rate, and the physical limitations caused by childbirth make travel on early postpartum days difficult for most women. Finally, the literature on follow up by home visits does not provide definitive information regarding the quantity or quality of visits needed to prevent adverse outcomes. While early discharge may be a viable alternative, it may not be appropriate for every mother and neonate. Britton and colleagues conclude that because the literature is flawed and definitive data are lacking, the timing of discharge should be an individual decision with careful consideration given to the AAP guidelines of 1980.8 The practitioner should consider the medical, social, and economic aspects of each case, as well as the ability to provide appropriate postpartum visits at home or in the office. Other researchers suggest shifting the focus of discharge timing from "rule-based" decisions (24 vs. 25 hours) to clinical judgment, arguing that hospital care is only one component in the overall postpartum health-care experience and that individual follow-up plans are more important to good outcomes than the exact time of discharge.9 One critical review of the literature, citing the relative lack of sound clinical research, appeals to future researchers to shift their focus from the duration of hospitalization to establishing the most effective means of achieving quality health education and social objectives.10 Evaluating outcomes of discharge timing An often studied outcome measure of nursery discharge is rehospitalization of newborns. However, using readmission, actually an infrequent event, to measure well-being requires large sample sizes to detect significant results. A 1% readmission rate requires 38,211 patients to detect a 25% difference, for example.9 Moreover, readmission criteria vary geographically between institutions and among physicians. An infant readmitted to the hospital for jaundice in one community may be managed with home phototherapy in another. The reason for rehospitalization may not even reflect the timing of discharge. A newborn readmitted in the first week of life because of a community-acquired infection, for example, would most likely have been rehospitalized regardless of the timing of discharge.11 Data on the effects of discharge timing on readmission are conflicting. Two large retrospective population-based studies demonstrate an increased risk of rehospitalization associated with early discharge.12,13 When Lee and colleagues reviewed the records of 920,554 healthy Canadian newborns weighing 2,500 g or more, they found that as the average length of stay decreased from 4.5 to 2.7 days, readmissions in the first 2 weeks of life increased from 12.9 to 20.7 per 1,000.12 Although there was no change in mortality, severity of illness in neonates readmitted with jaundice and dehydration increased. A state of Washington review of 310,578 live births also found an increased incidence of readmission within 28 days after birth, with the highest risk in the first week of life. Infants born to primigravidas, mothers less than 18 years of age, and mothers with premature rupture of membranes were at highest risk.13 In contrast to these two studies, Edmonson and associates found no independent effect of early discharge on the rehospitalization rate in 120,290 normal newborns readmitted for feeding difficulties, dehydration, or poor weight gain.11 Other studies also have found no significant differences in readmission rates between early and late discharges.14-19 The most devastating potential consequence of early discharge is delay in recognition and treatment of a life-threatening condition. A review of hospital discharge times of newborns weighing more than 2,500 g who died in the neonatal period found that 99% of the full-term infants who died were symptomatic within the first 18 hours of life. There was thus no association between early discharge and neonatal mortality.12,20 Other measurable maternal and newborn health outcomes include outpatient visits, breastfeeding success, parental satisfaction, and parental outcomes. A prospective study of the effect of a voluntary moderate reduction in hospital stay on these outcomes concluded that when "adequate" postpartum outpatient services were available, early discharge did not adversely affect any of the outcomes.14 A prospective study of health-care factors contributing to the success of breastfeeding in a highly motivated group of women found that the factors significantly associated with success were maternal education and age, appraisal by the mothers of the quality of the breastfeeding experience in the hospital, and the quality of the home visit. Health system support of breastfeeding women during the hospital stay and early postpartum period was more important in determining success than the actual length of stay.21 A recent prospective study by Britton and colleagues evaluated the effect of perinatal discharge on parenting outcomes during infancy, including breastfeeding, mother-infant interaction, and mother-infant attachment. Despite its small sample size, selection bias, and the fact that participants were low-risk and homogeneous, the study is the most comprehensive assessment of parental outcomes to date. The results failed to demonstrate any association between timing of discharge and parental outcomes over a 3- to 12-month period.22 Although methodologic differences may explain the contradictory conclusions in the literature regarding medical harm from shorter postpartum hospital stays, the actual explanation may be more substantive. First, while deaths during the first month of life account for two thirds of infant mortality in the first year, most infants and mothers are healthy. Second, for some mothers and families who are physically and emotionally well, feel educated in perinatal care, have adequate support, and are prepared for the arrival of an infant, discharge earlier than conventionally recommended can be successful and even advantageous. Third, desirable redundancies may exist in the systems available to care for newborns and families, including inpatient postpartum care; outpatient nursing visits to the home, pediatric offices and clinics; and social and financial resources. Reducing existing inpatient and outpatient postpartum services may eliminate some of these redundancies, which help to shield mothers and infants from demonstrable harm in the vast majority of cases, and make the system more fragile. Fourth, judgment about the timing of discharge is generally applied on a case-by-case basis. Thus, it is difficult to define an optimal length of postpartum stay for well newborns and mothers. The importance of follow-up services A pilot study conducted in the Pediatric Research in Office Settings Network of the AAP assessed the impact of mothers' and clinicians' perceptions of mother/infant readiness for postpartum discharge and ascertained some of the issues facing mothers in the first 2 weeks after discharge. This prospective cohort study of mothers and healthy infants during the first month of life showed substantial variation in the mothers' and clinicians' perceptions of readiness for postpartum discharge on the day of discharge and over time. The perceptions were significantly associated with maternal and infant health status, health-related behaviors, and health-care utilization. The study also found considerable morbidity during the first 2 weeks after postpartum discharge. The mean number of days a mother had any concern about her infant was 5.4, prompting a mean number of 3.4 phone calls and 1.4 health visits. During this same 2-week period, the mean number of days a mother had any concern about herself was 5.9, prompting a mean number of 1.4 phone calls but only a rare health visit. Interventions should be developed to minimize the impact of these concerns at this potentially vulnerable time. For limited nursery stays to succeed, reliable follow up is imperative. Current knowledge about newborn physiology and disease processes suggests that a clinical evaluation on postpartum day 3 to 4 is needed, regardless of the time of discharge. As previously mentioned, however, present guidelines and legislative efforts do not provide details concerning the quantity or quality of postpartum discharge services. Who should provide the care (nurse, physician, or a midlevel provider), for example? What type of follow up should occur (visit or telephone call)? Where should it occur (home, office, or clinic)? When should it occur? How often and for whom should the services be provided? How can we ensure follow up for indigent populations with high "no-show" rates, particularly for individuals with access issues and limited support at home? Unfortunately, answers to these burning questions are not readily found in the literature. Although a plethora of studies have examined a variety of postpartum programs that provide care after discharge, little information currently exists to help guide routine follow up.6,23 Most of the available studies are small, with voluntary enrollment and limited generalizability to community practice. More research is needed to lend scientific justification to refining follow-up guidelines. Potential alternatives to routine office follow up include using qualified nurse practitioners who can provide care for both mother and infant. Such a programGreat Starts, run by Kaiser Permanente in Northern Californiahas decreased use of urgent care services while successfully meeting the needs of both mothers and newborns.24 Kaiser Permanente in Colorado has developed a successful program in which nurses specializing in perinatal care provide home visits to members. Good communication with the primary-care provider is built into the system, and patients have embraced this alternative to an office visit. A pilot program in the Southeast employs staff nurses from a tertiary-care hospital and college of nursing to provide follow-up care at home to new mothers and their infants. Women in the project have been less likely than those not in the project to seek nonroutine health care, have used fewer health-care dollars, and are generally pleased with the program.25 Making a difference in clinical practice Since shortened hospitalizations seem to be here to stay, what can be done to assure that quality of care is not compromised? Decisions about readiness for discharge must be made jointly by the mother and the physician. By being familiar with current guidelines, anticipating potential problems, and reacting appropriately, we can all help to guard against bad outcomes. Focus should be placed on a smooth transition during the first 6 hours after birth when the neonate adapts to extrauterine life. Be aware of the subtle warning signs that may signal problems in subsequent days of life (Table 3).26 If in doubt, observe the baby longer in the hospital. Table 3 Signs of potential newborn problems during transition Central cyanosis Weak pulses Significant or persistent heart murmur Poor feeding or suck Pallor Tachypnea or increased work of breathing >1 hr after birth Stridor No stool or urine >24 hr after birth Excessive vomiting Poor urinary stream Bladder distention Hypothermia/hyperthermia Hypoglycemia ABO incompatibility, positive Coombs test, or jaundice on the first day of life Hypotonia or decreased activity Irritability or excessive crying Parents need to be provided with comprehensive newborn teaching before discharge. Although most parents receive a variety of reading materials and many take prenatal classes, reviewing the essentials of newborn care in the hospital after delivery can reinforce parental knowledge. This is also an opportune time to begin to establish the primary care relationship. An adequate newborn screen also has to be obtained before discharge. Screening done at less than 24 hours of life can affect the results of tests that depend on metabolite accumulation or rapid changes in hormone levels (phenylketonuria, maple syrup urine disease, tyrosinemia, galactosemia, hypothyroidism, and congenital adrenal hyperplasia). The appropriate arrangements for outpatient follow-up screens need to be made if such a system does not already exist at your hospital. Risk factors for development of jaundice in the infant must be reviewed. These include, but are not limited to, prematurity (<37 weeks' gestation); low birthweight (<2,700 g); ABO incompatibility, positive Coombs' test, or jaundice on the first day of life; breastfeeding; bruising or cephalohematomas; history of a previous infant with severe jaundice or anemia; infant of a diabetic mother; oxytocin use; Chinese, Japanese, Korean, or American Indian ancestry; poor feeding because of oral defects (such as clefts or micrognathia); and underlying hemolytic diseases, metabolic abnormalities, endocrinopathies, or conditions that increase the enterohepatic circulation.27 If any of these situations exist, some type of follow-up communication within 24 to 48 hours after discharge must be arranged. Carefully review the mother's chart for group B streptococcal status, maternal fevers, and timing and quantity of perinatal antibiotics if the mother was treated before delivery. Consider deferring discharge if there are risk factors for infection that warrant further evaluation as defined in the revised AAP guidelines.28 Timely discharge follow up must be arranged as well. Routine practice has been to schedule a follow-up visit for newborns at 2 to 4 weeks of age. The 1997 AAP/ACOG guidelines, however, suggest follow up within 48 hours after discharge if the patient is released early. A survey of 20 private pediatricians in a large community to evaluate how well pediatricians are responding to the guidelines in the context of shorter hospital stays produced discouraging results. Although follow-up practices have changed in response to the current guidelines, 38% of infants discharged at less than 48 hours were seen 4 or more days after discharge, and 33%, 14 days after discharge.29 Timely follow up for all infants should be scheduled, regardless of length of stay. An individualized approach Newborn discharge has been the subject of much change, debate, and controversy over the years. The focus has clearly shifted from the exact length of stay to the quality of postpartum services provided both in the hospital and after discharge. Recent federal legislation helps protect physicians' and mothers' joint discharge decision-making. An individualized approach must be taken, considering the physical and emotional health of the mother and infant, the ability of the mother and family to care for the baby, the unique social situation, and the availability of follow-up services in the community. Current research provides little concrete information on the safety of "early" discharge or the efficacy of longer hospital stays. In the meantime, individual physicians can do their part to make discharge a safe alternative to continued hospitalization by closely monitoring the infant during the transition; providing comprehensive discharge teaching in the hospital; ensuring a mechanism of proper metabolic screening; being aware of risk factors for jaundice and sepsis; and providing appropriate follow-up planning prior to discharge. Furthermore, although follow-up services appear to be the key to the safety and success of limited hospital stays, the exact nature of such necessary services remains unknown. Each community must develop standards of care applicable to its unique medical, social, economic, and geographic environment. A more complex conceptual model of postpartum care that incorporates the responsibilities of each partner (hospitals, insurers, providers, patients, social and health services agencies) is desirable. Further research is essential to develop specific practice guidelines necessary to achieve optimal health outcomes for all mothers and their babies. A thoughtful approach to newborn discharge is critical. REFERENCES 1. Centers for Disease Control and Prevention. Trends in length of stay for hospital deliveriesUnited States, 1970-1992. MMWR. 1995;44:335. 2. Annas GJ. Women and children first. N Engl J Med. 1995;333:1647-1651. 3. Thilo EH, Townsend SF, Merenstein GB. The history of policy and practice related to the perinatal hospital stay. Clin Perinatol. 1998;25:257-270. 4. American Academy of Pediatrics and American College of Obstetricians and Gynecologists. Postpartum and follow-up care. In: Guidelines for Perinatal Care. 4th ed. Elk Grove Village, Ill: American Academy of Pediatrics; 1997:164. 5. American Academy of Pediatrics Committee on Fetus and Newborn: criteria for early infant discharge and follow-up evaluation. Pediatrics. 1980;65:651. 6. American Academy of Pediatrics Committee on Fetus and Newborn. Hospital stay for healthy term newborns. Pediatrics. 1995;96:788-790. 7. Braveman P, Egerter S, Pearl M, et al. Early discharge of newborns and mothers: a critical review of the literature. Pediatrics. 1995;96:716-726. 8. Britton JR, Britton HL, Beebe SA: Early discharge of the term newborn: a continued dilemma. Pediatrics. 1994;94:291-295. 9. Kessel W, Kiely M, Nora AH, et al. Early discharge: In the end, it is judgment. Pediatrics. 1995;96:739-742. 10. Margolis LH. A critical review of studies of newborn discharge timing. Clin Pediatr. 1995;34:626-634. 11. Edmonson MB, Stoddard JJ, Owens LM. Hospital readmission with feeding-related problems after early postpartum discharge of normal newborns. JAMA. 1997;278:299-303. 12. Lee KS, Perlman M, Ballantyne M, et al. Association between duration of neonatal hospital stay and readmission rate. J Pediatr. 1995;127:758-766. 13. Liu LL, Clemens CJ, Shay DK, et al. The safety of newborn early discharge.The Washington State experience. JAMA. 1997;278:293-298. 14. Mandl KD, Brennan TA, Wise PH, et al. Maternal and infant health: effects of moderate reductions in postpartum length of stay. Arch Pediatr Adolesc Med. 1997;151:915-921. 15. Soskolne El, Schumacher R, Fyock C, et al. The effect of early discharge and other factors on readmission rates of newborns. Arch Pediatr Adolesc Med. 1996;150:373-379. 16. Kotagal UR, Atherton HD, Bragg E, et al. Use of hospital-based services in the first three months of life: impact of an early discharge program. J Pediatr. 1997;130:250-256. 17. Kotagal UR, Atherton HD, Eshett R, et al. Safety of early discharge for Medicaid newborns. JAMA. 1999; 282:1150-1156. 18. Cooper WO, Kotagal UR, Atherton HD, et al. Use of health care services by inner-city infants in an early discharge program. Pediatrics. 1996;98:686-691. 19. Pittard VVB, Geddes KM. Newborn hospitalization: a closer look. J Pediat. 1988;112:257-261. 20. Beebe SA, Britton JR, Britton HL, et al. Neonatal mortality and length of newborn hospital stay. Pediatrics. 1996;98:231-235. 21. Kuan LW, Britto M, Decolongon J, et al. Health system factors contributing to breastfeeding success. Pediatrics. 1999;104:e28. 22. Britton JR, Britton HL, Gronwaldt V. Early perinatal hospital discharge and parenting during infancy. Pediatrics. 1999;104:1070-1076. 23. Egerter SA, Braveman PA, Marchi KS. Follow-up of newborns and their mothers after early hospital discharge. Clin Perinatol. 1998;25:471-481. 24. Nelson VR. The effect of newborn early discharge follow-up program on pediatric urgent care utilization. J Pediatr Health Care. 1999;13:58-61. 25. Brown SG, Johnson BT. Enhancing early discharge with home follow-up: a pilot project. J Obstet Gynecol Neonatal Nurs. 1998;27:33-38. 26. Hurt H. Early discharge for newbornswhen is it safe? Contemporary Pediatrics. 1994;11:68. 27. Thilo EH, Townsend SE. Early newborn discharge: Have we gone too far? Contemporary Pediatrics. 1996;13(4):29. 28. American Academy of Pediatrics Committee on Infectious Diseases and Committee on Fetus and Newborn. Revised guidelines for prevention of early onset group B streptococcal (GBS) infection. Pediatrics. 1997;99:489-496. 29. Maisels MJ, Kring E. Early discharge from the newborn nurseryeffect on scheduling of follow-up visits by pediatricians. Pediatrics. 1997;100:72-74. Dr. Arnold is Associate Medical Director, Children's Hospital/Beverly Hospital Pediatric Program, and Instructor in Pediatrics, Harvard Medical School, Boston, Mass. Dr. Bernstein is Associate Chief, General Pediatrics, and Director of Primary Care at Children's Hospital, and Associate Professor of Pediatrics, Harvard Medical School, Boston, Mass. The authors are indebted to Sabrina Cimino for her editorial contributions. The critical reviews of Drs. Tracy Lieu, Harris Lilienfield, Edward Rothstein, and Carol Touloukian are also very much appreciated. Henry Bernstein, Sharisse Arnold. When should newborns be discharged?. Contemporary Ob/Gyn 2000;11:47-69.