BACKGROUND. Congress passed the Newborns’ and Mothers’ Health Protection Act in 1996, reversing the trend of shorter newborn nursery lengths of stay. Hope existed that morbidities would lessen for this vulnerable population, but some reports indicate that the timeliness and quality of postdischarge care may have worsened in recent years.
OBJECTIVE. Our goal was to determine risk factors for the potentially preventable readmissions because of jaundice, dehydration, or feeding difficulties in the first 10 days of life in Pennsylvania since passage of the Newborns’ and Mothers’ Health Protection Act.
PATIENTS AND METHODS. Birth records from 407826 newborns ≥35 weeks’ gestation from 1998 to 2002 were merged with clinical discharge records. A total of 2540 newborns rehospitalized for jaundice, dehydration, or feeding difficulties in the first 10 days of life were then compared with 5080 control infants. Predictors of readmission were identified by using multiple logistic regression analysis.
RESULTS. An unadjusted comparison of baseline characteristics revealed numerous predictors of readmission. Subsequent adjusted analysis revealed that Asian mothers, those 30 years of age or older, nonsmokers, and first-time mothers were more likely to have a readmitted newborn, as were those with diabetes and pregnancy-induced hypertension. For neonates, female gender and delivery via cesarean section were protective for readmission, whereas vacuum-assisted delivery, gestational age <37 weeks, and nursery length of stay <72 hours were predictors of readmission in the first 10 days of life.
CONCLUSIONS. Although readmissions for jaundice, dehydration, and feeding difficulties may be less common for some minority groups and Medicaid recipients in the era of the Newborns’ and Mothers’ Health Protection Act compared with nonminorities or privately insured patients, several predictors of newborn readmission have established associations with inexperienced parenting and/or breastfeeding difficulty. This is one indication that this well-intentioned legislation and current practice may not be sufficiently protecting the health of newborns and suggests that additional support for mothers and newborns during the vulnerable postdelivery period may be indicated.
The Newborns’ and Mothers’ Health Protection Act (NMHPA) was passed by the US Congress in 1996, heralding a new era in the health care surrounding newborn delivery.1 Coverage for a 48-hour hospital stay after a vaginal delivery and a 96-hour stay after a cesarean section became mandatory for insurance providers, and since then, shorter stays occur only if a physician elects to discharge a mother and infant early after consultation with the mother. In Pennsylvania, minimum maternity-stay legislation was enacted on July 2, 1996, and became effective 60 days later, >1 year before the NMHPA provisions were due to start in January 1998.2 More than 40 other states have passed similar laws.3 In large part because of these national and local laws, the trend toward decreasing maternity and newborn nursery length of stay (LOS), declining for decades, was halted,4–6 and hope existed among clinicians and researchers that morbidities would lessen for these vulnerable populations.
Shortly after the passage of the NMHPA, which primarily addressed the maternity and nursery hospitalization, the postdischarge care of infants also became a focus of the federal government. In 1998, the Secretary of Health and Human Services’ Advisory Committee on Infant Mortality reported to Congress that all newborns require clinical evaluation and health promotion services on the third or fourth postnatal day.7 However, since passage of the NMHPA 10 years ago and the report to Congress, disturbing data have emerged suggesting that postdischarge care may actually have worsened for these vulnerable populations in recent years.8–11 Follow-up seems to be less timely and may put newborns at unnecessary risk for morbidity, necessitating study of risk factors for postnatal morbidity in the current health care climate.
One of the most frequently used variables in large epidemiologic studies to assess the adequacy of the newborn hospital LOS and follow-up is newborn readmission, particularly readmission for potentially preventable causes such as jaundice and feeding issues. The severity of these common neonatal conditions can often be lessened and readmission can be prevented by appropriate hospital care and neonatal follow-up with early detection and appropriate intervention.12–14
Given the changes in policy that occurred with the passage of the NMHPA, the publication of criteria for newborn discharge and follow-up by the American Academy of Pediatrics (AAP) in 1995 and 2004, and the AAP recommendations related to the support of breastfeeding,15–18 it is appropriate to assess the risk factors for potentially preventable newborn readmissions because of jaundice, dehydration, and feeding difficulties in the period since the legislation and guidelines have been put in place by using a diverse population not limited to a single hospital or health care insurance plan. As such, we sought to analyze predictors of newborn readmissions for these common causes in term or late preterm newborns born in Pennsylvania from 1998 to 2002, with the hypothesis that neonates with short LOS, late-week nursery discharge, and Medicaid would be at greater risk for readmission. We also suspected that those born to first-time, minority, and unmarried mothers would be at risk for readmission.
A total of 521656 birth records from singleton newborns ≥35 weeks’ gestation born in Pennsylvania between 1998 and 2002 were obtained from the Pennsylvania Department of Health (PDOH) birth registry for retrospective analysis. A total of 418991 (80.3%) of these birth records were successfully matched with and then merged with the clinical discharge records for each newborn collected by the Pennsylvania Health Care Cost Containment Council (PHC4) through a matching procedure by using date of birth, gender, subject-reported race and ethnicity, zip code, death status, hospital, gestation, and birth weight as variables in the absence of a social security number. The only records selected for analysis were those with diagnosis-related groups 391 (normal newborn) and 388 (preterm newborn without major problems) at discharge. Newborns with significant problems that could result in a complicated, prolonged, or atypical newborn nursery stay such as respiratory distress syndrome, meconium aspiration syndrome, seizures, central nervous system anomalies, heart malformations, Down syndrome, and other chromosomal anomalies were excluded as was done in previous studies.9,12,19–22 Of note, phototherapy during the nursery stay did not change the newborn diagnosis-related group, was not coded differently, and therefore was not an exclusion criterion. Other variables that were exclusions included birth weight <1800 g, nursery stay longer than 7 nights, transfer to another hospital, or death during the nursery hospitalization. By excluding the neonatal morbidities described above, an atypical LOS would most likely be because of maternal rather than neonatal morbidity. This resulted in a cohort of 407826 newborns that were selected for subsequent analysis.
The Human Subjects Protection Office of the Pennsylvania State University College of Medicine approved this study.
The primary outcome for this study was readmissions related to jaundice, dehydration, or feeding problems in the first 10 days of life. These conditions were specifically chosen because as opposed to sepsis, congenital heart disease, or intestinal obstruction, appropriate postnatal follow-up and outpatient intervention can prevent a large majority of jaundice or feeding-related readmissions. Readmission diagnoses were determined from the data set through review of International Classification of Diseases, Ninth Revision, diagnosis codes. Ten days was chosen as the end point for this analysis, because the recommended assessment on days of life 3 and 4 is most likely to have an impact on readmissions in the first 10 days after birth.
The PHC4 data set also allowed for the identification of clinical discharge records for infants readmitted to the hospital in the first year of life for jaundice, dehydration, and/or feeding difficulties. Of 5064 total readmission records for jaundice and/or feeding problems in the first 10 days of life, 2540 (50.2%) were successfully matched to their birth record and hospitalization data, a percentage that is limited by the absence of social security numbers for neonates. For analysis purposes of the matched cohort of 2540 readmissions, 5080 controls (newborns not readmitted within 10 days of birth) were matched randomly to these cases based on the year and quarter so that there were 2 controls matched to every case.
In comparing the unmatched readmission records to the matched readmission records, no significant differences existed between groups in regard to patient gender, Medicaid status, or mean age at readmission. However, significantly fewer newborns of black mothers were in the matched cohort. This is likely because of the increased probability of these mothers living in a high population zip code, a variable that makes the matching of discharge records to birth records more difficult.
The merged data sets from the PDOH and PHC4 contained an extensive set of variables related to maternal health status and hospitalization, newborn health status and hospitalization, and socioeconomic variables. Data from the prenatal period included timing and extent of prenatal care, pregnancy weight gain, tobacco and alcohol use, and pregnancy complications including conditions such as hypertension, pregnancy-induced hypertension, and diabetes. Maternity and newborn hospital data included information on delivery type and complications, LOS in hours since delivery, infant birth weight, gender, gestational age, Apgar scores, and neonatal complications. Socioeconomic data included maternal race, ethnicity, age, parity, education, marital status, and insurance type. Information on newborn feeding type was not available.
To determine a set of independent variables to focus on in building a multiple logistic regression model for readmissions within 10 days of birth, a univariate analysis was conducted on the previous list of variables of interest from the matched data to test the unadjusted effect of each variable on readmissions within 10 days of birth. Categorical variables were summarized with frequencies and percentages. Their associations with readmissions within 10 days of birth were tested with χ2 tests and odds ratios (ORs) for nominal variables or with Cochran-Armitage trend tests for ordinal variables. Two-sample t tests with means and 95% confidence intervals (CIs) were used to test for differences in continuous variables in terms of readmissions within 10 days of birth. Pearson correlations, χ2 tests, and 1-way analysis of variance were used to determine what pairs of independent variables might be significantly correlated with each other and might be the cause of multicollinearity in a multiple regression analysis. Where possible, ORs and CIs are reported so that clinical significance can be judged with statistical significance.
A multiple logistic regression analysis was then performed for readmission within 10 days of birth by using a stepwise selection method to determine the most significant set of covariates. The initial set of possible covariates was comprised of all of the significant variables from the univariate analysis. Goodness-of-fit tests were used to assess the fit of the model at each step. Decisions regarding the inclusion of highly correlated variables such as birth weight and gestational age were made on the basis of their clinical utility with the most clinically meaningful or easily determined variable chosen for the final model. After finding a good model fit with the most significant set of covariates, which were all significant to the model at the .05 level, first-order interactions between pairs of the covariates remaining in the final model were also examined using the correlations found in the univariate analysis as an aid, but no interactions were determined to be significant to the model.
A total of 2540 newborns with readmissions in the first 10 days of life were identified who met the inclusion criteria for this analysis without meeting any exclusion criteria and had readmission discharge records that could be matched with the clinical discharge records from their birth hospitalization. The mean time to readmission was 111.4 ± 37.8 hours since birth and 61.9 ± 38.1 hours since nursery discharge. The primary diagnosis for readmission was jaundice-related for 2324 (91.5%) of the infants, with the remaining 216 (8.5%) primarily attributed to dehydration, feeding problems, and/or related electrolyte abnormalities.
Unadjusted Comparison With Control Group
When compared with the mothers of 5080 newborns that were not readmitted in the first 10 days of life, mothers of readmitted newborns were significantly more likely to be Asian or Pacific Islanders (OR: 1.60; 95% CI: 1.24–2.06) (Table 1). Maternal age was a significant predictor of readmission, with those <30 years of age less likely to have a readmitted newborn (OR: 0.85; 95% CI: 0.77–0.93). Primiparous (OR: 1.75; 95% CI: 1.59–1.93) and college-educated (P < .0001) women were also more likely to have a newborn readmitted in the first 10 days of life, but those insured by Medicaid were less likely than those privately insured to have a readmitted infant (OR: 0.78; 95% CI: 0.70–0.88).
Several features of the pregnancy history also were significantly associated with newborn readmission. Although the small minority of women with late or no prenatal care seemed less likely to have a readmitted newborn (P = .0001), mothers with diabetes (OR: 1.88; 95% CI 1.47–2.41), hypertension (OR: 1.56; 95% CI: 1.03–2.35), pregnancy-induced hypertension (OR: 1.67; 95% CI 1.33–2.09), and premature rupture of membranes (OR: 2.03; 95 CI: 1.61–2.57) were more likely to have their newborn readmitted in the first 10 days of life. Maternal tobacco use (OR: 0.51; 95% CI: 0.44–0.59) and alcohol use (OR: 0.59; 95% CI: 0.36–0.96) during pregnancy, however, seemed to be protective against this outcome.
Without adjusting for LOS, infants delivered by cesarean section were significantly less likely than those delivered vaginally to be readmitted (OR: 0.37; 95% CI: 0.31–0.43). Both forceps-assisted (OR: 1.60; 95% CI: 1.25–2.07) and vacuum-assisted deliveries (OR: 2.64; 95% CI: 2.28–3.07) were associated with an increased likelihood of readmission, but no data were reported regarding neonatal bruising or the presence of cephalohematomas.
For the newborns, females were significantly less likely to be readmitted than males (OR: 0.75; 95% CI: 0.68–0.82), and gestational age was a strong predictor of readmission. Compared with those delivered at 39 to 40 weeks, infants born at 35 to 36 weeks (OR: 5.96; 95% CI: 4.96–7.16) and 37 to 38 weeks (OR: 2.52; 95% CI: 2.26–2.81) were much more likely to be readmitted, whereas those born after 40 weeks were less likely to have a subsequent hospitalization (OR: 0.65; 95% CI: 0.54–0.79). Lower birth weight was also strongly associated with the likelihood of readmission, with those weighing <3000 g more likely to be readmitted for jaundice or feeding difficulties in the first 10 days of life (OR: 1.60; 95% CI: 1.43–1.79). Finally, although Apgar scores ≤8 at 1 minute (OR: 1.19; 95%: CI 1.08–1.31) and 5 minutes (OR: 1.39; 95% CI: 1.14–1.70) and day of the week of birth (P = .0149) were associated with an increased chance of readmission, the day of the week of discharge was not a predictor of readmission. More specifically for the day of the week of birth, those born on weekdays were significantly less likely to be readmitted than those with weekend (Saturday and Sunday) births (OR: 0.86; 95% CI: 0.77–0.96), but although 17.5% of newborns were delivered by cesarean section on weekdays, only 9.0% were delivered operatively during weekends (P < .0001).
LOS Analysis and Subgroup Analysis by Delivery Type
In examining the entire cohort, nursery LOS was a significant predictor of readmission in the first 10 days of life. Infants that remained in the nursery for <72 hours were significantly more likely to be readmitted (OR: 2.51; 95% CI: 2.09–3.02) than those with LOS of ≥72 hours. Similarly, those with LOS <48 hours were more likely to be readmitted than those staying ≥48 hours (OR: 1.17; 95% CI: 1.06–1.29). However, differences in the median LOS after vaginal delivery (46.5 ± 11.7 hours) and cesarean delivery (75.5 ± 16.9 hours) led to a subgroup analysis by delivery type, which reveals that mode of newborn delivery confounds the LOS variable (Table 2). After a vaginal delivery, newborns with LOS <48 hours are not significantly more likely to be readmitted (OR: 0.92; 95% CI: 0.83–1.02) than those newborns staying ≥48 hours; however, a stay of <72 hours after a cesarean section is a significant predictor of readmission (OR: 1.44; 95% CI: 1.07–1.96) when compared with those with stays of ≥72 hours.
The significant variables retained in the final multiple regression model were adjusted for the other variables remaining in the model and were all associated with an increased risk of hospitalization for jaundice and/or feeding issues in the first 10 days of life (Table 3). Although first-time mothers were significantly more likely to have a readmitted newborn compared with experienced mothers (OR: 1.58; 95% CI: 1.41–1.76), mothers <30 years of age were less likely to have a newborn readmitted than mothers 30 years of age or older (OR: 0.80; 95% CI: 0.72–0.90). Mothers of Asian or Pacific Island ancestry were also more likely to have a readmitted newborn than those from other races (OR: 1.51; 95% CI: 1.15–1.98).
Several pregnancy and delivery variables also were predictors of readmission. Neonates born to mothers with diabetes (OR: 1.92; 95% CI: 1.46–2.54) or pregnancy-induced hypertension (OR: 1.47; 95% CI: 1.15–1.88) were more likely to have a second hospitalization in the first 10 days of life, but those born to tobacco-using mothers were less likely to be readmitted (OR: 0.55; 95% CI: 0.47–0.65). After adjusting for other covariates including LOS, delivery via cesarean section remained protective for readmission (OR: 0.45; 95% CI: 0.36–0.56), whereas infants who had vacuum-assisted deliveries were more likely to require a subsequent hospital stay (OR: 2.08; 95% CI: 1.77–2.45).
Female infants were less likely than male infants to be readmitted for jaundice and/or feeding issues (OR: 0.76; 95% CI: 0.69–0.84), and preterm newborns <37 weeks’ gestational age at birth were much more likely to have a second hospitalization in the first 10 days of life (OR: 4.40; 95% CI: 3.65–5.31). Finally, after adjustment for all covariates, nursery LOS <72 hours remained a predictor of readmission for the often preventable conditions that are the subject of this study (OR: 1.35; 95% CI: 1.04–1.74).
The NMHPA may be the most well-known law affecting perinatal health care in the United States, and it has had a clear impact on the maternity and nursery LOS and the economics that surround the birth hospitalization.8,10,14,21,23–28 It has been argued that increasing the length of brief postpartum hospitalizations is a cost-effective intervention.26 Alternatively, the costs of these longer stays are significant as evidenced by the estimated increase of up to $20 million per year in hospital charges in the city of Philadelphia alone.24 Given the importance of maternal and newborn health care in our society, these costs are likely to be judged as reasonable and worthwhile if the quality of care that is delivered as a result of a longer hospital stay is improved. Although a recent report from California showed that newborn LOS has increased and all-cause newborn readmissions have decreased since passage of postpartum discharge legislation,29 unfortunately according to some reports, the quality of some aspects of perinatal health care seem to have worsened in the last decade, leading some to question whether the law is actually protective for newborns and mothers.8,30 Although few would argue that a law that guarantees insurance coverage for a minimum hospital stay is itself harmful, it alone is not sufficient to ensure the best possible health care in the perinatal period. Thus, it is relevant to examine aspects of the maternity and neonatal course during the NMHPA era that are risk factors for morbidities that might be prevented with appropriate intervention.
The current investigation evaluated risk factors for potentially preventable causes of newborn readmission in the first 10 days of life in Pennsylvania from 1998–2002, a period of relative stability in terms of legislation, policy statements, and insurance coverage affecting the maternity and nursery hospital stays. The population chosen for this evaluation was selected to represent a typical “well-infant” newborn nursery by excluding those infants with congenital anomalies, significant prematurity, or birth weight incompatible with sustained thermoregulation. This diverse population with a large sample size of mothers and newborns allowed us to identify numerous predictors of early life readmission, many of which are consistent with previous investigations that evaluated populations before the passage of NMHPA or consisted of populations from a single hospital or health care insurance plan. This analysis also benefited from having the nursery LOS in hours after delivery as opposed to most other studies with large data sets that had to approximate LOS.
Similar to other reports, we found that primiparous mothers were more likely to have a newborn readmitted to the hospital in the first 10 days of life for jaundice and/or feeding problems.12,19,22,31–33 Primiparous mothers are known to have delayed onset of lactogenesis, which could contribute to jaundice and feeding problems during early life.34–37 It is noteworthy that several other unadjusted predictors of readmission such as being married and more educated are associated with an increased likelihood of breastfeeding13,22,38–42 when compared with reference groups, whereas apparently protective variables such as being a Medicaid recipient or a smoker are inversely related to breastfeeding likelihood and have previously been shown as negative predictors of readmission.13,43–48 In addition to an increased likelihood of breastfeeding, newborns of married and more-educated mothers may have an increased risk of readmission because of ascertainment bias, because disparities in timely follow-up for disadvantaged and Medicaid recipients have been reported and may lessen their risk for readmission.9,32,33
Mothers who are of Asian or Pacific Island descent were also more likely to have a newborn readmitted within 10 days, as were those with diabetes and pregnancy-induced hypertension. Asian infants are known to have higher bilirubin values, and this has previously been associated with increased readmission rates.33,39,40,49,50 Infants of diabetic mothers have also been shown to be a high-risk group in the past.13,40,51–53 Particularly for large-for- gestational-age infants born to diabetic mothers, these newborns have been shown to have an increased risk of jaundice because of polycythemia and an increased amount of red blood cell turnover.51–54 In contrast, we are unaware of previous reports showing an association between pregnancy-induced hypertension and an increased likelihood for newborn readmission. It is possible that the treatment of this maternal condition using magnesium sulfate results in less effective breastfeeding and also slows gastrointestinal motility in the neonate resulting in an increase in the enterohepatic circulation of bilirubin.55
For the newborns, male gender and earlier gestational age at birth were associated with an increased likelihood of early life readmission. These data are similar to those previously shown and highlight the increased attention that should be paid to those born near term or late preterm.13,22,31–33,39,56 The current guidelines only provide guidance for infants born at ≥38 weeks’ gestation, and those born at ≤37 weeks are a group that is becoming increasingly prevalent and also are a group more commonly cared for in newborn nurseries as opposed to NICUs. Our findings of an increased association between vacuum-assisted delivery and an increased risk of readmission are probably related to neonatal bruising and/or cephalohematoma risk.39,57–60 Contrary to our hypothesis, day of the week of delivery and discharge were not predictors of neonatal readmission in the adjusted analysis.
For the nursery LOS, the current findings in the era of the NMHPA seem somewhat different from those previously reported. Although we report a protective effect in general for infants with LOS of >72 hours, subgroup univariate analyses by delivery type help to further characterize this variable. For those born vaginally, a LOS of <48 hours seems to be no different from a stay of ≥48 hours. One possible explanation for this is that a selection or indication bias occurs where those newborns and mothers with the least complicated courses and less feeding difficulty are more likely to be discharged <48 hours after delivery. In contrast, those with stays longer than 72 hours may be more likely to have phototherapy started during the nursery stay, which could decrease their risk for readmission.
In contrast to those delivered vaginally, our data suggest that newborns with stays <72 hours after a cesarean section are at greater risk for readmission than those staying ≥72 hours. Some of this increased risk may be attributable to breastfeeding difficulties, because lactogenesis was shown to occur later for women after cesarean delivery either for physiologic reasons or because of a delay in the initial feeding after surgery.34–36,61–64 In addition, longer inpatient hospital stays have been associated with breastfeeding success after cesarean delivery.65 Notably, and just as for the case of the late preterm newborn, the AAP guidelines on the care of the newborn also do not address infants born via cesarean section, a cohort that currently represents between 25% to 30% of the population. Future versions of this guideline should consider this large subgroup in their recommendations.
A major limitation of this analysis is the lack of information regarding infant feeding. As noted above, breastfeeding has a strong relationship with many of the variables included in the analysis and may be an important confounder when interpreting these data. The second major limitation of this report is the absence of data related to the timing and extent of newborn follow-up both in the outpatient and home settings. The timing of follow-up may either reduce readmissions if jaundice and/or feeding problems are caught early enough or increase readmissions because of the previously mentioned ascertainment or detection biases.10,11,14,22,23,50 In addition, although evidence-based guidelines for the management of hyperbilirubinemia existed during the period under evaluation and have recently been modified,66,67 selection bias certainly contributes to readmission rates for jaundice, as noted by Edmonson et al22, both because of provider preferences and family characteristics.
One additional limitation is our inability to compare readmission rates before and after passage of postpartum discharge legislation in Pennsylvania because of the nature of our data set. However, 2 other reports have examined this variable. Data from Ohio’s Medicaid recipients did show modest improvements in jaundice-related readmissions in the postlegislation period,23 but the recent report using California outcomes showed no significant change in readmissions for jaundice between the prelegislation and postlegislation periods.29
Finally, the use of neonatal readmission as a primary outcome has advantages and disadvantages, which have been extensively reviewed and discussed. As described by Malkin and colleagues,68 readmission is useful because it reflects morbidity, is costly, and can be identified relatively easily. It has also been described as a proxy for other perinatal morbidities.13,69 Madden et al11 have cautioned, however, that many factors, including follow-up services and outpatient management of jaundice and feeding problems, can affect rehospitalizations. Most agree, however, that preventing readmissions is a desirable goal and can prevent stress for a family during the particularly vulnerable period surrounding childbirth. In addition, because breastfeeding is a desirable outcome to most providers, it should be noted that readmission has been associated with cessation of breastfeeding.70
Despite the limitations of this study, several predictors of readmission have been identified that have established associations with inexperienced parenting and/or breastfeeding. These predictors persist despite well-intentioned legislation and guidelines for care, and future public policies and national guidelines could help newborns and their mothers by specifically addressing the common occurrences of late preterm birth and delivery by cesarean section. For individual practitioners, practice patterns can be modified with persistent reminders of the need for change as evidenced by the “Back to Sleep” campaign and the advice given to prevent sudden infant death syndrome. Unfortunately, practice related to newborn care and follow-up has been comparatively resistant to change. Although clinicians may prove to have the ability to modify their practices to accommodate the needs of newborns and new mothers, if practice patterns remain relatively unchanged, alternative practices such as routine use of the widely available service of home nurse visitation should be considered.14,27
Perinatal health care has likely benefited overall from the NMHPA, but deficiencies are evident as demonstrated by the results of this study. Ten years after this important legislation was passed, it is time to consider whether this law and the current guidelines for care are sufficient in fully addressing the diverse population of newborns and mothers and the preventable problems that may occur during the perinatal period. Given the increasing prevalence of breastfeeding, late preterm birth, and cesarean delivery, it is appropriate to now consider what practical and available early intervention strategies for newborns and families can be implemented to best reduce morbidity, increase the chances of successful breastfeeding, and reduce disparities in health care for newborns and mothers during the vulnerable perinatal period.
Dr Paul is supported by the Health Resources and Services Administration and the Maternal Child Health Bureau through grant R40 MC06630. A Penn State College of Medicine Dean’s Feasibility Grant awarded to Dr Paul also supported this work.
Coauthor Mr Lehman had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. We thank Edward Hain from the PHC4 for extensive technical assistance. Additional assistance was provided by Craig Edelman from the Pennsylvania Department of Health.
The PHC4, who provided the clinical discharge records, is an independent state agency responsible for addressing the problem of escalating health costs, ensuring the quality of health care, and increasing access to health care for all citizens regardless of ability to pay. The PHC4 provided their data in an effort to further their mission of educating the public and containing health care costs in Pennsylvania. The PHC4, its agents, and staff have made no representation, guarantee, or warranty, express or implied, that the data (financial, patient, payor, and physician-specific information provided to this entity) are error-free or that the use of the data will avoid differences of opinion or interpretation. The authors of this article conducted this analysis without the assistance of the PHC4, which bears no responsibility or liability for the results of the analysis.
- Accepted August 24, 2006.
- Address correspondence to Ian M. Paul, MD, MSc, Penn State College of Medicine, Pediatrics H085, 500 University Dr, Hershey, PA 17033. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
- ↵Newborns’ and Mothers’ Health Protection Act of 1996. Pub L No. 104–204, §601 (1996)
- ↵Health Security Act, Bill No 1977 (Session of 1995), General Assembly of Pennsylvania (1996)
- Kozak LJ, Hall MJ, Owings MF. National Hospital Discharge Survey: 2000 annual summary with detailed diagnosis and procedure data. Vital Health Stat 13.2002;153 :1– 194
- ↵Kozak LJ, Owings MF, Hall MJ. National Hospital Discharge Survey: 2002 annual summary with detailed diagnosis and procedure data. Vital Health Stat 13.2005;158 :1– 199
- ↵Eaton AP. Early postpartum discharge: recommendations from a preliminary report to Congress. Pediatrics.2001;107 :400– 403
- ↵Galbraith AA, Egerter SA, Marchi KS, Chavez G, Braveman PA. Newborn early discharge revisited: are California newborns receiving recommended postnatal services? Pediatrics.2003;1112 :364– 371
- ↵Madden JM, Soumerai SB, Lieu TA, Mandl KD, Zhang F, Ross-Degnan D. Length-of-stay policies and ascertainment of postdischarge problems in newborns. Pediatrics.2004;113 :42– 49
- ↵Maisels MJ, Kring E. Length of stay, jaundice, and hospital readmission. Pediatrics.1998;101 :995– 998
- ↵Paul IM, Phillips TA, Widome MD, Hollenbeak CS. Cost-effectiveness of postnatal home nursing visits for prevention of hospital care for jaundice and dehydration. Pediatrics.2004;114 :1015– 1022
- ↵American Academy of Pediatrics, Committee on Fetus and Newborn. Hospital stay for healthy term newborns. Pediatrics1995;96 :788– 790
- American Academy of Pediatrics, Committee on Fetus and Newborn. Hospital stay for healthy term newborns. Pediatrics.2004;113 :1434– 1436
- Gartner LM, Morton J, Lawrence RA, et al. Breastfeeding and the use of human milk. Pediatrics.2005;115 :496– 506
- ↵American Academy of Pediatrics, Work Group on Breastfeeding. Breastfeeding and the use of human milk. Pediatrics.1997;100 :1035– 1039
- ↵Meara E, Kotagal UR, Atherton HD, Lieu TA. Impact of early newborn discharge legislation and early follow-up visits on infant outcomes in a state Medicaid population. Pediatrics.2004;113 :1619– 1627
- ↵Malkin JD, Keeler E, Broder MS, Garber S. Postpartum length of stay and newborn health: a cost-effectiveness analysis. Pediatrics. 2003;111(1). Available at: www.pediatrics.org/cgi/content/full/111/4/e316
- ↵Datar A, Sood N. Impact of postpartum hospital-stay legislation on newborn length of stay, readmission, and mortality in California. Pediatrics.2006;118 :63– 72
- ↵Hyman DA. What lessons should we learn from drive-through deliveries? Pediatrics.2001;107 :406– 407
- ↵Danielsen B, Castles AG, Damberg CL, Gould JB. Newborn discharge timing and readmissions: California, 1992–1995. Pediatrics.2000;106 :31– 39.
- Dewey KG, Nommsen-Rivers LA, Heinig MJ, Cohen RJ. Risk factors for suboptimal infant breastfeeding behavior, delayed onset of lactation, and excess neonatal weight loss. Pediatrics.2003;112 :607– 619.
- ↵Chen DC, Nommsen-Rivers L, Dewey KG, Lonnerdal B. Stress during labor and delivery and early lactation performance. Am J Clin Nutr.1998;68 :335– 344
- ↵Cooper WO, Atherton HD, Kahana M, Kotagal UR. Increased incidence of severe breastfeeding malnutrition and hypernatremia in a metropolitan area. Pediatrics.1995;96 :957– 960
- ↵Maisels MJ, Gifford K, Antle CE, Leib GR. Jaundice in the healthy newborn infant: a new approach to an old problem. Pediatrics.1988;81 :505– 511
- ↵Adams JA, Hey DJ, Hall RT. Incidence of hyperbilirubinemia in breast- vs. formula-fed infants. Clin Pediatr (Phila).1985;24 :69– 73
- Linn S, Schoenbaum SC, Monson RR, Rosner B, Stubblefield PG, Ryan KJ. Epidemiology of neonatal hyperbilirubinemia. Pediatrics.1985;75 :770– 774
- ↵Escobar GJ, Greene JD, Hulac P, et al. Rehospitalisation after birth hospitalisation: patterns among infants of all gestations. Arch Dis Child.2005;90 :125– 131
- ↵Peevy KJ, Landaw SA, Gross SJ. Hyperbilirubinemia in infants of diabetic mothers. Pediatrics.1980;66 :417– 419
- Stevenson DK, Ostrander CR, Cohen RS, Johnson JD, Schwartz HC. Pulmonary excretion of carbon monoxide in the human infant as an index of bilirubin production: Iib. Evidence for the possible effect of maternal prenatal glucose metabolism on postnatal bilirubin production in a mixed population of infants. Eur J Pediatr.1981;137 :255– 259
- ↵Berk MA, Mimouni F, Miodovnik M, Hertzberg V, Valuck J. Macrosomia in infants of insulin-dependent diabetic mothers. Pediatrics.1989;83 :1029– 1034
- ↵Maisels MJ. Jaundice. In: MacDonald MG, Seshia MMK, Mullett MD, eds. Neonatology: Pathophysiology and Management of the Newborn. Philadelphia, PA: Lippincott Co; 2005:768–846
- ↵Evans KC, Evans RG, Royal R, Esterman AJ, James SL. Effect of caesarean section on breast milk transfer to the normal term newborn over the first week of life. Arch Dis Child Fetal Neonatal Ed.2003;88 :F380– F382
- ↵American Academy of Pediatrics, Provisional Committee for Quality Improvement and Subcommittee on Hyperbilirubinemia. Practice parameter: management of hyperbilirubinemia in the healthy term newborn. Pediatrics.1994;94 :558– 565
- ↵Management of hyperbilirubinemia in the newborn infant 35 or more weeks of gestation. Pediatrics.2004;114 :297– 316
- ↵Britton JR, Britton HL, Beebe SA. Early discharge of the term newborn: a continued dilemma. Pediatrics.1994;94 :291– 295
- Copyright © 2006 by the American Academy of Pediatrics