Early Discharge From the Newborn Nursery—Effect on Scheduling of Follow-up Visits by Pediatricians

DISCUSSION

Our educational intervention—a presentation at the monthly staff meeting in January 1995 and a letter sent to all pediatricians in April 1995—appeared to have an effect. By July 1995 there was a significant increase in the percent of short-stay infants scheduled to be seen by the physician within 3 days of discharge. (We cannot, however, rule out the possibility that other factors also influenced the observed change in practice although the most recent AAP statement on this subject was only published in October 1995⁶). Nevertheless, in November 1995 and March 1996, only 58% and 62% of short-stay infants respectively, were scheduled to be seen within 3 days of discharge.

These infants were patients of private pediatricians and part of a relatively affluent, suburban population where transportation and access to medical care are readily available. Although there was significant improvement, it is discouraging that, despite these advantages and our educational efforts, one third of short-stay infants discharged in March 1996 were scheduled for their first follow-up visit 14 days after discharge.

In June and July 1996, we contacted 12 pediatricians who did not consistently follow the AAP guidelines during November 1995 or March 1996. When asked why, they gave a variety of reasons. Four physicians indicated that (now) they were following the guidelines (although they had not requested appropriate follow-up at the time of the survey). Six said that if they knew the mother and family well, or had taken care of previous siblings, they were confident that the mother would notify them if she were concerned. In one practice, a nurse calls every mother within 1 to 2 days of birth and, in addition to other questions, specifically asks about the presence of jaundice. When necessary, an office visit is arranged. One physician, whose group had changed their practice from the traditional return visit at 2 weeks to a visit within 1 to 2 days of discharge after a short-stay, found this to be a rewarding experience. He felt that he and his partners had been able to identify problems early (particularly feeding problems and jaundice) and intervene successfully.

How effective are the AAP/ACOG recommendations likely to be? Even if desired, this type of follow-up in rural locations will be quite difficult, if not impossible, to achieve. Parents may have to travel many miles to visit physicians and, depending on weather and road conditions, this may not be possible. Compliance in indigent populations is also likely to be low. A home visit by a nurse is an excellent way to provide follow-up but is not routinely available in most communities and, in our experience, is not accepted by some families. It is likely that, nationally, there will be wide variations in compliance with the AAP recommendations. The most important question—whether or not failure to follow these guidelines leads to an increase in morbidity—cannot be answered until satisfactory studies are performed.

We have documented a significant change in pediatric practice although physicians are generally reluctant to change long-standing practice habits. In this situation some reluctance is not surprising, as in normal newborns the incidence of significant adverse events (such as severe hyperbilirubinemia and hypernatremic dehydration) is low. We reviewed 29 934 infants discharged from our well baby nurseries between December 1988 and November 1994. A total of 127 infants (4.2 per thousand) were readmitted with hyperbilirubinemia and 5 with dehydration or failure to thrive.⁷ Thus, the risk of these morbidities in any one physician's practice is very small.

Although one of the major risks of early discharge appears to be the development of significant neonatal jaundice, the presence of known risk factors for jaundice did not appear to influence pediatricians' decisions regarding the timing of a follow-up visit. It has long been a practice in this institution to do blood typing on the infants of all group O mothers, thus allowing the pediatrician to be aware of the potential for ABO incompatibility and possible hemolytic disease. In our chart review, we found that when the mother was group O only 27% of records indicated that the pediatrician was aware of the infant's blood type. Thus, this information seemed to play little role in the subsequent decision for follow-up. A recent study suggests that routine blood typing of cord blood is not cost-effective and is of questionable value⁸ and this practice has been discontinued at our hospital.

We recognize that this study surveyed practitioners from a single hospital and thus has limited generalizability. However, informal discussions with colleagues in other cities in the east and midwest suggest that these practice patterns are similar to those elsewhere. We need reliable data documenting the impact of shorter nursery stays on neonatal morbidity. If these data support an increase in morbidity, further attempts are necessary either to change existing practice patterns or to introduce new methods of providing surveillance for newborn infants who are no longer in the hospital.