April 1996

The Effect of Early Discharge and Other Factors on Readmission Rates of Newborns

Author Affiliations

From the Department of Pediatrics, St Joseph Mercy Hospital, Ann Arbor, Mich (Dr Soskolne and Ms Fyock); and the Departments of Pediatrics (Dr Schumacher) and Biostatistics (Drs Soskolne and Schork and Ms Young), University of Michigan Hospital, Ann Arbor.

Arch Pediatr Adolesc Med. 1996;150(4):373-379. doi:10.1001/archpedi.1996.02170290039006

Objective:  To assess the relationship between early discharge, breast-feeding, and other factors on hospital readmission of newborns.

Design:  Retrospective record review.

Setting:  An urban, private community hospital.

Patients:  All newborns born over a 1-year period who were readmitted to the hospital within the first 3 weeks of life (n=117). The control group consisted of a systematic sampling of newborns born over the same period who were not readmitted (n=147).

Results:  Early discharge, defined as discharge when younger than 24, 36, or 48 hours of age, does not seem to contribute to readmission. However, newborns whose initial stay was longer than 72 hours were at significantly lower risk for readmission (P=.02, χ2). Factors in the initial hospitalization associated with readmission included vaginal delivery and length of stay less than 72 hours (difference, 12 percentage points; 95% confidence interval [CI], 4% to 20%; P=.005), need for performance of a complete blood count (CBC) (difference, 16 percentage points; CI 6% to 26%; P=.002), presence of jaundice (difference, 17 percentage points; CI, 5% to 29%; P=.005), and gestational age 37 weeks or less (difference, 10 percentage points; CI, 2% to 18%; P=.02), discharge weight less than 3 kg (difference, 11 percentage points; CI, 0 to 22%; P=.05). However, almost all newborns delivered vaginally were discharged within less than 72 hours, so our ability to comment on the independent effect of delivery mode on readmission is limited. A trend toward significance was noted between breast-feeding and readmission (difference, 9 percentage points; CI, 0% to 18%; P=.07). However, when only vaginal deliveries were considered, this association was statistically significant (difference, 13 percentage points; CI, 4% to 22%; P=.02). A significant association was noted between breast-feeding and jaundice or dehydration. Of babies admitted with jaundice or dehydration, 94% were breast-fed, compared with 67% of babies admitted with neither jaundice nor dehydration (difference, 27 percentage points; CI, 13% to 41%; P<.001).

Conclusions:  These findings strongly suggest that early discharge (at younger than 24, 36, or 48 hours of age) from the hospital is not associated with hospital readmission within the first 3 weeks of life. Factors associated with readmission included breast-feeding, vaginal delivery and length of stay less than 72 hours, jaundice or need for a CBC, gestational age of 37 weeks or less, and discharge weight less than 3 kg.(Arch Pediatr Adolesc Med. 1996;150:373-379)