ATLANTA – An adjustment in the culture reporting schedule at Texas Children’s Hospital, Houston, helped reduce the average length of stay for neonatal fever from 48 to 43 hours, without increasing readmissions for serious bacterial infections, according to a review presented at the Pediatric Hospital Medicine meeting.
M. Alexander Otto/MDedge News
Dr. Huay-Ying Lo
Investigators there were working to meet the goals of the Reducing Excessive Variability in Infant Sepsis Evaluation project (REVISE), a national collaboration aimed at improving care. One of the goals is to reduce the length of stay (LOS) for neonatal fever to fewer than 30 hours for low-risk infants and fewer than 42 hours among high-risk infants.
The traditional standard is to keep children in the hospital for 48 hours to rule out sepsis, but that thinking has begun to change amid evidence that blood cultures generally do not need that long to turn positive, among other findings, said investigator Huay-Ying Lo, MD, a pediatrician at Texas Children’s.
“At our institution,” which admits more than 200 NF cases annually, “we have order sets for neonatal fever, and we’re actually doing pretty well” meeting most of the REVISE goals, “so we decided to focus on reducing length of stay,” she said at the meeting.
Evidence of the safety and cost savings of earlier discharge was presented to providers, and weekly emails reminded them of the early discharge goal and updated them on the current average LOS for NF.
Dr. Lo and her team also brainstormed with providers to identify problems. “One of the barriers they consistently mentioned was the timing of cultures being reported out from the microbiology lab. A lot of time, people were just waiting for the report to say no growth for 36 hours or whatever it was going to be,” she said.
That led to talks with the microbiology department. Blood cultures were already automated, so there wasn’t much that needed to be done. Urine cultures were read manually three to four times a day after the initial incubation period. However, after an initial Gram stain, CSF cultures were read manually only one or two times a day – whenever somebody had time. The hours were random, and sometimes results were not reported until the evening, which meant the child had to spend another night in the hospital.
The lab director agreed that it was a problem, and standardized procedures to read cultures twice a day, at 7 a.m. and 2 p.m. “The times we agreed upon; 7 a.m. works really well for morning discharge, and at 2 p.m., the day team is still there and can get kids out that day,” Dr. Lo explained.
. Among infants 7-60 days old admitted with NF – excluding ill-appearing children and those with comorbidities that increased the risk of infections – the mean LOS fell from 48 hours among 144 infants treated before the intervention, to 43 hours among 157 treated afterward (P = .001), and “we didn’t have any more readmission for serious bacterial infections,” Dr. Lo said.
“We want to reduce it further. If we get to 42 hours, we’ll be pretty happy.” Updating discharge criteria, and letting providers know how their LOS’s compare with their peers’ might help. “I’m sure some people are more conservative and some a little more liberal,” she said.
There was no industry funding for the work, and the investigators had no disclosures. The meeting was sponsored by the Society of Hospital Medicine, the American Academy of Pediatrics, and the Academic Pediatric Association.