Comparative Effectiveness of Intravenous v. Oral Antibiotic Therapy for Serious Bacterial Infections

Overview

Some children get serious bacterial infections that require hospitalization and then a long course of antibiotics to completely treat the infection. Examples of these serious infections include ruptured appendicitis (when the appendix gets inflamed and bursts, releasing bacteria into the abdomen), complicated pneumonia (when an infected pocket of pus forms either in the lung or between the lung and chest wall), and osteomyelitis (an infection of the bone). To extend the duration of antibiotic therapy after discharge from the hospital, doctors will often insert a long catheter called a PICC line in the child's vein, which can stay in the body for several weeks. However, PICC lines require a fair amount of maintenance and training of caregivers in their use, require children to restrict their activities, and can lead to serious complications, such as blood stream infections and clots. An alternative to PICC lines is extending the duration of antibiotic therapy with oral antibiotics (pills or syrup) that achieve high levels of medicine in the blood and do not have the extra work, inconvenience and risks of PICC lines. Unfortunately, there are very few high quality studies that have demonstrated that oral antibiotics are just as good as intravenous antibiotics delivered via a PICC line, and so many doctors still recommend the PICC line treatment option. Also, no studies have been done to compare the impact of these two treatment options on the quality of life of the child and their caregivers. In this proposal the investigator outline a series of projects to compare oral antibiotics vs. intravenous antibiotics delivered via a PICC line in children who require prolonged (at least 1 week) home antibiotic therapy after hospitalization for three different serious bacterial infections: ruptured appendicitis, complicated pneumonia, and osteomyelitis. To see whether oral antibiotics are just as good as PICC lines, the investigators will use data collected from over 15,000 children with one of these three infections who were hospitalized at one of 43 US children's hospitals during the years 2009-2011, and determine whether PICC lines resulted in fewer rehospitalizations for treatment failure than oral therapy.

Background: When clinicians want to prescribe prolonged antibiotic therapy for children who have been hospitalized with a serious bacterial infection they have two options: (1) discontinue the intravenous antibiotics and discharge the child to receive a prolonged course of oral antibiotics at home; or (2) insert a central venous catheter (usually a peripherally inserted central catheter (PICC)), train the parents in care of the PICC and administration of intravenous antibiotics, and discharge the child to receive a prolonged course of intravenous antibiotics via the central venous catheter at home. These two options have major implications for the overall experience of the child and their caregiver, but there is a paucity of evidence on their comparative effectiveness to help clinicians, patients, and their caregivers make an informed choice. Objectives: In this proposal the investigators outline a series of projects to compare oral antibiotics vs. intravenous antibiotics delivered via a central venous catheter in children who require prolonged (at least 1 week) home antibiotic therapy after hospitalization for three different serious bacterial infections: perforated appendicitis, complicated pneumonia, and osteomyelitis. Methods: To compare the clinical effectiveness of these two treatment options, the investigators will conduct three separate retrospective comparative effectiveness studies for each of these diseases using detailed electronic data on diagnoses, procedures, laboratory tests, and orders for drugs and devices supplemented by limited chart review from 43 free standing children's hospitals that are members of the Children's Hospital Association (CHA) (study years 2009-2011, total sample size will exceed 15,000 children). The primary outcome will be rehospitalization for treatment failure. The investigators will use conventional and propensity-score based methods, including matching, to estimate and compare the effect of the two treatment options on patient outcomes. These comparisons will have the goal of asking about the effect of therapy as if the patients had been randomized to one of the two treatment options.