Improvement in Early Recognition of Deteriorating Pediatric Bone Marrow Transplant Patients

Background: Hematopoietic stem cell transplant (HSCT) is a curative treatment for children with a variety of malignant and non-malignant disorders. Approximately 100 stem cell transplants are performed at Texas Children’s Hospital annually. HSCT is associated with a high rate of treatment-related complications. Some of these complications can lead to a sudden deterioration of the patient on the bone marrow transplant (BMT) unit. Early recognition of these patients could improve outcomes when codes and intubation on the BMT unit can be avoided. In 2010, 18 rapid response team (RRT) calls and three codes were called on the BMT unit.

 Purpose: The purpose of this quality improvement project was to reduce the number of codes on the BMT unit to zero by implementing the Pediatric Advance Warning Scoring system (PAWS) and the BMT PAWS algorithm for early recognition of the deteriorating BMT patient. 

 Methods: Quality improvement techniques, such as Plan, Do, Study, Act (PDSA) cycles were used to plan, implement and evaluate this quality improvement initiative. During the first PDSA cycle, the current hospital PAWS scoring and algorithm tools were reviewed; a BMT specific PAWS algorithm was developed; and an educational offering piloted. The educational offering consisted of six case scenarios reflective of the BMT patient population. The second PDSA cycle included finalizing and initiating training of BMT staff, launching the program and monthly monitoring of compliance with the scoring tool and algorithm. Subsequent PDSA cycles included monitoring compliance with use of the tools, debriefings with staff after RRTs and sharing outcomes with others throughout the BMT program and institution.

 Outcomes: The PAWS scoring tool and algorithm accurately reflected deteriorating BMT patients based on the initial pilot with 14 staff members of the BMT multidisciplinary team. The revised educational offering was piloted with 8 charge nurses who all passed the post-test. All multidisciplinary staff that attended the mandatory inservice met the post-test passing score of 80%. Monthly compliance monitoring of the inpatient charts revealed an average compliance rate of 85% by bedside nurses. RRT debriefings revealed multiple opportunities for improvement by various team members. Improvement opportunities included scoring accuracy, documentation of compliance with algorithm interventions and engagement by medical staff. Over time, documentation of events leading to RRTs was improved by the nursing staff.

 Conclusions: The PAWS scoring tool and algorithm accurately reflected the deteriorating BMT patient. Twenty-five RRTs were called for deteriorating BMT patients during the 11 month period following implementation of the program. Thirteen (52%) of the RRTs called resulted in patient transfers to the pediatric intensive care unit. During this same time period no codes were called on the BMT unit.