KEY POINTS
- 1.
The quality of clinical care and improvement is typically measured in six areas: safety, effectiveness, patient-centeredness, timeliness, efficiency, and equality.
- 2.
Healthcare administered during pregnancy, labor, and childbirth continues to show considerable center-to-center variability and needs careful study both for improving outcomes and also to reduce the possibility of harm.
- 3.
The care of premature and critically ill infants in neonatal intensive care units involves high-risk skilled procedures that need careful analysis.
- 4.
Quality improvement (QI) priorities need to be relevant to ill infants. The plan-do-study-act methodology is important to evaluate and refine treatment efforts.
- 5.
The distinction between research and QI needs to be understood, with a clear distinction between activities that aim to improve the quality of care delivered to patients and those that may improve quality of care by contributing to generalizable knowledge.
General Considerations
Quality improvement (QI) has become a well-accepted science for improving healthcare quality across all settings. The Institute of Medicine’s six aims for a healthcare system have influenced QI efforts in the hospital environment, and neonatal intensive care units (NICUs) are poised to facilitate QI projects that promote safety, effectiveness, patient-centeredness, timeliness, efficiency, and equality. Several aspects of newborn care combine to form ideal conditions for the application of continuous quality improvement (CQI):
- •
Healthcare administered during pregnancy, labor, and childbirth encompasses some of the most common processes of care delivery in the hospital setting. The consistent need for maternal and newborn care during the perinatal period lends itself to the systematic testing of rapid cycle changes of CQI.
- •
Pregnancy and delivery involve the interaction between two distinct but dependent patients whose health and safety require unique consideration. Therefore, close collaboration and communication between maternal and neonatal providers is essential for care optimization.
- •
NICUs are complex organizations with providers from various specialties. NICU care involves high-risk skilled procedures, specialized equipment, and relative uncertainty of patient clinical status in the immediate postpartum period. Patient safety can be adversely impacted in this type of environment if processes are not reliable.
- •
Despite evidence to support best practices for pregnancy, labor, delivery, and neonatal care, notable variation in practice exists among hospitals.
- •
Malpractice claims involving obstetric and pediatric care more commonly lead to higher awards than for other medical specialties. Interventions that can improve healthcare delivery while reducing patient harm present an opportunity to reduce malpractice costs.
Methodology
The methods used to guide systematic rapid cycle changes in the NICU have a long history in industries outside of healthcare. Modern healthcare improvement draws on prior quality work developed in industries such as telecommunication, automotives, and manufacturing. The commonly used plan-do-study-act (PDSA) cycle ( Fig. 97.1 ), which has evolved during the course of the century, remains the cornerstone of QI methodology. Although there are several approaches and philosophies used in QI work in general, all of these techniques use the PDSA cycle at their core. The Associates in Process Improvement have adapted the PDSA cycle in their Model for Improvement approach, an approach championed by the Institute for Healthcare Improvement and adopted by many in healthcare QI work.
In addition to having a working knowledge of PDSA cycles, the following factors should be considered in any QI project.
Defining and Selecting the Problem to be Addressed
Having a clearly defined problem is an essential element of QI work. Defining the problem will allow project teams to describe shared missions and targets, focus ongoing work, and allow the team to determine whether the goals were met. One example of planning QI project(s) is shown in Fig. 97.2 . There are many potential problems that a healthcare team may encounter every day during routine care delivery, and leaders must prioritize the problems to be addressed through QI work. The severity of the problem, the impact expected from a potential solution, and the anticipated effort to solve the problem should all be considered when prioritizing problems.
Establishing a Project Team
Project teams should be established early in the QI project and should minimally include, or be endorsed by, a leader who can ensure that changes are implemented. Given that NICU care is provided by a wide range of medical specialists and support staff, it is important to consider who is impacted by the problem and who would be impacted by the changes that may occur as a result of the QI work. The project team should include representation from the aforementioned groups. As the team examines the processes associated with the problem to be addressed, additional impacted groups may be identified and added to the project team. Project team members should have unique and well-defined responsibilities to prevent the team from becoming unnecessarily large or inefficient.
Establishing a Mission and Target
Developing a mission statement at the start of a QI project can be a way to disseminate the purpose of the QI work among front-line staff, who may not be part of the project team. Mission statements can be less specific than target statements, with the intention of uniting individuals around the project. An example might be “to reduce noise in the NICU for patients, families and staff.” Targets are considerably more specific and should be constructed in a way that addresses the current and a desired state ( Fig. 97.3 ). A timeline also needs to be defined. To continue the example, a workable target statement might be “to reduce the number of nonactionable patient alarms from 80 alarms per 12-hour shift to 70 alarms per 12-hours shift over the course of 4 weeks by modification of the programmed alarm limits.” When developing targets, consideration should be given to how realistic it is to achieve the desired state, to alter the involved processes, and to achieve the goal desired state within the specified time frame.
Measures
To determine both the current state and changes to the current state, valid and consistent measurement is needed. Relying on measurements that are difficult to obtain can be prohibitive to a successful QI project. In the example of trying to reduce the number of nonactionable patient alarms, it would be virtually impossible to determine the frequency of alarms by direct, real-time observation. However, if one could easily query the patient monitors’ alarm history on a daily basis and dissect the data into 12-hour shifts the project becomes more feasible. Additionally, it is important that project teams consider all potentially relevant measures including process measures, outcome measures, and balance measures from the start. Evaluating these measures in the context of time through the use of annotated run charts or statistical process control charts will help identify significant changes in relation to interventions.
Prioritizing Changes
Processes associated with healthcare delivery are frequently complex and should be clearly delineated before intervening. There are often a multitude of potential changes that can be made to a process, or its subprocesses, when evaluating for improvements. There may also be a need to stratify the needed changes to narrow disparities ( Fig. 97.4 ). Team members may disagree on which changes will lead to the best outcome. Alternatively, team members may be overwhelmed by the magnitude of the process and have difficulty developing change ideas. When deliberating change ideas, consideration should be given to whether a reactive or fundamental change is needed, whether a change was tried and failed in the past, understanding the reason for prior failures, the time and effort required to implement the change, the potential for adversely impacting competing processes, the anticipated benefit gained from the change, and whether similar changes have been successful in similar environments. Tools such as driver diagrams, Pareto charts, cause and effect diagrams, flow diagrams, impact effort matrices, and others can help teams develop and organize potential changes to a process.
