Patient Safety and Quality Improvement: An Overview of QI

As health-care delivery systems become increasingly complex, ongoing quality improvement becomes increasingly important. Pediatricians must become familiar with tools that have been developed to facilitate quality improvement, including the model for improvement, Lean and Six Sigma methodology, root case analysis, and failure modes and effects analysis.After completing this article, readers should be able to: This is the second in a series of articles reviewing the topics of patient safety and quality improvement in pediatrics. (1)Quality improvement (QI) is an essential component of 21st century medicine. As health-care delivery systems become increasingly complex, ongoing QI becomes increasingly important. QI and patient safety are inseparable disciplines. Health-care organizations often use QI methodologies to create safer systems and implement safer practices. The pediatric provider needs to be familiar with basic QI principles, tools, and methods to improve clinical care, patient safety, efficiency, and the overall patient experience.A central concept in QI is implementation of incremental change, followed by measurement of the effects of these changes over time. (2)(3) Changes in health care occur within complex organizational and social systems as well as in comparatively smaller office settings. The impact of a given QI initiative can therefore vary greatly across health-care environments. Data generated by changes in process need to be tracked to assess how both anticipated and unanticipated factors affect outcomes. Careful review of such data enables clinicians to assess which strategies are successful and which ones are not.At minimum, baseline and postimplementation measurements generally are necessary for determining the impact of any initiative. Many projects require multiple data points collected over time. Defining specific measures is critical to any QI project but can be challenging. For example, it is difficult to measure adverse events that were averted.There are two basic types of measures: process measures and outcome measures. A process measure quantifies compliance with the practices being implemented as part of an initiative. An outcome measure gauges the impact of the initiative on a patient population or health-care system. Process drives outcome. The process measures for the project can be viewed as the independent variables and the outcome measures as the dependent variables.Most QI projects strive to achieve high compliance with process changes to achieve the maximum effect on outcome measures. Understanding the benefits of continuous QI within health-care organizations relies on the ability to measure outcomes. (4) When possible, outcomes should be evaluated in terms of clinical effectiveness, financial impact, and patient and family satisfaction.For example, an office practice wants to improve the care they provide to patients who have asthma. Their goal is to reduce the frequency of severe asthma exacerbations. The clinicians plan to achieve this goal by providing a detailed asthma action plan to every patient who has asthma in their practice. A process measure for this project might be the percentage of patients who have an action plan documented in their medical records, with the numerator being the number of patients having an action plan and the denominator being the total number of patients in the practice who have asthma.A clinical outcome measure for this initiative might be the reduction of emergency department visits or inpatient hospitalizations for asthma exacerbations once adequate compliance is achieved; specifically, a threshold number or percentage of patients having asthma action plans in their charts.From a financial perspective, reduction in emergency department visits or inpatient hospitalizations could be translated into decreased costs to the health-care system as well as averted co-pays, which are becoming increasingly expensive. The parents’ comfort level with their child’s asthma management after implementation and review of the action plan could help to assess satisfaction.There are instances in which measuring outcomes is impractical or cannot be done for extended periods of time after an initiative has been implemented. In such cases, tracking process measures and improving compliance can be a reasonable approach. Figure 1 provides a simple template that has been successfully used to develop and implement QI initiatives. More comprehensive tools are available through organizations such as the Institute for Healthcare Improvement.Implementing change can sometimes introduce unwanted outcomes. A balancing measure is a variable that can be used to assess a potential negative consequence of a process change. For example, an inpatient unit implements a project to reduce the number of hospital readmissions. A balancing measure for this initiative might be the length of stay, which could increase if patients are observed longer to be certain they are ready for discharge and will not need to be readmitted.Just as a handyman’s toolbox must contain a hammer, a screwdriver, and a pair of pliers, so too must a QI toolbox contain some fundamental tools for the clinician to utilize. The choice of tools for a given situation will depend on the nature of the issue to be addressed. Although every clinician does not necessarily need to be expert with each tool, it is important that she or he at least be familiar with the basics of each and knows in what scenarios each is applied best.The Model for Improvement, based on work by W. Edwards Deming, is used commonly as the basis for many QI initiatives. (3) The model begins by asking a few key questions: What are we trying to accomplish; how will we know that a change is an improvement; and what change can we make that will result in improvement? A Plan-Do-Study-Act cycle, or PDSA cycle, is then initiated.In the Plan phase, the QI team states the project’s objective and creates a detailed plan that specifies who will be responsible for the changes; what the specifics of the changes will be; where the changes will be implemented; and when the changes will occur.The tests of change begin during the Do phase, and the team documents any obstacles that are encountered, so they can be addressed. During the Study phase, the team begins to analyze the process and outcome data, and a record of lessons learned is kept.In the Act phase, the QI team considers adopting, adapting, or abandoning changes based on the results of the Study phase and integrates the findings into the next PDSA cycle. The number of PDSA cycles needed for a given QI project varies and depends on the individual nature of the initiative and the environment in which it is implemented.Run charts provide an easy, practical way of tracking and displaying data generated from a QI initiative. Time intervals are plotted along the x-axis, and a process measure (eg, compliance rate) or outcome measure (eg, infection rate) is plotted along the y-axis. A line is drawn to connect data points and display trends. Annotations often are added to show how a process change affects the variable being measured over time. Figure 2 illustrates a simple run chart.Lean and Six Sigma methodologies can be used to improve quality of care, efficiency, and the financial performance of health-care organizations. (5) These techniques originated in the manufacturing industry to increase productivity by eliminating steps that do not add value to a process. Lean methodology focuses on reducing waste and improving workflow. Six Sigma strives to decrease variation through detailed data collection and analysis. These two strategies have subsequently been combined into a methodology known as Lean Six Sigma.A commonly used Lean Six Sigma tool is a value stream map that outlines an entire process from start to finish. Each step in the process is analyzed in three domains: personnel, information flow, and time. Teams of health-care providers then address workflow and find ways to increase value at each step.Some ways to increase value include reducing variation, streamlining processes, eliminating redundant information, and increasing employee engagement. Lean Six Sigma can be used, for example, to improve efficiency in an outpatient office or an emergency department. When embarking on a Lean or Six Sigma project, it is generally helpful to have a coach (ie, a facilitator formally trained in these methodologies) to help guide the endeavor.Root cause analysis (RCA) is a retrospective tool often used to determine the factors contributing to an error or adverse event so improvements can be made to prevent a recurrence. (6) RCA has its origin in industrial psychology and human factors engineering. It enables health-care organizations to identify system-based flaws and therefore develop system-based solutions.Because RCA is system focused, it helps to avoid blaming an individual whose actions are rarely responsible for an error or adverse event alone. An RCA begins with a detailed account and time line of events leading up to the incident. A multidisciplinary meeting is convened to review this summary and identify key contributory factors, or “root causes,” for the error or adverse event. Once the root causes have been established, risk reduction strategies are developed with a time frame for implementation. RCA can be used to evaluate incidents such as medication errors and wrong-sided procedures.Reason describes two major categories of errors: active error and latent error. (7)(8) An active error occurs at the point of human interface with a complex system. A latent error represents failures of system design. RCA is well suited to identify latent errors related to an event. In 1997, The Joint Commission mandated the use of RCA in the investigation of adverse events in accredited hospitals. (9) In 2003, the Agency for Healthcare Research and Quality described factors identified by using RCA as contributing frequently to errors and adverse events in health-care organizations. (10) These factors included communication failures, organizational transfer of knowledge, patient-specific issues, staffing patterns, workflow, technical failures, and inadequate policies and procedures.Failure modes and effects analysis (FMEA) is a prospective methodology used to anticipate system vulnerabilities and potential adverse outcomes related to failures in human, process, and mechanical interactions. The analysis then identifies areas most in need of improvement. (11) In FMEA, risks are evaluated in terms of severity, occurrence, and detection. The risks are given a priority value assessing overall risk level; actions for risk reduction or elimination are then determined.FMEA works well for processes that are relatively linear and has been used for many years in the engineering field. The Institute for Safe Medication Practices began using FMEA in 1990 and continues to promote the tool in the analysis of potential medication errors. FMEA has been used to guide implementation of computerized order provider entry, reducing ordering errors in pediatric chemotherapy. (12) Similarly, the process could be used to help mitigate risk in the transition to a new electronic medical record in an office setting.Evidence-based medicine can be viewed as a QI strategy, although not routinely thought of as such. Evidence can serve as a powerful tool to support and drive QI, as in the implementation of evidence-based guidelines. The use of best-practice guidelines to reduce adverse medical events is well documented. The guidelines can improve the quality and safety of care delivery to children, particularly in high-risk situations. (13) Best-practice guidelines reduce variation in care delivery, thereby reducing risk of errors. Many hospitals and pediatric practices integrate guidelines into routine care through the use of physician order sets, electronic templates, and alerts built into computerized provider order entry, as well as electronic documentation systems. Atlas et al (14) found that a strong connection between patients and their primary care providers is correlated with the use of guideline-driven care.The Muir Gray classification typically is used when employing evidenced-based interventions to reduce adverse medical events. This classification system uses a rating scale, with randomized controlled trials being ranked as the best or highest level of evidenced-based intervention. (15) Most clinical guidelines use evidence-based interventions and stress the application of interventions that have the strongest support from research studies.Shojania et al (16) summarized evidence-based practices that support patient safety. However, applying principles of evidence-based medicine to patient safety practices is not always an exact science. The following factors should be considered: level of evidence, anticipated clinical outcomes and risks of harm, generalizability to other institutions, and patient populations. A small dose of common sense also should be added.When applying evidence-based interventions to reduce medical errors and adverse events, the context of each practice setting must be taken into account. Berwick (2) describes the importance of context when applying recommendations from high-level studies. Each health-care system is unique. The complexities of each individual system need to be considered when deciding how best to apply evidence for reduction of adverse events.Like a handyman’s tools, QI tools sometimes are used best in combination. For example, an 8-month-old child who has influenza is admitted to an inpatient pediatric unit. During the hospitalization, she experiences a significant adverse event. An RCA can be performed to determine the factors contributing to the event. One of the root causes identified is that the seasonal influenza vaccine was not offered during the child’s recent health maintenance visit. Further investigation reveals that the primary care office does not have a standardized practice for offering influenza vaccine to infants. The Model for Improvement, by using several PDSA cycles, can be used to improve the process of offering the vaccine during well-child visits. The evidence-based guidelines for pediatric influenza immunization can be used to help drive the initiative, achieve better compliance rates, and, it is hoped, reduce the number of acute cases of influenza and resultant complications.The prospect of embarking on a QI initiative can be overwhelming. However, it is important that pediatric providers be involved in such work. The recent requirement by the American Board of Pediatrics for physicians to partake in QI activities as part of Maintenance of Certification underscores the importance of these endeavors. The following section provides some guidance to assist pediatricians in developing and implementing QI projects.For those with limited experience and resources, collaborating with other clinicians provides a rich opportunity to perform QI work. Collaboratives facilitate exchange of knowledge, experience, and ideas, and allow for pooling of resources. Organizations such as the Institute for Healthcare Improvement promote national and international quality initiatives that often span the medical specialties.The American Academy of Pediatrics’ Quality Improvement Innovation Network and its Chapter Alliance for Quality Improvement provide support and resources to engage pediatricians in QI efforts. Organizations such as the National Association of Children’s Hospitals and Related Institutions and the National Initiative for Children’s Healthcare Quality facilitate initiatives to improve health care for children.There are organizations that focus on pediatric subpopulations, such as the Vermont Oxford Network, which has facilitated pioneering work within neonatology. Pediatricians also may want to check locally for foundations that support health-care QI projects within their communities.The acronym SIERRA offers some simple guiding principles for implementing a successful QI initiative (17):Clinicians have a natural tendency to try to solve too large a problem or address too many issues with one project. Often, less is more. Simple initiatives focused on one key goal with one key change often are more achievable and increase the likelihood of a successful outcome. Even focused strategies can affect a wide range of health-care professionals, including physicians, nurses, and pharmacists. Creation of an interdisciplinary team increases the scope of expertise brought to the table, creates a collaborative environment, and helps promote more widespread adoption within an office or across an institution. Choose early adopters when recruiting members for your team. These are individuals who are respected by their peers, are open to change, and can help spread new ideas. (18) Colleagues who resist change and do not accept new ideas readily can impede progress, create divisiveness, and derail a project.QI projects rarely are perfect out of the starting gate. Rapid cycles of change within a small, defined patient population or geographic area make implementation more manageable. Consider making changes one provider or even one patient at a time. If the change is successful, the breadth for the next cycle can be expanded. This approach also will prevent suboptimal practices from becoming embedded or too widespread.A few weeks of data collection between interventions often is sufficient time to assess the effect of the initiative and to make any necessary adjustments. To maintain the momentum of the project, it is important to provide rapid feedback to your team as well as any other staff who have been affected by the initiative. Once positive outcomes have been achieved, it is helpful to advertise gains. This action provides positive reinforcement and can motivate colleagues to participate in future endeavors.Health-care providers often work in teams that include many different roles, and it is that these teams these teams do not always as a unit. The Joint Commission found that communication clinicians is with with to across of care, will be in in the next of this et al that the and industry have a large of time and resources into the ways in which teams and the best ways to is being used in many medical team employing can be used to improve quality of care and develop strong patient safety environments. The goal of these is to improve team and providers to and high-risk medical situations. has been added to of et al performed a randomized controlled in which were to as part of team or as individual The who in team more communication those in may a new for The of in QI has been is also being in risk management Many reduce the of for providers who participate in a major to communication in the health-care setting is the medicine In this an for example, not the actions of an In she or he might even be for This and is to and team members of a health-care team should and must to up they an situation or to being of the a and that the likelihood of critical thereby patient safety. The Agency for Healthcare Research and Quality is one of many available team that can be used to and improve within office or institution. can be used by medical teams to increase during key processes in health-care The use of in health care is a to reduce error in at risk of adverse medical events but is not and organizations must drive a that quality and patient safety. have been developed to engage hospital of in setting and for Quality must be evaluated and as medicine and new evidence becomes Pediatricians are in a to support and QI to improve health-care outcomes for quality and patient safety financial often use initiatives to motivate health-care and Although some such might be in improving health care, and that the changes made to measures might not necessarily improve an will QI in the future as the for begin to health-care and providers based not on the care they provide but the quality of care they Many of the performance measures used to motivate improvement are focused on individual patient and provider as well as process measures. and using measures that focus more on patient outcomes on a of This approach opportunity to clinicians who high value and care and better across the care have been made in health-care The increasing national on quality and patient safety the transition from blaming individuals to system-based and the increasing focus on and care are of positive a more clinical perspective, to reduce improve medication safety, and reduce errors have decreased and However, we cannot on is work to be health-care system needs to be into one that the principles of continuous