The Vermont Oxford Network

Abstract

After completing this article, readers should be able to: “The nation’s health care system lacks… the capabilities to ensure that services are safe, effective, patient-centered, timely, efficient and equitable… Between the health care we have and the care we could have, lies not just a gap but a chasm.” IOM 2001 (1)“More people die in a given year as a result of medical errors than from motor vehicle accidents, breast cancer, or AIDS.” IOM 2000 (2)In 2000 and 2001, the Institute of Medicine (IOM) of the National Academy of Sciences issued two landmark reports that presented a clear and compelling challenge to all health care professionals. (1)(2) We must improve the quality and safety of the medical care for the patients and families that we serve. In this article, we describe the Vermont Oxford Network (VON) and the quality improvement (QI) activities it has undertaken to meet this challenge.The VON is a not-for-profit organization whose mission is to improve the quality and safety of medical care for newborns and their families through a coordinated program of research, education, and quality improvement. (3)(4) VON currently has more than 440 member neonatal intensive care units (NICUs) from North America and around the world. Members participate in a range of Network activities, including randomized controlled trials, outcomes research, and a variety of QI projects. The focus of this article is QI. Descriptions and references for other VON activities are available on the VON Internet site (http://www.vtoxford.org).In support of its mission, VON maintains a database that includes information about the treatment and outcomes for high-risk infants receiving neonatal intensive care. Importantly, all members adhere to a uniform set of procedures and definitions. (5) Data are submitted either on paper forms or digitally using a standardized file format.All members participate in the very low-birthweight (VLBW) database for infants weighing 401 to 1,500 g. The growth of VON, in terms of the number of NICUs participating and in the number of VLBW infants enrolled each year in the database, is shown in Figure 1. Since its inception in 1989 until the end of 2002, the database has enrolled more than 214,000 VLBW infants. In 2003, we anticipate that member units will submit data for more than 35,000 additional VLBW infants. Members also have the option of participating in an expanded database for all NICU infants regardless of birthweight (approximately 70 units are participating in 2003).The database provides members with confidential, customized quarterly and annual reports that document their performance, identify trends over time, and provide comparisons of their NICU with the entire network and with subgroups of similar NICUs. The reports are intended for use in local QI initiatives. The database also is used for outcomes research and to provide core data for VON randomized trials.The information in VON reports lays a foundation for QI efforts at member NICUs. It assists members in identifying opportunities for improvement and allows them to track changes over time. As important as this information is for supporting QI, by itself it is not sufficient to create change. Health professionals also need specific knowledge, skills, tools, and resources for improving neonatal care.The VON has organized a series of multidisciplinary improvement collaboratives designed to provide necessary knowledge, skills, tools, and resources for initiating change. These collaboratives, known as Neonatal Intensive Care Quality (NICQ), have adapted QI methods and approaches pioneered by the Institute for Healthcare Improvement and by the Northern New England Cardiovascular Disease Study Group. (6)(7)(8) They involve multidisciplinary NICU teams working intensively together under the guidance of expert faculty and trained improvement facilitators to develop measurable improvement goals, identify potentially better practice ideas, and test and implement these ideas with methods that are logical for their local setting. (9)The specific aims of the NICQ collaboratives are to make measurable improvement in the quality and safety of medical care for newborns and their families; to develop NICU-specific knowledge, tools, and resources for improvement; and to disseminate these broadly within the neonatal health care community.The NICQ collaboratives have two face-to-face meetings each year that include training in QI and evidence-based medicine, facilitated small and large group exercises, and extended discussion. Participants receive feedback based on the VON database. They choose the improvement topics on which to work and form multi-institutional, multidisciplinary working groups to examine these topics. Between meetings, participants communicate through scheduled facilitated conference calls and dedicated e-mail discussion lists. Round robin site visiting among the teams and selected site visits to benchmark institutions that have exhibited superior performance in specific areas are another component of the NICQ collaboratives. Social networking is encouraged and is an important feature that contributes to the success of the improvement work.The first NICQ Collaborative included 10 self-selected institutions and was funded in part by a grant from the David and Lucile Packard Foundation. There was no fee for participation, but teams were responsible for their own travel expenses to the meetings and internal personnel costs, which were estimated at approximately $68,000. (10) Teams participating in this collaborative demonstrated improvements in incidences of coagulase-negative staphylococcal infection and chronic lung disease, (11) which were associated with significant reductions in the cost of care. (10)Based on the successes of the first collaborative, the VON organized the NICQ 2000 Evidence-Based Quality Improvement Collaborative for Neonatology. The fee for participation was $15,000 per year. Comprised of multidisciplinary teams from 34 institutions, this collaborative originally was scheduled for 2 years, ending in 2000. However, at the request of the participants, the collaborative was extended for a third year to focus on patient safety issues. NICQ 2000 addressed a range of improvement topics, including lung injury, family-centered care, nutrition and growth, unit culture and multidisciplinary collaboration, brain injury, and nosocomial infection, in addition to a broad range of patient safety topics. The work of the NICQ 2000 collaborative is documented in an online supplement to Pediatrics that appeared in April 2003. (12)In 2002, the VON convened its third improvement collaborative, NICQ 2002, with a total of 48 institutions, 27 of which had participated in NICQ 2000. The list of improvement topics has continued to expand and now encompasses a wide range of clinical, organizational, and operational topics, including pain and sedation, staffing, discharge planning, high-risk perinatal care, respiratory care, nosocomial infection, and patient safety. Although originally scheduled for 2 years, this collaborative also has been extended at the request of the participants and will continue until December 2004.An international study of QI collaboratives noted little evidence of the ability to sustain specific changes or the continued use of improvement methods when a collaborative ends. (13) We have taken specific steps in NICQ to address this deficiency, thereby moving beyond improvement as a project to create improvement as a habit.Based on his work with the original NICQ collaborative, Paul Plsek, codirector of the NICQ collaboratives, identified four key habits of clinical improvement that formed a foundation for improvement work. (9) These habits have provided a foundation for all subsequent VON collaboratives. Multidisciplinary teams apply these four habits to improvements in a broad range of clinical, operational, and organizational domains of NICU care (Fig. 2).Good care depends on the complex coordination of many factors and the efforts of many people. Teams in NICQ begin from this premise and move beyond the more restrictive, discipline-based view of care. Further, our understanding of systems suggests that systems improvements require coordinated changes in structure (eg, new policies, committees, role descriptions), processes (eg, modifying the sequence of care events in the first hour after birth, changing the process of ordering medications), and patterns (eg, addressing communications, conflict, leadership, and other organizational culture issues).The IOM reports (previously cited) noted that a significant proportion of the care delivered daily is not consistent with what is known to be most effective. Although some variation is necessary to address the unique needs of patients, NICQ collaboratives stress the need for a continuous effort to bring the daily practice of neonatal care more in line with evolving knowledge of what works. Participants learn tools of evidence-based medicine, such as the critically appraised topic, (14) and hear directly from experts in the field through presentations and involvement in focus groups at twice-yearly meetings and on dedicated e-mail lists.Although substantial knowledge for improvement is available in the literature, even more knowledge is potentially available in the currently unexamined variation in practice. The only avenue for obtaining this knowledge may be through collaborative learning with others. Working together openly, curiously, and nondefensively allows NICQ centers to examine carefully the variations in outcomes and practices in a continuous search for superior care. This habit is played out within the focus groups, through round robin visits and visits to superior performers, and more widely through the posting of specific questions on e-mail lists.Improvement is more than simply knowing what to do; improvement requires change. Unfortunately, in many health care organizations, maintaining the status quo –“the way we have always done it” –is more of a habit than is improvement. NICQ centers are encouraged to have multiple teams operating simultaneously to make changes on several fronts. The goal is to make change a more normal event in the life of the NICU. Teams apply methods for rapid-cycle improvement using plan-do-study-act (PDSA) cycles (15) and have shared their experiences with these techniques through hundreds of poster presentations and cycle reports over the years. The poster presentations, or case studies, are available for others to review via the collaborative’s Web site at NICQ.org.The VON sponsors an innovative series of Internet-based collaboratives for its members known as iNICQ. The goal of these collaboratives is to disseminate the knowledge, tools, and resources developed in the more intensive NICQ collaboratives to a broader audience and to spread the concept of collaborative improvement widely within the neonatal community. The first series focused on quality, and the second focused on patient safety. The iNICQ conferences are financed by subscription fees. Teams from more than 60 NICUs have participated in these online conferences.Participating teams connect to the audio portion of the iNICQ conference via standard telephone conference call and view the slide presentations over the Internet. Teams usually meet in a conference room, with the audio presented via speaker phone and the slides projected using a digital projector. Each online conference includes formal presentations by expert faculty, questions and discussion time, and interactive exercises for the local teams. Between scheduled conferences, participants have access to a dedicated e-mail discussion list.The iNICQ conferences are designed to be action-oriented. Prior to each scheduled conference, teams are given specific work assignments to prepare for the conference and assist their improvement efforts. For example, the prework for the conference on the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) safety goals held in November 2003 included a guided self-assessment of the institutional- and NICU-specific plans and actions related to each of the JCAHO safety goals for 2004. Teams were asked to come to the conference having completed this assessment. During the online conference, teams were given time to prioritize their future action plans for addressing the JCAHO safety goals.The VON Internet site, NICQ.org, provides improvement tools and resources to participants in the NICQ and iNICQ collaboratives. This site includes resources developed by multi-institutional focus groups in the NICQ collaboratives related to specific improvement topics such as nosocomial infection, family-centered care, nutrition, safety, and organizational culture as well as plenary presentations from VON meetings and a growing collection of case studies of specific improvements in quality and safety made by individual NICU teams. NICQ.org also supports the voluntary, anonymous reporting of errors and near-miss errors in NICU care. (16) More than 1,600 errors have been reported and demonstrate the wide range of errors that occur in the NICU.Each December since 2000, the VON has held an Annual Quality Congress for Neonatology. Open to all interested health professionals, these Congresses are intended to disseminate knowledge about QI and foster collaboration among teams involved in improvement. Topics addressed at the Congresses have included safety lessons for health care from the aviation industry, teamwork and team learning, staffing for quality, involvement of families in patient safety, and many other issues of relevance to NICU quality and safety.With funding from the Agency for Health Care Research and Quality, the Center for Patient Safety in Neonatal Intensive Care has been established as a partnership among the University of Vermont (Jeffrey D. Horbar, MD, PI, Jerold F. Lucey, MD, Gautham Suresh, MD), Dartmouth College (William Edwards MD), Children’s Hospital, Boston (Donald Goldmann, MD), Beth Israel Deaconess Medical Center, Boston (James Gray, MD, Robert Ursprung, MD), and the VON (Kathy Leahy, RN, NNP, Pam Ford, Patricia Shiono, PhD, Paul E. Plsek, MS, Julianne Nickerson, MSW). The Center is dedicated to examining medical errors in the NICU from multiple perspectives, including provider reports, medical record reviews, and parent reports. The VON has provided the NICQ and iNICQ collaboratives as forums to advance the Center’s research. Initial research has included analyses of the medical errors reported on NICQ.org (16) and parent reports of NICU errors submitted to the Internet site, Howsyourbaby.com, developed by William Edwards and John Wasson at Dartmouth College. This site, originally developed to improve communication with families at the time of NICU discharge and in use at a number of NICUs in the NICQ 2002 collaborative, has been expanded to address issues in patient safety from the perspective of parents.Evidence for the effectiveness of QI collaboratives has been slow in developing, but is beginning to emerge. (17) O’Connor and colleagues (7) have shown in a longitudinal observational study that a multidisciplinary team approach that includes audit and feedback, QI training, and site visits can reduce mortality from coronary artery bypass surgery. A longitudinal study of the NICUs in the first VON NICQ collaborative, which used a similar multidisciplinary team approach, demonstrated decreased incidences of nosocomial infection and chronic lung disease as well as reductions in the overall cost of care. (10)(11) Grol and Grimshaw, (18) in a recent review of the evidence for a variety of initiatives and approaches to changing medical practice, concluded that although substantial evidence suggests that changing practice behavior is possible, change requires comprehensive approaches at different levels tailored to specific settings and target groups. The NICQ and iNICQ collaboratives incorporate these features.In a recently completed cluster randomized trial of a multifaceted QI intervention based on the four key habits, designed to promote the use of evidence-based surfactant therapy for preterm infants, the VON demonstrated that multidisciplinary teams exposed to the intervention treated infants with surfactant significantly sooner after birth than teams in a control group. (19) Further research is needed to determine how generalizable this multidisciplinary QI intervention will be to other target practices and patient populations.The improvement work of the VON is just beginning. A total of 119 NICU teams have participated in the NICQ and iNICQ collaboratives (Appendix). Many units have yet to be engaged. To make the collaborative improvement model available to these units, the VON plans to initiate a new iNICQ series in 2004 and to offer a new intensive face-to-face collaborative, NICQ 2005, beginning in January 2005. Teams participating in both the iNICQ and NICQ collaboratives will contribute to NICQ.org and build a comprehensive online archive of NICU improvement knowledge (Fig. 3). By promoting the four key habits for improvement, the VON hopes to assist members in improving the quality and safety of medical care for the patients and families that they serve.Supported by grants from the Centers for Disease Control (H50/CH121553) and the Agency for Healthcare Research and Quality (P20 HS11583). NICQ.org was supported by grants from Ross Products Division of Abbott Laboratories and a grant from the David and Lucile Packard Foundation.We thank the teams and faculty members of the NICQ and iNICQ collaboratives for their dedication to improving the quality and safety of NICU care that made this work possible.