Introduction

What is a QI collaborative?

A quality improvement collaborative (QIC) is defined as a structured framework that amalgamates groups of practitioners from different healthcare organizations to improve a specifically identified aspect of the quality of their service [1,2,3]. QICs facilitate multidisciplinary teams to implement evidence-based knowledge by learning from one another, collaborating, testing changes, and utilizing data driven changes. Using their collective experiences, QICs can accelerate the translation of evidence into practice, resulting in reduced unnecessary variation and improved clinical care and outcomes [4,5,6]. A QIC usually consists of policy leaders and health stakeholders working toward a common goal of improving performance on a well-defined quality measure or set of measures related to a safety issue or clinical outcome [7]. A QIC takes a multifaceted approach to quality improvement (QI) that involves five essential features- a target topic, clinical and quality improvement experts, multi-professional teams from multiple centers, a model for improvement, and a series of structured activities geared towards improvement [8]. Multidisciplinary teams include key stakeholders at the individual centers and across the centers, including patients and families [6, 9, 10].

Do QI collaboratives work?

QICs have reported significant improvements in targeted clinical processes and outcomes as described in a systematic review by Wells et al. [3]. Zamboni et al explored the relationship of QIC outcomes with context domains (health facility setting, project-specific factors, organizational and external factors) and mechanisms (intra- and inter- organizational changes). They reported that participation in a QIC improves knowledge, problem-solving skills, attitude, teamwork, shared leadership, and generates opportunities for capacity building [9]. They also identified that QIC outcomes are successful when the collaborative goals are aligned with national patient safety goals. Two other systematic reviews of QICs in healthcare have identified improvements in patient care and organizational performance [11, 12].

Success of a QIC depends on five general factors: the team’s ability to work together; their ability to learn and apply quality methods; the strategic importance of their work to their home organization; the culture of their home organization; and the type and degree of support from management [1]. Collective learning and an evidence-based approach help the collaborative models accelerate learning and process change amongst participating hospitals [2, 13].

However, QIC reports in adult literature have described limitations in design and methodologies for any conclusive evidence and the effects cannot be predicted with certainty [11, 12]. Often, it is difficult to disentangle the different components of an intervention or to assess interactions between longitudinal activities of the collaborative components. Hence, these encouraging results must be tempered by the limitations of persistent gaps in QIC design, quality of reporting, and publication bias.

Collaborative learning

The earliest QIC activities and methodology were first described by Northern New England Cardiovascular Disease Study Group, established in 1986 [14], and the Vermont Oxford Network, established in 1988 [15, 16]. This methodology for QIC became popularized as the Breakthrough Series developed by the Institute of Healthcare Improvement in 1995. Many examples in the literature of the QIC initiatives are based on the Breakthrough Series [13], a collaborative improvement model which outlines key elements including multidisciplinary teams working together to improve performance on a chosen topic supported by experts using evidence-based best practices [2]. QI Collaboratives have described improvement in a broad spectrum of topics in adults and children [17,18,19]. Furthermore, pediatric collaborative improvement networks have demonstrated improved outcomes for children and their families across multiple subspecialities [6, 20].

QI collaboratives in neonatology

There are several regional, national, and international collaborative networks established in neonatology. These include, but are not limited to, the California Perinatal Quality Care Collaborative (CPQCC), Ohio Perinatal Quality Collaborative (OPQC), Tennessee Initiative for Perinatal Quality Care (TIPQC); Children’s Hospital Neonatal Consortium (CHNC), The Pediatrix Center for Research, Education, Quality and Safety (CREQS), The National Network of Perinatal Quality Collaboratives (NNPQC), the Solutions for Patient Safety (SPS), and Vermont Oxford network (VON); Canadian Evidence-based Practice for Improving Quality (EPIQ) [21], Australia-New Zealand Neonatal Network (AZNN) [22], International Network for Evaluating Outcomes (iNeo) of neonates [23], and Chinese Neonatal Network (CHNN) [24].

In this review, we aim is to evaluate national QI collaboratives primarily based in US (described below in alphabetical order). We will discuss the QI framework used by these collaboratives and provide examples of specific outcomes that their studies addressed.

Children’s Hospital Neonatal Consortium

The Children’s Hospital Neonatal Consortium (CHNC) provides a platform in QI for participating children’s hospital neonatal intensive care units (NICUs) by means of Collaborative Initiatives for Quality Improvement (CIQI) [25, 26]. This platform provides leadership, collaborative structure, resource access, QI education, team feedback, data driven change, and several other resources to assist participating institutions in completing QI projects. CIQI Steering Committee is a multi-disciplinary group of QI leaders involving physicians, nurses, and advanced practice clinicians which provides oversight of the CHNC CIQI Program. A CIQI Program Manager is responsible for operations, data management, and coordination of QI activities. At the local level a project planning and management team together with faculty advisors help oversee the collaborative efforts.

Children’s Hospitals Solutions for Patient Safety

Launched in 2009, Children’s Hospitals Solutions for Patient Safety (SPS) [27] was founded as a partnership between providers and the business community to improve quality and reduce costs. The mission is to work to eliminate serious patient harm from children’s hospitals by partnering with families and frontline staff. The tenets that are the driving force for SPS include executive leadership, focus on outcomes through HRO concepts and QI methodologies, focus on sharing successes and failures transparently, learning from one another instead of competing, “All Teach, All Learn” principle, and finally a commitment to building a “culture of safety”. SPS aims to decrease rates of the most common serious patient and employee harms, to identify and eliminate safety disparities, and to create robust learning systems for ambulatory safety. The network has now grown to include 140+ children’s hospitals across US and Canada.

National Institute for Children’s Health Quality

National Institute for Children’s Health Quality (NICHQ) [28] is committed to achieving better health outcomes for children and their families. It is a mission driven non-profit organization that focuses on providing equitable change for the health of children and their families. The National Network of Perinatal Quality Collaboratives (NNPQC) [29] provides resources and expertise to state-based perinatal quality collaboratives to improve maternal and infant health outcomes, particularly within marginalized populations disproportionately affected by adverse perinatal outcomes.

Pediatrix- center for research, education, quality and safety

With the goal to utilize evidence-based medicine to advance clinical care, the Pediatrix Center for Research, Education, Quality, and Safety (CREQS) engages in clinical research, education, continuous quality improvement, and safety initiatives. The specific objectives of CREQS are to contribute to better patient outcomes and reduce long-term healthcare costs. Pediatrix also provides [30, 31] a High Reliability Organization (HRO) program that combines characteristics of HROs with Six Sigma, TeamSTEPPS (Team Strategies and Tools to Enhance Performance and Patient Safety), and Crew Resource Management.

Vermont Oxford Network

The Vermont Oxford Network (VON) [32] is a nonprofit voluntary collaboration of health care professionals and families at more than 1400 member hospitals, primarily in the United States, working together as an interdisciplinary community to improve the quality, safety, and value of care for newborn infants and their families. iNICQ is an Internet-Based Newborn Improvement Collaborative for Quality coordinated by VON which brings teams together by using local and collective data to guide improvement on a designated topic. Participating centers are supported with access to expert faculty, VON measurement and reporting, an interactive online toolkit of evidence-based practices, modules, and online classrooms. Newborn Improvement Collaborative for Quality (NICQ) are groups of 8–10 teams that work together creating a culture of collaboration allowing for data-transparency, data-driven and evidence-based approaches, and rigorous quality improvement methodology to achieve and sustain improvement. Each homeroom faculty consists of a QI leader, a family/parent leader, and a NICU clinician. The VON NICQ collaborative includes families as partners to ensure that the family perspective is represented, valued, and enhanced during each improvement cycle.

Approach and outcomes of QI collaboratives in neonatology

Table 1 shows some examples from each of the 5 collaboratives described. We identified papers that described improvement work within each QIC and did not include individual team reports, or case reports. In addition, we chose a sample of convenience, up to 6 for each QIC that represents different methodologies and various types of clinical issues that QICs have addressed. This sample serves as a representation of the work of each OIC. The number of centers varies based on the size of the collaborative. The number of centers in each of the QIC publications varies from 7 to 330. These collaboratives demonstrate a wide range of topics have been addressed via QIC in neonatology. A clearly defined aim statement was identified for most of these QIC initiatives. However, the operational definitions for outcome, process, and balancing measures have not always been defined in these projects. Although all manuscripts have reported a collective improvement by the collaborative, it is apparent that not all the individual participating centers demonstrate improvement. Some centers contribute more data than others to the outcomes leading to overall improvement in the collaborative. In our evaluation, we identified only one QIC which addressed the economic implications of a QIC, specifically on low-birth-weight infants (NIC/Q- Neonatal Intensive Care Collaborative Quality).

Table 1 Examples of QI papers published from each of the 5 collaboratives.
Full size table

The implementation and mechanism of impact of a QIC varies widely and are briefly summarized here. CHNC utilizes the Model for improvement [33] methodology in their QI efforts. The collaborative provides best practices through clinical practice recommendations, but each center develops their own tools to conduct the project. IHI Breakthrough Series Collaborative Framework [13] concepts are used through regularly scheduled monthly meetings and huddles which provide opportunity to educate, share challenges, find solutions, report progress and data. Shared learning helps each center achieve their goals. CHNC also successfully utilized orchestrated testing [34], an application of planned experimentation that allows simultaneous examination of multiple practices (bundle elements) to determine which intervention or combination of interventions are the most effective. This technique has been used to reduce central line–associated bloodstream infections [35, 36].

100,000 Babies Campaign, first launched by Pediatrix in 2007, utilizes a QI infrastructure based on the Kotter organizational change model [37, 38]. QI projects in this collaborative utilize electronic health records- Pediatrix BabySteps Clinical Data Warehouse which provides a valuable resource for population and outcomes data [39]. The 100,000 Babies Campaign used QualitySteps [39, 40], a system developed by Pediatrix to assist development of project design, worksheets, and annotated run charts for CQI projects. This program continues to provide QI opportunities through its version 2.0.

NICHQ developed a change package and provided technical assistance to incorporate QI principles to improve the systems of care for premature babies during the Neonatal Outcomes Improvement Projects (2007–2010) in several states to reduce the burden of mortality and morbidity associated with premature birth and low birth weight. Some of the neonatal related initiatives by the organization includes but not limited to Best Fed Beginnings [41], a national QI initiative to help increase the number of “Baby-Friendly” designated hospitals in the United States; Infant Mortality Collaborative Improvement and Innovation Network, safe sleep, smoking cessation, social determinants of health, perinatal regionalization, and NewSTEPs 360, a national collaboration that aims to improve newborn screening programs.

The SPS network utilizes QI Science & HRO Concepts to drive the culture to reduce harm. SPS has developed change packages that inform best practices to reduce Hospital Acquired Conditions (HACs) such as Central line associated blood stream infections, unplanned extubations, adverse drug events, pressure injuries, and others. The change packages also include measurement strategy, care bundles, toolkits, and action items to inform improvement. Another key element SPS incorporates is partnering with and engaging patients and families in their QI initiatives.

VON utilizes the Model for Improvement [33] and experience-based co-design in their QI efforts. VON develops toolkits that consist of change ideas and evidence based potential better practices that are available to the teams. The term “potentially better practices” indicates that a practice is not considered better or best until adapted, tested, and shown to work in the local context. VON QI collaborative provides resources including clinical examples, measurement plans and reporting, foundational QI resources, consultation and collaboration with care experts and other teams. Teams test and implement evidence-based changes and apply continuous measurement to demonstrate improved outcomes. The collaborative uses the approach Learn + Measure + Share = Improve. Online courses related to QI foundations, VON Grand rounds and webinars provide evidence-based education to participating centers.

Discussion

In this review, we have described the framework and initiatives of 5 neonatal QICs that have demonstrated improved outcomes for neonates. These QICs are part of a broader network of collaborative efforts throughout the USA. At the grassroots level, the collaboratives offer a systematic approach towards improvement work, identification of common specific aims, and a standardized approach towards data collection, data analysis, benchmarking, reporting of results, and sharing best practices. Collaboratives are designed to have a broader impact across multiple centers by promoting collective learning. Our review indicates that the strategies of implementation of interventions vary with each QIC initiative. The most common approach is to develop ‘best practice’ focused interventions or bundle elements allowing centers to choose the interventions that are best suited for their local context. Other QICs offer specific interventions which centers must implement. Examples of more specific planned interventions include orchestrated testing, which was used to reduce central line–associated bloodstream infections. The perceived strength of a QIC is the collaboration of a group of experts and committed individuals in improving an outcome, a structured approach, sharing of best practices, collective learning, undertaking rapid cycles of change thereby leading to improvement [6, 42, 43].

Multiple factors contribute to the success of a QIC. At the QIC level, a structured framework, fostering a collaborative environment, data abstraction and data access, transparent sharing of data and interventions, identification of outcomes with clear operational definitions, effective use of technology, utilizing sound QI methodology, engagement of multidisciplinary stakeholders and families are paramount. In our review, we have identified that improvement is not uniform across all centers. At the individual institution level, participating smaller institutions lacking resources can perform studies using resources provided by the collaborative. Additionally, irrespective of whether an initiative is at individual center or part of a collaborative, at the individual center level, addressing local barriers, local leadership, organizational factors, team dynamics plays an important role [42]. One of the strengths of a QIC is that it enables less resourced hospitals to perform effective quality improvement initiatives.

Provider roles and strategies in single-center QI projects vs QICs has not been clearly described in the literature [44]. Single-center QI projects are often designed within the scope of an institution’s requirements, policies, and procedures. In single-center QI projects, since the improvement team is aware of the context, the interventions are targeted towards the local context. Since micro/macro environments and resources differ amongst every healthcare institution, one might argue that the lessons learned may not be generalizable in other settings.

On the other hand, QICs include multiple centers with different systems and resources. QICs offer a benefit over individual QI initiatives since the improvement work has been conducted in multiple different contexts and on a larger patient population. Hence, the interventions and results in a QIC may be more applicable in multiple settings as opposed to single institution QI initiatives. If nonparticipating individual institutions want to incorporate a similar improvement in their own healthcare system, the strategies from a QIC may be more adaptable. Additionally, challenges to sustaining improvement, to spreading change, and identifying stable funding sources exist in QIC much like single-center QI.

QICs have limitations that should be considered as areas of opportunity for further work. While all the QIC publications describe a positive outcome, there were institutions that participated in the collaboratives that did not show improvement. The individual contributions of each participating center towards the collaborative outcome varies but this variability is not always taken into consideration in QIC analysis. Perhaps, when reporting the QIC results, recognition can be made of the unequal contributions of the individual institutions. Furthermore, a detailed documentation of local processes is beneficial [44] to compare the outcomes and adaptability of QIC results so that readers can decide what approaches to incorporate in their own context. In addition, there could be publication bias, with positive studies being published more often, which may influence the conclusions one draws from a QIC [9, 45,46,47]. A study by Olson and colleagues [48] suggests that journal editors may not have a strong bias towards publishing positive results once manuscripts are submitted, but rather that researchers are more inclined to submit studies with positive results. We suggest that prospective registration followed by publication of the QIC efforts irrespective of whether the QIC had positive or negative results will help with accurate reporting of the effectiveness of QICs [47]. Although most QICs report on outcomes, further improvement work related to cost effectiveness and value [49, 50], evaluating the data and targeted interventions of a QIC from an equity lens [51, 52], and engaging families [53] are important next steps. As described in a recent commentary, local change, QIC, and equity focused QI are all necessary strategies to improve outcomes [54]. In relation to academic advancement, in QICs oftentimes the local project leader may not receive credit for the publication while the credit goes to the QIC management team. This factor may deter some from joining a QIC which could introduce a sampling bias. Our review is limited since we did not perform a comprehensive review of all collaborative neonatal QI but rather selected examples to illustrate their process and outcomes. Lastly, by limiting this review to collaboratives primarily based in the United States, some methods used in other countries may have been excluded.

Conclusions

QICs play a significant role in identification of target topics, developing best practices, improving provider knowledge, building QI capacity, and improving outcomes in neonates. QICs offer a benefit over individual QI initiatives since the improvement work has been conducted in multiple contexts. QICs are valuable since the knowledge gained is more generalizable and adaptable. Simply by enrolling, a center is not guaranteed improvement. In the description of the QICs, the individual contributions of each participating center towards the collaborative outcome varies. Although the results from the QICs we chose for this review are quite promising, these encouraging results must be tempered by the limitations of persistent gaps in QIC design, quality of reporting, and publication bias. Finally, the knowledge gained from our qualitative synthesis is valuable and could be taken into consideration for strategic planning of future QICs.