Collaborative Model

An “improvement collaborative” is a shared learning system that brings together a large number of teams to work together to rapidly achieve significant improvements in processes, quality, and efficiency of a specific area of care, with intention of spreading these methods to other sites.

Improvement collaborative seek to adapt and spread existing knowledge to multiple sites. This existing knowledge may consist of clinical practices based on scientific evidence, proven practices that are widely considered as “good” or even “best” or any other changes to the existing way of doing things that have been shown to result in better health care. Such knowledge is the collaborative’s “implementation package” the changes in processes and organization of care that the collaborative seeks to introduce,refine, and spread.
In a collaborative, site teams work out and test ways to operationalize or put in practice the concepts included in the implementation package and to over come barriers to making them work in their local settings. Collaboratives are intended as a time-limited improvement strategy, typically achieving significant results in 9–18 months, although improvements are often seen earlier. However, in cases of redesigns of complex systems (for example, involving multiple chronic diseases), collaboratives have continued beyond this timeframe,usually in a phased approach.
Teams within a collaborative use a common set of core indicators—ideally the smallest number of indicators that can inform the improvement and tell the story of the collaborative’s efforts and achievements—to measure the quality of the care processes the teams are trying to improve and, where possible, the desired health outcomes. Each team collects data on the indicators to measure whether the changes it is making are resulting in improvement. Local health care providers are the improvement “experts” who develop action plans to test and implement changes at their local level to achieve collaborative goals.
What differentiates improvement collaboratives from other improvement methods is shared learning. In a collaborative, multiple (10, 20, 50, or more) teams all try to make improvements in the same topic area. They simultaneously test and implement process redesigns and changes and share their experiences while doing so.
Through this shared-learning mechanism, facilitated by the collaborative, teams communicate the results of their tests and their solutions, and all teams can benefit from the knowledge of both successful and unsuccessful changes implemented by any team. In this way, teams learn from other teams’ experiences and can avoid “re-inventing the wheel” in discovering successful changes.
Frequent (usually, monthly) monitoring of results (i.e., process and outcome indicators) and regular sharing of successful changes help to spur the pace of improvement, creating a sense of friendly competition among teams to see which one can achieve the best results. The network of shared learning results in rapid development and testing of innovations to solve problems, rapid dissemination of effective changes, and rapid development of effective models of care, enhancing the original implementation package of evidence-based standards with operational learning.
A distinguishing feature of the Improvement Collaborative approach compared with traditional QI methods is that it seeks to spread improvements beyond the initial teams, to be applied throughout the organization(s) participating in the collaborative.
Typically, a collaborative will conclude with the definition of a final package of interventions that have been field-tested and proven to yield results in a particular setting complemented by a set of organizational learning that facilitates achieving those results. This package, which may be thought of as a refinement of the implementation package, is then ready for spread to other sites. This emphasis on intentional spread of the improvements not only distinguishes collaboratives from other QI methods but also makes the approach an attractive scale-up strategy.
An Improvement Collaborative begins with a preparatory period when the collaborative’s objectives and technical interventions are refined and a structure developed to support the collaborative’s implementation. The “implementation period,” when site teams develop and test changes to put in practice the technical interventions that make up the implementation package promoted by the collaborative, is generally divided by three to six learning sessions that are separated by periods of one to four months when teams test changes. These intervening periods are sometime referred to as “action periods.” Once teams know how to operationalize the interventions and have achieved the collaborative’s objectives, a workshop or conference may be held to review the teams’ collective experience to decide which changes were the most effective and to share results with stakeholders outside the collaborative.
Once a collaborative has been completed and an enhanced implementation package developed, several different strategies may be used to spread that operational knowledge to new sites. The initial collaborative—sometimes called a “demonstration” collaborative—may then be followed by a second,or “spread” collaborative whose purpose is to spread the enhanced implementation package from the demonstration sites to the rest of the parent health system. Members of the original collaborative often serve as change agents and advisors during a spread phase.


From The Improvement Collaborative: An Approach to Rapidly Improve Health Care and Scale Up Quality Services. USAID Health Care Improvement Project. 2008. Bethesda, MD: University Research Co., LLC (URC).