Health systems are eager to learn about better ways to deliver care. This requires innovation—doing something differently from how it is currently done. In a recent Health Affairs article, Elizabeth McGlynn and Mark McClellan noted that innovations often fail to meet expectations, particularly when they are spread from the initial site that piloted the innovation. McGlynn and McClellan pointed to the absence of evidence as to what makes these innovations work as a source of these failures and encouraged health systems to evaluate innovations. In addition to the critical role evidence plays in making decisions of whether to adopt an innovation, health systems factor in other important criteria.

Even if there is solid evidence for the innovation, such as from a systematic review of research, health system leaders will need to decide whether the innovation is likely to work in their own systems. What worked in one, or even many places, will not necessarily achieve the same results elsewhere. Furthermore, leaders need to consider not only whether they can achieve those results but also how important it is for their systems to do so. Making a wrong decision is costly; either leaders miss out on an improvement, or they spend valuable time and resources on a failed implementation attempt.

McGlynn and McClellan described what they call the assessment phase—the point when an organization has identified an innovation that might improve care or reduce costs but before the innovation is introduced. They implied that there may be reasons for an organization not to adopt an innovation that has been successful elsewhere (for example, baseline performance is already high, so there is not much room for improvement), but since their focus was on evaluation, they didn’t dwell on helping organizations think through whether they want to implement the innovation at all.

Picking up where McGlynn and McClellan left off, this article addresses an important component of innovation diffusion – whether to adopt an innovation. As an organization that is in the business of helping health care systems learn how to improve, the Agency for Healthcare Research and Quality (AHRQ) published a guide, Will It Work Here, to assist health system leaders in making adoption decisions. AHRQ contracted with RTI International to conduct research to inform the guide, including a review of the literature on innovation adoption and case studies of organizations that had adopted innovations that had been developed elsewhere.

Does The Innovation Fit?

First, health system leaders need to learn how the innovation worked in the past. Where has it been used, and what’s the evidence that it worked? When evaluating the evidence, understanding the context in which the innovation operated and how it is similar or different from their own environment is critical. They also need to understand how the innovation achieved its results. As suggested by McGlynn and McClellan, a logic model can be a valuable tool to capture how the innovation’s inputs and activities are expected to produce outcomes. In other words, a logic model makes explicit which processes will be used and what results they will generate.

Second, health system leaders need to determine whether the innovation will solve any of their system’s problems or contribute to achieving their system’s goals. If there’s a problem that needs fixing, carefully define the problem and honestly assess whether the innovation will address the root cause.

Third, think about whether the innovation is compatible with the mission, values, and culture of the organization. A clash with a system’s mission and values is likely to be a fatal flaw. The occurrence of a cultural clash is less cut and dry. Organizational culture, that is, the norms that guide behavior in the organization, is not monolithic. Several organizational cultures—such as patient-care and patient-safety cultures, business and management cultures, and professional and interpersonal cultures—come into play. Research indicates that certain kinds of organizational cultures—such as those with strong leadership, clear strategic vision, good managerial relations, comfort with experimentation and risk taking, and effective data systems—are more conducive to adoption of innovations in general.

But beyond general receptivity is the question of the match of a particular innovation with a health system’s organizational culture. If the innovation is not congruent with how the organization operates, can the innovation be adapted to improve compatibility? For example, a practice adopting a care management innovation might decide to hire another nurse and integrate care management functions across the nursing staff instead of hiring a separate care manager if nurses view the addition of a care manager as a threat to their relationships with patients. Whenever an innovation is adopted, there is some reinvention to fit the local context, but the innovations are not infinitely malleable. And while organizational culture can be transformed, it is a long-term proposition. If implementing the essential elements of an innovation runs counter to a system’s way of doing things, it may not be a good fit.

Should We Do It Here?

If the innovation isn’t rejected as unsuitable, the organization will want to consider the arguments for and against adoption. Some innovations aim to increase efficiency or decrease costs. Potential adopters may calculate the return on their investment, taking into consideration the upfront and maintenance costs of the innovation along with savings or revenues the innovation may produce. Even if there is a good return on the investment, it is important to compare that gain with alternatives. Opportunity costs of adopting an innovation might include delaying, precluding, or interfering with other initiatives.

It’s not only about dollars and cents. Hard-to-quantify aspects of a business case for adoption include the benefits to patient and families, staff, and other stakeholders. These might include increased patient involvement in health care decisions, better health outcomes, reduced stress on the workforce, or enhanced reputation. An innovation may be responsive to requirements of insurers, regulators, or accreditation organizations. Non-financial factors, such as a mission-driven system’s imperative to satisfy its charge, have to be weighed along with financial matters.

Any change entails uncertainty, so the risks—both the risks of adoption and of inaction—figure into the adoption calculation. Risks are not just financial; they include political, medical, and operational risks. It takes imagination to anticipate risks, and each system’s appetite for risk will vary. Consider the following:

  • What are the best and worst case scenarios?
  • What can go wrong?
  • How can we mitigate these risks?
  • How likely is it that the innovation will fail or that we will be worse off than we are now?
  • What risks are we unwilling to take?
  • What risks would we be taking by not adopting the innovation?

Can We Do It Here?

Just because adopting an innovation would be advantageous, doesn’t mean that the health system will be able to make the changes necessary to make it successful. First, an organization has to be ready for change. Organizational readiness for change has been described as “a shared psychological state in which organizational members feel committed to implementing an organizational change and confident in their collective abilities to do so.” Lack of a perceived need to change and confidence the change can be made, or widespread resistance to the proposed change, will diminish the likelihood that implementation of the innovation will be successful.

Leaders have to take stock of the magnitude of structural, process, workforce, and other changes that would be needed. Structural changes might mean centralizing (or decentralizing) certain functions or adding new teams. Changing processes, such as workflow or communication, can be hugely disruptive. Furthermore, some alteration to the workforce invariably will have to be made to accommodate the innovation. Even small changes in staff roles or the need to train staff in particular skills are likely to trigger a reaction. Once an inventory of needed changes has been generated, leaders will need to honestly appraise whether and how the changes could be made.

Determining whether the system has the capability to integrate the innovation into its operations also entails examining whether it has the ingredients for successful adoption. For example, having champions—key individuals, such as opinion leaders, in favor of the innovation—is an important means to generate support and overcome resistance. A learning health system will also reflect on past experiences with innovation adoption. It should use past experience to inform the current adoption decision by asking:

  • What were the major factors responsible for the success or failure of the innovation?
  • How is the proposed innovation similar to or different from past innovations?
  • Are any of the elements that were critical to success in the past missing this time? Is there any way to compensate for this absence?
  • What can be done differently this time? Is this enough to make the innovation succeed when others have failed?

How Will We Do It Here?

The final phase of the adoption decision involves envisioning how the organization will implement the innovation. The adage “the devil is in the details” comes into play here, as systems that should and could adopt an innovation may find that there are still obstacles. For example, lack of capacity to monitor and evaluate the innovation could trigger a red flag. Conversely, the ability to try the innovation on a small scale or for a short period of time may increase willingness to pursue adoption. Developing a change management plan can further increase confidence in a successful outcome if the innovation is adopted.

McGlynn and McClellan asserted that health systems have given way to pressure to adopt innovations that are not always evidence-based. As this post demonstrates, the decision regarding uptake of an innovation requires more than evidence of the innovation’s effectiveness. Researchers can help organizational leadership decide whether an innovation is a good fit—or an appropriate stretch—by not only producing evaluation findings but also fully describing the particulars of the innovation and its context when writing up results. When formal evaluations have not been conducted, accounts of innovations that contain credible evidence that they will be effective should include information about context. Contextual data, coupled with using the heuristics in the AHRQ guide “Will It Work Here?” can help health systems make better adoption decisions and save both time and money.