May 2020, Volume XXXIV, Number 2

cover story One

Implementation science

The systematic uptake of change

ccording to Martin Eccles and Brian Mittman in their introduction to the new journal Implementation Science in 2006, implementation science is “the scientific study of methods to promote the systematic uptake of research findings and other evidence-based practices into routine practice, and, hence, to improve the quality and effectiveness of health services.” Implementation science was born out of the fact that evidence-based medical treatments do not tend to organically make their way into standard clinical practice. Many are never fully implemented or take years to move from research into practice.

A current example

Given the intense focus on the opioid epidemic, the state of treatment for opioid use disorders provides just one example of the type of gap in clinical care that highlights the need for implementation science. Evidence-based medications (methadone, buprenorphine/naloxone) that are highly effective for treating opioid use disorder have been available for decades. Yet it is still not the norm for patients with opioid use disorder to receive these medications. The reasons for this include health care system, organizational, and provider-level factors. Substance use disorder treatment has traditionally been provided outside of the health care system such that many substance use disorder treatment facilities do not have medical providers on staff. These traditional systems also continue, in some cases, to maintain a philosophy that frowns on psychiatric medications. Within the health care system, primary care providers generally were not trained to identify and treat substance use disorders. Providers are wary of medication treatments for opioid use disorders because of Drug Enforcement Administration requirements for training and licensing to prescribe them. They express concerns about the time to provide care management for complex patients in already overburdened practices.

Evidence-based medical treatments do not tend to organically make their way into standard clinical practice.

Implementation science can, and has, worked to identify strategies to overcome many of these barriers and works to spread these solutions to additional health care and substance use disorder treatment settings.

A new discipline

Implementation science is relatively early on in its lifespan as a scientific discipline. Theories, models, and frameworks have proliferated to guide and inform successful implementation. To date, there is not one prevailing theory, model, or framework. However, many commonalities exist among them and some essential ingredients for successful implementation are starting to become evident. The essential ingredients described below are based on those commonalities as well as 20 years of in-the-trenches implementation work.

The first step to successful implementation is to identify a clearly defined and measurable gap in practice. This requires the ability to measure the practice of interest, which has become easier with the expansion of electronic medical records. So, one might identify all patients with a diagnosis of opioid use disorder and the percentage of those patients who are receiving evidence-based medication treatments. While it might not be possible to identify the “correct” percentage, comparison to rates of guideline-recommended prescribing for other medical conditions may serve as a guide. For example, rates of guideline-recommended prescribing for hypertension and for depression well exceed those for opioid use disorder in most settings. The metric can also be used to provide continuous feedback on progress toward implementation goals, another key ingredient for successful implementation.

Second, it is important to have an in-depth understanding of current practice and the barriers to change in the local environment. This type of local information is generally gathered through interviewing key individuals. It is important to talk both to those that are likely to be supportive and not supportive of the effort in order to get a full range of perspectives. Through these local interviews, key stakeholders who are interested in directly participating in an implementation effort are likely to begin to emerge. Stakeholders should include representation from all levels of the clinic, such that high-level leaders are included as well as providers and support staff. Any role in the clinic that will be impacted by the implementation effort should have representation when discussing how the implementation plan will unfold. Local stakeholders must include someone who is in a position of leadership with the power to set the agenda for the clinic, as well as individuals who have the interest and time to invest in the day-to-day work of moving implementation forward.

Next, it is important to define an individualized plan for the clinic. There are many different paths that a hospital or clinic could take to reach the same end goal. For the medication treatment for opioid use disorder example, there might be one provider identified that wants to treat these patients in a specialized team that may have some additional resources dedicated to it, or the clinic may decide that every provider should be able to provide this treatment to the patients on their panel. Patient care management may be done by each provider, by a care management nurse, or by a clinical pharmacy specialist. The details depend on the resources available and what is acceptable to the staff. By individualizing the plan to the specific clinic environment, using input from key stakeholders and others within the clinic, the clinic develops a sense of ownership over the plan that is key to maintaining focus.

It is also essential to have a “facilitator.” This refers to an individual whose role it is to manage the implementation intervention. They are separate from the local stakeholders in that they are usually external to the clinic and serve as a manager of and a resource for the implementation team. They need to have training or expertise in implementation. They do not have to have specific clinical expertise, but they should have a basic understanding of the evidence base and connections to clinical experts whom they can call on for advice. They can provide the local stakeholders with the resources they need or track down resources or even create them, for example, if local stakeholders request an informational brochure for patients or a newsletter to share their progress with organizational leadership. They also are the glue that maintains the focus on the implementation goals. Any stakeholder, whether they are in a leadership role or a clinical role, is constantly bombarded by the “crisis of the day.” The facilitator maintains contact with the team and focus on the implementation goals on a regular basis.

Implementation science is relatively early on in its lifespanas a scientific discipline.

Challenges and responses

As mentioned earlier, implementation science is a relatively new discipline and as such, there are ongoing challenges facing the development of the field. Key among these is the development of reliable and valid measures of the constructs that are theorized to impact implementation success as well as the outcomes of implementation efforts. Many quantitative measures have been developed and used in individual projects. Only a few have been rigorously evaluated. The added challenge is that measures must be feasible for completion by busy health care providers. While some quantitative measures are promising, at this time, qualitative methods are essential to confirm and supplement the information gained from quantitative measures, as well as to understand the unique challenges and strengths of a particular local environment.

A second challenge is defining what aspects of an evidence-based clinical intervention are core and, therefore, cannot be adapted without compromising effectiveness versus those aspects that can be adapted to meet the needs of the local environment. This is relatively easier for interventions such as pharmacological treatments. However, for complex behavioral interventions, this becomes more challenging. Are aspects such as the format (group vs. individual; face-to-face vs. virtual) and timing (weekly vs. bi-weekly) adaptable or not? For these reasons, it is essential to define the presumed core elements in advance, preferably in consultation with the intervention developer, and also have a methodology in place to document adaptations and, if possible, their impact on patient outcomes.

Translating methods

While most implementation science work has been conducted by implementation researchers working in collaboration with large health care systems, the field is also concerned with how to translate implementation science methods more broadly into health care settings. This has led to an interest in “applied implementation,” or the practice of implementation outside of the research context. The question is what is needed for health care settings to scale up and conduct this work independently? First, the highest levels of health care leadership need to buy in to the idea that this type of work is of value to their systems and to back this up with resources. They will need to support individuals trained in implementation research and practice to serve as facilitators within their system or support growing that knowledge within their practice. They will need to express a willingness to hear from staff about gaps in evidence-based

practice, allow time for staff to develop and test organic solutions, and reward and recognize innovation.

Whether a large health care system or a small private practice, the key is to have a continued focus on learning and improving, and this requires dedicated time to consider such issues. Clinicians are often stressed, overworked, and focused on getting through each day. Change is always difficult but under these circumstances it is almost impossible. Wide uptake of applied implementation will require a restructuring of health care system and practice priorities to dedicate resources to the endeavor.

Hildi Hagedorn, PhD, LP, is a principal investigator and Director of the Implementation Core at the Veterans Health Administration Health Services Research & Development Center for Care Delivery & Outcomes Research located at the Minneapolis VA Medical Center. She is also an associate professor in the Department of Psychiatry in the School of Medicine at the University of Minnesota. Her research focuses on implementation of evidence-based practices for the treatment of substance use disorders. 


PO Box 6674, Minneapolis, MN 55406

(612) 728-8600

follow us

© Minnesota Physician Publishing · All Rights Reserved. 2019