A practical approach for health system leaders to implement early detection of mild cognitive impairment and dementia in primary care

Abstract

Primary care offers an important pathway for timely diagnosis and treatment of mild cognitive impairment and dementia. Transformative changes require a structured approach to implementation of early diagnosis workflows. Planning, designing, implementing, and sustaining are important aspects of these improvements. A practical approach for health system leaders is provided.

Background

Despite the increasing prevalence of dementia in the US [1], healthcare systems fail to provide high quality, effective, and equitable dementia care [2]. Early detection in primary care offers a pathway for timely diagnosis to care and treatment [3]. Yet, Primary Care Providers (PCPs) face constraints, including limited time, lack of comfort with dementia screening, testing, and care, and inadequate reimbursement for early detection and care [4].

Improvement in early detection requires transformative changes in primary care (PC) culture and processes of care to create sustainable, dementia-capable workflows [3]. Planning and implementing such changes require health system leaders to use a systematic approach for implementing a model at their institution.

The aim of this commentary is to provide a framework for that approach.

Planning and gathering evidence

Creating a narrative of the need for change is the first step. Combining evidence-based data from the literature with tangible local stories of the current system is the most compelling. For example, this might include the story of a patient not diagnosed until late-stage dementia, paired with the data the describes the outcomes of late diagnosis, such as more severe behavioral and psychological symptoms of dementia, with local estimates of the number of patients who might currently be undetected at the location. Selection of an informal and well-respected leader as deliverer of the narrative is optimal.

Evidence should also be gathered, if available, about potential solutions in other health systems, including successful early detection implementations or tools developed to support early detection in primary care [5]. Identifying similarities and differences between exemplars and the local environment can be insightful.

Designing and defining requirements

Innovations are best designed by a team that includes the people who perform the day-to-day work in the setting where the implementation will be performed (“front line team”), as they are in the best position to know how change will impact other processes, identify the biggest opportunities, and measure outcomes. Involving front-line team members creates engagement in the success of the implementation. A typical improvement team may consist of two medical assistants, a front office team member, one provider, and a naïve participant to inquire about how and why the steps in a process exist. The naive participant does not need to be a member of the clinical team. Choosing a provider who is collaborative, open to change, and willing to function as a team member rather than as a leader of the team is key to this aspect. An executive leader who is invested in the success of the project should be identified as an executive champion.

The role of health system leaders is to support the change effort by freeing up the team members to do the work of the change, defining the system boundaries and constraints, defining the available resources for the effort, and creating a safe space for trying things that may fail. Senior leaders should be asked to define success for the project. In early detection, success factors might be to increase detection rates by at least 50% without adding more than 5 min of work for team members to the process. Leaders should define what the team needs to achieve, but how the team achieves it should be determined by the group. The executive champion should define whether or not the implementation project meets the success/endpoint criteria [6], check in with the team regularly, and assist in removing barriers.

Spreading to new locations requires identifying the minimum similarities or minimally specific criteria that are necessary for the improvement to retain its effect across new sites [6]. For example, the minimums may be using a specific cognitive assessment tool or documenting results of the assessment in a specific location in each patient’s electronic health record (EHR). New sites could change the timing of the assessment in relation to rooming the patient but must use that tool and must document it in that specific EHR location.

In advance of testing the new workflow, operational clarity is necessary for each step in the new process, such as when that action will be performed, what resources or information is needed to perform that action, and who will perform it.

Implementation and iterative improvement

Performing incremental tests of change is a well-established method of health care system innovation. While a variety of successful improvement methods exist [6, 7], iterative change is an essential element across methodologies. A decision framework that balances certainty and degree of support for the need to change can be useful in determining the scale of cycles of change.

Implementation cycles should be driven by frequent measurement of metrics that relate to the success criteria for the project; e.g. decreasing the number of missed assessments each day and not adding more than 5 min staff time. Plans to capture and frequently review the metrics should be developed prior to implementation start up. Brief daily improvement team standing meetings (“huddles”) can be very effective for facilitating the change process.

At the end of each iteration, full adoption, further iteration, or termination must be decided based on success and endpoint criteria previously set by leadership.

Operating plan sustainment and spread

Once success criteria have been met, measurement should continue until the process is stable and embedded into routine workflows. Stability can be identified by reduced variability in day-to-day performance metrics. The frequency of measurement may progressively decrease over time if stability is maintained.

Focusing on the minimum requirements for the project is critical to spreading successful change across multiple locations. Because the context of local environments varies with respect to patient, staff-, layout-, and space-characteristics, each work unit must use improvement methodology to adapt the innovation. The innovation can continue to be spread further as long as the minimum project requirements are retained. Iterative cycles of change should be used to achieve the success targets at new locations. Local adaptation is the key to successful implementation in complex primary care systems.

Conclusions

Health system leaders can use commonly recognized concepts and tools of health system improvement methodology to implement early detection of dementia in primary care.

Abbreviations

US

United States

PCPs

Primary Care Providers

PC

Primary Care

EHR

Electronic Health Record

Biographies

Deanna R. Willis

Dr. Willis was the Principal Investigator of the Indiana University/IU Health Early Detection Flagship program which was funded by the Davos Alzheimer’s Collaborative Healthcare System Preparedness Program.

Diana Summanwar

Dr. Summanwar completed the Indiana University Center for Health System Innovation and Implementation Science, Innovation and Implementation Science Graduate Certificate Program.

Authors’ contributions

DW: Design, implementation, manuscript writing, editing, coordination, and submissionNF: manuscript writing, editingJB: manuscript writing, editingDS: manuscript writing, editingDH: manuscript writing, editing.

Funding

Funding for the Early Detection flagship project was provided by the Davos Alzheimer’s Collaborative Healthcare System Preparedness program. 

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.