Specialty Corner: Application of Implementation Science to Improve Uptake of Electronic Tools

Electronic capture and storage of health information is a potentially powerful tool. When electronic information is captured in a way that is translatable across multiple platforms and stored in an easily accessible platform, it provides longitudinal information regarding patients and their health histories that can be used for clinical medicine as well as for quality improvement and research. This information can also be leveraged to uncover medication adherence patterns which can then be used to identify and target patients for intervention. (1) Interventions such as pharmacy-led programs or multidisciplinary lipid clinics (2) that use this data to improve patient care have been tested and have been shown to increase adherence to lipidlowering medications, but implementation and sustainability of these programs within U.S. healthcare systems has not been robust. In this article, we describe an application of implementation science to the deployment of electronic tools to improve care of individuals with lipid disorders.

Electronic databases and tools have been used to improve the care provided for patients with lipid disorders. Several health systems have implemented algorithms into the electronic health record to help identify patients to be screened for familial hypercholesterolemia (FH).(2,3) These algorithms rely on multiple sources of data including laboratory values, prescribed medications, medication adherence measures, and others to predict the likelihood of any individual patient having FH. In addition, specialized lipid clinics can also improve identification of patients who are at high risk for future atherosclerotic cardiovascular events, or patients with severe or difficult-to-manage lipid disorders. Geisinger recently created a multidisciplinary lipid clinic that is staffed by a cardiologist/Clinical Lipid Specialist, a clinical pharmacist and a cardiovascular genetic counselor. In addition to the creation of this clinic, an electronic registry was created for our clinic to track clinical outcomes (the degree to which patients met their lipid goals, outcomes of genetic testing, and medication approvals). Our patients’ average reduction of LDL-C during our first year was approximately 36-47% mg/dL reduction (individuals with dyslipidemia and familial hypercholesterolemia, respectively).(2) Even though electronic algorithms improve identification of at risk individuals and multidisciplinary clinics have shown great promise in closing care gaps, their full potential has yet to be recognized due to lack of widespread use.

Health information stored within electronic databases is not always easy to obtain, and optimizing care based upon the findings is the ultimate goal of their creation. In fact, it is the implementation of changes in clinical care based upon the results that remains the greatest challenge. In addition, decades of research have shown that evidence-based interventions are often slow to be implemented into clinical care, if at all.(2) Thus, there is a critical need to understand how best to integrate and utilize this data within existing clinical workflows and clinical teams need to prioritize this as part of their mission.

The tools offered by the emerging field of implementation science may help address this critical need. Implementation scientists have developed evidence-based frameworks, models, and theories which can guide the development of interventions and implementation strategies.(2,3) An example of an implementation strategy that can be deployed is a clinical decision support tool in the electronic health record to improve adherence to lipid guidelines. The use of implementation strategies is not new, however these strategies are often poorly described in the literature, limiting their ability to be widely used. (2,3) By using a common language and comprehensively describing the strategies we use to implement evidence-based practices, we may be able to facilitate research related to implementation strategies and their ultimate adoption into clinical practice. 

The ability to utilize an implementation strategy goes beyond descriptions, however; clinicians need to be able to fit the appropriate implementation strategy to their specific context or problem. An example of this is illustrated as part of two research studies mentioned above. Our team conducted focus groups to determine the acceptability, appropriateness, and feasibility of using algorithms (FIND FH) to identify patients with FH in clinical practice. We found that patients and providers were accepting of this concept and offered potential solutions of how algorithms were currently being used for other medical conditions and how an FH algorithm could potentially be integrated into practice in the future. Another example is the use of an implementation science framework called RE-AIM (Reach, Effectiveness, Adoption, Implementation and Maintenance) (2) to evaluate the rollout of the multidisciplinary clinic after 1-year.(2) The RE-AIM framework allows process, clinical, and service outcomes to be reported in a standardized fashion that is able to be replicated by others. The definitions of the RE-AIM domains are often tailored to individual program planning or evaluation. In general, reach is the number of individuals that received the intervention, effectiveness is the impact on patient outcomes, adoption is the proportion of providers that participated, implementation describes the extent to which the elements of the program are implemented, and maintenance is the measure of sustainability over time of the program. We found during our initial year that we were only able to care for a small portion of patients with lipid disorders.

There are many opportunities for the field of lipidology to integrate implementation science to advance individual and population health. We provide one example for its utility for implementation of electronic tools and hope to build on this practice to enable us to deliver at scale for the health system.

Disclosure statement: Dr. Jones has no financial disclosures to report. Dr. Kann has no financial disclosures to report. 

References: 

1. Jones LK, Pulk R, Gionfriddo MR, Evans MA, Parry D. Utilizing big data to provide better health at lower cost. Am J Health Syst Pharm. 2018;75(7):427-435. doi:10.2146/ajhp170350
2. [1] Dixon DL, Khaddage S, Bhagat S, Koenig RA, Salgado TM, Baker WL. Effect of pharmacist interventions on reducing low-density lipoprotein cholesterol (LDL-C) levels: A systematic review and meta-analysis. J Clin Lipidol. 2020;14(3):282-292.e4. doi:10.1016/j. jacl.2020.04.004
3. Harris DE, Record NB, Gipson GW, Pearson TA. Lipid lowering in a multidisciplinary clinic compared with primary physician management. Am J Cardiol. 1998;81(7):929-933. doi:10.1016/ s0002-9149(98)00027-7