Center for Knowledge Management Research
CKM's research interests include integrating evidence into the patient care and research workflow using informatics applications; assessing various stakeholder groups' interaction with health information; education, skills development, and leadership development of librarians; optimal selection and use of information resources; organizational knowledge management; and digital library development.
Previous research projects include:
- Using Health Literacy and Learning Styles to Guide Oncology Patients through the Pharmacogenetic Maze of Personalized Medicine (IMLS LG-06-13-0180-13; completed November 2015)
Evidence regarding the effect of genetic variations on health is evolving at a rate that outpaces traditional educational and decision support approaches to knowledge dissemination. The goal of this 2-year project was to develop a reusable model for delivering patient-specific, consumer-friendly pharmacogenetics information. The central hypothesis of this proposal was predicated on the understanding that informed patients are more likely to take a proactive, participatory role and engage with their healthcare team.
The research team selected the freely-publicly available web-based tool My Cancer Genome (MCG), http://www.mycancergenome.org, to use as its testbed. Developed at VUMC, MCG provides up-to-date information on tumor-based genetic mutations that can impact cancer treatment. Written primarily for clinicians and researchers, patients can also access the website but are likely to be overwhelmed by the complexity of the content. This project, which addressed the IMLS strategic goal of placing individuals as "learners in the center," used MCG as a platform for developing a reusable model for delivering consumer-friendly pharmacogenetic information by adapting to appropriate learning styles and literacy level. Key components included the following:
- Define criteria and guidelines for developing consumer-friendly educational material about pharmacogenetic principles. The project team developed guidelines and criteria that incorporated best practice by pooling reported suitability guidelines from the literature to construct a gold standard matrix of components for creating consumer-friendly pharmacogenetic content.
- Create consumer-friendly pharmacogenetic information based on defined guidelines, learning style characteristics, and written at a 6th grade reading level. The team created a consumer-level dictionary of genetic pearls, which are complex words and clusters of words relevant to genetics, and a consumer-level version of content for the melanoma profiles available on the MCG website. The pearls each address four different learning style formats (visual, aural, reading, and kinesthetic).
- Detail an iterative multi-level process for content refinement and evaluation. The project team conducted a series of focus groups to gather patient feedback about the genetic pearls. Study team members then meet with melanoma patients individually to assess changes in patients' knowledge after viewing the consumer-level melanoma profiles and embedded genetic pearls. In Year 2 of the proposal, the team created genetic pearls and consumer-level MCG profiles for general precision medicine information, applying all lessons learned from developing the melanoma content. The ability to easily understand precision medicine in cancer content among patients viewing the materials was evaluated using similar methods as for melanoma. This phase allowed the team to establish whether the lessons learned from the melanoma profile refinement process were easily transferable and scalable to another condition, thus making the consumer-level customization process replicable for subsequent additional topics.
The project team combined multidisciplinary expertise in knowledge management and librarianship, pharmacogenetics, biomedical informatics, health literacy and learning styles, clinical care, focus group methodologies, study design and statistical analysis. The team selected two distinct content types, melanoma and a general precision medicine overview, to promote broad applicability of the model framework developed.
Results from this initiative have been reported in two publications:
- Giuse NB, Kusnoor SV, Koonce TY, Naylor HM, Chen S, Blasingame M, Anderson IA, Micheel CM, Levy MA, Ye F, Lovly CM. Guiding Oncology Patients through the Maze of Precision Medicine. J Health Commun. 2016;21(Supplement 1):5-17. doi: 10.1080/10810730.2015.1131772).
- Kusnoor SV, Koonce TY, Levy MA, Lovly CM, Naylor HM, Anderson IA, Micheel CM, Chen SC, Ye F, Giuse NB. My Cancer Genome: Evaluating an Educational Model to Introduce Patients and Caregivers to Precision Medicine Information. AMIA Jt Summits Transl Sci Proc. 2016 Jul 20;2016:112-21.
- Using Patient Literacy Levels and Learning Style Preferences to Optimize the Delivery of Health Information (IMLS LG-06-10-0186-10; completed November 2012)
The goal of this 2-year proposal by the Knowledge Management/Eskind Biomedical Library team at Vanderbilt University Medical Center (VUMC) was to create a workable model for generating patient-specific information prescriptions. This project integrated both literacy levels and learning style preferences to improve the usefulness of condition-specific health information provided to patients during healthcare encounters. Central to this effort was the team's hypothesis that this combined approach will improve patients' ability to learn and retain their health information.
This project created highly individualized information prescriptions guided by factors such as patient literacy levels and learning styles; prescriptions were generated dynamically via a web tool that maps patient attributes against hypertension and diabetes information to produce a customized information packet. The process was comprised of three implementation and evaluation phases with ongoing research data dissemination.
The project work in Phase One, included the following key components:
- creating, using both English and Spanish, a bank of hypertension information in various literacy levels and learning styles (visual, kinesthetic, etc.);
- selecting and evaluating literacy and learning style assessment tools;
- developing a combined literacy/learning assessment tool suitable for the emergency medicine department;
- creating a web application to reconcile assessment scores with appropriately targeted hypertension information to yield a customized information prescription; and
- evaluating the effectiveness of individualized prescriptions via a randomized controlled trial with adult hypertensive patients at VUMC's emergency medicine department.
Phase One results were published in the October 2012 issue of the Journal of Health Communication (Giuse NB, Koonce TY, Storrow AB, Kusnoor SV, Ye F. Using health literacy and learning style preferences to optimize the delivery of health information. J Health Commun. 2012;17 Suppl 3:122-40. doi: 10.1080/10810730.2012.712610.)
In Phase Two, to test the reusability of the hypertension information prescription in a setting outside the emergency department, the team conducted a randomized trial with adult hypertensive patients at a local community health center, the Vine Hill Community Clinic (VHCC). VHCC, operated by University Community Health Services (UCHS), provides extensive primary and preventive care services to Nashville's medically underserved community. Data gathered during this research phase informed the team on how to best replicate the personalized information prescription model in a community health center outpatient setting.
At the completion of Phase Two, the investigators had a clearer understanding of the effectiveness of a mature personalized hypertension information prescription for patients in two different clinical settings. Essential to the generalizability of the model's approach to other healthcare settings was an understanding of how best to implement it for various health conditions. In Phase Three, the investigators tested the reusability of the individualized information prescription model by developing and evaluating personalized health education materials tailored to learning styles and literacy levels for adult diabetic patients at VHCC.
The project team combined multidisciplinary expertise in clinical medicine, information and knowledge management, health literacy and learning styles, informatics, and evaluation. The project also included diverse care settings to encourage inclusion of hypertension- and diabetes-prone minority and uninsured patients, and to promote broad applicability of tools developed. Modular design of project materials facilitated widespread dissemination. Focusing on hypertensive and diabetic patients maximized the impact of the project's findings given the disease burden and their high potential for negative sequelae.
By testing whether customized information prescriptions positively affect patients' health knowledge and retention, this project established an infrastructure for developing customized information prescriptions that can be broadly adapted for use in varied care settings and with varied health conditions.
The health-literacy and learning-style tailored educational materials developed during this project are available online.
- Planning Multidisciplinary and Multicenter Strategies for Understanding Optimal Health Information Delivery to Children (IMLS LG-52-08-0074-08; completed March 2009)
The EBL was awarded an Institute of Museum and Library Services National Leadership Planning Grant in October 2008 to develop strategies for assessing children's interactions with health information. The team of librarians, psychologists, child development experts, pediatricians, informaticians, evaluators, and educators employed the planning period to develop techniques to assess health information needs and perceptions in children.
The team formalized a number of research aspects, including which potential interventions may offer the greatest benefit for study; the appropriate age range for study; and optimal venues for testing assessments and interventions. To broaden thinking and factor in elements of generalizability and dissemination from the outset, project partners include the Meharry Medical College, Vanderbilt Children's Health Improvement and Prevention Program (VCHIP), Cumberland Pediatric Foundation, organizations that offer expertise in areas key to long-range work.
- Go Local Tennessee (completed November 2008)
EBL was awarded an 18-month contract from the National Library of Medicine to develop a Go Local site for Tennessee in July of 2007. Go Local aimed to link MedlinePlus users to community health services and to direct Go Local users to relevant health topics on the MedlinePlus website. In 2010, the National Library of Medicine phased out the project.
Presentations and Publicity
- The team presented a poster describing Go Local TN workflow at the 2007 Southern Chapter of the Medical Library Association annual meeting.
- A Health Sciences Librarian Pre-Professional Integrated Learning Model (IMLS RE-01-03-0052-03; completed September 2007)
- The Clinical Informationist: Does the Model Work? (NLM 5 R01 LM07849; completed May 2007)
To evaluate the role of the Clinical Informatics Consult Service (CICS) in clinical decision-making, the EBL was awarded a 3-year grant from the National Library of Medicine (5 R01 LM07849). The CICS is a well-established program that facilitates evidence-based practice by delivering targeted information in answer to complex, patient-specific questions. The CICS places librarians with specialized training in medical subject areas and in information retrieval ("clinical informationists") on clinical teams in intensive care settings. Acting as expert consultants, they analyze the biomedical literature to identify, filter, and present the best examples of each clinical viewpoint expressed about key problems.
Working in cooperation with the Center for Evaluation and Program Improvement (CEPI) at Peabody College of Vanderbilt, the investigators formally evaluated the role of CICS in clinical decision-making and evidence-seeking behavior using a randomized pre/post design. While past studies have evaluated clinical medical librarians' roles, few if any studies have examined the effectiveness and utility of the new clinical informationist approach in sites where informationists are well-established.
Through a combination of observation and interviews, investigators examined librarian involvement in three existing Vanderbilt CICS intensive care units, collecting detailed information about the ways in which clinicians incorporate CICS-provided information into their workflows. The project also included a randomized trial to evaluate the effects of CICS on clinical decision-making. Investigators compared clinicians' intended course of care vs. actual care when CICS results are provided and when clinicians seek information themselves.
In the randomized trial, CICS questions were randomized to receive either a full evidence synthesis as is customarily provided, or no results (in which case clinicians could look for information themselves). Requestors completed a pre-request form, detailing their intended use of the information requested (support treatment decisions, etc.) as well as a post-request evaluation form, detailing how the information (if provided) affected the patient's treatment course or whether/how they sought information themselves for the question. To ensure patient safety, clinicians had the option of overriding a "no results" randomization in cases in which they felt the CICS results were critical for patient care decisions and they had no other means of locating information. The randomized phase of the study ended in May 2006, and results of the study appeared in the March 2008 issue of the Journal of the American Medical Informatics Association (Mulvaney S, Bickman L, Giuse NB, Lambert EW, Sathe NA, and Jerome RN. A randomized effectiveness trial of a clinical informatics consult service: Impact on evidence-based decision-making and knowledge implementation. J Am Med Inform Assoc 2008; 15(2):203-211) Abstract.
The project also included the development of "virtual cases" incorporated into training tools that can be used to export elements of the CICS model to other locations. The Evidence Based Knowledge Portal incorporates cases that set tenets of EBM in the context of a clinical setting to facilitate understanding and application of techniques such as NNT and absolute risk. The tool also contains assessments to allow users to gauge their level of readiness to adopt EBM practice and an EBM tutorial.
Demonstration of the utility of the clinical informationist approach can foster widespread adoption nationally and increase the degree to which clinical practice becomes evidence-based.
Results of the randomized trial appeared in the March 2008 issue of the Journal of the American Medical Informatics Association:
- Mulvaney S, Bickman L, Giuse NB, Lambert EW, Sathe NA, and Jerome RN. A randomized effectiveness trial of a clinical informatics consult service: Impact on evidence-based decision-making and knowledge implementation. J Am Med Inform Assoc 2008; 15(2):203-211. Abstract
Further information on the CICS program appears in the May 2005 issue of the Journal of the American Medical Informatics Association:
- Giuse NB, Koonce TY, Jerome RN, Cahall MC, Sathe NA, Williams A. Evolution of a mature clinical informationist model. J Am Med Inform Assoc 2005; 12(3):249-255. Abstract
The project team has also presented results from a pilot study funded by the grant at the May 2004 Medical Library Association Annual Meeting; results from this pilot study were published in Academic Medicine in 2005:
- Rosenbloom ST, Giuse NB, Jerome RN, Blackford JU. Providing evidence-based answers to complex clinical questions: evaluating the consistency of article selection. Acad Med. 2005 Jan;80(1):109-14. Abstract)
The team also presented a poster outlining the preliminary development of the virtual case tool at the 2005 annual meeting of the Medical Library Association ) and tool content at the 2006 Medical Library Association annual meeting:
- (Cahall M, Sathe NA, Todd P, Jerome RN, Giuse NB. Capturing Knowledge Management Practices of Local Expert Librarians to Inform the Development of a Virtual Case Training Tool. Presented at Medical Library Association Annual Meeting, San Antonio, TX, May 2005. (Poster)
- Sathe N, Todd P, Clark J, Gregg W, Giuse NB. Virtual Cases as a Tool To Promote the Incorporation of EBM Techniques into Housestaff Training. Medical Library Association Annual Meeting, Phoenix, AZ, May 2006.
In addition, the team has also conducted pilot studies to assess the effects of targeted communication strategies on expert information provision uptake:
- Jerome RN, Giuse NB, Rosenbloom ST, Arbogast P. Customizing implementation of a novel evidence request feature in an electronic medical record system. Submitted for publication to the Journal of the Medical Library Association (JMLA). May 2007.
- Jerome RN, Cahall M, Potter S, Todd P, Williams AM, Giuse NB. Scaling evidence-based services through integration with informatics tools. Medical Library Association Annual Meeting, Phoenix, AZ, May 2006 (poster).
The team has also discussed elements of the study's results regarding clinician-perceived barriers and facilitators of EBM:
- Mulvaney S, Bickman L, Lambert W, Sathe N, Jerome R, Giuse NB. Barriers to and Facilitators of Evidence Based Practices in Physicians. Rapid Communications Poster presented at the Society for Behavioral Medicine 28th Annual Meeting. Washington, DC. March 21-24, 2007.
- Envisioning the Information Specialist in Context: A Multi-Center Study to Articulate Roles and Training Models (MLA Informationist in Context contract; completed September 2005)
- Giuse NB, Sathe NA, Jerome RN. Envisioning the Information Specialist in Context (ISIC): A Multi-Center Study to Articulate Roles and Training Models. Medical Library Association (2006)