Examples of Current Projects
Predicting Hospital Readmissions Utilizing Health Information Exchange Real Time Data
- CPHIT is currently conducting a research project involving CRISP, Maryland’s health information exchange, and readmission risk predictive models (RRPMs). CRISP has an Encounter Notification System (ENS) to alert primary care providers (PCPs) of sentinel events regarding patients; the ENS lacks a tool to identify risk of readmission.
- The goal of this project is to develop and evaluate the feasibility of implementation of a real-time system to identify persons discharged from the hospital who are at high risk for readmission.
- This project, in part, will base its algorithms on the Johns Hopkins ACG® System, which has been used for over two decades across the US and in 15 other countries to generate predictive models of various healthcare events with a high degree of accuracy.
- As part of this research, a qualitative study will be conducted to evaluate the potential effectiveness and usability of the HIE-derived-and-delivered RRPM among participating PCPs.
Evaluation of Stage 3 Meaningful Use within Eligible Hospitals in Two States
- CPHIT is participating in research to evaluate the feasibility of Stage 3 Meaningful Use (MU3-CC) measures. One key deliverable will be the set of MU3 guidelines and final reports that will serve to guide both policy makers and hospital administrators regarding implementation of MU3-CC measures.
- This project will be administered and managed by the Johns Hopkins Armstrong Institute for Patient Safety and Quality, with Johns Hopkins CPHIT providing its HIT domain experts. Other contributing expertise are the Johns Hopkins School of Nursing and the Division of Health Sciences and Informatics at the Johns Hopkins School of Medicine.
View the final report here.
Identifying High-Risk Pregnancies through NLP
- In collaboration with Johns Hopkins Health Care and the Johns Hopkins Whiting School of Engineering, the goal of this project is to identify patients who are at high risk for premature birth and low birth weight and may qualify for the “Partners with Moms” program.
- Applying natural language processing and other computer science techniques, we are using physician notes and other unstructured parts of an Electronic Health record (EHR) to identify high risk markers and flags.
- This project and the techniques used will help demonstrate how different parts of an EHR can be used to identify populations for care management programs. Read more here.
Developing Next Generation EHR-Supported Predictive Modeling: Developing the Johns Hopkins “e-ACG” System
- CPHIT and the Johns Hopkins ACG R&D unit (housed in CPHIT) has a major project underway to use new clinical digital data streams to enhance current predictive and analytic models. This project is being done in collaboration with faculty from the Johns Hopkins School of Medicine and the Department of Computer Science.
- Some of the EHR elements that are being incorporated in advanced models include vital signs, lab values, cardiovascular data, clinician notes and patient reports.
- The goal of this project is to advance the state of the art of EHR based predictive modeling tools for high-risk case detection and management for populations. We will identify EHR and other HIT elements amenable to incorporation with traditional claims-based ACG Measures, and then test how to best integrate a combination of elements with the ACG System to enhance our predictive modeling ability.
- Over the various phases of this project, we will not only apply structured readily available EHR/clinical data sources, we will also apply Natural Language Processing (NLP) text mining approaches to capture information from unstructured data sources. We also will explore other types of “Machine Learning” techniques to develop prediction models that can be applied on a dynamic real time basis to augment clinical and population decision support systems.
Assessment of the Feasibility of CMS/ONC Quality Metrics Using Office- Based EMRs
- The integration of electronic health records (EHRs) into ambulatory care offices has created an opportunity to develop and test quality measures at the point of care. The Johns Hopkins CPHIT team and others from the Bloomberg School of Public Health are working with Mathematica (MPR) to evaluate the feasibility and accuracy of quality measures based on processes of care for potential use as part of the CMS/ONC Meaningful Use physicians’ incentive program.
- Johns Hopkins was an alpha test site for evaluating the measures associated with Medicare's Adult Wellness Visit benefit. Johns Hopkins is also a testing site for evaluating overuse measures associated with dexa scans for osteoporosis screening and use of imaging for uncomplicated headaches.
The Development and Testing of the Frailty Component of a Novel EHR-Based “Geriatric e-risk” Measure for Predictive Modeling
- The goal of this project is to develop advanced predictive modeling tools for high risk case detection and management for geriatric populations. This effort will rely on EHRs, other advanced electronic data sources, existing claims and administrative data, and existing functional status instruments/surveys.
- We hypothesize that evidence of the presences of frailty factors and underlying constructs are more likely discussed and described in various structured and unstructured portions of an EHR.