Population Health

What statistical methods and models are supplied?

Response: We have internal access to millions of de-identified hospital records in both the inpatient and outpatient settings. These training/test data are key to addressing the predictive analytics demands of clients and site customization. Data modeling and algorithm development are performed using industry-leading tools for data mining and supervised machine learning such as Weka, Orange, Spotfire, and R. Ongoing efforts include decision tree and regression models for a generalized predictor of hospital readmissions using variables such as length of stay, age at admit, APRDRG severity of illness, discharge day, discharge time, total number of admissions, etc. Our regression based global readmissions algorithm currently approaches 80% accuracy. This is a significant metric to forecast, and accuracy will improve as specific patient populations, such as heart failure, continue to be targeted.

How do you determine which diagnoses and procedures are disproportionately driving costs?

Response: Health Catalyst’s Key Process Analysis (KPA) helps organizations see where dollars are being spent on different patient populations. Using the Pareto principle, it enables organizations to better understand how 20% of their clinical process families drive 80% of the cost of care. This tool enables organizations to dynamically explore the cost of the care they deliver to identify opportunities of highest value for interventions to improve care and decrease cost of delivery.

Describe your case management workflow tools to support a patient population

Response: To improve case management services, subject-specific marts are developed around case management, care coordination, or populations. Data from multiple sources across the continuum can be included if available (such as inpatient, ambulatory, and claims). This gives teams the data to drive improvement in workflow, operations, and clinical care. These implemented improvements are then tracked to ensure effectiveness and sustainability. Within each subject-specific mart that is developed (such as heart failure, pregnancy, and so on), case management components can be incorporated into the data that drives the process improvement teams. For example, a heart failure team can evaluate the data across the continuum to recognize key opportunities for improvement such as discharge effectiveness, appropriate follow-up, medications, readmissions, high-risk patients or members, prevention, indications for setting and provider, and so on. As teams identify opportunities, interventions are planned, implemented, and measured.

Can your solution support a patient-centered medical home?

Response: The Health Catalyst solution helps an organization manage their Population Health initiatives through clinically developed condition-based registries and by applying metrics and measures to identify gaps in care and other quality measures. This is especially important for long-term care centers whose residents have challenges with multiple conditions and whose caregivers are trying to track adherence to best practices during their care. The Catalyst platform collects the data available in the organization and provides the tools and services for analysis to help drive the correct care decisions. The caregivers can use the information presented on the application dashboards and reports to determine patients that are in need of preventive care services and screenings for a range of conditions. The platform can also track and report on patient injury prevention programs such as falls, bed sores, and infections. The data is aggregated for overall performance reporting and to help the organization track to targets and improvement initiatives.

How do you identify patient populations?

Response: Health Catalyst can categorize patients under two methods: 1) Clinical Integration Hierarchy: All data are assigned to a multilevel care structure developed by Health Catalyst based on key work processes. Then the population can be analyzed for high cost and high variation. 2) Cohort Builder: This tool analyzes a number of variables, enabling non-technical users to identify specific populations with clinical and demographic criteria. For example, a client could analyze a variable such as socioeconomic issues, abnormal lab values, percentage of left ventricular ejection, and history of readmission. Patients are assigned to one of five categories, and patients needing specific interventions are flagged and appropriately treated by clinicians. The Health Catalyst EDW produces lists of patients and treatments.

What solutions do you provide for readmissions?

Response: Readmission Explorer is a tool that displays system-wide readmission metrics that are easily stratified and searchable by clinical area. Trends, comparisons, and detailed, patient-level data are all available to non-technical users through an uncluttered user interface built on data visualization best practices. Users can easily filter readmission performance by data range, ICD-9 code, attending provider, nursing unit, hospital, etc. Readmission Explorer can help quickly create a single source of truth for this important outcomes metric within your organization.

How do you support CMS Core Measures?

Response: Regulatory Explorer provides a framework for incorporating definitions of various regulatory metrics as well as a dashboard for viewing compliance with the metrics and exception reporting. This framework can support the measurement of current and historic results of regulatory and externally defined metrics such as CMS, HEDIS, PQRS, and other Professional Societies. This application allows drill-down into non-compliant cases.

Regulatory Explorer helps chart abstraction staff achieve higher levels of efficiency, accelerating external reporting processes and improving job satisfaction by automating as much repetitive data collection as possible. Regulatory Explorer allows busy clinical abstractors to focus on exceptions in patient care.