Turn Research Into Care Delivery Improvements Using the Research Analytics Adoption Model
Health Catalyst introduced the Healthcare Analytics Adoption Model as a strategic guide for health systems adopting analytics to drive healthcare transformation in quality, safety, and operations. The Research Analytics Adoption Model—consisting of 8 levels described below—is a strategic roadmap designed to help health systems achieve analytics-driven research capabilities.
Historically, research data capabilities have been separate from those used by operations personnel. And for good reason: using health data (i.e. patient records) for research is strictly regulated. But researchers involved with clinical trials have long understood the important overlap between research and operations. As health systems work to make precision medicine a reality, their research and healthcare operations must align.
The Research Analytics Adoption Model is a strategic roadmap for adopting and implementing comprehensive data and analytics strategies, encompassing everything from molecular data to clinical and financial data. The model offers a concise method for identifying current competencies, and better positions health systems to achieve competencies in a systematic way.
Organizational and Technical Roadblocks to Research
The research process is complex and involves many steps: hypothesis generation, cohort exploration, grant and IRB applications, patient recruitment, data collection and analysis, publication, and translation to clinical practice. Research is often hindered by roadblocks that arise from organizational and technical inefficiencies:
- Lack of tools to support cohort exploration
- No single source of truth for data
- Insufficient tools and processes to release data to authorized researchers
- Poor infrastructure for combining clinical and experimental data
The Research Analytics Adoption Model: A Strategic Roadmap
The Research Analytics Adoption Model consists of 8 levels designed to remove research roadblocks, and improve clinical research data management and care delivery:
Level 1: De-identified tools and data marts
Provide access to de-identified tools and data marts, allowing researchers to access data to power hypothesis generation, cohort exploration, grant applications, and IRB application generation. De-identified tools protect patient identity and typically minimize regulatory processes required for accessing protected health information (PHI) data for research.
Level 2: Delivery of customized data sets, including clinical notes
- Establish a research analytics team to develop customized data sets that answer complex research questions.
- Set clear guidelines for the data release process.
- Involve data stewards and dedicated research analysts who understand both the data and how PHI should be handled for research.
- Deliver data sets in an agile, consultative manner.
- Use a workflow tool that supports the data request workflow.
Level 3: Study recruitment facilitated by EDW
- Provide tools that allow researchers to define a population of eligible patients. Combine cohort criteria, scheduling data, and physician data to facilitate study recruitment.
- Centrally manage recruitment contact lists to prevent study fatigue.
- Provide recruiters with updated scheduling information via mobile devices.
- Capture “do not contact” requests.
- Provide an option for electronic consent.
Level 4: Research-specific data collection is centralized
- Avoid using Excel and Access as data collection tools; they aren’t secure and can’t handle multiple users.
- Use a data collection tool that pulls master data from the EDW and pushes collected data into the EDW.
- Provide the infrastructure to support self-service data collection form creation.
- Apply appropriate security to the data collection tool and data that ends up in the EDW.
Level 5: Automated reporting of research operations
- Pull key research systems into the EDW (clinical trials management system, patient recruiting system, and electronic IRB system) for deeper insight into research operations.
- Automatically generate key performance metrics (i.e. number of active studies, number of patients in trials, number of grants submitted/funded).
- Highlight areas for performance improvement.
Level 6: Biobank/Genomic data integration
- Provide a platform for genomic discovery by integrating biobank and experimental genomic data with the EDW (easily answer questions like, “what populations are enriched for a given set of gene variants?” and “what gene variants are enriched for a given population?”). Because biorepositories can be linked to a more detailed clinical record, they are no longer limited by data collected at the time of sample acquisition.
- Understand genomic data regulations and apply appropriate security measures.
Level 7: Multi-site data sharing
Provide the infrastructure necessary for coordinating centers to quickly and easily pull in data from multiple sites. Coordinating centers leverage data warehousing techniques for data collection, have automated intake or federation of data from participating sites, and provide exploratory tools to analyze a combined data set.
Level 8: Translational analytics
- Learning health systems routinely transition research data into care delivery guidelines.
- Use care delivery analytics to power research.
- Take discoveries made at the molecular level, and incorporate them into clinical practice.
- Leverage cohorts and outcome measures from care improvement while adding in new data.
The Research Analytics Adoption Model, when paired with engaged, supportive leaders, aligns research and clinical priorities and transitions data-powered research discoveries into care delivery improvements that can be deployed system wide.
Watch the Webinar: Powering Medical Research With Data: The Research Analytics Adoption Model http://healthcatalyst.wpengine.com/webinar/powering-medical-research-with-data-the-research-analytics-adoption-model/
Would you like to use or share these concepts? Download this presentation highlighting the key main points.