Analytics are supposed to provide data-driven solutions, not additional healthcare analytics pitfalls and other related inefficiencies. Yet such issues are quite common. Becoming familiar with potential problems will help health systems avoid them in the future. The three common analytics pitfalls are point solutions, EHRs, and independent data marts located in many different databases. An EDW will counter all three of these problems. The two inefficiencies include report factories and flavor of the month projects. The solution that best overcomes these inefficiencies is a robust deployment system.
Choose by Topic
- Latest Insights
- Accountable Care
- Analyst & IT Roles
- Best Practices
- Big Data
- Business Intelligence
- Care Management and Patient Relationships
- Clinical Analytics and Decision Support
- Dashboards & Reporting
- Data: Quality, Management, Governance
- EMR Fit
- Enterprise Data Warehouse / Data Operating system
- Financial Alignment & ROI
- Health Catalyst Overview
- Health Catalyst University
- Healthcare Analytics Summit
- Healthcare Transformation
- Implementing Health Catalyst Products & Services
- Leadership, Culture, Governance
- Machine Learning / Predictive Analytics
- MACRA / Regulatory Measures
- News Roundup
- NLP & Text Analytics
- Operations and Performance Management
- Outcomes Improvement
- Patient Experience, Engagement, Satisfaction
- Patient Safety
- Payers & Providers
- Physician Engagement
- Population Health
- Precision Medicine
- Quality & Process Improvement
- Value-Based Purchasing / Risk-Based Contracting
Outlined for practical use in the healthcare industry, this eight-level, industry-specific roadmap acts as a guide for organizations striving to be truly data driven. Developed by an independent, cross-industry group, this white paper explains the Analytics Adoption Model, its history and purpose. Organizations can then use the model to determine direction, highlight goals and gauge progress, as well as see the next steps needed to achieve goals associated using data to drive and assess clinical improvements.
Healthcare has remained entrenched in its cottage industry-style of operation, even within huge medical centers and significant medical innovation. The result, as documented by Dr. John Wennberg’s Dartmouth Atlas of Health Care project , is unwarranted variation in the practice of medicine and in the use of medical resources including underuse of effective care, misuse of care, and overuse of care provided to specific patient populations. The root of the problem, Wennberg concludes, is that there is no healthcare “system.” At Health Catalyst, we agree. Healthcare needs to be systematized and standardized in three key areas:1) healthcare analytics or measurement, 2) adoption or how teams and work are organized, and 3) best practice or how evidence/knowledge is gathered, evaluated, and disseminated for adoption.
Analytics - Additional Content
Spend time reading content for you
Health systems have vast amounts of data, but frequently struggle to use that data to solve strategic problems in a timely fashion. A healthcare analytics team, made up of the right people with the right tools and skillsets, can help address these challenges. This article walks through the steps organizations need to take to put an effective analytics team in place. These include the following:
- Recognizing the need for change.
- Demonstrating the value of an analytics team.
- Conducting a current state assessment.
- Identifying solutions.
- Implementing a phased approach.
- Building a roadmap.
- Making the pitch.
- Putting the roadmap into action.
The transition to ICD-10 in 2015 saw the number of available procedure codes increase from roughly 3,000 to more than 70,000. This change gives clinicians the ability to code procedures to a much higher degree of specificity and provides health systems the ability to unlock powerful clinical insights into how inpatient procedural care is delivered. This article covers the benefits and drawback of ICD-10 PCS, as well as concrete ways health systems can use these procedure codes to provide new clinical insights. The article also walks through the anatomy of the seven-digit alphanumeric codes and provides specific clinical examples of how healthcare organizations can slice and dice this data.
A robust data analytics operation is necessary for healthcare systems’ survival. Just like any business, the analytics enterprise needs to be well managed using the principles of successful business operations. This article walks through how to run an analytics operation like a business using the following five-question framework:
- Who does the analytics team serve and what are those customers trying to do?
- What services does the analytics team provide to help customers accomplish their goals?
- How does the analytics team know they’re doing a great job and how do they communicate that effectively to the leadership team?
- What is the most efficient way to provide analytics services?
- What is the most effective way to organize?
Opportunity analysis uses data to identify potential improvement initiatives and quantifies the value of these initiatives—both in terms of patient care benefits and financial impact. This process is an effective way to find unwarranted and costly clinical variation and, in turn, develop strategies to reduce it, improving outcomes and saving costs along the way. Standardizing the opportunity analysis process makes it repeatable and prioritizes actionable opportunities. Quarterly opportunity analysis should follow four steps:
- Kicking off the analysis by getting analysts together to do preliminary analysis and brainstorm.
- Engaging with clinicians to identify opportunities and, in the process, get clinician buy in.
- Digging deeper into the suggested opportunities to prioritize those that offer the greatest benefits.
- Presenting findings to the decision makers.
Health systems feel mounting pressure to demonstrate ROI from analytics investments but are faced with inefficacies and delays. Fortunately, the Rapid Response Analytics Solution delivers a 10x increase in analytics productivity and a 90 percent decrease in the time required to develop new analytic insights. The Rapid Response Analytics Solution solves these tough analytics problems through two primary elements: curated, modular data kits called DOS Marts; and Population Builder, a powerful self-service tools that lets any time of user, from physician executive to frontline nurse, explore data and quality build cohorts of patients without relying on IT staff and with no need for sophisticated and customized SQL and data science coding.
As healthcare systems are pressured to cut costs and still provide high-quality care, they will need to look across the care continuum for answers, reduce variation in care, and look to emerging technologies. This article walks through how to evaluate the safety and effectiveness and of emerging healthcare technology and prioritize high-impact improvement projects using a robust data analytics platform. Topics covered include:
- The importance of identifying variation in innovation.
- Ways to improve outcomes and decrease costs.
- The value of an analytics platform.
- The reliable information that produce sparks for innovation.
- Identifying and evaluating emerging healthcare technology.
- Knowing what data to use.
- The difference between efficacy and effectiveness in evaluation of emerging healthcare technology.
While many industries are leveraging digital transformation to accelerate their productivity and quality, healthcare ranks among the least digitized sectors. Healthcare data is largely incomplete when it comes to fully representing a patient’s health and doesn’t adequately support diagnoses and treatment, risk prediction, and long-term health care plans. But even with the obvious urgency for increased healthcare digitization, the industry must raise this trajectory with sensitivity to the impacts on clinicians and patients. The right digital strategy will not only aim for more comprehensive information on patient health, but also leverage data to empower and engage the people involved. Health systems can follow five guidelines to digitize in a sustainable, impactful way:
- Achieve and maintain clinician and patient engagement.
- Adopt a modern commercial digital platform.
- Digitize the assets (the patients) and the processes.
- Understand the importance of data to drive AI insights.
- Prioritize data volume.
Despite the complexity of healthcare analytics, one key strategy for effective, sustainable analytics stands out: designating an executive sponsor to oversee the program. This sponsor is a C-suite level leader who’s committed to championing analytics throughout the organization and has the influence and relationships to drive widespread outcomes improvement. Healthcare executives can use four criteria to identify a great executive sponsor for their analytics programs:
- Have a single accountable leader.
- Find a sponsor with passion for and knowledge about data.
- Choose organizational clout and a vision for analytics over a specific title.
- Build a partnership with the CIO.
Healthcare organizations know that they need to an effective clinical data analytics strategy to improve and survive in today’s challenging environment. In order to make these necessary improvements, healthcare leaders need to establish clear goals for their clinical data analytics initiatives. Achieving these goals requires clinical teams to clearly identify problems and plan for how to achieve them. This article walks improvement teams through sometimes confusing process of identifying problems, setting clear, achievable goals, and common pitfalls along the way. Topics covered include:
- Six categories of clinical data.
- Three types of goals: outcome, process, and balance.
- How to write an outcome goal.
- Internal vs. External Benchmarks.
- Mitigation strategies.
- Getting clinical buy-in.
Many health systems are eager to embrace the capability of natural language processing (NLP) to access the vast patient insights recorded as unstructured text in clinical notes and records. Many healthcare data and analytics teams, however, aren’t experienced in or prepared for the unique challenges of working with text and, specifically, don’t have the knowledge to transform unstructured text into a usable format for NLP. Data engineers can follow four need-to-know principles to meet and overcome the challenges of making unstructured text available for advanced NLP analysis:
- Text is bigger and more complex.
- Text comes from different data sources.
- Text is stored in multiple areas.
- Text user documentation patterns matter.
Data-driven quality improvement is propelling healthcare transformation. The ability to strategically leverage healthcare data is essential, making highly effective data analysts more valuable than ever. So, what attributes differentiate a good data analyst from a great analyst? Stephen Covey’s well-known book “The 7 Habits of Highly Effective People,” has long had far-reaching impacts in the business world. These same principles are relevant today and applicable in the world of healthcare analytics. Learn how Covey’s second habit, “Begin With the End in Mind,” drives great healthcare data analysts.
Smartphone applications, home monitoring equipment, genomic sequencing, and social determinants of health are adding significantly to the scope of healthcare data, creating new challenges for health systems in data management and storage. Traditional on-premises data warehouses, however, don’t have the capacity or capabilities to support this new era of bigger healthcare data. Organizations must add more secure, scalable, elastic, and analytically agile cloud-based, open-platform data solutions that leverage analytics as a service (AaaS). Moving toward cloud hosting will help health systems avoid the five common challenges of on-premises data warehouses:
- Predicting future demand is difficult.
- Infrastructure scaling is lumpy and inelastic.
- Security risk mitigation is a major investment.
- Data architectures limit flexibility and are resource intensive.
- Analytics expertise is misallocated.
Employers that offer robust employee health plans at affordable costs are more likely to attract and retain a great workforce. Healthcare, however, is often a top expense for organizations, making balancing attractive benefits with attractive costs a complex undertaking. Employers need a deep understanding of employee populations and opportunities to manage health plan costs without sacrificing quality. An analytics-driven approach to employee population health management gives employers insight into two key steps to lower healthcare costs and enhance benefits:
- Manage easily fixed cost issues.
- Use healthcare cost savings to fund expanded benefits.
By committing to transforming healthcare analytics, organizations can eventually save hundreds of millions of dollars (depending on their size) and achieve comprehensive outcomes improvement. The transformation helps organizations achieve the analytics efficiency needed to navigate the complex healthcare landscape of technology, regulatory, and financial challenges and the challenges of value-based care. To achieve analytics transformation and ROI within a short timeframe, organizations can follow five phases to become data driven:
- Establish a data-driven culture.
- Acquire and access data.
- Establish data stewardship.
- Establish data quality.
- Spread data use.
Every hospital and health system has to juggle significant IT needs with a limited budget. In the middle of these demands and possibilities, hospital executives have to prioritize and decide which technology solutions are the most critical to the health of their organization. I call these most critical IT solutions “survival software.” Advanced clinical analytics solutions are the survival software of the near future, as they really hold the key to achieving the triple aim and survive value-based purchasing.
Healthcare data models are the backbone of innovation in healthcare, without which many new technologies may never come to fruition, so it’s important to build models that focus on relevant content and specific use cases. Health Catalyst has been continuously refining its approach to building concise yet adaptive healthcare data models for years. Because of our experience, we’ve learned five key lessons when it comes to building healthcare data models:
- Focus on relevant content.
- Externally validate the model.
- Commit to providing vital documentation.
- Prioritize long-term planning.
- Automate data profiling.