Change is necessary in healthcare because, in many ways, the provision of healthcare is less efficient, less safe,...
and less sustainable than ever before. Healthcare organizations continually fight this reality by improving quality, safety, efficiency and patient satisfaction. As I mention in my book Healthcare Analytics for Quality and Performance Improvement, this transformation in healthcare is occurring as the result of both necessity and opportunity. The opportunity arises from the potential effects of clinical and health information technology on clinical decision-making, healthcare management and administration.
For health IT to become a transformative force, it will take more than adopting EHRs, supply chain management systems and business analytics tools to improve quality and performance. Healthcare providers must find a way to harness the data at their disposal to improve clinical and organizational performance. Clinical analytics takes the data collected or generated by clinical and administrative information systems and provides the information necessary to understand, evaluate and improve healthcare clinical and business operations. Data analytics is critical to gaining knowledge, insight and actionable information from growing health data repositories.
Analytics is a collection of many tools, technologies, and techniques providers can employ to leverage the data amassed from the increasing number of health IT systems in operation. These innovations in combination with competent, effective leadership enable healthcare organizations to become more efficient and adept at achieving, evaluating and sustaining improvement.
There is a danger of losing focus of why healthcare is improving because of the plethora of available technologies and tools. As new technology, innovative data management approaches, and advanced analytics capabilities are pursued and developed, it is easy to think of these tools for their own sake. In doing so, it's possible to lose sight of the principles involved in the delivery and use of clinical analytics.
To deliver the information and insight needed by healthcare leaders, quality improvement teams, and other decision-makers, analytics in healthcare must be focused toward the clinical and administrative questions and issues facing the organization. One approach to maintain this business focus is to adhere to the following six principles.
Clinical analytics for accuracy
Analytics are commonly used to provide information and insight about past, current and future performance. Accuracy is important in each of these three areas. For all types of analytics, accuracy means ensuring that source data is high-quality and has few missing or incorrect data points. Advanced analytics -- such as machine learning and predictive models -- are reliant on data quality as well as the context of how the data was generated. This requires careful stewardship to ensure that algorithms are as current as possible and are generating valid output. Simply put, if the information is not accurate, it cannot be trusted and is unusable.
Deliver data with timeliness
The information absolutely must be available to the decision maker when making a final evaluation. This requires source system data to be up to date, to be extracted from source systems in a timely manner, and to be processed within a timeframe that puts the insight in front of the administrator or clinician at the time they need it.
Make data relevant with clinical analytics tools
Relevance is reflected in how well the information and insight generated by analytics systems meet the clinical, administrative decision-making, and quality improvement needs of a healthcare organization. The insight uncovered by an analytics system must be tailored to the individuals who are using the information, and must be relevant to the important administrative and clinical decisions in question. Otherwise, the data is trivial and will contribute to a sense of information overload -- which is a barrier to analytics adoption.
Appropriately analyze data
Analysis is the process of turning raw data from clinical or administrative source systems into something that is more useful by processing it through statistical models or algorithms that answer clinical and business questions. The type of analysis or algorithm must correlate with the particular business or clinical issue under examination.
Present clinical data visually
Data visualization is the practice of representing data, such as raw numbers or statistics, graphically. It can range from using charts and graphs to more elaborate infographics. Visualizations are a compact way of presenting information that allow viewers to spot trends or issues at a glance, instead of trying to make sense of complex tables.
Convenience is key
This means information is easy to find for the appropriate decision makers and it is located where it is most useful and needed. Analytics portals must be designed intuitively and logically for all that to be true. Rather than making users seek out data and information, it is becoming more common for clinical analytics to be pushed to the point where a decision is made (for example, embedded in EHRs).
In the future, I will cover best practices and how healthcare organizations can use these principles to inform their decision making. For additional information about strategies to leverage analytics and data within your healthcare organization, visit the Analytics Primer section of my blog.
About the author:
Trevor Strome, M.S., PMP, leads the development of informatics and analytics tools that enable evidence-informed decision making by clinicians and healthcare leaders. His experience spans public, private and startup-phase organizations. A popular speaker, author and blogger, Strome is the founder of HealthcareAnalytics.info; his book, Healthcare Analytics for Quality and Performance Improvement, was recently published by John Wiley & Sons Inc.
HR Analytics and EHR systems headline recent conferences
HIMSS 2014 attendees share stories of clinical analytics success
Essential Guide to health IT analytics and business intelligence