- For hospitals wanting to increase their analytics capabilities and understanding, a good place to start is with the data in your EMR system. By aggregating data points from multiple patient files you can find a lot of useful insights about your operational efficiency and the quality of the care you give patients. With the right analytics application, you can tap both the structured and the unstructured data in these records.
While aggregating data from multiple hospitals can give you the kind of data set needed for more global insights, starting with the limited data in one hospital EMR offers a chance to quickly learn how to ask the right questions, which is a critical skill for analytics. It also offers another avenue for realizing greater value from your EMR investment.
Learn analytics process with real-world projects
While the projects you tackle in the beginning may not be of earth-shaking importance, they can provide real value in two ways. First, they can help provide clarity to complex problems based on actual data describing your patients. Second, the process of using analytics on a small project will help teach your staff how to make data-driven decisions and reinforce the importance of having a culture of measurement in healthcare.
Data driven technologies and decision support have fundamentally changed practically all industries dealing with customers or clients. For example, in the insurance industry predictive models based on historical data are routinely used to anticipate the cost of claims, automatically route claims to more experienced claims adjusters, or to refer claims for further investigation regarding possible subrogation or fraud. Such methods and systems are equally common in the financial services industry, marketing, telecom industry, etc. Often, new companies who were first to adopt the data driven culture completely changed the competitive landscape and “rules of the game,” creating new winners and losers.
Use the analytics opportunities in your EMR data
There are great opportunities in EMR data. For example, your EMR contains a wealth of data that can help you better understand the factors that contribute to medication errors. An analytics application can mine the EMR to find patterns that indicate contributing factors, such as time of day the medications were ordered, filled and administered; route of administration; location in the hospital; type of medication; age of patient; medical status of patient; physician ordering the medication; staff members giving the medication, etc.
You might find that errors are occurring most often at the busiest time of the nursing day, across several locations in the hospital. Or there may be one nursing unit or department that has a high number of errors, but only for particular kinds of medications. Or you might see that dosing errors are related to the late shift in the pharmacy, or a physician who is unfamiliar with the prescribing function in your EMR.
By pinpointing the trends, you can begin to see where your quality improvement efforts can have the most effect. By predicting risk, potential for errors, or quality issues before they occur, your organization can now drive better quality and efficiency through preemptive measures and policies, rather than react to issues or unexpected trends and events.
Combine EMR and financial data and look at risk prediction
You can also combine data from your EMR with financial data to help you see variations in cost for similar procedures or illnesses. This kind of insight will be critical if your hospital engages in outcomes-based payment models.
Because physicians seldom know what a particular medication or procedure costs, most don’t take cost into consideration in treatment decisions. And since they don’t see the total bill for a hospital stay, they may not be aware of how their decisions affect the final cost.
You can combine your EMR and financial data to run a cost-effectiveness comparison across all physicians who perform a particular kind of surgery. You can look at factors such as rates of complications and readmissions to help gauge patient outcomes, and detail patient charges to show where cost variations are occurring. These cost details, such as the difference in price between a brand-name drug and its generic equivalent, can often be a wake-up call for physicians.
Not only is this information useful for the hospital, it’s useful for physicians, too. As health plans narrow their physician networks based on cost-effectiveness, physicians will need to know how well they rank on cost as well as the quality of care. Providing this kind of transparency can help physicians understand how their treatment decisions affect cost and give them the opportunity to alter their habits to meet both cost and quality goals.
Your EMR may also hold data that can be used for more sophisticated predictive analytics, such as identifying patients at risk for sepsis and other serious complications. By predicting these risks and focusing the right kind of care on these patients, you can improve patient outcomes and lower treatment costs.
Use a consultant to guide your analytics journey
If you don’t have analytics expertise within your staff (few hospitals do), you can engage an analytics consultant to help you select a good analytics platform, guide you in your first efforts and help your train and mentor staff. A good consultant can also advise you on the kind of in-house expertise you will need as your analytics program grows.
As your staff gains experience, they will begin to see opportunities to use analytics to improve patient care, lower cost and work more efficiently.
As managing director and global head of healthcare consulting for Dell, Dr. Cliff Bleustein leads a team of healthcare professionals across management consulting, MEDITECH, Cerner, Epic, McKesson and ICD-10/revenue cycle. Dr. Bleustein helps customers by providing an in-depth understanding of the current problems facing healthcare executives and recommends innovative solutions to prepare for future business needs.