In the second chapter of our special report on healthcare business intelligence (BI), consultant Lou Agosta details some real-world uses of healthcare BI software and provides
advice on how to develop an effective healthcare BI system.
Table of Contents
business intelligence systems: an IT laggard no more?
Healthcare BI software in action: real-world examples, best practices
Business intelligence project management tips for a healthcare BI team
Key IT roles on healthcare business intelligence project teams
It has been said that the most expensive piece of equipment owned by hospitals is a physician’s pen.
For example, if medical records aren’t accessible in a timely manner, impatient doctors might decide to re-order tests. Meanwhile, billing systems may not accurately indicate who is ordering what if they simply list the admitting physician as the default doctor. Costs often can’t be tracked back to activities, so cost reduction and quality improvement efforts remain a high bar for many healthcare organizations. Into this data mess comes the U.S. Department of Health and Human Services (HHS) with a requirement to report more than 100 healthcare quality metrics tied to 30-plus care criteria, starting in 2011.
But just as the enterprise resource planning (ERP) revolution in manufacturing delivered automated inventory control and an end-to-end view of the materials supply chain, today in healthcare we have the prospect of an end-to-end information workflow supporting patient hand-offs between different doctors, specialists and clinics – and the ability to capture and analyze data in a business intelligence (BI) system in order to improve planning, internal operations and medical care.
To create an effective healthcare BI system, providers should have a critical mass of at least three years of medical records, billing transactions and information about doctor-patient encounters with which to work. An electronic health records (EHR) system can generate much of that data; in addition, numerous healthcare IT vendors are offering ERP-type systems that provide a common information architecture and backbone for automating the process of running hospitals and large physician practices.
Abundant case studies show how leading-edge healthcare enterprises are using BI and analytics tools to improve revenues, reduce costs and optimize both business and clinical operations.
Operational healthcare BI software strengthens profits at medical institute
For example, according to the Institute for Musculoskeletal Health & Wellness at the Greenville Hospital System in Greenville, S.C., its investment in an operational BI system enabled it to improve annual profits by more than $10 million by helping administrators identify operational shortcomings and ways to correct them. Other reported benefits include an average monthly increase of more than $1,700 per doctor in physician-generated revenue and a reduction in accounts receivable turnover time from 62 days to 44 days. This success was made possible through a template approach that integrated the operational BI software with the institute’s EHR system and helped solved rampant data quality issues.
The Total Cancer Care program at the H. Lee Moffitt Cancer Center & Research Institute in Tampa, Fla., includes a multidimensional data warehouse containing clinical data on nearly 40,000 patients. By designing different views of the data for clinicians, researchers and the patients themselves, the data warehouse and BI system enable in-depth analysis of large amounts of data about the diagnosis, genetic markers, lab results, treatment and lifestyle of patients. The goal is to find ways to better target treatments to meet the specific needs of individual patients, and the system also serves as a source for further research on best practices – for example, what treatment works best in a particular scenario.
In retail, the 360-degree view of the customer is designed to drive more sales. In healthcare, the main goal should be to improve the quality, consistency and timeliness of medical services. Of course, that may indirectly result in more business by building up the relationship between patient and provider. No one wants to change doctors if they’re satisfied with the care they’re receiving.
Being able to pivot healthcare data for different BI uses is also a valuable capability, allowing administrators and business analysts to transform questions such as what doctors, diagnoses and treatments has an individual patient received into inquiries such as what patients has a particular doctor treated and what have been the outcomes of the treatments. In each case, the value of BI and “clinical intelligence” is a faster understanding of complex scenarios, ideally resulting in improved healthcare quality with fewer medical errors plus actionable lessons-learned about treatment trends.
Business intelligence best practices not always rewarded in healthcare BI
Not all BI best practices have found a smooth path in healthcare. For example, a leading large physician practice that I know of began using a data warehouse and BI tools to improve its performance long before HHS issued its quality metrics mandate. It now turns out that the organization may not be eligible for the financial reimbursements being offered by regulators to improve IT infrastructures, since the data warehouse hasn’t been “certified” by a recognized entity. Another “dirty little secret” is that an EHR system won’t deliver a true 360-degree view of a patient unless it captures and aggregates information from all of his or her doctors – both inside and outside a particular healthcare organization.
Nevertheless, experience shows that process improvements are promoted by measurement and analysis of key clinical (and business) drivers. In many organizations, including community hospitals and small physician practices, healthcare BI is early enough in its developmental trajectory that quick wins with significant cost savings are available. For example, capturing metrics on which physicians refer the most patients to a hospital’s emergency room may surface an area where services provided in an urgent care clinic could substitute for expensive ER visits. A quality metric such as the percentage of hypertensive patients with their blood pressure under control requires tracking their annual exams in order to measure and analyze the results. Yes, this can be done with paper charts; but the opportunity is at hand to go to the next level in using information to improve healthcare delivery.
Coordination-of-care initiatives got a bad name in the 1980s, when that term came to signify a reduction of healthcare services instead of what it really ought to mean – namely, empowering doctors to work with one another and with patients to improve diagnoses and the results of medical treatments and interventions. This time around, true coordination of care requires accurate, integrated data. BI and clinical intelligence systems, along with high-bandwidth networks and well-managed databases, create a powerful set of tools for managing the complex healthcare needs of patients in the modern world. This is where CIOs and IT or BI managers can make a significant difference in putting the hands of clinicians and business executives on the levers and dials that drive improved patient care and reduced costs.
About the author: Lou Agosta, Ph.D., is an independent industry analyst specializing in data warehousing, data quality, data mining, business intelligence and healthcare IT. Keyword: data. Agosta's book, The Essential Guide to Data Warehousing, was published by Prentice Hall PTR. He can be reached at LAgosta@acm.org.