Recently, I attended the Digital Healthcare Conference in Madison, Wisconsin. Last year, the theme of the conference was simple – measure everything! Measure patients and their outcomes, processes and process quality, providers and their productivity. Measure for value. Measure for opportunity. Measure for protection.
This year, the theme of the conference was just as simple – what do we do with all of this data now that we have collected it? We now have crates of patient data, process and quality data, DICOM images, performance data, etc. How do we get value out of the data?
The Challenge
Many organizations face the first problem (i.e., what to measure). Other organizations face the second problem (i.e., what to do with the data). Both are problems that can be solved at least to some degree with business intelligence. But, either way, these organizations face the same challenge: how to prioritize which business intelligence applications to invest in to answer business and clinical questions to make the organization stronger, faster, smarter and better.
Over the years, I have worked on business intelligence projects in nearly two dozen industries and have heard a common statement on every project. Each industry feels that their analytical needs are markedly different from any other industry. To some extent, this is true. But in many cases, there are business questions, goals, priorities and challenges that are common to all industries and businesses.
The Healthcare Exception
Healthcare, on the other hand, may be one industry that is the exception to this rule, and this difference affects how to prioritize analytical applications. Consider this: What other industry combines biology, physics, mechanics, business, bureaucracy, governmental regulation, public pressure, community service, quality concerns, emotion and competition to the degree that healthcare does? This situation alone should be enough to put immense weight and complexity on the challenge of prioritizing business intelligence applications, but healthcare organizations have added complexity due to the political forces currently driving them to change – and change fast – how they operate and how they use information.
There are special considerations when prioritizing business intelligence applications in a healthcare environment, but fortunately there are only a few important ones.
Prioritizing Business Intelligence Applications
The practice of prioritizing IT application projects in most industries is fairly mature. Manufacturers, retailers, banks, distributors, etc., all use pretty much the same three key factors to determine what applications to pursue, in what order and why. At a high leve, those three factors are value, feasibility and risk. Value is the return, feasibility is the investment and risk is the chance of investing and not getting a return. There are a number of ways to measure each of these factors and variations in calculating them as well as using them in the ranking process; but, overall, these three cover the main variables involved.
Healthcare business intelligence projects, by contrast, require an analysis that stretches each of the three key factors a bit. Some of these “stretches” are due to the complex nature of healthcare organizations, while other stretches are due to the nature of business intelligence applications themselves.
One factor that is specific to healthcare project prioritization comes right out of the name of the field (i.e., care). As one physician told me recently, if faced with the choice of two equally fruitful projects, choose the one that takes the higher road. Business intelligence applications in healthcare provider organizations that truly promote patient care should be given preference over equally valuable projects from a business viewpoint. Examples of applications that promote patient care include patient registries, quality improvement and measurement, outcomes analysis support, access and affordability improvement, and so forth. Emphasizing care promotes the long-term mission of the organization.
Self-View
Right along with this added wrinkle in prioritizing analytical applications is understanding the unusual nature of how healthcare organizations, and especially the people in them, view themselves. This self-view has some built-in conflict that needs to be taken into account when ranking business intelligence projects. For instance, as another doctor told me, some days he lets his softer, emotional side take over, and he just wants to fulfill his mission and be a hero to his patients. Other days, his hard-nosed business sense kicks in. Either way, most doctors emphasize two driving forces in their work – wanting to help their patients get and stay well, and not wasting precious time.
The same is true of nurses, physician assistants, specialists, technologists and others in healthcare. Help me, they say, to do right by my patients and not waste time. It is essential that we in healthcare organizations use the data we already own to make our organizations more efficient, more effective and more attractive, as well as more financially viable.
Thus, prioritizing healthcare business intelligence applications has special factors that stem from the industry, but there are also some factors that stem from the applications being analytical in nature. The two most prominent decision factors in this process are the crossover value of business intelligence applications and the added risk that can come from the analytical culture in an organization.
Many IT applications can be viewed as separate systems, even if they are modules of a larger supersystem. For instance, a payables application pays the bills and posts the appropriate journal entries. A receivables application collects the money and posts the appropriate journal entries. A claims processing application collects the required info and transmits the claim. Most operational applications do one main task; and, in many cases, adding a lot of other complicated tasks just adds risk and maintenance costs.
Crossover Value
Business intelligence, on the other hand, derives the bulk of its value from crossover uses of the same data. I believe that crossover value is a project prioritization factor that is often overlooked and just as often not fully understood. As a consequence, many highly valuable business intelligence applications are passed by because their immediately visible return on investment is not high enough to justify the initial investment.
One can visualize this as the difference between individual boxes on a conveyor belt versus nested boxes. If completed systems are the boxes and the conveyor belt is time, then the task in prioritization is to line up the boxes, hopefully placing the most valuable ones first. However, business intelligence applications are not individual boxes. They are actually nested boxes. The smaller boxes inside may not be as valuable as the larger box that surrounds them, but they may be the most feasible place to start building analytical applications.
A perfect example of an application whose value is often underestimated is quality analysis. Usually, the organization looks at this business intelligence application as a reporting system. If all it could do is collect measures, calculate results and produce reports, then perhaps it should be ranked lower than other projects. But consider the crossover value of measuring quality. Pay-for-performance contracts are generally written based on quality measures. This is crossover value. Effectiveness analysis and outcomes measurement use quality measures as inputs. Efficiency improvement and waste reduction efforts use quality measures to determine what is important to the organization and its constituents. Additionally, today marketing and service-line planning in healthcare are based on the areas of the organization that show the greatest promise in business terms as evidenced to a great degree by quality measures.
Crossover value is incredibly important, but is often overlooked. This is true in other industries as well, but it is critical in healthcare due to the increased pressures to catch up with other industries in technology and information usage.
Analytical Culture
Another factor that is specific to business intelligence application prioritization is the risk associated with the degree to which an organization has an analytical culture. Guessing incorrectly can derail business intelligence projects and ruin the chances of getting the full value from business intelligence. Some organizations, for instance, have an operational nature and are, therefore, more likely to accept reports that just tell them what to do to get the job done. Other organizations have a highly exploratory nature when it comes to data analysis. They do not want to be told what to do. Rather, they want to be given data to determine for themselves what to do and why.
For instance, two organizations I worked with were developing patient registry applications. The first emphasized using the application to prioritize their staff work and patient calls to provide the best care possible at the right time in the right place. Motion was their watchword. Their culture emphasized reports that told them what to do and why.
Another organization developed a registry application with two analytical questions in mind. These were:
1.Where do we put our next facilities and why?
2.What are the patterns in the population trends and how do we staff to anticipate those trends?
Their culture emphasized analytical exploration of the data to tell them what was happening.
Both of these types of value can come from the same application, but the structure of the system and its outputs depends on the analytical culture of the organization. This can be a key factor in prioritizing business intelligence applications and is often overlooked.
Next Steps
Business intelligence applications must go through the same project prioritization process as any other technology, business or clinical project. Understanding some of the factors that are specific to healthcare business intelligence applications will hopefully help to make them more attractive to the organization and could also keep the organization from making bad decisions on dead-end applications.
Consider the four factors described in this article when planning business intelligence applications. Think about the high road. Think about applications that help to bridge the conflict between the softer, mission-side of the organization with the harder, business side of the organization. In addition, consider the crossover value of business intelligence applications as well as the analytical needs that are consistent with the organizational culture.
No matter what, using data wisely is essential in today’s healthcare organization to make what we do more effective, more efficient, safer and, of course, more patient-centered.
Thanks for reading!