Research puts the evidence in evidence-based medicine. Business intelligence puts the evidence in evidence-based business decisions. One of the best business intelligence applications for healthcare organizations is the patient registry repository. This one application provides clinical value, as well as business value to these organizations. But it is generally overlooked when it comes to providing value for healthcare research. This article describes the various ways your organization can get this additional value from patient registries, which is data you already own.
Healthcare Research and Business Intelligence
Healthcare research is thousands of years old and has become an enormous component of the total healthcare industry pie. It is conducted by and for virtually every type of player in the industry—clinical organizations, payer/purchaser organizations, pharmaceutical organizations, public health organizations and so forth.
At its core, healthcare research involves:
•Formulating a scientifically interesting question (hypothesis).
•Gathering, preparing and storing data used to test that hypothesis.
•Finding patterns and trends in the data, as well as changes in those patterns and trends.
•Presenting the findings so they can be used to make evidence-based medical decisions.
Business intelligence is the commercial equivalent of evidence-based clinical decision-making. It too is thousands of years old, with roots in the first commercial transaction that took place between buyer and seller. Only recently has it become a large component of the business information technology market, with the advent of powerful databases and data manipulation capabilities.
At its core, business intelligence involves:
•Formulating a commercially interesting question (business driver).
•Gathering, preparing and storing data used to address the business drivers.
•Finding patterns and trends in the data, as well as changes in those patterns and trends.
•Presenting the findings so they can be used to make evidence-based business decisions.
Business Intelligence and the Patient Registry Repository
The best business intelligence application for clinical, as well as other types of healthcare organizations is the patient registry repository. Patient registries have been in use for over six decades, providing clinical value to the organizations that use them, such as:
•Who the patient is as well as their health history.
•What healthcare conditions or changes to those conditions caused the encounter.
•Where the patient encounter took place.
•When the encounter took place. This includes when during the day and more importantly when during the patient’s health history.
•Why the patient had the encounter (emergency, routine visit, hospital admission, etc.).
•How the encounter took place. Increasingly, remote care such as phone or email is used.
•Who provided the care.
The information in the patient registry is used to support clinical activities such as:
•Analyze patient population patterns and trends.
•Give clinicians a timeline of an individual patient’s illness and the care provided.
•Support patient reminder programs by phone, email, etc.
•Reporting to public health authorities for disease control purposes.
An emerging use for the patient registry is for business decision-making purposes, in which you can:
•Anticipate patient activity, what care they will need and the effect on your revenue.
•Anticipate the resources needed to provide that care, and the costs involved.
•Find ways to continuously improve the efficiency of your operations in providing care.
•Promote your strengths in providing care for marketing purposes.
•Support decisions that have a long-term, strategic impact on your organization and how resources are allocated to achieve your goals.
The Value of Patient Registry Repositories to Healthcare Research
Despite the tremendous value that the patient registry repository has for business intelligence in general, its value for healthcare research has been for the most part overlooked.
Registries can be used to support activities at virtually every step in the healthcare research process, including:
•Identifying scientifically interesting questions that are worthy of study.
•Defining the scope and strategy of the study.
•Speeding the processes involved in gathering, preparing and storing the data used in the study.
•Improving the administrative and financial aspects of the study.
•Providing technical solutions to support the data and information handling processes from end to end.
Patient Registry Repositories Support Research Scope and Strategy Decisions
Patient registries provide a rich source of real-world data to help you identify and define the scope and strategy of the study, telling you what to study and why.
•Improve Hypothesis Discovery. Mining the patient registry is a good way to discover real-world patterns and trends in populations, as well as to see shifts in those trends and patterns. This mining can be done either formally (i.e., systematically) or informally (exploration for novel hypotheses).
•Validate the Value and Feasibility of Studies. Research studies require sufficient populations in order to make them statistically valid, and to make them meaningful. Having readily available data to test this validity can circumvent a great deal of wasted work. For example, a hypothesis may seem like a good one on the surface, but if the researcher can only find one or two real-world examples demonstrating the hypothesis, then it may not be feasible or worthwhile to conduct the study at all.
•Prioritize Studies. Often the problem is not a lack of potential study topics, but an abundance. You need to prioritize them before conducting them, so as not to waste valuable time, effort and expense. A patient registry can give you ballpark numbers on which to base your prioritization.
•Narrow Research Topics. You may find that the scope of your research question is too wide. With patient registry data, you can look at the population from a number of angles which can help you define the scope more appropriately.
Patient Registry Repositories Support Research Data Handling Processes
Once the research strategy is set, the study itself requires sophisticated tools and techniques to support the actual processes involved.
•Consolidate Data from a Number of Sources. Rarely do research efforts involve information from a single subject area. Even rarer is to find all of the data in a single database, or even a single system. Patient registry repositories are built by consolidating data from a number of sources, such as encounter systems, lab systems, financial systems, pharmacy systems, etc. This data from various sources allows the researcher to answer questions from a variety of perspectives.
•Prepare and Organize Data. Business intelligence repositories require raw data to be cleansed, transformed and organized into logical categories before it can be used for analysis purposes. The same is true for research data. The same types of processes used to prepare patient registry data for clinical and business analysis purposes can serve the researcher in preparing it for hypothesis testing purposes.
•Slice, Dice, Sort and Sum. The central value proposition of business intelligence is that the same data can be used in a number of ways, by various decision makers for a number of analysis and decision-making purposes. A single registry, for instance, can be analyzed in healthcare research along patient lines, provider lines, procedural lines, demographic lines, etc. Having several different ways to look at the data allows the researcher the flexibility to find patterns and trends not otherwise visible to them.
Patient Registry Repositories Support Research Financial and Administrative Decisions
Research requires money, time, people, processes, tools and technology.
•Improve Funding Requests. Using the data in your patient registry repository gives you a head start in refining the budgets for the required resources more precisely, leading to a higher percentage of accepted requests.
•Define Appropriate Resource Needs. Not only are budget requests more refined by using the patient registry, but the types of resources needed are more refined by using this data. For instance, the real-world patient data in the repository can help you determine what types of specialists you need, what types of equipment you may need, what types of data gathering methods you need, etc.
•Reduce the Cost of Data Collection. The cost of a study may prohibit doing it at all. Data you already own in the patient registry may provide many, if not all, of the data points you need to conduct a valid study. This saves time and money.
Patient Registry Repositories Support Research Technology Decisions
Business intelligence technology processes and components already tackle some of the same issues encountered by researchers when conducting studies, such as:
•Create Repeatable Data Management Processes. The combined process of extraction, transformation and loading (ETL) of data is a well-worn path in business intelligence, which can be exploited for use in supporting healthcare research processes.
•Handle Massive Amounts of Data. Business intelligence technical abilities in handling enormous quantities of data have grown geometrically, approaching or even exceeding the data storage and manipulation requirements of healthcare research initiatives.
•Support Multiple Studies through Segmentation. In business intelligence, a subset of data used for a specific purpose is known as a data mart. This too is a well-worn path, since each department or function of a typical company needs data segmented to meet their particular needs. A research topic often has a number of segments, or scientifically interesting sub-questions, which require data to be “marted” in a similar way.
•Presentation Tools. And last but not least, the presentation of data is critical. All of the research effort would be a waste if the results are not understandable, are confusing or are not usable in some way. Business intelligence technologies offer the researcher various ways to present and disseminate their findings, including tables, graphs, downloadable datasets and so forth.
The potential value of business intelligence in general and patient registry repositories in particular to healthcare research is tremendous. Plus, business intelligence practices, processes and technologies are similar in many respects to those used in healthcare research. The first step is to determine which research topics might be best helped by using business intelligence and patient registries, and then map out a strategy for using these powerful resources in your total research effort.
Thanks for reading. I look forward to your comments.