In today’s healthcare environment, the subject of data governance is at the top of the list for hospital executives. Within the framework of strategic planning, quality, big data and regulation, your data reflects everything you do: patient care, cost practices and outcomes. In fact, the very sustainability of your organization relies on quality data. Dr. Elliot King, Department of Communication Chair, Loyola University, MD, makes the powerful point that “Data is an organization’s most valuable asset.”
Data governance is a proactive process that provides the necessary framework to ensure data can be trusted and that there is clear accountability in place should an adverse event occur as an outcome of poor data quality. This structure is vital in managing the use, monitoring, maintenance and protection of an organization’s data. Without it, you may be missing crucial information needed to align clinical and business goals while reducing risk.
Data governance is not the routine maintenance of your data with oversight by the Information Services/Technology department. Data governance is a commitment organizationally to a process that manages the integrity of data. In turn, there is increased consistency and confidence in clinical and business decision making. There is better organizational planning, data transparency and optimized staff effectiveness through less re-work and decreased risk of regulatory fines.
In 2001, Congress passed The Data Quality Act (also referred to as the Information Quality Act). Embedded within a two-sentence rider was the requirement that the information you provide to any Federal agency must be consistent, of high quality and have data integrity. In other words, you are required by law to provide quality data.
Consider these notable comments:
- When asked about the biggest challenges and opportunities associated with the significant influx of data in healthcare, Anil Jain, MD, Vice President with (IBM) Watson Health responded: “The biggest challenges relate to the interoperability of data and then standardizing and cleansing the data.” “Just because you can get the data, that does not mean you can make sense of it.”
- “Accurately measuring costs and outcomes is the single most powerful lever we have today for transforming the economics of health care” Dr. Robert Kaplan, Emeritus Professor of Leadership Development, Harvard Business School
The American Health Information Management Association (AHIMA) has been a leading authority on health information management since 1928 and was one of four parties responsible for ICD-10 Coding Guidelines as well as a contributor to the AMA’s CPT terminology. Some important AHIMA tenets are that Information Governance (IG) should ensure information is trustworthy, actionable, and aligned with organizational strategy. It is needed for all information, whether it is in electronic or paper form. EHRs are assets. Like other critical assets—people, capital, inventory, etc.—information is a strategic asset that requires high-level oversight in order to effectively use it for organizational decision-making, performance improvement, cost management, and risk mitigation. Information Governance requires a system-wide, multidisciplinary team (clinical, finance, decision support, materials management, risk management, pharmacy & IT) with executive-level sponsorship.
How “Bad Data” Can Increase Risk
The Controlled Risk Insurance Company (CRICO) is the patient safety and medical malpractice insurer for the Harvard medical community. The Massachusetts-based company has expanded its proprietary coding system to capture EHR-related problems that have contributed to patient harm, and to guide the hospitals, physicians, and other providers it serves toward addressing vulnerabilities in their systems.
Results from a 2013 analysis of 147 medical malpractice cases were found to have an EHR-related contributing factor:
|Top Issues in Claims w/EHR factors||% Cases|
|Incorrect Information in the EHR||20%|
|Hybrid health records/EHR Conversion issues||16%|
|System failure – electronic routing of data||12%|
|System failure – unable to access data||10%|
|Pre-populating/copy & paste||10%|
|Failure of system design to meet the need||9%|
|EHR user training/education||7%|
|Lack of integration/incompatible systems||7%|
|EHR user error (other than data entry)||7%|
These errors can range from faulty data entry (inches vs. centimeters can distort BMI calculation), unexpected automatic data conversions (2.5 becomes 25 impacting medication error), accessing wrong patient file or field, and repeated errors (mistakes that persist in patient records for years). Which EHR vulnerabilities are most troubling? CRICO’s early analysis reveals that incorrect information in the EHR was a factor in 20% of the 147 medical error cases reviewed. Half of the cases resulted in severe injury with $61 million in direct payments and legal expenses. This risk can and should be eliminated.
How Good Data Goes Bad
EHR system conversions are one of the most prevalent sources of data issues. Whenever data from different sources (multiple hospitals, physician office, and outpatient clinics) is merged, the result can be inconsistency, duplication and questionable validity. Data may not be compatible when different systems are linked/interfaced together. Additionally, formal data standards can be old or not consistently applied over time. Without standards, you may end up with data that can’t be used for accurate comparisons or other analytics.
Deferring ownership of clinical data to IT is all too common; everybody uses data but nobody really knows who “owns” it. Your IT department needs to know how to resolve technical issues with data; don’t assume they know what the content should be. Data content issues fall into the overview of clinical and informatics staff, working hand in hand.
What is Good Data?
Your data must be available and easily accessible. This requires integration to enable all caregivers to see patient data in the right location at the right time while supporting patient care decisions and optimum utilization of resources.
You must have standardization as a bedrock of your data tables/files.
- Any information that is recorded using free text is not reportable.
- Non-standardized procedures, for example, can result in duplicates or incorrect selection while documenting a patient’s surgical procedure. Procedures that cannot be tied to industry standards may introduce insufficient specificity.
- A recent study published in the Journal of the American Medical Informatics Association found that a lack of standardization of observable markers of diabetes being collected in the EHR resulted in a significant variation from the standards created by the American Diabetes Association to diagnose the disease. The findings point to the need for standardization to avoid confusion among symptoms related to other conditions the patient may have.
Data must be complete. You must be able to show improved outcomes, regardless of the acuity mix of your patients. Data must be consistent. Necessary information is collected for all patients, all the time and data must be accurate. A single source of truth ensures that each time a data element is used, it means the same thing everywhere it appears.
Introducing Data Governance
The most important point of data governance is to recognize that senior leadership and stakeholders must be engaged to ensure that your data meets the requirements of “good data;” not only within your EHR but also with every other system with which it interacts. Governance means putting a strategy and process in place for the proper proactive management of good data practices.
Some example considerations in the development of a successful strategy for the management of data are:
- Put audit and validity checkpoints in place to keep your data accurate. Any interaction with humans will generate data errors – your aim is to identify and remediate as part of everyday activities across your hospital/health system.
- Institute a schedule of routine audits of information that crosses from one system to another to find items that impact accuracy of cost and charge for supplies used during patient care. Missing supply, implant or pharmacy charge coding and inaccurate mapping to codes can have a significant impact on revenue and case costing analytics.
- Challenge the content of your reports. “Apples-to-apples” comparisons aren’t always what they seem to be. Do you collect data that does not result in action? If so, why are you collecting it? Can you see your data in real time to allow proactive action to mitigate unplanned barriers to efficiency?
- Find any free text that can be converted to a standardized list of selections.
- Review all of the lists of options selected by staff during the documentation process to search for duplicates, missing options and “home grown” areas that need to be updated to today’s clinical and regulatory standards.
- Ensure that you are meeting all regulatory requirements by examining the data used to report to outside entities.
Introducing and establishing data governance is a challenging endeavor that takes discipline and time. Check out The American Health Management Association for more information.