- Healthcare organizations are often told that the importance of a strong data governance plan cannot be overstated, especially as their big data assets grow exponentially in this primarily digital era.
With EHR data, lab data, the Internet of Things, patient-generated health data, socioeconomic and community data, performance data, financial information, and imaging studies all contributing to a murky, swirling pool of potential insights, providers that want to succeed with quality improvement may benefit from focusing on quality and integrity from the very beginning.
But putting a data governance framework in place is a major undertaking that requires complete cultural and organizational commitment – and many providers simply aren’t sure where to start.
The challenge is complicated by the fact that the idea of data governance may seem somewhat abstract for clinicians and other front-line staff, who may not see the direct connection between high-quality analytics on the back end and their daily priorities of patient care.
It is often difficult to motivate these key players to improve their data generation habits without concrete evidence that altering a workflow or spending extra time on ensuring complete documentation will measurably enhance some aspect of their day-to-day responsibilities.
According to the eHealth Initiative Foundation and LexisNexis Risk Solutions, however, the connection between data governance and downstream patient care is clear.
In a new report, health information management (HIM) experts outline four of the most important direct benefits of data governance on patient engagement, big data analytics, and population health management.
Enhancing patient matching and identification
Even the most comprehensive and complex risk scoring, predictive analytics, and care planning is irrelevant if a provider is looking at the wrong patient’s chart.
Patient matching and identification issues are worryingly common in the healthcare industry.
In 2016, approximately half of HIM professionals reported encountering patient matching problems on a regular basis, AHIMA found. Nearly three-quarters spent time every week resolving discrepancies with one or more mismatched patient records.
HIM departments typically rely on matching multiple demographic elements, such as addresses and birthdates, to ensure they are identifying information with the right individual.
However, “not every system has the same number or type of variables and not every provider has a high enough level of data integrity to feed the algorithms required for matching,” points out the eHealth Initiative report.
“A record that is incomplete, incorrect, or outdated can negatively impact a patient’s treatment, affect reimbursement rates and patient satisfactions scores, contribute to denied claims, and require staff to take time to correct records. Coordinating patient matching across registries is also critical. An inability to do so diminishes the potential capacity of databases and creates silos of data.”
Organizations may consider developing unique identifiers for each individual, such as a numerical code, that does not rely on highly variable demographics to create and match patient records. Unique identifiers can reduce the incidence of duplicate files and conflated records that may have significant negative impacts on care.
Bolstering patient engagement and data access
CMS has made patient access to personal data a top priority for the healthcare industry, and is strongly urging providers to expand their data-driven patient engagement capabilities.
“When we go to the doctor’s office, we want to be able to have the information about what happened there,” said CMS Administrator Seema Verma to HealthITAnalytics.com.
“We want to be able to build our record from birth through your entire life. You want to be able to aggregate information from your medical record and from devices so you can put that story together and see what your health looks like.”
In order to keep patients engaged, build trust, and ensure their data is meaningful for collaborating around care decisions, providers must ensure that the information is accurate, up to date, and complete.
“When patients experience transparency, accuracy, and integrity of information, they will have greater confidence in their providers and become more engaged in their care,” the report asserts.
Patients must be able to control who can access their data through patient portals, decide which providers can see sensitive information such as mental health records, and understand how privacy and security regulations can work in their favor.
To create a meaningful data ecosystem that revolves around the patient and his or her needs, organizations should assess their interoperability and health information exchange capabilities, the eHealth Initiative and LexisNexis suggest.
“Organizations need data and information governance to exchange data with their patients, intra-organizationally, and between various entities” to support care coordination and create a comprehensive, longitudinal record of care accessible to individuals, the report notes.
A complete and accurate record that includes input from all members of an individual’s care team can equip patients to actively participate in making important care decisions.
Developing tailored workflows that include meaningful analytics
The explosion of digital information has irreversibly altered the way clinicians interact with data to make decisions – but not all providers are convinced that the changes have been positive.
Providers continually struggle with creating streamlined, intuitive workflows within their electronic health records, and many may be hesitant to add analytics insights to processes that already require dozens of clicks to navigate multiple screens.
Data governance planning is a critical component of successfully integrating data-driven risk scores, alerts, and insights without overwhelming EHR users.
“Even if data is correct, information delivery needs to be precise to avoid overwhelming providers at the point of care,” the report states.
“Care will be most effective when physicians receive the most relevant information in the right places at the right time. Consistently providing incorrect or irrelevant information to providers creates alert fatigue and may cause doctors to ignore information that could actually be useful in treating their patients.”
Delivering information correctly requires collaboration between data scientists, HIM professionals, workflow experts, and clinicians themselves.
Interdisciplinary committees that include physicians and nurses can help organizations develop strategies that meet the needs of all end-users, says Lisa Grisim, RN, MSN, Associate Chief Information Officer at Stanford Children’s Health.
“Governance is a shared opportunity and a shared responsibility,” she said. “Make sure that you’re being interdisciplinary in terms of job function in addition to bringing together different specialties.”
“If you can create a space where all stakeholders can share ideas and be heard by their peers, you are going to be able to create workflows that are much more suited to the needs of users across your organization.”
Careful and comprehensive planning before implementing a new risk score or care gap alert can ensure that providers understand the reasoning behind the notification and trust the data that supports it, making them more likely to engage with the message instead of ignoring it.
Managing populations and the social determinants of health
More and more organizations are recognizing the importance of non-clinical factors for patient care. The social determinants of health, such as transportation barriers or food insecurity, can significantly impact success with chronic disease management, patient engagement, and utilization rates.
“Leading studies indicate social and environmental factors account for nearly 70 percent of all health outcomes, making the addition of data around social determinants of health critical to the equation,” says the report.
“There is also value in accessing and sharing clinical data for quality reporting, measurements, registries, and patient care.”
But socioeconomic data is typically highly unstructured, and is often found in free-text clinician notes if it is recorded at all.
Data governance efforts can bring natural language processing into the equation, allowing for the automated extraction of meaningful data elements related to socioeconomic needs.
Organizations can also enlist their health information management teams to create structured templates to capture key social determinant data, which may make it easier to integrate the data into risk scoring algorithms.
As a result, providers may be able to proactively address these important challenges and connect patients with community services that could alleviate socioeconomic pressures.
Providers are just beginning to develop the analytics strategies to manage the social determinants of health, and their approaches are likely to develop as tools become more sophisticated and financial incentives change to support a greater focus on non-clinical factors.
No matter how the healthcare environment evolves, however, robust and comprehensive data governance will remain a core component of success with patient engagement, data access, analytics, and population health.
“As healthcare continues to amass huge volumes of data, a robust data governance culture is necessary for industry to thrive,” the brief concludes. “Data governance creates knowledge for common understanding and facilitates conversations around technology and data that ultimately enable healthcare stakeholders to work together to improve interoperability, patient care, and outcomes while promoting data access.”