How the right data analytics diminish administrative burden on clinicians

Data flooding the healthcare industry has the potential to completely revolutionize patient care and drive improved health outcomes. Yet when left inadequately structured or under-automated, the deluge of data is one contributing factor to administrative burden — a pervasive issue affecting clinicians across most specialties.

Eighty percent of physicians today are professionally overextended or at capacity, leaving them with no time to see additional patients, according to the 2016 Physicians Foundation survey. This finding is troubling for many reasons, one of which is the fact that the majority of physicians consider patient relationships the most satisfying part of their job. In short, administrative tasks or other work that takes clinicians away from patients directly contributes to their overall morale, or lack thereof.  


Physicians are not data-averse — they just need data that is prioritized, meaningful and able to inform decisions. When faced with a wealth of unstructured data, physicians may become overwhelmed and have difficulty driving actionable insights. Research shows prolonged task saturation yields cognitive overload, leading to more stress and anxiety.


"Clinicians willingly provide detailed data when it is perceived to be enabling care, but not to satisfy some top-down bureaucracy," says John Middleton, MD, vice president and CMIO of Broomfield, Colo.-based SCL Health.

As data grows more sophisticated in healthcare, hospitals and health systems are devoting more attention to the caregiver experience and how data can bolster it. From operational analytics to genomic data, care teams are exposed to a vast amount of information each hour of the day. CMIOs increasingly see the need to preserve their clinicians' cognitive energy and equip them with the data most likely to yield the greatest returns on efficiency, patient safety, health outcomes and patient satisfaction.

Below are overviews of how health systems deploy three specific forms of data analytics —  automated intelligence, predictive analytics and precision medicine — to optimize their care teams, reduce administrative burden and reinforce the patient-physician relationship.

Automated intelligence

Automated intelligence solutions help hospitals simplify a range of daily tasks, from notification management to EHR documentation.


Physicians face EHR notifications in mass quantities. On a daily basis this onslaught of alerts creates a palpable but avoidable sense of administrative burden. This gradually takes its toll on clinicians' cognitive acuity and takes them away from patient care and interactions.


A study published in JAMA provides some insight regarding the number of EHR notifications clinicians receive per day. Researchers found 92 physicians across three Texas group practices received 276,200 EHR message notifications over 125 workdays. On average, that amounts to 2,210 notifications per day.

Health systems across the country are deploying analytics to stratify those alerts and better understand how physicians can most effectively interact with them. SCL Health is one of them.


"We are trying to use analytics to understand which alerts are firing, used, ignored or effective in order to manage the portfolio better," says Dr. Middleton, CMIO at the system. "We are developing modern dashboards with patient-level drill down for a host of activities, and think this will make managing a number of patient care and quality activities much easier."


Natural language processing one another automated intelligence tool designed to reduce the time-consuming manual data entry many clinicians face in daily use of the EHR. Thesoftware interprets language in clinician documentation to suggest more specific descriptions for coding purposes. By catching discrepancies in the beginning, providers don't spend time on clinical note clarification later.


Golden Valley Memorial Healthcare, a nonprofit system based in Clinton, Mo., and anchored by a 56-bed hospital, uses a cloud-based platform featuring speech recognition software. The system deployed the AI to speed up documentation production as well as collect more relevant data.


Dr. Middleton at SCL Health says the system does the same for key service lines and departments, including the emergency department and cardiology.  "We have also developed EMR templates and commands to be used with the basic speech recognition. In such cases, the workflow is often much faster than baseline," he notes.


Predictive analytics 


Data-powered forecasting empowers clinicians and hospitals' operational teams to arrive at more informed care decisions and properly allocate staff and resources, respectively.


On the clinical side, predictive analytics have the power to transform disease management. Sepsis is one prime example of a condition data analytics can counter or completely prevent. By using predictive analytics through data platforms, hospitals can implement early detection of high-risk sepsis patients. Armed with a better idea of who is most at risk in the hospital, clinicians are able to target and deploy effective care interventions for the patients most in need. This ultimately saves time, costs and patient outcomes by preventing further readmission, other hospital-acquired infections and gaps in care.


For clinicians and care teams, data analytics are most helpful when integrated into the care team's workflow, says Paul Fu, CMIO at Harbor-UCLAMedicalCenter. Analytics tools should be easily accessible and well annotated.

On the operational side, predictive analytics can help hospitals develop more efficient staff schedules and align resources with expected patient flow. This reduces the ongoing administrative tasks physicians face on a daily basis by allowing them to maintain their focus on patient care and clinical decisions versus capacity or patient flow problems.

A lack of forecasting for patient volume may result in ED crowding, extended length-of-stay, prolonged discharge processes, poor staff communication and a disruptive care experience for the patient. When armed with data to leverage beds and staff for the expected patient volume depending on the season, staffing levels or other external factors, hospitals are less likely to confront these trying circumstances and can better preserve their clinicians' mental sharpness for care decisions.


Precision medicine

As far as data goes, precision medicine is in a league of its own. The practice involves a molecular diagnosis of a disease and a medical plan personalized for effectiveness based on the patient's DNA and unique genetic variations. Through the analysis of enormous data sets, care teams can glean life-, time- and cost-saving insights, such as whether a person faces a high risk of certain diseases or how that individual might respond to certain medications.

Precision medicine is one of the youngest fields in medicine, emerging shortly after the human genome was mapped in 2003. Given its novelty, hospitals and health systems are still in the early stages of finding the right technology to best support genomic datasets, translate them to biomedical insights and — most importantly — support the resulting clinical decision-making.

Academic medical centers are pioneering precision medicine: 71 percent of AMCs said the new field of medicine will play a significant role in their organizations in the next five years. The challenge is most EHRs and legacy health IT infrastructure are ill-equipped to deploy genetic data for clinical decision support or meet other demands, some as basic as data sharing and storage. One patient's sequenced genome requires 1 terabyte of data; multiple that by 1,000 patients and systems need a perabyte-friendly enterprise data storage system.

Sequenced genomes bring a vast amount of data at clinicians' fingertips. If not deployed correctly, the sense of administrative burden care teams will experience is imminent. Healthcare providers must frame perabytes of genomic data in a way that empowers physicians to find the most pressing genetic variations and tailor care decisions appropriately. Otherwise the value of personalized medicine is largely lost in a sea of bytes, and physicians are disappointed to see the same data once anticipated to alter the course of medicine result in little action.  

To avoid either scenario, health systems are working to find the right custom solutions to glean actionable, meaningful insights from quadrillion bytes of data and thousands of unique genetic variations. Danville, Pa.-based Geisinger Health System is one system that has garnered great attention over its precision medicine pursuits in the past year.

Geisinger has secured 100,000 people to participate in its genomics biobank, and the system invested in technology that integrated gene scans into its EHR. The system deliberately screens patients for 27 conditions and pushes easy-to-interpret gene variations to the clinical team. David Ledbetter, PhD, Geisinger's chief scientific officer, told NPR genetic testing results in actionable information for 3.5 percent of patients, although he expects that figure to grow to 5 to 10 percent as scientists continue to correlate more diseases to certain genes.

John Kravitz, Geisinger's CIO and interim chief data officer, notes increased attention around population health via advanced analytics and genomics over the last year, "especially as genome sequencing is beginning to identify risks to patients that save their lives," he said. "The integration of genomics into the provider workflow as this information becomes available is an item that a lot of providers are wrestling to accomplish successfully."

Copyright © 2024 Becker's Healthcare. All Rights Reserved. Privacy Policy. Cookie Policy. Linking and Reprinting Policy.


Featured Whitepapers

Featured Webinars