How to Leverage Big Data for Patient Care Management
The healthcare industry is in the midst of a rapid structural evolution, with a growing emphasis on “big data” gathering and utilization. The Centers for Medicare and Medicaid Services (CMS) is a primary catalyst in promoting many of these changes, specifically regarding quality and clinical data reporting. Though many CMS reporting programs lead to an increase in care quality, it is often challenging for providers to constantly adapt to new guidelines and parse which data points are worthwhile to collect and analyze.
Along with the stress of collecting and using big data comes the question of which new technologies would work best for a unique facility. More than 90 percent of hospitals currently employ an electronic health record (EHR) system and nearly half of the American Hospital Association’s (AHA) Most Wired hospitals used telehealth services last year. This increase in technology utilization means more data to gather and analyze—but how can hospital leaders decide what information is most useful for their facility?
Top Health Systems Using Big Data Management
|Health System||Vendor||# Staffed Beds|
|Universal Health Services||Conduent||18,169|
|Catholic Health Initiatives||Conduent||14,353|
|Trinity Health||WhiteCloud Analytics, IBM||14,023|
|Providence Saint Joseph Health||Quantros, Strata Decision Technology||10,100|
|Kaiser Permanente||GetWellNetwork, Conduent||8,913|
|Dignity Health||Conduent, Optum||8,060|
|Prime Healthcare Services||Infor||6,429|
|University of Pittsburgh Medical Center||Nuance||6,288|
Fig 1 Data from Definitive Healthcare. The table consists of health systems using big data applications for performance management purposes, ranked by number of staffed beds.
Tracking the Right Data
Visual, interactive data is becoming a more popular tool for healthcare providers. Whether through charts and graphs or comprehensive infographics, visualizing data can be helpful in understanding trends in patient care. However, these dashboards and graphics are useless if they’re focused on the wrong data. Rather than relying on new artificial intelligence (AI) and other technologies to assess problems, use data a care facility or health system already has to identify obstacles.
EHR systems offer various methods for collecting and sorting through patient data. EHR systems even keep track of how often physicians follow up with high-risk patients—and how often the follow-ups slip through the cracks. Physician leaders can also use claims and readmissions data to analyze trends: Has there been an increase in patients diagnosed with hypertension? Was there a sharp decrease in patients being treated for kidney disease? Have patients been visiting urgent care or retail clinics rather than making an appointment with a primary care physician?
Providers are probably collecting the most useful data for their facilities already. However, an initial collection and sorting of the most essential data could reveal that a facility isn’t tracking the right data for what goals hospital leaders would like to achieve. Perhaps some physicians are better at maintaining contact with patients after they receive care, and the goal is for all physicians in the facility to perform at the same level—but the facility doesn’t record patient communication properly in an EHR or other record-keeping system. This could be a wake-up call that a healthcare facility is focusing on collecting the wrong varieties of data.
Additionally, it’s important to collect data at regular intervals. Though some data categories can be tracked quarterly, others should be analyzed monthly to most accurately measure changes. Quarterly data tracking is useful for assessing diagnosis and treatment trends and care delivery improvements, while monthly tracking is more beneficial for monitoring patient behavior and 30-day readmission rates. Once providers have gathered the right data for a given facility and found specific trends—like lapses in patient communication or high readmission rates—healthcare facility leaders can prioritize what goals they’d like to achieve.
Fig 2 Data from Definitive Healthcare
How to Integrate AI and Technology
New technologies, including AI, are not all-knowing solutions to every data-based problem a care provider may face. These tools are not replacements—only enhancements. Telehealth, for example, cannot replace an in-person physician consultation for every patient; it should be used to supplement in-person visits and provide access for rural patients. This is especially true when it comes to high-risk assessments. An EHR system can alert providers to a potential problem, but it is up to physicians and specialists to verify the significance of an issue.
Care providers can be easily overwhelmed by the number of patients they see every day and forget to follow certain procedures. The right technology can recognize when procedure has fallen through and signal providers to remedy the situation. EHR systems often have safety-net capabilities built in. If a patient’s test results are abnormal, even by a small fraction, AI can record this and send providers a notification to check in. These systems also record when physicians and other providers do not follow up with patients, holding them responsible and improving performance.
Healthcare facilities can use this real-time, interactive data to improve care quality and performance. Kaiser Permanente health system used EHR alerts and data collection to reduce the prescription of antibiotics by 22 percent in just 7 months. Kaiser has nearly 9,000 staffed beds across 44 hospitals, 59 urgent care clinics, 34 ambulatory surgery centers, and more according to Definitive Healthcare data.
In order to fully utilize technologies a facility is already using, like an EHR system, the entire staff should be knowledgeable of its best applications. This means continuous education for nurses and doctors, as well as other staff members who may be inputting or categorizing data in the system. Intelligent use of EHR systems enables nurses to stay engaged with patients—which keeps patients involved in their own healthcare experience.
Top Health Systems Using Inpatient EHR
|Health System||Vendor||# Staffed Beds|
|Department of Veterans Affairs||Cerner||39,243|
|Universal Health Services||Cerner||18,169|
|Community Health Systems||MEDHOST||16,997|
|Catholic Health Initiatives||MEDITECH||14,353|
|Providence Saint Joseph Health||Epic||10,100|
Fig 3 Data from Definitive Healthcare. Table consists of health systems using EHRs, ranked by number of staffed beds.
AI Best Practice
Though artificial intelligence and other technologies can have a positive impact on patient care and experience, there is a limit to what they can do. Providers may rely, at least partially, on EHR systems for alerts on the most high-risk patients. However, if the algorithm for risk prediction is inaccurate, providers could overlook some high-risk patients and prioritize those are at a lower risk.
If physicians receive alerts regarding abnormal results or risk factors, they have the option to override the AI alerts. It is always wise to double check technological assessments because they are so new. Machine learning is a great way to enhance patient care, but it is not a substitute for specialists or physicians.
Visit the Definitive Blog to read more about the top hospitals using an Epic EHR system.
Definitive Healthcare is the leading provider of the most up-to-date, high-quality, and integrated data on the healthcare market. Our database tracks financial, clinical, and newsworthy intelligence on over 8,800 hospitals, 1.5 million physicians, 600 payors, and dozens of healthcare industry players.