2443 Fillmore St #380-17156
San Francisco, CA 94115
Copyright © 2021 Carta Healthcare, Inc. All rights reserved.
Health systems and individual hospitals and clinics see thousands of patients annually. There are countless physicians and nurses who dedicate themselves to providing and documenting care to this inpatient and outpatient visitors.
The sheer volume of patients and the corresponding data produced can make it challenging for hospitals and healthcare systems by making clinical data abstraction for patient registries manually tough, as the patient registries have proven themselves to be very crucial for patient outcomes.
The recent COVID-19 pandemic gives us a deeper insight on the role of data registries and the impact they make, which raises an important question, which is how to improve the quality of data and speed up the time?
Clinical data abstraction is the process of searching through medical records—electronic and/or paper—to identify the data required for secondary use. This process results in the summary of information about a patient for secondary use.
Data abstraction relates to direct matching of information found in medical records to the data elements required. This practice also involves additional operations on this data as categorizing, coding, interpretation, summarizing and then calculating.
Healthcare organizations can use data from clinical registries to measure their outcomes and perform against other organizations. The abstraction and reporting required for registry can be quite a painstaking process for which hospitals often have a dedicated team.
Clinical data abstraction is a primary method of data collection in clinical research, which helps in surveillance and identifying trends of disease, improve quality and safety of care provided, performance measurement, and determine cost of providing the care, among other uses.
Conventionally, clinical data abstraction has been a manual process, in which we enter relevant data that is not collected electronically into pre-defined fields so that every appropriate person or organization who needs access to the data can do so easily. The entire process is significant because this practice of abstraction culminates in summarizing information about a patient.
Patient files are reviewed and key data points are abstracted, which are then fed into electronic files. The sources of this data can vary depending on the measure or purpose, ranging from paper medical records, EMRs, administrative databases, etc.
Thorough review of both large and small data sets and documenting of information is critical because of the light it sheds on future decision making.
This conventional process revolves around collecting organizationally defined, clinically relevant data elements from a repository of documents describing each patient’s medical record. The process makes detailed patient data available in the electronic chart and thus facilitates access to care without having to refer to paper documents or an EMR.
It requires accurate information for clinical registries to be actually useful for advancing healthcare, and the data and records are essential for communicating patient and clinical information. But providers often find it difficult to get the time and resources to pull data from medical charts. The pressure of keeping up with this monumental task can be quite burdensome, if registry abstraction doesn’t get the attention it deserves.
Manually abstracting this huge chunk of information threatens to swamp healthcare organizations. Patient data buried in different systems and varied departments presents quite a challenge to hospitals in calculating and reporting core measures in a timely fashion.
This turning of unstructured data into structured data that is actionable is the primary goal of the exercise. And when it is required that patient care issues be identified, pinpointing the responsible contributing factors can be quite difficult.
It may take a considerable amount of time, which significantly slows down and hampers a hospital’s ability to correct quality problems and successfully impact quality improvement. Also, it has a financial impact on the facility as your top rated staff might be swamped with this manual labor, instead of focusing on other aspects of clinical care.
With automation healthcare providers can now treat this process of data collection not as an ancillary matter, rather as information evaluated simultaneously with the provision of care. It also helps address the time lag between provision of care and reporting of measurement on the quality of care. Feedback loops are a key component of any robust quality improvement framework, and they only work when this loop is timely.
Artificial intelligence (AI) effectively reduces this burden on healthcare facilities and enhances the reporting process. If you are wondering why AI, well the reason is quite straightforward, it can perform just and–often better than–humans in carrying out abstraction processes when dealing with complex and spread out data. It requires a large volume of data from different repositories between different departments for compliance. This can overtax the dedicated staff which can lead to abstraction and reporting becoming a delayed task. This can be bad because to reduce lag in reporting quality data, it is necessary to adopt the approach of better capturing and reporting quality measures simultaneously with the provision of clinical care.
With their efforts drawn away from the stress of abstracting, clinicians can focus on utilizing the data got from registries to perform quality improvement projects and improve patient care. Hospitals then save on cost and resources dedicated to abstracting, and produce more accurate data in a time efficient manner where clinicians have access to relevant information right when they need it.
In the best of circumstances, hospitals and clinicians find delayed quality data to be less meaningful and less accurate for improvements. However, in times of crises, much like the ongoing pandemic, this effect of data and time lag is magnified.
It directly translates into an absence of timely information on performance of the healthcare system. And this is precisely what many healthcare systems saw as the wave of COVID patients began. By the time the data is available, significant opportunities for learning and improvement have been lost.
Understanding the benefits of automating this process, healthcare organizations are pivoting to solutions like Carta Healthcare’s Atlas, which focuses on improving quality care and patient care by taking away the stress of abstraction.
Carta Healthcare’s Atlas data abstraction solution for registries, harnesses artificial intelligence to translate your data from clinical notes and medical records into recommendations.
Atlas automates the clinical abstraction process by using AI to translate data from clinical notes and medical records into content recommendations. Following up by auto-filling fields, it assists abstractors so that they can process registry cases with greater efficiency. Atlas makes the job easier by taking the job of the abstractor from author to editor.
One of the bigger advantages is that it just doesn’t make the process more efficient by cutting down on time and expense, abstractors can also enjoy a full view of the entire process with the ability to trace data back to its source and view it in context. Providers are also working with Atlas to manage submissions and normalize data to industry standards (FHIR).
By maximizing the use of IT and AI in the clinical data abstraction process, healthcare organizations can reap many benefits. These include
Looking for a solution to automate and simplify clinical tasks, freeing your clinicians to focus on patient care? Carta Healthcare works with leading cardiac and pediatric healthcare providers.