For decades, the healthcare sector has generated a wealth of data, driven by record-keeping, compliance and regulatory requirements, as well as patient care. While most of the information is stored in hard copy form, the current trend is toward holistic digitization.
By definition, big data in health IT applies to electronic datasets so vast and complex that they are nearly impossible to capture, manage, and process with common data management methods or traditional software/hardware.
My friends at Belitsoft have prepared an overview of how big data analytics can be used for the benefit of healthcare providers and patients alike.
According to the IDC (International Data Corporation) study, the volume of big data will increase faster in healthcare than it will in other fields like manufacturing, financial services, and entertainment. Thus, the industry is projected to see a CAGR, or a compound annual growth rate of 36 percent.
Big data in healthcare is overwhelming, not only because of its size but also of the variety of data types and the speed at which it must be captured and processed. It comes from sensors, medical devices, smartphones, networks, log files, transactional apps, web, and social media - much of it generated in real time and in huge amounts.
The rise of AI-powered chatbots, virtual assistants, and the Internet of Things (IoT) are driving data complexity, new forms and sources of information.
“In the future, in-home robotic healthcare assistants will monitor elderly patients and provide notification if an individual requires assistance, ensure medication is taken, and even perform simple tasks,” the report said.
Big data analytics: solutions to the industry challenges
The term refers to the use of advanced analytic techniques against extremely large, diverse data sets that include structured, semi-structured and unstructured information from different sources and in different sizes.
In that regard, what are the most urgent challenges for healthcare organizations dealing with big data and how to solve them effectively?
- Big data capturing
Capturing data that is clean, complete, and formatted properly for use in multiple systems is a constant struggle for organizations, many of which are failed. Recent research at an ophthalmology clinic found that just 23.5 percent of EHRs contain exactly the same info as reported by patients.
“Symptom reporting was inconsistent between patient self-report on an ESQ (Eye Symptom Questionnaire) and documentation in the EHR, with symptoms more frequently recorded on a questionnaire. These results suggest that documentation of symptoms based on EHR data may not provide a comprehensive resource for clinical practice or “big data” research,” the study described.
Providers can improve their data capture operations by prioritizing valuable types for their specific projects, involving the data governance and integrity expertise of health information management professionals.
Plus, they should develop clinical documentation improvement programs that train medical staff to determine what stats are useful for downstream analytics.
- Big data storage
Healthcare organizations require more storage space for big data analytics and the volume of unstructured information needed to be stored for analytics initiatives. It is a critical cost, security, and performance issue for the IT department.
As the volume of health-related data grows exponentially, some providers are no longer able to manage the expenses and limitations of on-premise data centers.
Cloud-based health IT infrastructure is becoming an increasingly popular option because of nimble disaster recovery and more flexible and scalable environment. As organizations adopt mobile apps, storing clinical data in the cloud gives users more complete access.
This method also saves organizations money by allowing them to purchase more storage space, rather than investing in additional on-premise servers.
A SADA Systems survey found that 89 percent of healthcare organizations are currently using cloud-based infrastructure. However, the major concern they have when moving to the cloud is the lack of control over where their data is kept.
Moreover, organizations are reluctant to trust a third-party vendor to host PHI, fearing that unknow security could lead to a data breach. That’s why discussing how their solution complies with HIPAA regulations is so important for cloud vendors.
“Organizations should always be leary of any vendor selling a HIPAA-compliant solution,” Jeff Thomas, CTO of Forward Health Group, said. “Even if a cloud solution enables you to use it in a compliant manner doesn’t mean it solves the compliance problem for you […] Is the vendor you choose willing to sign a business associate agreement? If they hesitate or don’t know what that is, they aren’t the right vendor to choose because they don’t understand your healthcare compliance needs when it comes to HIPAA.”
- Big data updates
Healthcare data is not static, most components require relatively frequent updates to remain current and relevant. For some datasets, like patient vitals, these changes may occur every few seconds. Other records, including a home address or marital status, may be modified a few times throughout life.
Providers must understand clearly which datasets need manual updating, and which can be automated, how to ensure this process without downtime for end-users, and how to allow updates to be conducted without demanding the quality and integrity of the datasets.
Organizations should ensure that they are not creating unnecessary duplicate records when attempting an update to a single element, which may make it difficult for clinicians to access necessary information for patient decision-making.
- Big data sharing
Exchanging data with external partners is essential, especially as the healthcare industry moves towards population health management and value-based care. However, data interoperability is a perennial concern for organizations of all types, sizes, and niches.
Fundamental unspecified differences in the way EHRs are designed and implemented can severely curtail the ability to move data between disparate organizations. So clinicians often go without the records they need to make key decisions, follow up patients, and develop strategies to improve overall outcomes.
The community is currently working hard to improve the sharing of big data across technical and organizational hurdles. Arising tools and strategies such as FHIR, public APIs, and partnerships like CommonWell and Carequality, are simplifying the data exchange between developers.
However, implementation of these methodologies has not yet reached a turning point. Thus many organizations are still cut off from the potentials inherent in the seamless sharing of patient data.
Analyzing big data enables researchers and business users to make precise and fast decisions by harnessing data that was previously inaccessible or unusable. However, big data poses great challenges.
In order to develop a healthy big data analytics ecosystem that connects all members of the care continuum with reliable, timely, and relevant information, providers have to overcome numerous problems (far more than we’ve listed). It will take time, commitment, and funding but success will relieve the burdens of all those concerns.