- What are the data sources which provide data?
- Why data standardization is important
- Brief list of key healthcare data integration standards
- Healthcare data integration case study
The healthcare industry is technology-focused and eagerly adopts the most cutting-edge technologies. The adopted technologies are aimed at cutting costs, advancing care, and improving outcomes. Yet, they also bring with them new problems to be solved. We’ll discuss some of the challenges related to new technologies, which healthcare providers and facilities have to overcome.
Safe data collection and storage are only part of the problem. Healthcare data integration and interoperability are still one of the greatest challenges when it comes to utilizing data in profitable ways. In an interview for Health Data Management, Karen DeSalvo, MD, National Coordinator for Health IT, said, “Despite the widespread adoption of electronic health records, the integration of healthcare data remains a critical challenge for the industry as it strives to achieve interoperability”.
What are the data sources which provide data?
Data can be collected by various departments of a single medical facility or by various medical facilities such as hospitals, doctor’s offices, nursing homes, etc. Each healthcare facility has its own structure and unique features.
The classifications below are compiled according to the Top Master in Healthcare Administration:
Hospitals provide short-term care for people with severe health issues resulting from injury, disease, or genetic anomaly. They are open 24/7 and employ physicians of assorted specialties, highly skilled nursing staff, medical technicians, and administrative staff. Hospitals are equipped with specialized equipment to provide care for people with acute and chronic conditions.
Ambulatory surgical centers
Ambulatory surgical centers are designed to provide relatively simple surgeries that do not require hospital care after undergoing them. Such centers provide a safe environment for the surgery and basic monitoring during the initial post-operation hours and are a cheaper alternative to hospitals.
Doctor’s offices are medical facilities where treatment is provided to patients by one or several doctors. These offices provide routine care and the health providers operating there are either general practitioners or practice a common specialty. As a rule, apart from doctors, physician assistants and nurses are also on the staff of these facilities.
Urgent care clinics
Urgent care clinics offer immediate outpatient care and basic medical care, though they are not the same as hospital emergency rooms and they do not deal with complex health issues or true medical emergencies.
Nursing homes are intended for patients who need constant care that cannot be provided at home, but there is no need to hospitalize them. Usually, nursing homes are associated with seniors, but they can provide care to patients of all ages on a 24/7 basis. The staff of nursing homes typically includes physicians, specialized nurses, and therapists.
All these different facilities collect and generate medical data. In fact, a huge amount of data - which can in turn be divided into several types.
Healthcare data types according to the U.S. Department of Health & Human Services are:
This type of data is the data on the characteristics of the population served by a medical facility. Data usually includes information on the use of services and charges for those services. The sources of the data are claims, encounters, enrollment, and provider systems. Common data elements are a type of service, a number of units (e.g., days of service), diagnosis and procedure codes for clinical services, location of service, and amount billed and the amount reimbursed. The data is available in electronic format and generated in accordance with uniform coding systems and practices. The problem areas are completeness, timeliness, and limited clinical information.
Patient medical records
Patient medical records are the documents containing information about the medical history of a patient and the care provided. The adoption of electronic health records (EHRs) has made patient files more accessible. This type of data gives a lot of clinical details, yet, efforts to compile data are costly and time-consuming due to the various places where the data is generated and different record formats.
Patient surveys represent self-reported information on the care, service, or treatment received and outcomes of care. The challenge here is the standardization of information and its correctness.
Why data standardization is important
Careful and consistent data collection and processing using standardized definitions and procedures are critical for measuring the actual performance of medical facilities and the outcomes of care provided. According to NCBI, “data standards encompass methods, protocols, terminologies, and specifications for the collection, exchange, storage, and retrieval of information associated with health care applications, including medical records, medications, radiological images, payment and reimbursement, medical devices, and monitoring systems, and administrative processes”.
Data interchange standards facilitate interoperability. This interoperability is ensured through the use of common encoding specifications, information models for defining relationships between data elements, document architectures, and clinical templates for structuring data as they are exchanged. That is the software solutions in the healthcare domain must embody these standards to facilitate smooth and safe data collection, processing, and exchange.
Brief list of key healthcare data integration standards
The list of data integration standards is not exclusive as there is a wide variety of standards applied for various healthcare organizations and different types of data and data sources. Anyway, you can find six of them below.
Health Level Seven is an international community to provide a framework for the exchange, integration, sharing, and retrieval of electronic healthcare information. HL7 sets standards for exchanging clinical data.
Digital Imaging and Communications in Medicine is the international standard to transmit, store, retrieve, print, process, and display medical imaging information. This standard ensures medical imaging data interoperability and integrates image-acquisition devices.
Fast Healthcare Interoperability Resources (FHIR) is an open health data standard describing data formats and elements (known as "resources") and an application programming interface (API) for exchanging electronic health records. The standard is to ensure semantic interoperability of data. It also should minimize the need for metadata translation services, which is fundamental to most data integration technologies.
Logical Observation Identifiers, Names and Codes is the universal code system to medical terminology related to the Electronic Health Record. This system assists in the electronic exchange and gathering of clinical results of all kinds of observations and measurements from many independent systems.
Systematized Nomenclature of Medicine Clinical Terms is the comprehensive nomenclature of clinical medicine, which determines global standards for health terms for the purpose of accurately storing and/or retrieving records of clinical care in human and veterinary medicine.
Statistical Data and Metadata eXchange provides standard formats for data and metadata, together with content guidelines and an IT architecture for the exchange of data and metadata. The version of the standard for healthcare - SDMX-HD (Health Domain) allows medical facilities to share and exchange medical indicators and metadata between medical organizations.
How do healthcare organizations integrate disparate data from multiple sources on a practical level? Here are two cases of healthcare organizations that integrated data from different sources to obtain standardized and manageable data warehouses capable of providing consistent and actionable data.
Healthcare data integration case study
The goal of healthcare data integration is to provide communication between different EHR systems and the ability to derive clinical meaning from data. Integrated clinical data enables adaptive and purpose-driven analytics to improve and streamlines operations and allows insights to be derived from the data. How do healthcare organizations achieve this goal?
Health Catalyst reports on the success story of Orlando Health, a Florida-based, not-for-profit health system made up of eight hospitals and 50 clinics, which has built its own analytics platform. It took six months to load 10 data sources into the platform and only one week to deploy dashboards, visualizations, and analytic insights.
Orlando Health had a legacy system which employed a rigid enterprise model that was cumbersome to manage. The transition from a legacy data warehouse solution to the new analytics platform took over 90 days and involved extensive resources to implement the changes. The team of professionals included a data modeler, QA tester, analyst, subject matter experts, and product engineers.
To have the ability to rapidly acquire and link disparate healthcare data sources in various ways in order to answer critical clinical and business questions, the system was built on a DOS platform, which uses shared data marts delivering standard data elements that are populated from multiple sources. The data warehouse allows all units of the organization to use interactive dashboards, visualizations, and analytic insights with organization-wide standards on data for reporting. The system provides meaningful and actionable data all throughout the healthcare system, leading to actual business value at a lower cost.
Another example cited by Harvard Business Review is Sanford Health, a $4.5 billion rural integrated health care system, which delivers care to over 2.5 million people in 300 communities. The problem to be solved by the organization was the underutilization of rich data resources – the data on admission, diagnostic, treatment, and discharge data to online interactions between patients and providers, as well as data on providers themselves. The organization aimed to improve care delivery, patient engagement, and care access through data-driven innovations.
The healthcare system is made of its own mature data infrastructure including a centralized data and analytics team, a standalone virtual data warehouse linking all data silos, and strict enterprise-wide data governance. Sanford Health reached out to potential academic partners in data science to create a collaboration, the Sanford Data Collaborative, which would allow them to improve healthcare quality and lower costs.
For now, the collaboration has developed prescriptive algorithms, and leveraged advanced machine learning analytics, to provide targeted management and predict the risks of unplanned medical visits. Another algorithm developed is a patient-engagement score algorithm for people with multiple chronic conditions using pre-existing patient behavior data. The data collaboration also allows operators to make insights on provider turnover, which is helpful in identifying predictors of provider turnover and retention and developing policy recommendations to retain front-line providers.
The problems with medical data integration can be solved effectively and can be solved using different approaches, such as effective data warehouses, healthcare APIs, authentication protocols, and algorithms of data processing, etc, to create powerful and flexible systems.
If the problem of underutilization of data keeps down your business processes, it’s a good time to reach out to professionals and give rise to new opportunities.
Contact the Cprime Studios experts at email@example.com to assess your problem and we’ll tell what can be done to implement your own data integration ideas.