Population healthcare management

Improve patient outcomes and minimize cost

Effective population healthcare management (PHM) offers many benefits—to health systems, practitioners, patients, communities, and payers. It can help improve health outcomes, elevate the patient experience, reduce health disparities, identify inefficiencies and care gaps, and control the cost of care. However, to be effective, PHM requires access to vast amounts of accurate, near real-time data from multiple sources, the tools to connect disparate datasets and use that data to develop actionable insights, and the safeguards to ensure that data remains secure at all times.

To overcome these challenges and support effective PHM strategies, health systems need a data platform that can capture, store, manage, and derive insights from data collected from admin, omics, clinical, and operational systems using a data lakehouse. The platform’s infrastructure must have a robust information governance framework that includes a citizen opt-out standard as well as interoperability, data, and cybersecurity standards. The platform must also provide advanced analytics, machine learning, and cognitive analytics, which are essential to designing effective, robust risk stratification methodologies and for monitoring the health of the population over time. With the ability to consolidate patient data spread across numerous healthcare information systems into a single record, healthcare teams can develop actionable insights that will improve patient outcomes while lowering costs for the patient, provider, and payer.

Make use of all the data

New approaches are required to help identify current healthcare needs and predict what patients will need in the future. The following architecture demonstrates how Oracle Data Platform is built to support population healthcare management by capturing, storing, curating, and analyzing data to better understand patient needs and deliver actionable insights.

There are three main ways to inject data into an architecture to enable healthcare organizations to optimally understand their patients.

• To understand patients, it’s imperative to know their medical history. To do this, healthcare organizations must obtain all their admission and transfer records from across healthcare information systems. This data can be enriched with patient data from third-party sources, such as unstructured data from social media and so on. Frequent real-time or near real-time extracts requiring change data capture are common, and data is regularly ingested from admission and discharge operational systems using OCI GoldenGate. OCI GoldenGate is also a critical component of evolving data mesh architectures where “data products” are the central data objects.
• We can now add streaming data from medical devices that will be ingested in real time using a streaming service/Kafka. Not only does this include data from all the devices and monitors a patient may wear while under the direct observation of their clinician, but it also includes data from all the personal wearable devices a patient uses, which can help their care team better understand their symptoms and lifestyle when they’re outside the treatment center. This streamed data (events) will be ingested with some basic transformations/aggregations that will occur before storing in cloud storage.
• While real-time needs are evolving, the most common extract from healthcare systems is a kind of batch ingestion using an extract, load, and transform or extract, transform, and load process. Batch ingestion is used to import data from systems that can’t support data streaming (for example, older mainframe systems). To get a holistic understanding of each patient, we also need data from operational systems, such as billing systems, and data arriving via the Fast Healthcare Interoperability Resources (FHIR) protocol from electronic medical record or electronic health record applications. The data is sourced across products and geographies. Ingestions can be frequent, as often as every 10 or 15 minutes, but are still bulk in nature. Groups of transacted datasets are extracted and processed rather than individual transactions.

Data persistence and processing options for collected data are built on three (optionally four) components.

• Ingested raw data is stored in cloud storage for batch processing, which will do the necessary cleansing, enriching, and so on to put the data into the necessary state to be consumed by downstream users, which can be people, applications, or machine learning platforms. Though some data may be directly placed in the serving data store, this data is also simultaneously placed in cloud storage. This data will be processed using Spark. Processing can be performed directly using OCI Data Flow or as part of a larger pipeline using the orchestration capabilities in OCI Data Integration. These processed datasets are returned to cloud storage for onward persistence, curation, and analysis and ultimately for loading in optimized form to the serving data store. Alternatively, depending on architectural preference, this can be accomplished with Oracle Big Data Service as a managed Hadoop cluster.
• We have now created processed datasets that are ready to be persisted in optimized relational form for curation and query performance in the serving data store. This will enable healthcare practitioners to better respond to a patient’s needs with the goal of avoiding having to admit or readmit them as an inpatient.

The analysis part is built on two technologies.

• Analytics and visualization services deliver descriptive analytics (describes current trends with histograms and charts), predictive analytics (predicts future events, identifies trends, and determines the probabilities of uncertain outcomes), and prescriptive analytics (proposes suitable actions, leading to optimal decision-making.). Together, they can be used to predict patient needs and offer suitable interventions. For example, analytics can predict whether a cluster of patients of a certain age, living in a specific area, subjected to varying environmental impacts, and prescribed a specific combination of treatments could develop a certain chronic illness.
• Alongside the use of advanced analytics, machine learning models are developed, trained, and deployed. These models can be accessed via APIs, deployed within the serving data store, or embedded as part of the OCI GoldenGate streaming analytics pipeline.
• Our curated, tested, and high-quality data and models can have governance rules and policies applied and can be exposed as a “data product” (API) within a data mesh architecture for distribution across the healthcare organization.

Reduce the complexity and costs of managing healthcare with the right data platform

The right data platform can help health organizations get the greatest value from all the available data to design better-aligned, more cost-efficient, and sustainable healthcare services. This includes better-tailored care plans and support for individuals, including early and proactive interventions such as routine population screenings, which can improve health outcomes as well as patient engagement and satisfaction.