Governments make thousands of policy decisions each year that affect the health and well-being of millions of Americans. Take babies, for example, where Medicaid covers one in five babies born in this country. Startlingly, we suffer from some of the worst infant and maternal health outcomes in the world. Better policy could make a profound impact in this area.
To address the rise in health inequities and budgetary pressures, state leaders are reexamining their health and human services (HHS) programs. The goal is to ensure they improve health outcomes, not just provide more health care. Given the magnitude and importance of these programs, policymakers are constantly looking for solutions to ensure the right care and social supports are provided at the right time and at the right cost.
Advanced analytics has shown tremendous promise in this area. Governments produce, store and consume large amounts of data. But deploying this data to create insights has historically been a challenge for health agencies. Too often, simple reporting is used to create policies that only guess at solutions. What our policymakers and health leaders need are stronger insights that allow for evidence-based decision making.
Ready to make data-driven health policy decisions?
SAS offers industry-leading software to integrate your data, generate insights and share compelling visualizations for evidence-based decision making across the health care spectrum.
Improve outcomes with health analytics
A derivative of Newton’s first law, data at rest needs an external force to create value. Data is most valuable when it is open, complete, reliable and centralized. By implementing data management and analytics technology, HHS agencies have proven they can:
- Create a 360-degree, whole-person view of each member to identify high-risk populations and target interventions where they are needed most.
- Assess quality and outcomes over time and across multiple dimensions – including demographics, disease progression, program aid category and managed care organizations – to ensure program investments are addressing population needs while controlling costs.
But acquiring these capabilities can be challenging. Mounting data volumes and complexity make it difficult for policymakers to harness the desired insight. In addition, outdated data management technologies force analysts to spend all day aggregating data across siloed programs, leaving little time for analytics. While simultaneously, legacy Medicaid management information systems are hard-coded and monolithic – making small upgrades or adopting new technology time-consuming and expensive.
A proven approach to analytics
Innovative HHS agencies are overcoming these challenges and achieving evidence-based policymaking through the use of enterprise analytics platforms. These flexible and robust platforms enable the rapid adoption of technology and faster innovation cycles, opening the door to advancements in value-based care and population health – even while reducing costs.
The most successful agencies focus on three related fields: data management, analytics and visualization. By aggregating and managing data, applying sophisticated analytics and making those analytical insights easily consumable through visualization, health policy leaders and their IT groups are unlocking transformational insights that can improve health outcomes within a single generation.
Let’s explore each of these technologies as steps in the analytical journey and see how each is being used to facilitate high-value care nationwide.
Step 1: Prepare your data for analytics
Government agencies are awash in data. But integrating data from an ever-increasing number of sources is no simple task. To further complicate the issue, data provided by various source systems is often of mediocre quality and outside direct agency control.
When dealing with multiple source systems, data quality issues arise, thus triggering the need for an enterprise data management approach. San Bernardino County uses a SAS® data warehouse to integrate disparate data sources and ultimately connect people to the right type of behavioral health services. For example, the organization maps data about its consumer population and uses these maps to determine the best locations for different behavioral health services that will best meet the location needs of the people they serve.
This enterprise data management approach not only benefits the community, it saves the county time. “We start with a question or hypothesis and mine our data warehouse for answers,” explains Sarah Eberhardt-Rios, Deputy Director for Program Support Services at San Bernardino County. “If we find that our hypothesis was wrong, we ask additional questions. This approach saves us months or years of manual or other types of analysis.”
Step 2: Apply analytics
As health agencies strive to make evidence-based decisions, it’s important not only to properly manage data assets, but to use that data to make mission-critical decisions.
Advanced analytics enable health care agencies to reveal the hidden insights and trends in health data to inform data-driven policy. In this step, analytical models are used in many ways such as predicting future population health needs, streamlining health care operations and optimizing financial management to improve beneficiary results.
One innovative state agency is using advanced analytics to improve its Medicaid program. The agency has adopted a data-centric strategy to embrace evidence-based decision making at all levels of the organization. In particular, the agency uses the SAS Platform to:
- Analyze geographic and demographic disparities in fee-for-service and managed care program coverage.
- Modernize and promote new models of care delivery that focus on access to care and improve the coordination of quality of care.
- Improve and execute value-based pricing models, which analyze episodes of care, provider effectiveness and disease management within the Medicaid population.
To expedite the development of analytical models, the agency uses machine learning – a form of artificial intelligence – giving analysts an intuitive interface to access and code predictive models in any programming language. In addition, the use of machine learning teaches the system to automatically prioritize workload based on prior successes.
This groundbreaking analytics project – which is garnering praise from CMS and industry peers – was implemented on time and under budget, providing a quick win for the agency and Medicaid beneficiaries across the state.
Step 3: Use data visualization to bring insights to life
Data visualization provides the link between analytics and action. Interactive dashboards bring insights to life, allowing policymakers and program analysts to easily see the results of advanced analytics, be it social determinants of health, drug overdose deaths or hundreds of other health-related datasets.
In New Jersey, the Interagency Drug Awareness Dashboard (IDAD) allows state agencies to exchange and view opioid-related data from disparate sources to mitigate the path to addiction for certain populations. The dashboard includes data from the New Jersey Prescription Monitoring Program, law enforcement, naloxone administrations, drug overdoses and other treatment sources.
Under the IDAD program, the Department of Law and Public Safety integrates the data into one centralized dashboard to create a clearer picture of the opioid epidemic – ranging from street drug activities to prescribing abuse. According to the state, information gleaned from the IDAD will help create a holistic picture of New Jersey’s opioid environment that will aid state agencies in developing and analyzing data that can be used to target intervention initiatives, enhance public outreach and education efforts, and develop other data-driven solutions to the opioid epidemic.
We need to make good decisions about the community we're serving, and the best way to do that is to collect, manage and analyze data. Sarah Eberhardt-Rios Deputy Director for Program Support Services San Bernardino County
How SAS can help
Across the country, HHS agencies are unlocking tremendous value in their data by using advanced analytics technology. Policymakers can now make timely and innovative policy decisions that have a profound impact on society.
The evidence is piling up. With analytics, those with a propensity for addiction can be supported by services that limit their suffering. Behavioral health patients can lead more fulfilling lives with targeted services. Children born today can experience a more responsive, efficient and resilient health care system than their parents. And this is just the beginning.
As the gold standard in analytics, SAS has become the shared language of health care data – used by every state public health agency, 100 percent of Fortune 500 health care companies and over 240 payers. SAS provides a robust analytics platform for our government clients that is the most cost-effective and efficient path to evidence-based decision making.
For a deeper dive into health care analytics, check out the following resources:
- Continuous monitoring: Stop procurement fraud, waste and abuse nowProcurement fraud, waste and abuse silently robs businesses an average of 5% of spend annually. And even when organizations invest in detection methods, they’re often let down by their techniques. Learn what continuous monitoring is and why this proven analytical method is key to fighting back.
- ModelOps: How to operationalize the model life cycleModelOps is where analytical models are cycled from the data science team to the IT production team in a regular cadence of deployment and updates. In the race to realizing value from AI models, it’s a winning ingredient that only a few companies are using.
- Next generation anti-money laundering: robotics, semantic analysis and AIAnti-money laundering taken to its next level is sometimes referred to as AML 2.0 or AML 3.0. What does this next wave of AML technology look like? What can it do that you can’t do with traditional AML? See the results innovative financial institutions around the globe are already getting.