Identifying Risky Opioid Prescriptions

hr hr hr

Created in response to the Medicare Modernization Act of 2003 (MMA), this system supports a mission of improving the cost and quality of care for Medicare beneficiaries – the reason why a group from SAS Institute were recently permitted access to the data, to analyze opioid prescription patterns that might indicate potential abuse.

SAS Leverages NewWave-Managed Database to ID Risky Opioid Prescriptions

For more than three years, NewWave has managed the world’s single largest healthcare research database, with over 327 billion records and rising. The system contains over 3 petabytes of data and 2.5 million unique datasets which are made available to approved researchers, who last year made a combined 1,750 data usage requests.

These strategies to integrate leaner, smarter technologies and best practices with an agile development approach, as well as NewWave’s previous success at supporting the main database infrastructure used to combat Medicare waste, fraud, and abuse, also recommended the company to play a major role as the system integrator.

The Importance of Location Data

In 2015, the number of opioid-related deaths in America for the first time surpassed 33,000, of which nearly half involved prescription opioids. In a recent analysis the client internally identified approximately 225,000 beneficiaries who have received potentially unsafe opioid dosing.

For its part, the newly released study by SAS presents clear evidence that location analytics, blending health and socio-economic data with geographic data available through the infrastructure managed by NewWave, can reveal the location of opioid abuse among at-risk Medicare populations.

In the economically distressed Appalachian region, where pain management and medication for blue collar workers injured on the job has been a common aspect of daily life, the opioid epidemic has hit hard. The region was allegedly targeted specifically for distribution by opioid manufacturers, according to widely published reports.

In several places such as Mingo County, a glut of opioids has occurred as the result of a high number of prescribers involved with “pill mills,” where doctors or unscrupulous providers hand out prescription drugs “like candy,” according to the SAS report.

Social factors contained in the NewWave-managed database were used by the SAS research team to begin identifying geographic hotspots where health or policy interventions might be most needed. Hotspots can help investigators determine where to allocate their resources in the attempt to uncover potential instances of opioid prescription abuse.

As part of its informative discussion around the opioid abuse data research, SAS provides what it calls an “analytics lifecycle” as a framework for researchers seeking to gain insights from large data sets such as those managed by NewWave:

  1. Data Management. In this phase, data is extracted from one location to another, for example using API calls to a central repository, transformed “long to wide” for analytic processing, and loaded into memory for fast access to reports.
  2. Data Explorations. This is the first step of actual analysis and typically involves summarizing the main characteristics of a dataset, which can include size, amount, completeness, correctness, and possible relationships between data elements or variables.
  3. Model Building. The process of developing a probabilistic model which describes the relationship between the dependent and independent variables. The main challenges at this stage are finding the key relationships. It’s important to test and compare models to choose one that best fits the purpose.
  4. Deploying Models and Reports. The model identified from the previous step as the best one for a particular use-case can be applied to a data set in a scoring process, enriching the data with additional variables that result from applying the model. The enriched data set can then be used to produce reports and/or dashboards for distributing insights to relevant stakeholders, ultimately supporting critical policy decisions.

Armed with the above analytics lifecycle, researchers accessing large datasets have a framework to analyze and understand opioid prescribing rates within a geographical area to target potential abusers efficiently. Insights produced from this approach to leveraging intelligence from massive amounts of data promise to help policy makers more fully understand the nature of the challenges they face, and to apply their limited resources more wisely.

Related Capabilities

img

Systems Integration

Successful systems integration requires a holistic approach toward technology and business processes where NewWave excels. It suits our culture to…

img

Cognitive Computing

NewWave is experienced with training machine learning (ML) algorithms to understand complex problems and deliver nuanced responses. We also leverage…

img

Software Development

Developing new applications of high quality that solve specific problems in today’s dynamic economy — and doing so within the…

img

Big Data

NewWave knows that good data is an invaluable resource. Everything we do reflects a practical understanding of the importance of…

Ready to start your next project with us?

Awards

A Top-10 IT Consulting Firm and a Top-5 Software Developer

NewWave is proud to have achieved many innovative accomplishments while supporting our customers and creating solutions for the greater good.

View Our Awards

Careers

Join our team to solve
for the greater good!

Be a part of the team that transforms the world of business solutions by bringing cutting-edge information technology and services to government and commercial companies in a range of industries.

Join The Team