Discovery Health turns big data into better healthcare

Rapidly developing predictive models to improve services for health plan members and eliminate fraud

Published on 25-Jun-2013

"With the IBM solution, we can get the right information, develop the right models, ask the right questions, and provide accurate analyses that meet the precise needs of the business." - Lizelle Steenkamp, Divisional Manager, Risk Intelligence Technical Development, Discovery Health

Customer:
Discovery Health

Industry:
Insurance

Deployment country:
South Africa

Solution:
PureData System for Analytics (powered by Netezza technology), PureSystems, BA - Business Analytics, BA - Predictive Analytics, BA - Risk Analytics, Big Data, Big Data & Analytics, Big Data & Analytics: Operations/Fraud/Threats, Big Data & Analytics: Risk, Business Integration

Smarter Planet:
Smarter Healthcare

IBM Business Partner:
BITanium

Overview

Founded in Johannesburg more than 20 years ago, Discovery now operates throughout South Africa, with offices in most major cities to support its network of brokers. It employs more than 5,000 people and offers a wide range of health, life and other insurance services.

Business need:
Discovery Health wanted to be able to predict members’ health problems and put programmes in place to prevent them, in addition to finding better ways to identify and combat fraud. With enormous volumes of claims data, pathology results and demographic information to process, how could Discovery analyze the results fast enough to support business decision-making?

Solution:
The company faced the big data challenge by building a new analytics infrastructure that now delivers results in minutes, rather than days.

Results:
Reveals new areas of investigation for fraud teams: in 2012, it recovered over R250 million in fraudulent claims for the schemes that it manages. Identifies possible “phantom drug” prescriptions by mining data from pharmacies and health providers. This query was too complex for the previous infrastructure to calculate; it now runs in just two minutes. Cuts development time for predictive models from weeks to days by enabling analysts to design and test prototypes iteratively in a much shorter period of time.

Benefits:
Provides more accurate predictors for chronic conditions, helping Discovery initiate better preventative programmes to address members’ health problems before they become more serious.

Case Study

Smarter healthcare analytics

Instrumented
Millions of daily claims transactions are gathered into a central data warehouse and combined with demographic information, clinical information, member surveys and other data.

Interconnected
The data is fed into sophisticated predictive models and processed by analytics-specific hardware to provide insight to analysts and decision-makers across the business.

Intelligent
The ability to predict which health issues members are most likely to experience enables the development of preventative programmes to keep members healthy and costs low.

When your health schemes have 2.7 million members, your claims system generates a million new rows of data daily, and you are using three years of historical data in your analytics environment, how can you identify the key insights that your business and your members’ health depend on?

This was the challenge facing Discovery Health, one of South Africa’s leading specialist health scheme administrators. To find the needles of vital information in the big data haystack, the company not only needed a sophisticated data-mining and predictive modelling solution, but also an analytics infrastructure with the power to deliver results at the speed of business.

Business context
Founded in Johannesburg more than 20 years ago, Discovery now operates throughout the country, with offices in most major cities to support its network of brokers. It employs more than 5,000 people and offers a wide range of health, life and other insurance services.

In the health sector, Discovery prides itself on offering the widest range of health plans in the South African market. As one of the largest health scheme adminstrators in the country, its is able to keep member contributions as low as possible, making it more affordable to a wider cross-section of the population. On a like-for-like basis, Discovery’s plan contributions are as much as 15 percent lower than those of any other South African medical scheme.

Big data analytics
By building a new accelerated analytics landscape, Discovery Health is now able to unlock the true potential of its data for the first time. This enables the company to run three years’ worth of data for its 2.7 million members through complex statistical models to deliver actionable insights in a matter of minutes.

Discovery is constantly developing new analytical applications, and has already seen tangible benefits in areas such as predictive modelling of members’ medical needs and fraud detection.

Predicting and preventing health risks
Matthew Zylstra, Actuary, Risk Intelligence Technical Development at Discovery Health, explains: “We can now combine data from our claims system with other sources of information such as pathology results and members’ questionnaires to gain more accurate insight into their current and possible future health.

“For example, by looking at previous hospital admissions, we can now predict which of our members are most likely to require procedures such as knee surgery or lower back surgery. By gaining a better overview of members’ needs, we can adjust our health plans to serve them more effectively and offer better value.”

Lizelle Steenkamp, Divisional Manager, Risk Intelligence Technical Development, adds: “Everything we do is an attempt to lower costs for our members while maintaining or improving the quality of care. The schemes we administer are mutual funds – non-profit organisations – so any surpluses in the plan go back to the members we administer, either through increased reserves or lowered contributions.

“One of the most important ways we can simultaneously reduce costs and improve the well-being of our members is to predict and prevent health problems before they need treatment. We are using the results of our predictive modelling to design preventative programmes that can help our members stay healthier.”

Identifying and eliminating fraud
Estiaan Steenberg, Actuary at Discovery Health, comments: “From an analytical point of view, fraud is often a small intersection between two or more very large data-sets. We now have the tools we need to identify even the tiniest anomalies and trace suspicious transactions back to their source.”

For example, Discovery can now compare drug prescriptions collected by pharmacies across the country with healthcare providers’ records. If a prescription seems to have been issued by a provider, but the person fulfilling it has not visited that provider recently, it is a strong indicator that the prescription may be fraudulent.

“We used to only be able to run this kind of analysis for one pharmacy and one month at a time,” says Estiaan Steenberg. “Now we can run 18 months of data from all the pharmacies at once in two minutes. There is no way we could have obtained these results with our old analytics landscape.”

Similar techniques can be used to identify coding errors in billing from healthcare providers – for example, if a provider “upcodes” an item to charge Discovery for a more expensive procedure than it actually performed, or “unbundles” the billing for a single procedure into two or more separate (and more expensive) lines. By comparing the billing codes with data on hospital admissions, Discovery is alerted to unusual patterns, and can investigate whenever mistakes or fraudulent activity are suspected.

Transforming performance
To achieve this transformation in its analytics capabilities, Discovery worked with BITanium, an IBM Business Partner with deep expertise in operational deployments of advanced analytics technologies.

“BITanium has provided fantastic support from so many different angles,” says Matthew Zylstra. “Product evaluation and selection, software licence management, technical support for developing new models, performance optimisation and analyst training are just a few of the areas they have helped us with.”

Discovery is an experienced user of IBM SPSS® predictive analytics software, which forms the core of its data-mining and predictive analytics capability. But the most important factor in embedding analytics in day-to-day operational decision-making has been the recent introduction of the IBM PureData™ System for Analytics, powered by Netezza® technology – an appliance that transforms the performance of the predictive models.

“BITanium ran a proof of concept for the solution that rapidly delivered useful results,” says Lizelle Steenkamp. “We were impressed with how quickly it was possible to achieve tremendous performance gains.”

Matthew Zylstra adds: “Our data warehouse is so large that some queries used to take 18 hours or more to process – and they would often crash before delivering results. Now, we see results in a few minutes, which allows us to be more responsive to our customers and thus provide better care.”

From an analytics perspective, the speed of the solution gives Discovery more scope to experiment and optimise its models.

“We can tweak a model and re-run the analysis in a few minutes,” says Matthew Zylstra “This means we can do more development cycles faster – and release new analyses to the business in days rather than weeks.”

From a broader business perspective, the combination of SPSS and PureData technologies gives Discovery the ability to put actionable data in the hands of its decision-makers faster.

“In sensitive areas such as patient care and fraud investigation, the details are everything,” concludes Lizelle Steenkamp. “With the IBM solution, instead of inferring a ‘near enough’ answer from high-level summaries of data, we can get the right information, develop the right models, ask the right questions, and provide accurate analyses that meet the precise needs of the business.”

Looking to the future, Discovery is also starting to analyse unstructured data, such as text-based surveys and comments from online feedback forms.

About BITanium
BITanium believes that the truth lies in data. Data does not have its own agenda, it does not lie, it is not influenced by promotions or bonuses. Data contains the only accurate representation of what has and is actually happening within a business.

BITanium believes that one of the few remaining differentiators between mediocrity and excellence is how a company uses its data.

BITanium is passionate about using technology and mathematics to find patterns and relationships in data. These patterns provide insight and knowledge about problems, transforming them into opportunities.

To learn more about services and solutions from BITanium, please visit bitanium.co.za

About IBM Business Analytics
IBM Business Analytics software delivers data-driven insights that help organisations work smarter and outperform their peers. This comprehensive portfolio includes solutions for business intelligence, predictive analytics and decision management, performance management, and risk management.

Business Analytics solutions enable companies to identify and visualise trends and patterns in areas, such as customer analytics, that can have a profound effect on business performance. They can compare scenarios, anticipate potential threats and opportunities, better plan, budget and forecast resources, balance risks against expected returns and work to meet regulatory requirements. By making analytics widely available, organisations can align tactical and strategic decision-making to achieve business goals.

For more information
For further information please visit ibm.com/business-analytics

Products and services used

IBM products and services that were used in this case study.

Software:
SPSS Modeler, IBM Netezza Analytics, PureData System for Analytics (powered by Netezza technology), SPSS Statistics

Legal Information

© Copyright IBM Corporation 2013. IBM South Africa, 70 Rivonia Rd, Sandton, 2146, South Africa. Produced in South Africa. June 2013. IBM, the IBM logo, ibm.com, Cognos, PureData and SPSS are trademarks of International Business Machines Corp., registered in many jurisdictions worldwide. Other product and service names might be trademarks of IBM or other companies. A current list of IBM trademarks is available on the web at: www.ibm.com/legal/copytrade.shtml. Netezza is a registered trademark of IBM International Group B.V., an IBM Company. IBM and BITanium are separate companies and each is responsible for its own products. Neither IBM nor BITanium makes any warranties, express or implied, concerning the other’s products. This document is current as of the initial date of publication and may be changed by IBM at any time. Not all offerings are available in every country in which IBM operates. The client examples cited are presented for illustrative purposes only. Actual performance results may vary depending on specific configurations and operating conditions. THE INFORMATION IN THIS DOCUMENT IS PROVIDED “AS IS” WITHOUT ANY WARRANTY, EXPRESS OR IMPLIED, INCLUDING WITHOUT ANY WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND ANY WARRANTY OR CONDITION OF NON-INFRINGEMENT. IBM products are warranted according to the terms and conditions of the agreements under which they are provided.