Research: Recent Trends in Health Care Data Analytics

Data analytics can and does change the way health care does business—yet there are still key challenges to adoption.

To develop and implement effective business strategies and plans for its clients, FrogDog continually reviews industry trends. In these articles, we share recent insights that affect key industry sectors.

Data analytics capabilities offer a number of benefits for all players in the health care continuum, including payers, care providers, patients, and product and service providers. Through data analytics, organizations are improving patient care, reducing readmission rates, and improving health outcomes.

Worth $4.8 billion as of late 2014, experts predict the health-analytics industry sector will see substantial growth and exceed a worth of $20 billion by 2020. At present, North America makes up the largest market share for this industry.

Key Drivers for Health Care Data Analytics

Increased technological sophistication has made data analytics more viable and more powerful than ever before.

At the same time, changes in the health care climate have begun to drive all industry participants toward improving the overall quality of care provided and the cost at which it is provided.

In recent months, across all players in the health care spectrum, below are some of the factors that have served to drive health care organizations’ prioritization of data analytics:

• Rising costs: Everyone involved in every facet of health care is investigating solutions to the rising cost of patient care. Data analytics helps through assessing trends and areas of improvement and by providing algorithms that reduce readmissions, improve the efficiency and efficacy of clinical operations, and identify the most at-risk patients for targeted treatments.
• Budgeting and forecasting: Through data analytics, health care organizations can track trends in cost, use, and risk measures so that they can understand and forecast future needs. This has helped organizations move away from reactive decisions and toward using real-time data to make more effective budgeting adjustments.
• Risk management: As health care organizations move to more outcomes-based business models, risk stratification and managing population health have never been more important. Using data analytics, providers can now pinpoint which patients are at high risk for readmission and develop preventative processes that improve the health of the patient as well as reduce cost.
• Fraud: Health care organizations can now use predictive analytics models to detect fraud. Examples of types of predictive models in use today include the following:
  • Rules-based models: This model highlights specific charges automatically. For example if a charge originates from a stolen Medicare ID number, it’s immediately flagged as fraudulent.
  • Anomaly models: These models flag unusual behavior, such as more procedures billed in a day than a provider could reasonably handle.
  • Predictive models: Predictive models detect when billing patterns indicate fraudulent activity.
  • Social-networking models: These models monitor previously fraudulent providers or businesses.
• Physician performance: Data analytics can measure physician performance against certain targeted patient and health outcomes. These assessments help payers and providers give physicians incentives to provide long-term health solutions and focus on the overall health of patients.
• Tracking effectiveness: Many employers and payers adopt programs to increase staff and member engagement with their health care and to improve overall health and wellness. Employers and payers can now use data analytics to measure the effectiveness and success of these programs.
• Patient and member education: Payers and providers can use data management to provide better health service to patients and members. By analyzing past health activity and recommendations, patients can gather information on their states of health and take preventative measures to improve their long-term health outcomes.

Factors Slowing Data Analytics Adoption in Health Care

Health care has more slowly adopted data analytics than other industries. One 2013 report from Gartner showed that market penetration within health care is relatively low, at between 5 and 20 percent.

Larger health care networks, systems, and organizations tend to have higher rates of data-analytics adoption, likely because they can realize more savings from these programs, which offset the high costs of implementation.

A few key factors affecting the health care industry’s slow adoption of data warehousing and analytics systems include the following:

• Data security: Health care executives are concerned with the security of their data once they have made them digital and accessible, whether hosted on the cloud, in hosted servers, or in on-site server networks. Recent high-profile data breaches at several health care companies have heightened this concern.
• Data integration: Typically, health care organizations store data in multiple formats. Data-analytics systems must be able to collect and integrate these disparate sources. Lack of true interoperability between data sets and technologies poses a huge challenge.
• Lack of historical investment: Rather than updating existing solutions, many health care organizations need to implement data warehousing and analytics for the first time. Launching data warehouses and analytics systems is an intensive investment of time and money. Taking on this level of cost must be justified by potential savings that can be realized relatively quickly. To move forward, top levels of management must be on board.
• Data use: While technology has made it possible to collect an enormous amount of data, the sheer amount of information can be overwhelming, and companies are struggling with how to put it to best use.
• Staff knowledge: Some organizations cite a lack of qualified staff as the primary reason they have been unable to take advantage of the opportunities presented by health care analytics. The health care industry is looking to upskill and hire more staff in this area.
• Industry standards: The disparate data types and formats and collection systems shows the lack of standardization in the health care industry. Working together to identify a data standard will take the industry considerable time.

For further information about FrogDog’s recent research into trends in health care data analytics, reach out to us directly.

FrogDog continually researches and monitors industry trends for its clients. Does your business know what is happening in your industry and have strategies to complement it? If not, contact us.

Image courtesy of cooldesign/freedigitalphotos.net