Tag Archives: behavioral analytics

Transforming Claims for the Digital Era

As insurers undertake digital transformation programs, many rightly turn to the claims function. Claims is a very good candidate for such initiatives because of its importance to the relationship between customers and their insurers. Claimants and insurers both want speedy and fair resolution, based on clear lanes of direct and personalized service. A data-driven, analytics-enabled claims process can satisfy the objectives of all parties.

Continuous improvement to customer experience in claims is critical to any strategy. After all, claims are a real “moment of truth” for insurers, with meaningful impacts on outcomes and customer loyalty. Insurers that craft the right strategies and deploy the right mix of digital technologies will be able to turn their claims operations into a source of competitive advantage, market differentiation and brand perception. While advanced technologies such as robotic process automation (RPA) and artificial intelligence (AI) are very much part of the long-term transformation story, there is much insurers can do that will generate immediate benefits.

What matters to claimants — and how to deliver

EY’s insurance consumer research confirms that speed, efficiency and transparency are among the most important characteristics of a quality claims experience. Better data and analysis can help streamline steps in the claims process, setting the foundation for an enhanced experience. Those analytics also set the foundation for the future where many claims will be resolved via “no-touch” processes.

See also: 4 Ways That Digital Fuels Growth  

Insurers seeking to automate their claims processes or to achieve straight-through processing for basic claims have multiple options, including:

  • Advanced telematics data (including video imagery) can be instantaneously captured during an automobile accident and downloaded from the cloud to automatically trigger a first notification of loss (FNOL) entry. Underwriters can “score” the data to determine the extent of loss relative to the automobile’s current value.
  • Drones and satellites can survey damage and collect information about property damage to initiate claims before a homeowner makes contact.
  • Via intuitive apps or other interfaces, insureds can submit photos of damage to their homes or vehicles to initiate the claims process, provided there is no sign of fraudulent behavior (which analytics programs can evaluate).
  • Property and casualty (P&C) insurers may use historical repair data to dramatically decrease estimating times for different types of vehicles and homes. They may also better manage repair costs and quality based on deeper analysis of these data sets.
  • AI may be used in combination with social media and other data to scan claims for the likelihood of fraudulent behavior.

Insurers also have good options when it comes to personalizing service, which include:

  • Voice analytics that can assess customer sentiment during phone calls, with appropriate classification and prioritization of resolution.
  • Behavioral analytics that can be applied to model likely customer needs and identify high-value policyholders or those likely to dispute a claim.
  • Analyses of customer records that can identify claimants facing renewal as well as good candidates for purchasing additional products.

A redesigned claims experience can pay immediate dividends (e.g., lower processing costs, improving claims resolutions or higher renewal rates). In all of them, insurers can engage at key points during the claims life cycle, with accurate and consistent information delivered on a timely and transparent basis. At the same time, claims teams can focus on high-value interactions, high-risk claims and other exceptions.

The path toward a better claims experience

No matter where insurers fall on the maturity curve today, there is much they can do to transform the claims process. The path to success begins with a series of well-thought-out steps designed to produce useful learning and incremental value. Huge investments in new technology or large teams of data scientists are not required for substantial improvements. Organizational and cultural factors are also part of the claims transformation equation.

Insurers should endeavor to integrate third-party data (such as medical claims, consumer credit and weather data) with existing records. They also have the opportunity to pilot the use of automated notifications via chatbots and to encourage customers to submit photos of damage. While taking these initial tactical steps, they can begin building the business case for, and perhaps even pilot, more advanced capabilities, such as “no-touch” claims handling for specific products, regions, claims types or payments.

Insurers in the intermediate phases of their digital transformation journey should consider expanding automated claims handling to more claims types and larger amounts, broaden their use of chatbots for communication and seek to integrate more external data sources. They can also deploy drones as “adjusters” and establish analytics Centers of Excellence in claims.

More mature organizations will look to leverage new data storage and management technologies as the basis for advanced analytics and real-time visualization. They may also strengthen antifraud efforts by implementing machine learning. The most forward-looking insurers may build out data science teams to probe large and diverse data sets stored in analytics ecosystems. Similarly, they may expand claims volumes handled via RPA-enabled straight-through processing and evaluate medical treatments or repair effectiveness against leading practices.

See also: Digital Transformation: How the CEO Thinks  

As claims organizations become more digital, the benefits of additional data and more effective analytics should extend beyond the customer experience. Machine learning and visualization techniques can help assess and predict claims risk with greater accuracy and certainty. They also provide a consistent claims handling approach relative to unbiased reserving, litigation, subrogation and other claims processes.

It is worth noting that technology enhancements alone will not produce a claims organization for the digital era. A cultural willingness to embrace change also matters. Many insurers must overcome risk-averse cultures to encourage experimentation and “fast failures” in the spirit of learning what works best for their culture and customers.

How do they do that? Test-and-learn approaches are a good start for insurers with limited digital capabilities. Pilot programs for automated claims processing and bot-driven notification systems are an ideal place for many organizations to start.

Customer experience is everywhere

In the digital era, where customers have been trained to expect real-time access to data and personalized service, the stakes for the claimant customer experience have been raised. Insurers must learn to deliver what customers want and expect — and deliver it efficiently, accurately and quickly. Digital transformation makes it possible, while offering insurers significant upside in terms of lower costs, increased customer loyalty and reduced risk of fraud.

The True Face of Opioid Addiction

It’s likely that when people hear about the growing opioid addiction problem in America, the face that comes to mind is the one commonly shown on TV and in the movies, which is a very broad generalization : the young, strung-out heroin addict living on the streets. Or dying of an overdose.

Heroin abuse is definitely a growing problem in America. But it’s not the only opioid-related issue we’re facing. In 2012, an estimated 2.1 million people were suffering from substance abuse disorders from prescription opioid use, and deaths from accidental overdoses of prescription pain relievers quadrupled between 1999 and 2015. Sales of prescription opioids also quadrupled during this period.

While prescription pain killers are often seen as a gateway drug to heroin among the young, the issue is much broader than just one demographic group. The reality is that the face of opioid addiction could be the soccer mom down the block who has been experiencing back pain. It could be the marathon runner who is trying to come back after knee surgery. It could be your grandmother baking cookies as she works on recovering from hip replacement surgery.

In fact, it could be anyone. And that diversity is what has made prescription opioid addiction so difficult to manage.

Drivers of addiction

What is driving this explosive growth of such a potentially dangerous substance? Part of it, quite frankly, has been the incredible improvements in healthcare over the last 20-some years. Hip replacements, knee replacements, spinal surgery and other procedures that were once rare are now fairly common. More surgeries mean more patients who need pain relievers to help them with recovery.

The greater focus on patient satisfaction, especially as the healthcare industry shifts from fee-for-service to value-based care, has also had some unintended consequences. Physicians concerned about patient feedback from Healthcare Effectiveness Data and Information Set (HEDIS) measures or Medicare Star ratings have additional incentive to ensure patients leave the hospital pain-free. Physicians may prescribe opioids, particularly if patients request them, rather than relying on less addictive forms of pain management.

See also: In Opioid Guidelines We Trust?  

Here’s how that translates to real numbers. An analysis of 800,000 Medicaid patients in a reasonably affluent state showed that 10,000 of them were taking a medication used to wean patients off a dependency on opiates. This particular medication is very expensive and difficult to obtain – physicians need a specific certification to prescribe it. So it is safe to assume that the actual number of patients using prescription opiates is two to three times higher.

Those numbers aren’t always obvious, however, because the prescriptions may be obscured under diagnoses for other conditions such as depression. Indeed, more than half of uninsured nonelderly adults with opioid addiction had a mental illness in the prior year and more than 20% had a serious mental illness, such as depression, bipolar disorder or schizophrenia, according to the Kaiser Family Foundation. The result is that, without sophisticated behavioral analytics, it can be difficult to determine all the patients who are addicted to opioids. And what you don’t know can have a significant impact on care, costs and risk.

Complications, risk, and prioritization

Opioid addiction tends to interfere with the treatment of other concerns, especially chronic conditions such as depression, congestive heart failure, blindness/eye impairment and diabetes. As a result, physicians must first take care of the addiction before they can effectively treat these other conditions.

That is what makes identifying patients with an addiction, and prioritizing their care, so critical. Failure to do so can be devastating, not just clinically but financially – especially as healthcare organizations take on more risk in the shift to value-based care.

Take two patients with an opioid addiction who are on a withdrawal medication. Patient A also has eye impairment while Patient B is a diabetic. If the baseline for cost is 1, analytics have shown that Patient A will typically have a risk factor of 1.5 times the norm while Patient B, the diabetic, will have a risk factor of 5 times.

Under value-based care, especially an Accountable Care Organization (ACO) where the payment is fixed, the organization can lose a significant amount of money on patients who are costing five times the contracted amount. For example, if the per member per month (PMPM) reimbursement for the year is $2,000, this patient — who is using this medication for withdrawal from an opiate dependency and is a diabetic — will end up costing $10,000.

It is easy to see why that is unsustainable, especially when multiplied across hundreds or thousands of patients. Yet the underlying reason for failure to treat the diabetes effectively – the opioid addiction – may not be obvious.

Healthcare organizations that can use behavioral analytics to uncover patients with hidden opioid dependencies, including those on withdrawal medications, will know they need to address the addiction first, removing it as a barrier to treating other chronic conditions. That will make patients more receptive to managing conditions such as diabetes, helping lower the total cost of care.

They can also use the analytics to demonstrate to funding sources why they need more money to manage these higher-risk patients successfully. They can demonstrate why an investment in treating the addiction first will pay dividends in the long term with a variety of chronic conditions.

See also: How to Attack the Opioid Crisis  

Many faces

It’s easy to see that opioid abuse in all forms has reached epidemic levels within the U.S. What is not so easy to see at face value is who the addicts are — or could be.

Despite popular media images, the reality is that opioid addition in America has many faces. Some of them may be closer to us than we think. Behavioral analytics can help us identify with much greater clarity who the likely candidates are so we can reverse the trend more effectively.