Skip to main content
Webinar

What Will MPL Look Like After COVID?

  • Join MPL experts as they bring their unique perspectives to the challenges and opportunities of potential changes in healthcare, cyber risks, telehealth, and more.
Government Relations

COVID-19 Information Center

  • Senate Leadership Unveils COVID-19 Legal Reform Proposal
  • HCLA Submits Written Statement to Senate HELP Committee
  • Bipartisan Congressional Letter in Support of H.R. 7059

Inside Medical Liability

Fourth Quarter 2020

 

 

Cover Story

Optimize MPL with Data Analytics

Leverage the transformative power of data for actionable business insights and improved patient safety

BY AMY BUTTELL

 

There’s no shortage of medical professional liability (MPL) data these days—in fact, insurers and providers are swimming in it. From internal sources, collaborative databases like the MPL Association Data Sharing Project, and the National Practitioner Data Bank Data Analysis Tool, and third-party providers, there is a wealth of data on MPL available.

However, this abundance of data can create analysis paralysis, leading to indecisiveness in regard to using this data. If that happens, potentially valuable insights can remain locked inside databases, never to see the light of day. Such a situation creates an opportunity cost for MPL stakeholders, depriving decision-makers of vital intelligence that others may be capitalizing on.

Advances in data analytics mean that it is more possible than ever before to mine data for emerging vulnerabilities that can facilitate increasingly robust risk management practices. Today, MPL companies are employing analytics to improve patient safety. On the macro level, organizations are sharing data with practitioners to improve care in general, and, on the micro level, to undertake proactive steps that can actually result in lower premiums.

Entities in the MPL universe are quite varied, including large companies that operate across the country as well as smaller insurers that confine their activities to a single state. And then there are self-insured hospitals, health systems, and medical practices. While larger organizations may seem to have the advantage in the race to optimize data for actionable business insights, there are strategies that companies of every size can employ to improve their ability to collect data, analyze information with appropriate tools, and deliver that data for the purposes of patient safety and MPL operations and underwriting decisions.

For MPL insurers and the clinicians they support to benefit from data analytics, organizations must already have or build a culture where such practices can thrive, said Larry Van Horn, PhD, founder and CEO of Preverity, a data analytics company specializing in MPL industry data. “Investing in analytics and not actually taking action on the analytics is a bad decision that is value destroying,” he said. “You must start with a baseline of deciding that once you trust the analytics that you have, you will use them across the board.”

Collecting data

No MPL company has to start from scratch when it comes to collecting data, because they all have a certain amount of data at their disposal already. In fact, Chad Karls, principal and consulting actuary at Milliman, describes the data possessed by MPL companies as “a distinct competitive advantage.”

While this data may not have been used to its maximum potential, it still offers powerful promise. “If your company is the primary MPL provider in X state, that means that you have more data about MPL in state X than any other company,” Karls said. “Even if you’re not a very big company, this data still is a significant asset that can be leveraged going forward.”

Until recently, the tools and knowledge didn’t exist for MPL companies to take advantage of this data. Fortunately, data science has progressed to the point where useful intelligence can be extracted from both structured and unstructured data through algorithms and artificial intelligence.

One of the first steps in leveraging data is understanding what data you already have, and what you are already measuring, according to Brigitta Mueller, MD, executive director for patient safety, risk and quality at ECRI in Plymouth Meeting, Pennsylvania. “Starting with getting the right data from within your own organization is very important,” she added. “It doesn’t help you as a first approach to know what anyone else is doing—first you have to know what you are doing.”

Van Horn believes that MPL stakeholders need to leverage much larger data sets than those available to individual companies. That’s because the rarity of claims requires greater visibility around the number of procedures that are performed that eventually yield specific claims. Otherwise, he noted, the claims data in and of itself is of limited utility. “Understanding the frequency in which specific procedures are done that produce claims is what facilitates actually under standing how the physician activity relates to risk,” Van Horn said.

Fortunately, there are a variety of third-party data sets that MPL companies of any size can successfully access. Whether you operate nationally, regionally, or in one state, you can obtain relevant data to provide context for your internal data. Then, you can successfully analyze that data and use it to make actionable business decisions.

Robert Hanscom, vice president of risk management and analytics at Coverys, suggested that MPL insurers forge partnerships with related institutions in their area to access more robust data sets. “For example, if you are an insurer operating in a specific state, you could collaborate with the major hospital systems in that state to access information gathered by their incident reporting systems, in a HIPAA-compliant way, to get a better sense of what is going on in real time,” he said. “You could then link those coded incidents to your coded malpractice claims to bridge the past with the present for actionable insights.”

MPL Association member companies can participate in the Data Sharing Project (DSP), the largest independent ongoing MPL claims database. During its 35 years of collecting data, the DSP has gathered information on more than 350,000 physician and dental professional claims.

Analyzing data

To successfully analyze data, you need data scientists who can build useful tools that reveal the actionable intelligence that exists in the structured and unstructured data available to you. “In an insurance operation, data can be applied in many ways,” said Karls. “You can use it in sales and marketing functions, in underwriting, risk selection, risk pricing, fraud detection, and claims.”

Within claims, there are many opportunities to apply data analytics to areas such as identifying outlier claims earlier in the process, selecting the best partners in the claims process, and understanding which options to pursue within the claims lifecycle, Karls noted.

At Coverys, Hanscom and his team are mindful of centralizing their data in order to capture an integrated view that can be fed into tools designed to improve both business decisions—and patient safety. “We developed a risk scoring tool that applies advanced analytics to a combination of several key data sets that we could leverage,” he said. “These include our own comparative benchmarking database of 25,000 claims, national profile data on approximately 1.3 million healthcare providers, and medicare data. This work has provided intelligence that is helping us predict whether individual providers are likely to be named in claims within the next three to five years.”

Employing data in this way gives Coverys the opportunity to engage in a dialogue with their policyholders about proactive steps they can take that might lower their premiums. “To me, that’s critical—tying these worlds together so that lower premiums hinge on proactive mitigation steps,” Hanscom continued. “If we as an industry can get there, I think we have the ingredients of the model that we really want.”

For Van Horn, the benefits are almost immeasurable. “If you fully embrace the potential for analytics in MPL, providers can be classified into appropriate rate groups better than ever before, meaning that risk can be discriminated within rate classes in ways never done before,” he commented. “The visibility and the analytics then also create operational efficiencies based on rules and procedures, driving down the cost of operations.”

Communicating data insights

You have to have strong organizational leadership and commitment to prioritizing data analytics and to making the changes that the data dictates, whether they align with your views or not.

Van Horn agreed. “You have to have strong organizational leadership and commitment to prioritizing data analytics and to making the changes that the data dictates, whether they align with your views or not,” he noted. “It’s useful also to have a burning need because without the need, there is little will to change.” The hardening market is what Van Horn calls “a forcing function” that is inspiring insurers to embrace data and data analytics because financial performance has slipped, providing an incentive to engage in different behavior.

At ECRI, Mueller believes that the patient stories behind the clinical mistakes are what ultimately inspire change in practitioners and insurers. “With providers, you need evidence, and even then, they have a natural inclination to resist making changes, so you have to convince them that your approach is sound,” she said. “There’s always a patient behind a claim, a patient that was hurt. When you can use the stories of patients that illustrate what went wrong that needs to be prevented from happening again in the future, that’s very powerful.” Coverys has had success with turning data analytics insights into educational modules that practitioners can access from any device and use to obtain continuing education credits. These modules are very specific and action oriented, designed with clear take-aways in mind that mitigate risk, Hanscom said.

Employing data insights

Ultimately, the goal for the MPL industry is to gain insight into claims and clinician behavior to move the needle measurably on patient safety. With an improved safety environment, claims and premiums will fall, and the practice of medicine will ultimately be safer. Using analytics in this pursuit is a highly worthwhile endeavor.

Today’s MPL Industry Data Environment

Across the organizational and business landscape, data analytics is big business. While many property and casualty lines of insurance have successfully leveraged data analytics to inform business decisions and incentivize customer behavior, the MPL segment lags in these efforts. Some of the issues that the MPL industry faces are unique to its position within the property and casualty universe because MPL claims are relatively rare, even in specialties that tend to generate the most claims.

That means that extracting valuable insights from MPL data is more challenging than, say, in the auto insurance segment, where more than 225 million Americans drive an average of 13,747 miles a year.1, 2 In this segment, readily available data has translated into actionable insights that influence policyholder behavior for positive change. Such changes can also create a safer driving environment for all drivers, when incentives succeed in lowering rates for accidents, speeding, and other adverse driver events and behavior.

“Within the property and casualty insurance industry, the auto segment is an example of how data can be tied together,” said Robert Hanscom, vice president of risk management and analytics at Coverys. “This segment uses a central source of intelligence that prongs out to pricing, risk mitigation, and other areas, which is a highly holistic approach.”

“MPL has struggled to achieve the same model,” he continued. “To succeed, I think we need to employ artificial intelligence, advanced analytics, and whatever other tools at our disposal to create not just an identical model to the automobile industry, but to go beyond it because we’re dealing with an imperative—patient safety—like Coverys is doing with a number of clients.”

Hanscom’s views align with financial rating agencies, which are encouraging the industry to innovate to improve financial results. In an article published in the second quarter 2020 IML, AM Best authors David Blades, associate director of industry research, and Jason Hopper, associate director of research, described the impact of large verdicts and increasing annual indemnity losses on the industry in these terms: “Claims settlements and legal costs have been rising at alarming rates and underwriting results are weakening in response.”

“Solutions driven by medical research, advanced analytics, and predictive modeling can help enhance patient safety, promote better outcomes, and avoid future claims and legal costs,” they wrote later in the same article. “Companies have also reported the potential use of artificial intelligence to make better use of unstructured data—for example, scanning public records could reduce the amount of information needed from patients in the application process.”

References
1. “How Much Do Americans Drive?” PolicyAdvice. net, Dec. 7, 2020, https://policyadvice.net/insurance/ insights/how-much-do-americans-drive/.
2. “How Many Licensed Drivers are there in the US?” HedgesCompany.com, 2019, https://hedgescompany. com/blog/2018/10/number-of-licensed-drivers-usa/.


 

   
 


Amy Buttell is the editor of Inside Medical Liability.